Scientific Computing

S-Plus 8 emerges from Beta

2007-05-22T09:03:00.000Z

Back in April 2006, I wrote about S-Plus 8 going to public Beta and predicted therefore that it would be out in Q4 2006.

Well it has finally appeared in May 2007. With such a long release cycle one might expect dramatic changes, but the new features list is largely as I described a year ago:

- Improvements to the Eclipse based Workbench (debugging and profiling)
- New graphic enhancements
- A new package mechanism
- "Over 100 new functions" in the S language

Since there is no information on their site about exactly what is in any of these changes, I can say no more!

Mathematica 6 released

2007-05-08T16:10:00.000Z

Wolfram Research have released a new version of Mathematica with the claim that "Mathematica's been reinvented."

This is certainly a huge upgrade with around 500 items on the new features summary, all of which (from a quick scan) appear to have substance.

To get a sense of the major directions, you are better off with the marketing version which summarises the most exciting things on the interface side as new application building tools with a free "Mathematica Player" which is some kind of runtime version. Depending on the details, this may be important for deploying tools developed in Mathematica.

On computation, the most important seem to be related to data (I/O, and built in data sources), and new optimization tools.

There is a lot of new visualization stuff which seems to cross over from computational features to interface features.

Labview adds Vista support

2007-04-22T15:06:00.000Z

Story title says it all, LabView is now Vista compatible.

I was amused by two paragraphs of the press release. First was the hint of resentment at Microsoft for the way Vista works...

"...this version of Windows requires 64-bit hardware drivers. National Instruments has invested considerable time and energy in providing updates to existing 32-bit drivers to support Windows Vista and in creating new 64-bit hardware drivers for Windows Vista x64 Edition, the 64-bit version of Windows Vista."

And second...

" “National Instruments continues to be among the first to market with compatibility for the latest PC technologies, including Windows Vista,” said Tim Dehne, NI senior vice president of R&D. "

Obviously Tim doesn't read my blog where I listed some of the first to market over a month ago.

Link: The death of computing

2007-04-11T09:34:00.000Z

Here is a link to a very well considered piece "The death of computing" from Neil McBride at the British Computing Society, which deserves to be read. I suspect that much of his arguments also apply to scientific computing.

If you can't be bothered to read the whole article, settle for this snippet: "If the gap between public knowledge and academic curriculum isn't large enough, the gap between academia and industry practice is a gaping hole. While academic departments concentrate on developing new computer systems in an ideal organisational environment, a lot of industry has moved away from in-house development to a focus on delivering a service."

O-Matrix sales shake-up

2007-04-02T13:11:00.000Z

Developers Harmonic Software have decided to give up selling their product, and will have someone else do it for them. Exclusive world wide distribution rights have been granted to TecPlot.

Lots of software companies choose to use partners to distribute in fields our regions outside of their expertise, but it is a bold decision to hand over the entire world.

TecPlot certainly seem a competent outfit, but the problem is that the product is a bit too close to their own. Just looking at the front page of www.tecplot.com at the side-by-side listing of products finds that O-Matrix is for "analyzing data, creating simulations, visualizing results" while TecPlot 360 is "Simulation Visualization Software... analyze and explore complex datasets, arrange multiple XY, 2D and 3D plots...".

It is pretty hard to imagine TecPlot will push O-Matrix too hard against their own products.

It looks like it is not good news for the customers either. As far as I can tell O-Matrix has jumped in price from $365 (as listed at www.sciencesoftware.com) to $950 on the TecPlot site. The developer kit from $420 to $2500.

MathCAD drops Maple

2007-03-23T09:45:00.000Z

I have finally confirmed what a reader pointed out to us a couple of weeks ago. The new MathCAD 14 has dropped the Maple engine which previously powered its symbolic calculations.

I wrote about plans to do this back in Jan 2006, though I predicted the wrong replacement. The MathCAD gig goes to Mupad. PTC's only explanatory comment was this "improves robustness". While anyone who reads comp.soft-sys.math.maple knows that Maple bugginess is a recurrent theme, the truth is probably more likely to be a cost cutting exercise.

Mupad also replaced Maple as the engine in Scientific Workplace a few years earlier and one must speculate that Mupad will now have its sights set on the one remaining Maple OEM deal, the Matlab Symbolic Toolbox. Perhaps this helps to explain Maplesoft's recent re-implementation of the toolbox, to put it under their own control?

Stable income from this deal will be a much needed boost to Mupad, who had seemed in trouble a year ago. Likewise, this must represent a significant loss to Maplesoft. A few weeks ago I wrote a piece called "Maple loses three million users" which was a dig at press release bullshit. But now, it seems, it was partly true. If PTC is to be believed, this will remove Maple technology from between 1.4 and 1.8 million desktops, over the upgrade cycle.

FORTRAN developer dies

2007-03-20T09:49:00.000Z

John Backus, who lead the IBM team that developed FORTRAN in the 1950s has died at the age of 82.

While I am no fan of FORTRAN (probably because I had already learned some more recent languages before I was taught the (then current) 1977 version), there is no doubt that it revolutionized computing and has had a long lasting effect on scientific computing.

With a more subtle effect, he also worked on early functional programming languages FP, and FL.

The NYT obituary can be read here.

Upgrade: Quality Companion 2.2

2007-03-19T16:32:00.000Z

Minitab have released an upgrade to their Six Sigma product, Quality Companion.

The upgrade is a modest collection of interface improvements, (new toolbar actions, auto-filling forms etc), new import export options (always good to see) and increased flexibility (better search and editing tools, use of images within documents etc).

There is nothing that you won't already be used to seeing in other products, but they are filling out the regular feature space at a rate that seems fair for the 9 months that have passed since Version 2 was released.

I did like "Automatically Convert 2.1 projects to version 2.2 projects". A self fulfilling feature that would appeal to people who like to have a car so that they can drive it to the gas station!

The full list is here.

Oddly it looks as if Minitab skipped a version as I never noticed, and can't find any information on Quality Companion 2.1. It must have existed though as there were maintenance releases to 2.1.1 and 2.1.2.

Maple NAG connector - a link to the past

2007-03-12T11:23:00.000Z

I was going to pass on commenting on the new Maple NAG connector because adding the ability to call NAG C subroutine libraries directly from Maple seemed a quaintly old-fashioned idea.

I didn't realize how old fashioned until I was reminded of Intercall. I Googled InterCall and it looks like exactly the same product except connecting Mathematica to the NAG libraries. The entertaining fact being that Intercall was something that was created at least 10 years ago!

Compare elements of the press release for the "new" Maple NAG connector with an eight year old Intercall page in the NAG website:

Maple:"Easy access to the powerful NAG C Library numeric routines from inside the Maple environment"
Mathematica:"easy access to all the routines in the NAG subroutine library."

Maple: "Appropriate default values for many parameters are provided automatically"
Mathematica: " straightforward declaration of default settings for arguments in external routines."
Maple: "NAG C Library documentation is integrated into the searchable, indexed, Maple Help System, and includes hundreds of examples coded in Maple"
Mathematica: " a detailed TeX manual describing how to use InterCall with Notebook examples"

As far as I can tell, Intercall is no longer available. I can't particularly see any reason why it should appeal to Maple users more than Mathematica except that Maple's home grown numerics are weaker than Mathematica's and Maple already has some NAG libraries included (for the same reason).

But I don't really see the appeal. O think that one is either a C/FORTRAN libraries and subroutines kind of person, or an application using person. Bridging the gap technically, doesn't bridge the gap psychologically.

Vista support in scientific software

2007-03-06T09:57:00.000Z

Here is a quick survey of the current state of Vista support, in scientific software, based on information on suppliers web sites.

Where Vista did not appear on the supported platforms page, I have taken that to mean that it is not supported, though it is quite possible that the product works, but that the supplier has not yet formally tested it.

Supporting Vista:
Comsol Multiphysics
Matlab
Mathematica
Mathtype
Mupad
Origin (32 bit only)
Scilab
Stata
Systat

Not supporting Vista:
Genstat
Labview
Maple
MathCAD
Minitab
O Matrix
PVWave
Statistica
Scientific Word
Scientific Workplace

And in a special category of its own, Endnote, which works under Vista, but you have to follow some manual steps listed in their support pages to make it work right.

Systat 12 released

2007-02-26T12:30:00.000Z

A new version of Systat is out. You can read the details here.

My stats knowledge isn't good enough to comment on the major features: extended mixed model analysis, robust regression. But they seem decent enough to be worth existing users upgrading to. Though, since it has been three years since the last upgrade, I am not sure they were worth the wait!

Some of the lesser features seem unimpressive. The improved plotting interface and cluster analysis sound rather like the claims from the Version 11 press release, and interface improvements sound more more like window dressing than real innovation.

What I was most drawn to were the removal of some limitations, only to be replaced with new limitations. Variable names can be longer to be more descriptive... but still only 256 characters. I know that 256 should be enough, and I can't think of a realistic reason why it shouldn't be. But history is full of such mistakes ("640k RAM is enough for anyone", "you only need 2 digits to store a date" etc). If you are going to fix that problem, why not fix it completely (or at least absurdly far above current expectations).

Likewise, a new multi-step undo... but only up to 32 steps. I am sure that I have undone more than 32 steps in a document before!

Link: Vista Spin Unspun

2007-02-22T09:44:00.000Z

I didn't yet have much to say about Vista when it came out, so since there is no interesting news today, here is a fun piece from The INQ, which is not so much about Vista, as a good summary of IT journalism and IT PRs.

Or, if you are in a hurry, today's Dilbert says pretty much the same in three pictures!

TI-NSpire advertising isn't the solution

2007-02-20T09:50:00.000Z

As directed by an irate reader of a previous TI-NSpire article, I was doing a little Googling to see if there was any news on its shipping status, when I found a fun thread on an HP site. It was pointing out that the calculation in an example screenshot of the nSpire, on TIs site, is wrong.

Perhaps more surprising is that one of the writers claims to have mailed TI to point out their mistake a month ago and it is still there. Perhaps the TI web people don't think correct answers are important.

Assuming, that they do fix it soon, here it is recorded for posterity. (Correct answer is 1/2 and -9/2).

[[Update 26 Feb: Within a few days of appearing on this site, the page has now been fixed. Here is the new, fixed, version of the image]].

MathCAD 14 released

2007-02-18T20:33:00.000Z

Its appealing to try and find a simple narrative to summarize new software - either it is total junk or some excellent advance. But the contents of the first release of MathCAD since PTC purchased the company is not so easily categorized.

There is unsubstantiated and weak claims, and there is real development.

Lets start with the lame...there are several items where one's natural response should be "what? didn't it do that before?" like the ability to enter more than one step of a calculation at a time in a single piece of input and support for UNICODE (so that the majority of the world can type their own non-ASCII based words into it) and they think that "Polar plot can now accept negative radii" is a significant advance! And perhaps the best "IMPROVED Worksheet Search – Search results are more complete, including collapsed and hidden regions of worksheets." - so, before, if you couldn't see the text, neither could the search feature!

There are some features with limited evidence "improved symbolic engine robustness" (I have a theory, which I will report if it turns out to be true [update]). And disingenuous features e.g. one of the "Highlights of the enhancements in Mathcad 14.0" was "Increased technical support" - they should license that code!

Then there is the solid... PTC exerting its more significant development resources, by providing translation into 9 languages (fitting in with the Pro-Engineer existing market) , integration of document difference highlighting (fitting in with the existing Arbortext document workflow products).

Really what I think this release shows is that PTC knows what PTC knows. The most significant features are the more prosaic ones that are about integrating with their existing business, and the more technically interesting, are no more solid that when MathSoft was a small company standing alone.

There is nothing wrong in that per-se, but you can also see where corporate culture can be negative. In the cheer leading over the improved symbolics, you find the phrase "Power users, research and design engineers, and other Ph.D. level users requiring symbolics will benefit from these improvements." Hold on. I was doing symbolic math when I was 13 or 14 (OK, it was only simple equations, and I didn't do integrals until 16, but it was useful and I didn't have a Ph.D!).

But PTC has always built numeric-only software and you can almost sense their confusion that there are people who need more..."Oh, it must be some Ph.D level thing"

New O-Matrix 6.1 Performance claims

2007-02-05T17:20:00.000Z

Matlab clone O-Matrix has released a new version, 6.1, with the summary: "This release has made many ease-of-use enhancements, added numerous application-oriented examples and made significant performance enhancements for newer multi-core machines such as the Intel HT and Intel Duo Core CPU."

I can't find a single reference to what these" ease-of-use enhancements" might be on their site or elsewhere.

As for the performance enhancements, there is plenty numbers to demonstrate 6.1 going much faster than other things, but as ever with such things, a little scrutiny reduces their credibility.

They have tables of O-Matrix 6.1 vs O-Matrix 5.62 --- but NOT against O-Matrix 6.0, so we can only assume that most of the improvements were in 6.0 not in this 'performance enhancing' release. They have also updated their comparison against Matlab (which I commented on here) to use 0-Matrix 6.1 but have still not upgraded the Matlab side of the comparison to anything newer than 7.01 which was released back in 2004.

Dodgy scientific methodology in marketing comparisons probably don't reflect poorly on you when you are selling shampoo or cat food but it surely does in technical software.

Having said that, I can't actually find any claims that Matlab has got any faster in the last three years, but still- show us the current comparison, if you want us to take it seriously.

Did MapleSoft censor their own blogger?

2007-02-01T11:06:00.000Z

Strange occurrences at the MapleSoft sponsored user forum, that are likely of little interest to anyone except me...

The site manager added a blog entry to the front page of the MaplePrimes site about an item that I wrote about calculators. It's always nice when people link to your site, so I went back a few days later to see if I could contribute anything useful to the discussion. However, in that time the links to scientificcomputing.blogspot.com had been removed and replaced with plain text.

Now I can't think of any logical reason why one would write a piece talking about a web page one read, and then decide to later remove the links to where it was, having initially made the effort to put it in.

The only hypothesis that I can come up with is that his superiors didn't agree him linking to a site that has been critical of them at times.

I thought I was exactly on-topic for MaplePrimes, which says that one of the purposes of it is "To offer a contrast and complement to the corporate Web content".

Kind of amusing. Perhaps in a few days I will get annoyed and edit out the links in this item!

[[Update 26 Feb: I decided to do the opposite, here is a link to his personal blog where the original version of the article still lives]]

Minitab 15 released

2007-02-01T10:49:00.000Z

Minitab have announced a new version of their flagship statistics package. Headline new features are listed as...

Ability to assign formulas to columns in the worksheet
Expanded Gage R&R capabilities
Power (OC) curves for power and sample size
Probability Distribution Plot
New Reliability methods for forecasting future warranty claims
More than 40 new calculator functions

When you read that the top new feature is "Ability to assign formulas to columns in the worksheet", you prepare yourself for a completely underwhelming upgrade.

But when you dig into the full list there is quite a lot more new stuff than the press release suggests, with modest improvements in many, many areas.

Where the release disappoints is by a lack of any clear direction to R&D, perhaps that is why it was hard to write a press release that made it sound interesting. There is no attempt to extend the reach of Minitab. They are happy with what they do, and just want to do the same a bit better.

The result is an upgrade that should please Minitab users, but is unlikely to attract anyone who had previously dismissed it.

Design Science releases news - MathFlow 1.7

2007-01-30T11:57:00.000Z

Back in October I was rather sarcastic at Design Science's expense over their lack of news output. In the comments, BruceV from Design Science, promised that news would start flowing again as he was hiring a new Marcoms person the next week.

So it has been with some anticipation that I have been waiting to criticize their first output! I have to report that it is a nicely written and hype free piece of news.

The item is an announcement of a new version of MathFlow, their XML oriented math typesetting tool.

Though the news was too late to prevent 2006 being empty of news, I wonder if one of the key new features is also a bit late. Headlining the new features, is compatibility with Arbortext XML authoring system. Now that isn't too late to be a sensible and useful feature, but it would probably have got a lot more traction from the users and makers of Arbortext if it had been a year earlier.

In the last year Arbortext was bought by PTC, who then also purchased MathCAD, which can also be used to write math, and outputs in XML. So it is likely that PTC will have little interest in making Arbortext users aware of Design Sciences solution and will instead push their own solution.

I should mention that I have previously argued that MathCAD is not a good way to author XML (though that would be easy to change).

The release also alludes to behind-the-scenes changes to MathFlow that might be more significant, but not yet of major benefit to users.

Site comment

2007-01-26T14:24:00.000Z

Apologies to those who subscribe to the RSS or ATOM feeds, which are all messed up.

I upgraded to the new engine for blogger.com and implemented some basic tagging on the archive of articles here.

For those who blog, the new engine is certainly an improvement with better layout editing and faster content management, as well as the tagging. However, the transition managed to convince it that a lot of old articles should be at the top of the feed, instead of the newest ones!

I guess the answer is just to write some new articles!

[Update: 12-Feb, the solution is to edit the recent articles, so I have hyperlinked a lot more of the text]

Guessing at MapleSoft's strategy for VSNI

2007-01-25T12:03:00.000Z

In a good start to this year, scientificcomputing.blogspot.com managed to break two world exclusives that have now both been confirmed. The announcement of Maple 11 and the appointment of MapleSoft's Jim Cooper to the board of VSNI.

Looking in more detail at the VSNI tie-in... the press release from VSNI is all very sensible, talking about using Mr Cooper's experience to help grow VSNI. But there is still no clue as to what's in it for MapleSoft to lend out their (presumably) most expensive employee for the benefit of another company, or if its in his own time, why Mr Cooper would want another job.

I now think the idea of shared sales channels seems unlikely, as VSNI have few major resellers that are not already selling Maple and are not, themselves, a large sales force. So MapleSoft must be interesting in accessing the statistics technology and expertise that they currently lack.

It would fit with previous strategy, since little of recent versions of Maple, except for the interface, are created by MapleSoft themselves - they have mostly been assembling parts from other organizations (numerics from NAG and GMP, algebra from INRIA, ORCCA etc) why not statistics from VSNI?

Two key choices would be needed: whether to do lots of work to interface existing data structures seamlessly, or just bolt it on the side and lets the users do the work. And whether to use the existing statistics language design of Maple as a user layer, or to deprecate all of that and expect users to re-write their work in a language more akin to Genstat. If you take the NAG integration as an example, then you could expect a middle compromise on the first, and the deprecation route on the second. Either way, there is enough to do to make this a Maple 13 or Maple 14 feature, not Maple 12!

Whatever the choices, I can see no downside for VSNI, who at the least get some advice from an old hand, and at best might be at the start of a negotiation for a big sale.

Maple 11 to be announced next week

2007-01-02T11:50:00.000Z

Some more news leaking out of MapleSoft...

On January 8, they will be pre-announcing an upgrade to their principal product Maple although the actual product is not anticipated to be ready until (at least) March.

I will comment again when the actual details are out, but I understand the headline features to be mostly improvements to their Java interface. Unlike the recent minor upgrades to the interface (see comment here) these look not so much bug fixes as design fixes. eg Allowing the typesetting to work within a graphic, equation numbering that works between documents, and the confusing "in place" menu driven operations that previously left no history as to what you did to get your result, now automatically insert a comment.

Features that are actually new include a "preview version" of new handwriting recognition for math. But if you recall my comments on the handwritten character recognition in current Maple, I think we can read "preview" to mean "doesn't work". There is also a Mathematica style "slide show" mode for documents and some kind of content "annotation" feature.

There are some new computational capabilities, mostly "speciality packages" or centred around some nice new Groebner basis capabilities, which they already announced back in March.

So with the central drive still being the document interface, it will be interesting to see if this is the release where they push people to move from the "Classic interface" or if that will still ship alongside the Java interface.

Maple CEO moonlights at VSNI

2006-12-31T15:54:00.000Z

I may have to retract this story, as I can't find any written reference, but I have heard this gossip from more than one source, so am assuming its true.

I am told that MapleSoft CEO, Jim Cooper, now has a side job as Chairman of the Board at VSNI, makers of the venerable statistics software Genstat.

Unless he has too little work to do at MapleSoft, this makes little sense to me.

The obvious guess, that this is leading to the inclusion of Genstat statistics within Maple, like the inclusion of NAG numerics (NAG are part owners of VSNI) seems unlikely. Any hybrid of the two would be an uncomefortable compromise for both. Perhaps we will see a sharing of sales channels and user lists?

All rather odd. (If true).

New Scilab 4.1

2006-12-15T11:13:00.000Z

The (Mostly) French, Scilab Consortium have shipped an update to their open source Matlab competitor, Scilab.

Version 4.1 appears to be just a maintenance release, though the bug fix list is reasonably significant.

Perhaps more interesting is the beta of a Scilab- Labview gateway. Scilab are already cosy with Mathworks' arch enemies National Instruments through their joint participation in the Numerical Math Consortium (more: 12/05, 1/06, 9/06).

Labview are more likely to be the winners of such an alliance, with customers who want to get rid of Mathworks licensing fees to move to Scilab, being left with spare cash to buy Labview! Existing Labview customers are less likely to be concerned about licensing fees, and already have a choice of professional computational engines to put behind Labview like Mathematica, and even Matlab!

New FEA tools for Mathematica

2006-12-12T13:37:00.000Z

Wolfram Research have announced a couple of surprise products for working with finite element analysis in Mathematica - AceGen and AceFEM.

AceFEM, is a general finite element environment designed to solve multi-physics and multi-field problems based around Mathematica.

AceGen is a code generation and optimization. Unlike the existing code generation tools for Mathematica, which are aimed at FORTRAN 90 and C++, AceGen is specifically generating code for FEA tools like ELFEN, FEAP, ABAQUS and AceFEM.

There are some entertaining example animations on their website.

Minitab fights Dilbert over Six Sigma benefits

2006-12-08T16:44:00.000Z

Minitab are highlighting a report that claims that those using technology based approach (read "using Minitab") to Six Sigma are more successful.

This refutes an article referred to in a recent Dilbert cartoon which really was in Fortune that claimed "of 58 large companies that have announced Six Sigma programs, 91 percent have trailed the S&P 500 since."

However with the claim being highlighted being based on the 16% of the 400 companies who took part in the study who have been judged to be "holding true to the rigorous program" doesn't seem to me to be too significant (though I have long since forgotten the appropriate math for significance tests to defend that claim). Perhaps this is just prejudice because the study was sponsored by Minitab themselves.

A few related comments on other sites: link, link, link

Maple loses three million users

2006-11-30T10:22:00.000Z

I very much doubt that the headline of this item is true, but I did get the story from MapleSoft press releases!

I made a sarcastic remark at the expense of Parametric a couple of months ago when they couldn't get their story straight on how many users MathCAD had. So I can't help but comment on a similar inconsistency at MapleSoft.

The boiler plate part of their press releases used to contain the claim "Over a million users have adopted Waterloo Maple products". The last time being in Nov 2002.

Around that time, there was some kind of management buyout, the company changed its name to MapleSoft and lots of other changes were made, including the boiler plate which from Jan 2003, read "Over 5 million users benefit from advanced Maple technology." It goes to show the rapid effect of new management on the success of a company!

Nothing changes until Jan 2005, when suddenly the claim drops to "Over two million users at thousands of organizations benefit from advanced Maple technology."

The truth? Who knows. But I doubt that I am going to find it in a MapleSoft press releases.

Derive to be discontinued

2006-11-23T11:25:00.000Z

According to Bernhard Kutzler writing in the Derive user group newsletter, the Derive computer algebra system will soon be discontinued as Texas Instruments focuses on its future nSpire platform. (Which I have previously written about here, and here.)

This radical change, rather than an upgrade evolution, will likely disturb many existing customers. Kutzler writes "There are many features in Derive 6 which are not (yet) available in TI-Nspire."

The main reason provided is the aim of unifying CAS, graphing, geometry, spreadsheet and text processing in one product. If true then this leads to the natural question of whether TI's existing geometry product, Cabri, is destined for the same fate?

Mupad not just available in Engrish

2006-11-21T10:45:00.000Z

Mupad is now available in a Japanese version. But in a cheap shot that I can't resist, I have to question whether I should trust the quality of a language product when the announcement on their website reads:

"The japaneese version of MuPAD Pro 4 is ready. It is exclusively available from our japaneese reseller LightStone Corp. "

(That was a particularly cheap shot considering I can't spell either and the Mupad developers are German). More serious comment next time.

Sun open sources Java

2006-11-13T13:05:00.000Z

It has been discussed for months, but today Sun made it official- Java is to be open source, with all libraries code to be available by March.

On the surface, Java would appear to have made little impact on scientific computing with few applications outside of a cheap way to add GUI layers to existing code.

Partly it comes from older packages having their roots and source code firmly embedded in FORTRAN, and partly from a reputation for Java being slow. The speed issue is not intrinsically true but the combination of JREs needing to be loaded before use and that everyone has experience of using sloppily written Java applets (and sometime sloppily written commercial applications) has left the impression.

But look behind the scenes, and a lot of the glue that links together parts of scientific packages and connects them to data sources etc turns out to be Java.

If Sun is right and this move adds momentum to Java, we might see more use of it.

UK health service IT project run by IT failure

2006-11-13T12:36:00.000Z

Somewhat off the science topic, but amusing enough to include:

As a lesson in not making enemies when you are in the public eye, especially family ones, and a reminder of the unique power to embarrass that ones mother possesses - here is an interview with the disgruntled mother of the man tasked with upgrading the entire UK health system IT infrastructure. She says:

"I can't believe that my son is running the IT modernisation programme for the whole of the NHS.

"He was disappointed when he failed his computer studies course at Bristol.

"It was pretty serious, so I had to write to Princess Anne, who at that time was "university visitor" there to appeal for him to be allowed to resit the exam, as initially he was refused permission."

I think a cosy family Christmas looks unlikely there!

Modelica gets the Simulink treatment

2006-11-07T16:04:00.000Z

The Modelica modelling language has got a boost with a new product from MathCore: MathModelica System Designer Professional.

This is a tool built on top of the Modelica language which seems to take the project onto the ground covered by Simulink by adding a graphical programming interface.

I was told once that Modelica's strength is the ability to model engineering systems that combine different fields. eg. A system that has electrical subsystems and hydraulic subsystems that are interconnected. The project seems to have a lot of academic and some serious industry support, with the likes of ABB on board, but seems like a largely European affair at the moment.

As well as the Modelica libraries the product also integrates Mathematica to give it the number crunching it will need to take on Matlab.

COMSOL Multiphysics 3.3

2006-11-03T15:02:00.000Z

I was a little slow to notice the release of a new version of COMSOL Multiphysics. The simulation software, which was until last year sold under the name of FEMLAB, was updated over a month ago.

The new features are quite numerous and you can read them here.

Broadly speaking, they mostly involve trying to widen the space of applications further away from the original finite element modelling background of the company, eg with optimization and signal processing.

There is also some work on improving core functionality with work on adaptive meshes.

All very sensible.

Yet another $2000 Matlab link

2006-11-01T09:32:00.000Z

This weeks $2000 link from Matlab connects to Cadence Incisive.

I said pretty much all I can on the subject the last two times (here and here).

O-Matrix sets Kalman Filter tool free

2006-10-26T10:12:00.000Z

Previously costing a little over $100 the O-Matrix add on "Kalman Filter Design Studio" is now available to O-Matrix users as a free download from here.

Presumably they were not selling many of them.

There is no comment on the site about the strategy of releasing it, so we will have wait to see if any of the other seven O-Matrix add-ons share the same future.

Site comment

2006-10-24T16:35:00.000Z

Warning: No real news in this post.

You might notice a new bit of Amazon advertising on the site. My grand ambition is that writing a piece for this site, might one day pay as much as spending the same time flipping burgers for McDonalds!

It's got quite a way to go, but I am committed to keeping adverts small and discrete, as I am really more interested in the computing news.

However, since (unlike Google) Amazon's conditions allow me to draw your attention to the advert, I would encourage you to go click on the link and go shopping as I get a little kick-back if you do. Unfortunately, unlike Google adverts, it only happens if you actually spend money.

In other site news- the release of the Firefox 2.0 browser today, with built in spell checking for forms input, should see an end to the bad spelling on this site, though probably not bad bad grammar. The www.blogger.com authoring interface is pretty good, but lacks spell checking.

Interestingly, in this first post using the new Firefox, I see that the spell checker highlights Firefox as misspelled!

Yet another version of Maple 10

2006-10-21T19:47:00.000Z

Disregarding my comments on Maple 10.05 being too many maintenance releases on a product, Maplesoft have released yet another in under two months - Maple 10.06.

As before the bug fixes are again interface related...

Scrolling behavior when inserting output near the end of a document
Palettes, including constructing higher-order derivatives and choosing insertion points
Export to RTF, HTML and LaTeX
Removing output from a selection

But here is the fun part... If you recall, 10.05 was only available to subscribers to the EMP maintenance program. This version is available to all, but on the Maplesoft website we read "Elite Maintenace Plan (EMP) customers only, please note: Maple 10.06 does NOT include all updates from Maple 10.05. You must install Maple 10.05 before installing Maple 10.06".

So it seems that if you have EMP you can have it all, but if not you can have the 10.06 fixes but not the 10.05 fixes. It doesn't say whether you need to have already picked up the 10.01-10.04 updates. And it is not clear whether the users of the Japanese version get the 10.05 fixes, as there was no 10.05 for the Japanese version, but there is a 10.06.

So now they have lots of different branches of code to support in the wild- this puts a huge demand on testing.

So it is not surprising to read comments on VersionTracker and the Maple user group that there are quality problems with this update with several people saying that installing the update destroyed their copy of Maple.

Mathworks doesn't announce Matlab release 2006b

2006-10-20T16:23:00.000Z

Mathworks seem to have somewhat lost interest in their core products.

This week I read a Mathworks news item on another expensive link product (well I think $2000 to connect Matlab more easily to Altium’s TASKING Compiler is expensive). In light of the other links that I discussed recently, this seems like a general strategy.

But I, understandably, managed to miss a new release of their flagship products. This is because Release 2006b of the Matlab/Simulink family has again been slipped out without making it to their news page. Like the 2006a release, the new biannual release cycle has not produced anything too exciting, so perhaps this is the reason.

The headline "New capabilities" for Matlab are listed as

Additional support for large data set handling in MATLAB
MATLAB application deployment to Java™
Distributed computing tools supporting Windows® Compute Cluster Server 2003, distributed arrays, and parallel math functions
Graphical interface for defining and solving optimization problems

But when you dig down, there is less to it than meets the eye. "Additional support for large data set handling" sounded exciting but the details turn out to be:

Elimination of temporary memory copies when calling M-file functions and some built-in functions in-place (i.e., x = myfunction(x)) for easier handling of large data sets and improve performance
Ability to read and write MAT-files greater than 2 GB on 64-bit platforms via new switches to the save function
Ability to view larger arrays in Array Editor
Full JIT/Accelerator support for 64-bit platforms, allowing faster code execution

IE removal of some internal limitations and finishing previous partial implementations of 64 bit addressing. Just maintenance issues really. Read the full lists in minute detail here.

So putting the two topics together it looks like Mathworks are currently focused on finding more things to sell to their user base by solving some system integration problems, than adding new core functionality to their main products.

Why no news?

2006-10-19T10:19:00.000Z

I have been a bit quiet lately, and human nature being what it is, I would like to lay the blame on someone else.

For this purpose, I will pick on the inactivity of Design Science, makers of MathType! While I am the first to criticize vacuous press releases, it is extraordinary that a company can find nothing new to say about itself in over a year. The last release they put out according to their news page was 10 August 2005, to announce that the NSF had given them some money.

Not far behind are Systat, who still have a 2003 item on their "Recent news" ticker. To be fair they have announced two things this year, but they are both about the Korean translation of Systat, so I hope the 72M Koreans will forgive me for suggesting that that isn't of general interest!

Upgraded toolboxes for Matlab and Mathematica

2006-10-05T15:50:00.000Z

This week sees upgrades to some of the toolboxes available from Mathworks and Wolfram Research.

From Mathworks comes Control System Toolbox 7, Simulink Control Design 2, and Simulink Response Optimization 3. The common theme of these upgrades is to reduce some of the barriers between them and Simulink.

Wolfram's upgrade is a new version of Mathematica Link for Excel, which adds control of Excel from within Mathematica (putting and getting spreadsheet regions) and typeset expressions in Excel, to their existing tools for calling Mathematica as a set of macros from Excel.

Geometry Expressions for Mathematica

2006-10-04T09:44:00.000Z

Yet another third party add-on for Mathematica this week- Geometry Expressions.

This educational tool describes itself as "An Interactive Symbolic Geometry Companion to Mathematica". It looks essentially like Cabri Geometry but with the advantage of symbolic formulas for the solutions to the geometry problems.

Looking at the author company's background that is not surprising, since they have previously created geometry software for HP and Casio calculators and so would have been up against TI's geometry tools for years.

StatSoft opens more offices

2006-09-26T12:06:00.000Z

If the number of countries you have offices in is a measure of success then StatSoft, makers of Statistica, are doing OK.

Scientific Computing World magazine (not related to my little site) are reporting that StatSoft have opened an office in Norway, their third new office this year.

While opening international offices is sometimes about impressing the money markets, this seems to be just an ongoing strategy of being local. This is the 20th office such office now.

Though there is no information on how substantial the offices are, though they claim to all be "full service" and since Statsoft offers statistical services as well as software sales, they are unlikely to be "one man" outfits.

MapleSoft news reveals little new

2006-09-25T10:52:00.000Z

Looking back I have been quite critical of some of MapleSoft's recent press releases, but sometimes they really do ask for it. They have just released four press releases in six days, but have very little worth announcing in any of them.

One was a repeat of the release that I covered two months ago about the Matlab toolbox. Except with the words "now shipping". Perhaps that is news because it didn't ship in August as was originally promised.

One was to say what I nice conference they had back in July. It might have been news then but two months later? Since you can still register for this event on their website it might be just incompetant timing!

One with real but minor news was for the Japanese version of their Blockbuilder product. But since that is quite small, I don't suppose there were many words to translate!

But the biggest piece of puff is the grand release of four "Engineering toolkits", which turn out to be already available products bundled together under new names.

In all, the only interesting thing I found was that with two of the bundles called "MATLAB Users’ Toolkit" and "Simulink Users’ Toolkit", there can no longer be any doubt about what I have been speculating since March and as recently as June, that MapleSoft are repositioning themselves as makers of Matlab add-ons.

A WorkLife FrameWork for Mathematica

2006-09-19T09:48:00.000Z

An unusual add-on to Mathematica this week- A WorkLife FrameWork claims to be a "creative workflow toolset for Mathematica".

That turns out to be a diverse set of tools such as ToDo management, blog creation, RSS reading, slide show creation, journal creation and more.

I can't see anyone buying Mathematica to replace Outlook or similar tools, but if you use Mathematica a lot, and want to stay in one environment, then it is probably useful.

Info on the Wolfram site and also on the original developers site.

Numerical Mathematics Consortium explains its problems

2006-09-11T16:19:00.000Z

Last time I wrote about the Numerical Mathematics Consortium project to create a standard for numerical math algorithms, I was impressed with the work rate, though still sceptical about its direction.

A new document released by the consortium this week, reveals a much less impressive rate of progress as the committee hits the swamp of problems that the project needs to resolve.

The document summarizes and illustrates some of the problems quite neatly, but doesn't present a compelling case for their solutions which seem pragmatic but quite arbitrary.

In the 8 months since agreeing the nearly 200 functions to standardize, they have so far agreed the standard for just 10. The document claims that we should expect and increased rate, though we should only expect biannual reports of the progress. So I think we can conclude that this project will take years not months.

An early compromize is a bad omen for the future- There are already two levels of 'standard' (required and optional). This is exemplified by whether a particular function needs to support both real and complex numbers. Real support might be required but complex support could be optional. It is suggested that other features of range or type might be also options for different functions.

Since the claimed purpose was algorithm portability, what sense does optional support make? The only way to expect portability will be to restrict yourself to required support functionality. And so the usefulness of this standard contracts down to the lowest common denominator between the partner systems. But who wants to buy expensive software, to then deliberately restrict ones usage to a small subset of the functionality?

If they had set a level 1 and level 2 support, and required that ALL of the extras in the higher level must be supported, to claim that level, I might see some sense in it. But as it stands, they may end up being hundreds of separate places where a "standard" algorithm will fail when moved to another system that supports the "standard".

Politically speaking, this kind of decision may be inevitable. If you have some functionality in your product, you want to convince the other members of the committee to include it in the standard. If you don't have that functionality, you want to exclude it, so that you are not forced to build it. Put a group of such vested interests in a room, and you either fail to agree or find some compromize.

Site comment

2006-09-11T15:52:00.000Z

Time for a new look to this site.

The colors are the blue half of the Summer Groove palette from DailyColorScheme.com in case you care.

TI Nspire calculator delays

2006-09-07T11:34:00.000Z

Back in May, at its announcement, I asked "Is the new TI Calculator good news for TI?".

What is certainly not good news for TI, is that they are having trouble delivering it.

German retailer Dynatech, which was until recently anticipating first shipments by the end of August, just in time for the new school year and in plenty of time for christmas, now says that it will not be available until Q2 2007.

TI's site is still suggesting that first shipments will be this year but only in Canada.

Software maintenance pricing strategies

2006-09-05T14:55:00.000Z

A few weeks ago I did an item on the variance in upgrade pricing, to see who suckers you in with lower license purchase fees but plans to make money out of you later once you are committed. At the time I said that I would do an item on maintenance fees, but researching that has turned out to be far harder.

First, it turns out to not be as popular way of offering upgrades as I thought. Most of the companies that I write about regularly do not offer any kind of maintenance program. You can't plan a budget in advance, you must wait for the upgrade and then react. Some companies do offer leasing schemes, where you lose the right to use the software if you stop paying. -- dare I say it, I will cover this another time!

Second, when they are offered, the prices are often hidden from public view. Prices are listed as "call for quote" etc. When more people are clicking on the adverts on this site, I might spend that kind of time on an article!

Here is some data gleaned from unreliable sources (from customers that might be under special arrangements). Annual maintenance fee as a percentage of initial purchase price:

Maple 40% (from UK site Chest).
MathCAD 26% (also from Chest).
Matlab 20-25% depending on the source.
Mathematica 25% from their site.
Origin 25% (minimum of 2 years, or 3 years for groups) from their site.

I couldn't find any maintenance schemes from the usual products that I look at. (Expect updates as readers correct me!)

So with the exception of Maple, the values are much more consistent than upgrade prices were. Comparing with that data, it seems like there is some correlation between those who offered lower upgrade prices and the availability of a maintenance scheme.

There wasn't much variation either in what was offered. Upgrades and technical support were standard. There were some "exclusive websites", access to beta versions and discounts off further purchases etc. Mathematica had a couple of significant extras- a home use licenses and a webMathematica licenses. Origin included a custom webinar of 2 hours (only for group licenses), which could potentially be a valuable benefit if they provide the right presenters.

There were a lot of dubious "benefits" - like "custom communications" (read more email!). My favourite is that paying Mathworks maintenance fee will give you "Ability to purchase additional products for your license". Really, see for yourself (it's item 6)!

Maplesoft seemed a bit confused on the subject. There are two conflicting pages about their EMP maintenance scheme, here and here. The former refers to the benefit of "exclusive access" to a website that is in fact public. The newer page contradicts itself by giving the benefit of a new exclusive web area, before listing it again further down the page as a future feature!

Endnote X for Mac ships

2006-09-01T12:29:00.000Z

I already covered the release of Endnote X back in July. Now two and a half months later comes the update to the Mac version of Endnote, which includes Mac Intel support.

I notice the lack of the word "Universal" in any of the product information or press releases. Either that is a little marketing oversight or, like the recent Maple Mac release, it indicates that, despite the release delay, a few shortcuts have been taken and the product does not meet Apple's specifications for use of the "Universal" tag.

IDL to include IMSL

2006-08-30T09:47:00.000Z

ITT Visual Information have been working with Visual Numerics Inc to embed the IMSL libraries into IDL.

While both parties presumaby consider this a win-win deal, it is hard to see who is doing best out of it.

ITT are a business monster with sales in the billions (of other products) and presumably could afford to build whatever features they want to into IDL. But instead this sales 'might' has allowed the development to stagnate. IDL was once quite advanced but now its feature list reads like the features that you take for granted on other products. This is a quick fix for that.

Visual Numerics have been more active in development but the market for libraries that you have to write your own code to connect must be diminishing so this is potentially a good OEM sale (depending on the price they got). But the potential cost is boosting a competitor to their main application product PV Wave. The press release makes reference to that, explaining that they are in different markets. Perhaps.

Maple 10.05 maintenance update

2006-08-24T12:12:00.000Z

It is usually a good thing when software companies produce maintenance releases to fix bugs in a shipping product. But news that MapleSoft are shipping a sixth release of Maple 10 does make you wonder if perhaps it could have been tested and finished a bit better before it was released.

If you look back over the releases, there is a general theme-- improvements to display, editing and interpretation of 2D math and document mode, appear in every update. It seems that a lot of work is being focused on getting the new interface, created for Maple 10, right.

A clue may be in a recent poll on the user group site, where nearly half of users are not using the new interface at all, most preferring the obsolete Maple 8 interface. And 1/3 of those that do use it, are not using any of the new features, preferring to put it in plain text mode.

It looks like MapleSoft think that quality is what is holding back adoption.

[[Update 6 Sept 2006- In a significant change in policy from Maplesoft this maintenance release is only available to subscribers to their EMP program.]]

New Six Sigma product from Minitab

2006-08-22T17:21:00.000Z

Minitab Inc are working to cement their position in the Six Sigma market with the release of Quality Companion 2.

While you wouldn't know it from the way that it is presented (apart from the "2" in the title) this is an upgrade to an existing product released around 2003 which provides tools to improve quality processes - mapping processes, reporting and communicating processes etc.

I couldn't tell you what is new in this version, as their presentation of it doesn't tell you, and it's not a product that I have used, but they must think it is a major upgrade as you can find Quality Companion 1 listed for sale on the web at $149 but with version 2 the price has jumped to $450. No upgrade options are mentioned on their site.

Round up of new products

2006-08-09T10:38:00.000Z

National instruments have released LabVIEW Toolkit for LEGO MINDSTORMS NXT.

I am sure that there are lots of professional uses for controlling the LEGO robotics platform from LabVIEW. But all I can think of is what a cool toy this would have been when I was a kid.

Mostly this week has been upgrades to existing software...

webMathematica 2.3 - Adding 64 bit and multi-core support to Wolfram Research's web server product.

BlockBuilder 1.1 - Linux support, bug fixes and making it work with their Dynaflex Pro product.
Dynaflex Pro 2.3 - The other half of the fix above.

PoochMPI Toolkit for Mathematica - Add on from Dauger Research to support gridMathematica use on their HPC platform.

Labview 8.20 released

2006-08-08T17:15:00.000Z

National Instruments released a signficant update to Labview today.

There is an extensive and diverse set of enhancements covering programming tools (.net and dll integratation), web based access, UI improvements with OpenGL support and touch screen control support, and improvements to real time data flow.

The subhead to the press release was interesting "LabVIEW 8.20 Delivers General Compatibility with The MathWorks, Inc. MATLAB® Software, FPGA-Based Rapid System Prototyping and New Modulation Toolkit" because it was hard to find what compatibility with MATLAB meant.

At first I thought it was going to be a friendly link, in the way that Maplesoft are trying to be closer to Matlab. But it looks more like a good old fashioned migration package. You can take your old Matlab .m files, throw away Matlab and run them in the NI "Mathscript" module.

There is no shortage of Matlab clones already but I think they mostly miss the point. Matlab is at its heart a bit old fashioned, its success is really about the Simulink graphical programming environment. Here NI stand out already the only serious competitor to Simulink, so adding .m code support is quite significant.

Four new control systems tools

2006-08-04T12:59:00.000Z

You go weeks without anyone releasing new control system design tools and then four come along at once...

National Instruments entry is NI Motion Assistant 2 for prototyping single and multi axis motion control systems.

Mathworks goes for top dollar with its $2000 tool (System Test) for checking if its other tools are giving the answer that you expected.

Wolfram wins the prize for most technical press release with lots of big words that I don't know anything about, for Polynomial Control Systems - a Mathematica add on that does what it says on the box.

Finally, Maplesoft wins the prize for niche product of the week with a Tire design add on to their Dynaflex add on to Maple. If you design tires you might find a use for it. But it did give them a chance to say "for use with Simulink" in a press release again!

Genstat 9.1 released

2006-08-01T10:49:00.000Z

With a modest change in version number, VSNI have provided a fairly broad set of improvements to the respected statistics package Genstat.

Once (in the 1960s) a leading (they claim "first") statistics package, Genstat has been completely overshadowed by newer packages.

But since a change of management a couple of years ago, the company seems to be getting itself sorted. With a steady development effort and a better attitude to marketing, illustrated by its recently redesigned website, they may start to regain some market share.

Mathworks links for revenue

2006-07-27T16:09:00.000Z

Mathworks has announced a couple of new links between Matlab/Simulink and third party software. One to TIs Code Composer Studio, an IDE for DSP design and another to ModelSim, the circuit design software.

Once a link was considered just a new feature rather than a whole product. I haven't studied the two links to know if they are extensive implementations, or basic. Either way, they are just plumbing. But at $1000 and $2000 respectively, it seems that Mathworks sees plumbing as big business.

That doesn't really happen much in hardware now, the cable costing as much as the equipment it plugs into. But hardware doesn't have so many proprietary interfaces, and internal secrets, meaning that anyone who cares to can make cables.

Extinction level events in software

2006-07-20T12:39:00.000Z

If you read this site for news, skip this item, it is just speculation.

I sometimes think that the software industry, scientific software included, can behave a bit like evolution. When times are good, lots of diversity occurs- the space of possibilities and choices is explored. When times get lean, the fit survive and the population (of users) dwindles for those that are not.

A software companies can exist on small user bases, as long as one person has the inclination to answer the phone and use a CD copier. But every so often, there is an extinction level event, a stepwise change that can kill off the weak (or just unlucky) and those that survive move into the territory relinquished by the dead. In evolution, think asteroid killing off the dinosaurs leading to the domination of mammals. In software it is things like the release of a new operating system.

Reasonably healthy software has died out in the past because it had been written for an outdated technology and wasn't successful enough to invest the development time to update to new technology. It happens less now than it used to. When I first learned to use a computer in the early 80s, it happened all the time. Every new technology was incompatible with the last. These days the platform providers make huge efforts to reduce this. I am sure most Windows XP software will work just fine on Vista, but some won't. Just as most OS X software runs on MacTel, but not all.

What got me thinking about this was the last piece about Multicore support. Is the coming of multicore computing a slightly slower extinction level event, like, say, global warming. At first it seems pretty harmless. Even more than the coming of Vista and 64bit platforms, multi-core promises complete compatibility.

But like global warming, it fundamentally changes the eco-system, and particularly for scientific computing. A lot of algorithms, have been studied and optimized in minute detail. Within a few percent, you should expect common algorithms, like say ANOVA, FFT, or basic linear algebra to be similar speeds on all of the systems that provide them. But optimizing algorithms for parallel computation is very different and is also somewhat less studied. There is a lot of work to be done by all the providers to make the shift. Nothing may be seen on the surface but lots of money will be spent behind the scenes.

It is not just that the weak will not keep up with improvements, but as multicore architecture is based around reducing clock speed, while adding more cores, those that don't shift may appear to get slower, while those that do will get faster.

Those that were once at least adequate, may very soon appear unfit. It may be in the small print of new features, but keep your eye on whether your prefered system starts claiming improved multi-core support over the next couple of years.

Stata prices by speed

2006-07-18T10:17:00.000Z

Statistics software Stata is now available in a new flavor- Stata MP. From the user point of view, this is exactly the same as Stata SE except a lot more expensive. The differerence is that its number crunching is multi-threaded over multiple CPUs, if you have them.

While the concept of paying more for a higher performing version is alive and well in hardware, where the same chip often comes at a variety of clock speeds, it has largely died out in software.

Many years ago, when a math coprocessor was an option on a PC, you got different versions of software that would make use of it or not. The problem was that with the 486 came on-CPU coprocessors, and they quickly became the norm rather than specialist extras. Very soon it stopped looking like there was a fancy version of the software that did something extra, but that the other version was a just a crippled version.

I think trying to revive this pricing model won't work. Multi-core computing will sweep in very quickly; Intel has indicated plans to reach 32 cores by 2010. But for most that is just a detail, what matters, as always, is that the computer you buy tomorrow will be faster than the one you have today- unless you have the crippled single thread version Stata SE.

Software upgrade pricing strategies

2006-07-14T10:25:00.000Z

Having been stung recently with a big servicing charge on my car, and being short on news this week, I thought I would do a survey of upgrade pricing strategies for different software companies.

Many people only look at the purchase price of software and do not consider the long term cost of buying into a technology. If you plan to stay up to date with new developments, compatible with other people using the same software, and compatible with future OSs and other technology, then you should also consider how much the software company will take from you tomorrow.

Today I will look at the upgrade route to staying up to date. (I will save service for another day). The survey is simple it is the upgrade fee from the most recently superseded version, as a percentage of the current purchase price. Where possible I have used US$ prices without special offers or discounts. When those were not available, I used Euro or GBP distributors prices.

The survey does not take into account how major the upgrade is or what the purchase price of the previous version was.

The results surprised me in their range. Here they are with the most painful first....

Matlab - not available (continous service renewal required)
TecPlot 360 80% (includes a year of maintenance sold separately for 15%)
TecPlot Focus 80% (includes year of maintenance sold separately for 20%)
O-Matrix 59%
Gauss 52%
Maple 50%
Mupad 50%
MathType 40% (from any previous version)
Scientific Word 35%
Scientific Workplace 35%
Endnote (download) 37%
Endnote physical box 33%
Minitab 33%
MathCAD 30%
Origin 27%
Mathematica 24%
Stata 18%

Caveat Emptor!

MathCAD security vulnerability revealed

2006-07-11T08:46:00.000Z

Like most interpretted systems, MathCAD suffers from the need to pass source code of your work to others for them to be able to use it. Like many systems, it tackles this with an option to encrypt the source code and give password access control to it.

An analysis of the vulnerabilities of the MathCAD implementation was released a couple of days ago on a Russian "security" site.

It turns out to be shockingly simple to bypass. Essentially, open in a text editor, find the "is-locked" attribute and type "false". The article also describes how to change content, fake the timestamp, and re-lock it.

If you were relying on this to secure your property or control modification, consider your work now open-source and meddled with.

Clearspeed accelerator goes mainstream

2006-07-05T09:42:00.000Z

At the risk of this turning into an HPC site, here is another story on high performance computing...

ClearSpeed Technology are a UK startup who make a neat co-processor board. Slot one of these 96 core boards into your PC and you get a boost of 50Gflops at the cost of 25W of power and a few thousand dollars!.

So far they seem to have mostly been involved in expensive specialist systems, partly because you need to optimize your code to take advantage of it.

This week though, they have signed an OEM deal with IBM which will put the boards into the IBM System Cluster 1350 product. This will be very important for raising their profile, driving support for the platform and making them some sales.

They have also made a start on cultivating third party software support, announcing this week, Mathematica support, with a claimed "quadruple performance". I think we can expect to see more such announcements in the coming months.

Mupad 4 released

2006-07-03T10:54:00.000Z

Last year I wrote quite a pessimistic piece about Mupad's problems and argued that they needed a compelling upgrade to make enough money to avoid last August's near bankruptcy but that they probably lacked the resources to produce one.

Well the upgrade is here, just in time for the summer, and I feel a little more optimistic, but not because of the upgrade.

The headline feature is "a new uniform user interface for all supported operating systems". Does this mean some backward incompatibility to try and draw users up to Mupad 4? It goes on "is platform independed [sic] and provides more formatting features" though none are explicitly listed.

On the computation side, top features listed are some new elliptic and Jacobi functions, improvements to symbolic factoring, anti-aliased screen graphics, and around 30 other changes that seem too trivial to note (eg "The option Column of output::tableForm has been renamed to Columns.").

It's not a compelling upgrade at all, so why do I feel more optimistic for them?

Well, at the same time as launching the upgrade, they re-designed their website and it shows a different attitude than before - much more commercial. There is a new web store (though you have to buy in Euros), a big list of new international distributors, and a price list focused on using the existing user base (eg upgrades at 50% of a new license fee). The shyness about making money that was exemplified by the previous free versions of Mupad has gone.

There is also a news item "SciFace cooperates with the university of Kassel" suggesting some new research backing, but there are no details and no mention on Kassel's site, so this may be mostly cheerleading to raise users confidence in future development.

Whether the new focus on the "bottom line" will be enough, time will tell, but at least there is fight.

Maple moves closer to Matlab

2006-06-30T10:55:00.000Z

This week's announcement of a new "Maple toolbox for Matlab" raises interesting issues, even though the product being announced is quite minor.

The product appears to amount to a new command for Matlab that sends computations to Maple. The press release claims that it is "bi-directional" though no examples are given of the reverse call being made.

Firstly, this is yet another example of the Maplesoft strategy being to position Maple as a companion product for control engineers using Matlab. See my speculation on this back in March. All the explanation for the link are written for the Matlab user, not the Maple user: features listed are mostly not the new link features, but a summary of what Maple is, only two of seven screenshots on the website show any interaction between Maple and Matlab, the others show Maple being used for control calculations .

Maplesoft's recent announcement to discontinue all of the platforms that it supports except Windows XP, Mac, Linux and Solaris makes sense in this context, as these are the exact platforms that Matlab supports.

However, it is interesting to note that this appears to compete with the existing relationship between Maplesoft and Mathworks.

Mathworks already offer two "Symbolic toolboxes" which are the same as this - Maple (or a subset of Maple) and a link to call it from Matlab. What's different? This one uses the current version of Maple, while the Matlab toolboxes use Maple 8 (three versions behind). This one includes the Maple interface. This one is sold by Maple. This one uses different syntax.

Mathworks has not made any announcement of this new toolbox and neither party has said anything about future support for the existing symbolic toolboxes.

So while the product strategy seems to move closer to Matlab, the business relationship seems to move away from Mathworks. I can only speculate to the reasons - lack of interest from Mathworks to upgrade the existing toolbox or perhaps Maplesoft is unsatisfied with the royalties that it gets from the Symbolic toolbox sales.

Government bodies and computing don't mix

2006-06-28T13:54:00.000Z

Something amusing today. A news item on Grid Computing Now says...

"THE TWO principal Grid standards bodies, the Global Grid Forum (GGF) and Enterprise Grid Alliance (EGA) have announced the completion of their merger, forming the Open Grid Forum (OGF)."

It starts with some exciting phrases about passion and action...

"This merger integrates the passion, expertise and experience of the EGA and GGF members to enhance our capability to deliver results faster, communicate more clearly and collaborate more effectively," said Linesch."

But then in a reminder that no government, that I am aware of, has ever managed the deployment of a major IT project without overspending and over-running, or ever written a successful piece of software, it then goes on to say...

"Over the next several months, OGF will focus on completing the merger integration. It will finalise the Board of Directors and day-to-day operations leadership team; transition current members and recruit new members into the organisation."

Look to industry for the action.

It reminds me of the People's Front of Judea in Monty Python's The Life of Brian- read the transcript here to understand why.

Eclipse IDE for Mathematica

2006-06-21T14:41:00.000Z

Wolfram Research have announced a new integrated development environment for Mathematica based on the open source platform Eclipse.

It always seemed that because Mathematica's language is very elegant and readable that Wolfram felt that development tools like debuggers were somehow unnecessary. While I have always liked the language (eg see comments on function naming), the lack of developer tools was definitely a weakness. It is good to see this deficiency finally corrected.

Wolfram have hooked up a large number of the Eclipse features: code editing (syntax highlighting, error reporting, local variable coloring, code outline etc), performance profiling, revision management, test management, a build system as well as step by step debugging.

As I commented previously, Eclipse has been gathering a lot of momentum. So as well as helping Mathematica users, this will make Mathematica more accessible to developers who use Eclipse for Java, C or FORTRAN development, who already use Eclipse.

PTC plans for MathCAD

2006-06-21T08:34:00.000Z

A few details of the plans that Parametric have for MathCAD, after the purchase of Mathsoft completes, were released at their user conference recently.

The headline feature is, as I predicted, a link between their Pro/Engineer products and MathCAD, which will allow you to link a parameter in Pro/Engineer to a variable in a MathCAD worksheet. Since they already have this capability between Excel and Pro/E it is not clear that they will push this 100% as the "right way" to work.

MathCAD will remain available as a stand-alone product- logical since the management of Parametric have stated their plans to reach a revenue goal, so as much of the reason for buying Mathsoft was to buy their revenue as to support any strategic technology plans.

One amusing quote from the PTC panel Q+A lead by Jim Heppelmann and Brian Shepherd was that there are 1.4 million MathCAD's out there. This is 400,000 less than the number that Mathsoft are quoting. Having just spent $60M you would think you would remember the right number- or perhaps Mathsoft have been guilty of just a little hype.

Endnote X released

2006-06-14T11:49:00.000Z

Thomson ResearchSoft have released a new version of their popular reference management software Endnote.

The main developments are in the handling of PDF files and less exciting, but very important for such a product, improvements to searching and organization.

I have always felt that PDF files were the wrong way to go for information, with their focus on visual presentation rather than semantic markup. But much as I would like to see XML type formats be developed further, there is no serious competitor to PDF right now, so like millions of others, I use them widely.

For Thomson to position themselves as the people to organize your PDF information, is a smart move.

Is Maplesoft MacTel support what it seems?

2006-06-13T12:57:00.000Z

I usually try to add personal experience and opinion to news, but last week with the Maple MacTel announcement, I reported it from the press release, like most of the Mac media, without question.

But queries on the Maple user forum hint at something more to the story. Two users have pointed out that while the Maple interface layer is claiming to be a Universal Binary, the Maple computational engine is reporting itself as a PowerPC application, and therefore running under emulation. Now since the computational engine is the purpose of the product, this rather undermines the claim of a native port of the product.

Furthermore, since the majority of Maple interface is written in Java and therefore is designed to run on the virtual platform of the JRE (likely slower than native PowerPC code running on Rosetta emulation), it is not clear that the native code can be doing anything very significant.

Perhaps this will turn out to be a misunderstanding or a bug is causing Maple to mis-report itself. But a thin layer of native code hiding an unported application would not endear Maplesoft to the Mac enthusiast community.

[[Update 15-Jun-6]]
So the issue turns out to be carelessness rather than a wholescale absense of Mactel code. Also, I had not noticed that the press release had carefully avoided the use of the word "Universal", as the Maplesoft statement that follows explains. Goes to show even press releases are sometimes worth deeper analysis!

"Hi there,

I'm a developer and a resident Mac expert at Maplesoft. I'd like to clarify some of the confusion about the state of the Intel version. The summary is, everything that counts in the kernel and GUI is universal. We have a two remaining vestiges of PowerPC code, but they should not impact performance in any significant way.

Here's a detailed explanation of what's going on.

1. A PowerPC process (mfsd) shows up when you start Maple. This process is a third-party license manager daemon and is not involved in Maple computations in any way. It won't affect the performance of computations done by the Maple kernel (mserver), which is a universal binary.

2. Maple 10.app shows up as "PowerPC" when you do a Get Info from the Finder. This is because Maple is packaged as a Java app, and such packages contain a launcher application stub (JavaApplicationStub). This is a tiny executable that launches Java and then quits. Unfortunately, we didn't update this stub (oops) with the rest of the update, so we're still using the PowerPC-only version. We'll address this in a future version, but in the meantime rest assured that the worksheet uses the Intel JVM and takes full advantage of Apple's latest and greatest Java technology for the Intel platform. (Read: it's a LOT faster than before!)

Because of this remaining PowerPC-only code, you may have noticed that we can't technically use Apple's "Universal" moniker for our product.

I'd be happy to answer any other questions you may have about Maple, whether on the Mac, or in general.

Yu-Hong Wang
Senior developer, Maplesoft"

Mathematica and Matlab add support for Microsoft CCS

2006-06-09T13:39:00.000Z

As I posted yesterday, Microsoft released their new HPC OS, Compute Cluster Server 2003 today.

Wasting no time at all both Wolfram Research and Mathworks have already released updates to their HPC products to support the new technology. Both have made the update free, though Mathworks appear to restrict to customers with active service agreements.

I am not sure if there is a recommendation hidden in the diplomatically similar comments provided by John Borozan, group product manager, Windows Server Division at Microsoft...

On Mathematica: "Microsoft is pleased to work with Wolfram Research to deliver a quality computing experience to our shared customers. gridMathematica is an essential application for high-performance computing, and with Wolfram's Cluster Integration Package for Windows Compute Cluster Server 2003 we are pleased to know that our customers will have a zero-configuration, seamless solution that meets their needs."

On Matlab: "We are pleased to work with The MathWorks to help facilitate the next wave of discovery in the technical computing community by delivering software that addresses some of the barriers engineers and scientists face. The MathWorks' Distributed Computing Toolbox and Windows Compute Cluster Server 2003 work together to simplify the process of developing distributed computing applications so that users can address more complex problems out of the gate."

Microsoft CCS takes on HPC

2006-06-08T16:38:00.000Z

Microsoft's new platform for high performance computing is rumored to be launching officially tomorrow(ish).

Anachronistically named Microsoft Compute Cluster Server 2003 (because it is an extension on the Server 2003 product range), it supports job scheduling, MPI, MPICH and MS MPI messaging, active directories, and other goodies for setting up and running a cluster.

While a little late on the HPC scene, Microsoft appears to be taking it seriously. And while they do not succeed in everything that they do, you have to expect this to make a big impact in driving the accessibility of (relatively) low cost HPC to a wider audience.

Also significant is that it only supports 64bit hardware. How long before we expect that of a new software release?

Mac Intel support for Maple

2006-06-06T13:02:00.000Z

Maplesoft announced today that they will shortly start shipping a native Mac Intel (Universal Binary) version of Maple. They have not said whether this will include the "Classic" interface version of Maple, still used by 30% of their users, or just the new Java based interface. It is likely to be just the latter.

Delivery has been rather tardy compared to, Maple developer, Paul DeMarco's estimate to the Maple user group, on Jan 17 that it would be "several weeks",and is nearly four months behind Mathematica.

Still this is well ahead of Mathworks who are now estimating that they won't even have a 'beta' of Matlab for Mac Intel until "late fall". Mathworks developer Brian Arnold, was surprisingly open about the situation to the Matlab user group, commenting, "It sucks that our Intel Mac transition is taking this long, in the face of Apple's accelerated hardware changeover, but we can't (currently) make this go any faster."

Other scientific suppliers who have managed to ship Mac Intel software include statistics softwares Stata and XLSTAT, visualization software Imaris, TeX editor LaTeXiT and ChemSpotlight.

[[14-Jun-06 Update: Is Maplesoft MacTel support what it seems?]]

O-Matrix version 6

2006-05-31T09:47:00.000Z

Harmonic Software have released version 6 of their Matlab-clone software O-Matrix. The new release adds some basic statistics functions and some performance improvements.

The statistics are mostly so basic (random number generation over different distributions, a small number of PDF and CDF functions, descriptors like mean, median, covariance etc) that long time users, have no doubt added them themselves.

The performance improvements are more impressive when compared to the previous version but don't seem out of line with other software on my machine. I have not done a proper comparison, and the comparison provided by Harmonic, rather misleadingly compares this new version to Matlab 7.01 (released in 2004). Either they don't have Matlab 2006a or that gives an unfavorable comparison!

But this is all a distraction from a more fundamental issue- O-Matrix is still fundamentally a numeric-only matrix manipulation system. In the 80's, when the alternative was gluing FORTRAN libraries together, this was a big step forward. But now there are much better simulation languages. Of course the same is true of Matlab. But Mathworks, I think, recognizes this and puts comparatively little effort into Matlab development. Their value is in the toolboxes and Simulink- Matlab is a just a rather dated and mundane component. Harmonic have little extra to add to O-Matrix.

The only Matlab customers they can attract with a faster cheaper version of Matlab, are those that think that the world begins and ends with matrix manipulation.

Is the new TI Calculator good news for TI?

2006-05-26T09:29:00.000Z

Texas instruments are about to launch a new model of handheld calculator, with the rather long name of the TI-Nspire™ CAS+.

At first glance this is an obvious and beneficial evolution of their product line. It appears to have a much higher resolution grayscale display like basic Palm's or Pocket PCs, and they have used this opportunity to tweak the software so that instead of having to switch between a spreadsheet view, calculator view or graph view, they can be displayed at once in a basic windowing system. It can also drive a full size monitor.

But I think that TI may be being drawn into dangerous ground. Consider the strength of their position, which as I commented at the start of the year, is based on calculators being supported in the classroom and exams but computers not. As the TI Calculators become more like handheld computers the pressure will increase to allow a wider range of computer solutions. As the conceptual gulf closes, TI's software (barely changed Derive from 15 years ago) will come under closer scrutiny. In the calculator world it is very impressive, in the PC world it would lose to almost every technical system available on capabilities and most of them on ease of use.

It is also only a matter of time before pocket PC's become routinely powerful enough to run PC based scientific software, blurring the distinction further(and perhaps only a few more years before your mobile phone can).

But what can TI do? They can't resist the pressure to advance towards this state, like lemmings drawn to the security of the lemming in front. Surely the solution is some serious investment in their software, but they haven't shown any interest in that yet.

Update 7 Sep 06: TI Nspire Calculator delayed
Update 23 Nov 06: Derive to be discontinued

Comparison of math typesetting software - Part 2

2006-05-23T13:37:00.000Z

Continuing last weeks examination of math typesetting software, following are some comments on the capabilities of the systems.

This is a big subject, so picked a few choice features to look at closely:
1) As examples of non-trivial expressions I looked at two nested constructs: how does sqrt(sqrt(sqrt(.... look, and how does 2^2^2^2^2 look?
2) Do the systems have any automatic layout? IE What happens if you can't figure out how to fit an expression on the page, what happens if you re-size the page or fonts so that it goes off the page?
3) Antialiasing. These days its embarrassing to give a presentation by printing on transparency and using an OHP, so expressions should look good on screen too.
4) Detail control. What can you do, if you don't like the default choices?

Maple:
Maple was the only system that made no attempt to handle the nested expression test in any special way. Because the sqrt sign gets bigger as it gets nested, the leading slope, which is attractive for a small sign, takes up a lot of room as the sign gets bigger, making this expression very wide. All the other systems used a steeper, more compact, sqrt sign for the outer placements. Likewise, there is a drop in font size to the first superscript, but none for the subsequent superscripts.
There is a layout control system, typing a very long expression next to the pi in the test caused it to line break at the page width without breaking the meaning of the expression. sqrt signs were replaced with new parenthesis and ^1/2. However, it rather unnecessarily chose to put a line break after the opening parenthesis each time, resulting in lots of unused space. Oddly the closing parenthesis all happily sit on the same line.
Maple managed to antialiase the hard parts just fine (parentheses even when they are stretched to accommodate large expressions, sqrt signs etc), but didn't antialias any characters, which one might expect to be provided by the OS for free. (Click on the image to see the full size version, to see the jagged pi and 2s).
There appears to be no control over the look of typeset expressions beyond font characteristics (size, font, color etc).
Overall pretty basic, considering the system has just had a major overhall.

MathType:
MathType does a little better on the nested test. First and second sqrt are different and first and second superscript are different, though third occurrences do not change.
There is no automatic layout in MathType, make the expression too large for the page and it just gets wider and wider, as if you are on an infinite page, so it is up to you to plan the layout of a large expression. Without a page preview or any indication of the page width on the ruler, you may well waste some paper.
Antialiasing is the way round I expected. Characters handled nicely, sqrt signs showing jagged edges. Spanning parenthesis are handled well.
There is some basic detail control s- you can set the relative size of certain fonts (eg the superscript is by default 58% of the size of the parent character), but not many of them.

Mathematica/Publicon:
Mathemetica performs well on the nested test. There are five levels of different sqrt sign, and nest superscripts keep getting smaller indefinitely until a floor font size is reached.
Automatic layout happens in two ways. Expressions break automatically as the page width is exceeded. In this example, Mathematica switches to a prefix version of a sqrt sign and parentheses, instead of the ()^1/2 favoured by Maple. You can also ask it to do an automatic layout to try and fit the expression the best way that it can, though this discards any character level settings and may replace linear notations like ^ or / that you have chosen with superscripts and fractions.
Mathematica fully antialiases.
One difference is that math symbols are provided in two new fonts, which switch depending on whether the surrounding font is mono-spacing like Courier or proportional like Times. Whether you think this is a good thing is a matter of taste, but you can use the Symbol font like the other systems, if you prefer.
The place where it seems to stand out is in detail control. There is an overwhelming number of options from how thick the horizontal line is in a fraction, to how high or far across the nth root value appears in a radical. There are perhaps a couple of hundred options and they can be set for a single character, the document or as default.

Scientific Word/Scientific Workplace:
Scientif Word came second in the nesting test. There are three levels of sqrt, though it makes these go further by using the third at the fourth level of the test, and two levels of superscript size.
There is no layout control, suffering from the same infinite page as MathType, though it does have a page preview.
Only characters are antialiased, and with the quite compact sqrt sign, this is probably the I think the worst screen representation from my examples.
Detail control was good, with access to lots of TeX tags, though the fact that these are applied through named tags was very frustrating. One doesn't appear to be able to select a character and make it 48 point. You have to edit a tag to have a 48 point and then apply the tag, being aware that any other use of that tag will inherit that value. Style sheets and abstraction like this are good, and ScientificWord provided a good range of pre-defined style sheets and tags. But to be forced to work like this is frustrating. I don't know if this is imposed by the use of TeX as the underlying language or if it is Mackitchin imposing good practice on us!

All systems claim to export into HTML, MathML, and TeX. I have not looked closely at this. Scientific Word and Publicon also support a range of target specific TeX or XML formats (eg specific journals).

So in conclusion, the Mathematica family performed best under these criterion, followed by the Scientific Word family. Though I repeat, there is a lot more to this subject, and this review was just a sampling of the feature space.

Putting this with my experience in the first part of the review, the Mathematica family are the clear winners (Mathematica if you have $1000+ and need computation, or Publicon if you have $150 and don't). For simple and easy to learn I still like MathType, though I was rather surprized by how basic it is considering its long pedigree and well known name. But at $130 its not expensive.

Silicon Graphics files for bankruptcy

2006-05-12T11:54:00.000Z

Once the leader in high end visualization, Silicon Graphics is now on its knees. Court approval for its bankrupcy status will keep SGI going for a while: "We are pleased with the approval of our 'first-day motions' by the Bankruptcy Court," said Dennis McKenna, chairman and chief executive officer. "This approval will enable SGI to operate globally and meet normal business obligations."

He describes plans to reconstruct the business having shed most of its debt in a deal that sees existing shareholders lose their remaining share value. But any remaining confidence in the long term adoption of SGI technology must now be gone, so it is hard to imagine him turning it around.

It's sad. Ten years ago I was given a tour of SGI offices in Switzerland by a friend and it seemed the greatest place work - superb offices, excited staff and everyone had the most amazing (for the time) SGI workstations on their desks.

But a decade is a long time in IT.

Comparison of math typesetting software - Part 1

2006-05-11T17:00:00.000Z

Since I haven't done much real work on this site for a couple of weeks, and "comparison" is the keyword that brings a lot of people onto the site, I took a close look at the typesetting features of several products. My review covers: Maple, MathType, Mathematica/Publicon, OpenOffice and Scientific Word/Workplace.

Scope:While mathematical typesetting is only one part of publising technical papers, this review compares only that aspect. Some of the products do much more than this (Maple, Mathematica, Workplace also offer computation, data analysis, graphics etc, and Publicon, Scientific Word,/Workplace handle things like bibliographies, references, indexes etc), but that is not covered here.

The comparison will come in parts and today I will start with input. How do you actually enter your equation.

The first thing you notice is that all but OpenOffice have essentially similar kinds of palettes or toolbars, organized by types of symbols or structures. eg a palette for greek letters, a palette for matrix input etc. I was somewhat shocked by the absence of this from OpenOffice, which requires a language based input, based loosely on TeX type conventions. While keyboard input is important (and varied much more between products), I found the complete lack of mouse input frustrating as I learned the system and dropped OpenOffice from the rest of the comparison.

Maple:The palette system was neatly organized in a single column format and was easy to use, though you have to tidy as you go, as opening more than one section, quickly pushes parts off the screen.

You cannot type a keyboard combination for any special characters during typesetting, though you can type an entirely text based version of the whole expression and then format it.

Basic structures like superscripts and fractions happen automatically as you type the 1D version (^,/) which initially seems appealing, but means that you cannot type an actual "^" or "/" so expressions like appear to be impossible. Instead you must settle for the somewhat less clear expression on the right.

There doesn't seem to be any way to customize input, either with keyboard shortcuts or new palettes.

Maple does have the unique handwriting recognition feature, but regular readers will know that I was not impressed with that.

MathType:The surprisingly small to download MathType (4Mb), is described as the "Professional version of Microsoft Word's Equation editor". It has a very intuitive palette layout, though it was a quite sluggish to respond, and since double-clicking on a button brings up a button-properties editor, I found it frustrating to enter nested constructs like Sqrt(Sqrt(..

However, apart from a keystroke CTRL-Shift-G to go into a greek character mode, none of the special characters or constructs had built in keystrokes, though you can customize most palette buttons with a keystroke of your choice. [[Correction: Bob Mathews of Design Science points out that most have keystrokes, which one can see in the status bar on mouse over. I took a while to look at these, and it is true, perhaps 80% of the characters have preset keystrokes. Many are rather cryptic (ctrl+shift+k then # to type an Or symbol) making it hard to imagine knowing many of them, but the ability to change them mitigates this]]. There was a nice array of more complex templates organized by category, and you can also create your own.

Mathematica/Publicon: Publicon (the no-computation version of the Mathematica typesetting system) has the more elegantly organized palettes, with the nice feature of being able to scroll through the different sections with the mouse wheel), the Mathematica incarnation had separate palettes for different types of input, which is probably preferable if you plenty of screen space.

Neither had keyboard shortcuts to the palettes, but all individual structure elements and characters have keyboard access. The approach is the most thorough of the systems that I looked at, with every character having both a full name and a short name. Full names are very consistent, eg "\[Infinity] \[Alpha] \[LeftRightArrow] "etc. Shortnames were sometimes well chosen, sometimes less well and delimited with the escape key. eg "esc => esc" for \[Implies]. New keyboard short names and palettes can be created for any typeset structure in Mathematica but not in Publicon.

Scientific Word: The palette system has a neat button type that lets you tear off submenus to make separate palettes, but was otherwise the least well designed. eg the Sqrt button only comes with two placeholders (for the nth root) so if you want the more common Sqrt you must delete one of the placeholders, and rather a lot of cryptic icons (an Italic E means "Emphasized", an "A" with an arrow pointing to a "B" then an arrow pointing up represents search and replace!). There are no high level compound templates, and while you can customize which characters appear on palettes, you cannot create templates. [[Correction 23 May, thanks to Fred Chapman] - Shortcuts exist for all constructs and keyboard entry is provided for all characters by holding down ctrl and typing the TeX name of the character and then releasing ctrl. This is a system that mostly works well, except where the name contains an upper case letter. Trying to press and release the shift key without releasing the ctrl key can lead to some contortions.] Judge for yourself from the screen shots, but I also thought the palette design was rather ugly - though that doesn't affect its usefulness, its nice to work in a nice place!

Overall, when I compare input methods, I preferred MathType for entry level and Mathematica/Publicon as I became more experienced. Maple comes next then Scientific Word/Workplace with OpenOffice in a clear last place.

Continue to Part 2 - Which actually creates the best output?

Roundup of non-news

2006-05-10T08:26:00.000Z

There seems to be a lack of serious announcements at the moment, so here are some of the items I noticed which did not deserve serious comment...

In the category of "new marketing initiatives":
New bi-weekly MapleSoft podcast - have someone else read their website out to you.
Mathworks sponsors some student engineering competition (mostly with software).
Mathworks gets some noteworthy user to speak at their conference - interesting if you have heard of Thomas Scharnhorst.
Wolfram Research are giving away some Mathematica to China, and sending some people there on Mathematica roadshow.
Wolfram Research offer free online seminars to learn Mathematica.

In the "business as usual" category:
Insightful Q1 results out - 'steady as she goes'
MathSoft hire a new distributor - but that was before the agreement to sell MathSoft (oddly still not announced on the MathSoft site)
MapleSoft buys an algorithm- but they have always bought in external code, so what's new?

Still, at least those companies have something to say about themselves. Design Science, SciFace, UNISTAT and VSN haven't put a news item on their sites this year. Come on guys, surely you have done something worth mentioning in the last five months?

Do scientists and engineers prefer Firefox?

2006-05-05T13:37:00.000Z

It's just anecdotal evidence, but I was reviewing the site statistics for this site, and one piece of data stood out as interesting.

The browser used to read the site is running, fairly consistently at about 70% Firefox, 20% Internet Explorer, and about 10% combined of Opera, Safari and (new to me) Galeon.

Now I know that this is not normal, Firefox penetration is claimed to be around 20%, and professional sites that I am involved with are not much different.

Of course, it's small data, (my site delivers only a couple of thousand pages per week), and the self selection is much more specific, so "people interested in scientific computing, who are interested in blog comment, and can find a site that has barely promoted itself - prefer Firefox"!

The rest was all what you might expect, the average reader is American, British or Canadian, reads two pages from the site, is 25% likely to return, 10% likely to stay for over an hour and 95% likely to have got here through Google.

Visual Numerics moves to Texas

2006-05-02T09:32:00.000Z

In a strange and slightly rambling press release, Visual Numerics explains that it is moving its headquarters from California to Texas, along with some comments about corporate strategy that are so vague as to be almost meaningless.

There are some suggestions as to the benefits of Texas, from density of Fortune 500 companies, to the value of the oil industry and presence of universities. But no real coherence to the case or conviction in any of the points.

I think that this may just be a case of reducing office rent costs and hoping that if you talk enough, people will think you have said something interesting about why.

Parametric to buy MathSoft

2006-04-27T11:18:00.000Z

Only a month ago I reported on some MathSoft financial news. Now we have the answer to how the makers of MathCAD are going to stop making losses - they are selling up and going home!

CAD/CAE software giant Parametric is to buy MathSoft for the bargain price of $63M. This is a price per earnings ration of just 3. Compare that to Parametric themselves with PPE of 22, or the somewhat less realistic valuation on Google of 70. It looks like the backers, Edison Venture Capital, are pretty keen to get their recent $10M investments back.

Some of the information released confirmed my suspicion that the "Managed installations" in the previous news, were not so impressive. They work out to have an average value of $3000 per company. Not exactly the corporate site licenses that the press release implied.

What will happen next is known only to the senior management of Parametric. But what usually happens in takeovers is costs are cut by getting rid of people that do jobs already covered by the buyer. The recently opened MathSoft offices, that now look more like an expensive advert to Parametric, than a serious business step, may close. And at least some of the remaining distributors in other countries will be replaced by Parametric's existing sales teams.

I think we can safely expect the technical direction to focus on integration and interoperability with Parametric's other products. MathCAD has long focused on engineering more than science and any lingering interest in science may now be gone.

Matlab virus reported

2006-04-26T08:37:00.000Z

Security specialists F-Secure are reporting the existence of a functioning Matlab virus known as "Bagoly".

This is a harmless proof of concept virus that replicates itself into all of the infected users Matlab ".m" files. It does little to hide its presence and so should be easy to detect and remove and Symantec added it to their virus files within 24 hours.

However, proof of concept code is likely to be modified by others with more malicious intentions, making the sharing of models and computations in Matlab much more dangerous than it was before.

64 bit scientific computing

2006-04-20T17:18:00.000Z

A press release from MathWorks tells me that Matlab now supports 64bit Windows, not in previous versions as I mistakenly commented in previous news. This prompted me to do a quick survey of software which should benefit from 64bit support to see which does.

The real benefit of 64 bit to scientific computing is the large memory allocation that is made possible. No 32 bit application can use more than 4GB of RAM, and OS issues often reduce that to 2GB. So products that should go 64bit are those which might be applied to large data or complex data structures, my survey picked a collection of such products.

The only two important 64bit platforms are Linux and Windows. The first already has large penetration, and the other is, well, Microsoft, and so automatically important!

Supports Linux 64 and Windows 64
Matlab, Mathematica , Stata

Supports Linux 64 but not Windows 64
GAUSS, IMSL, Maple

Supports neither
AutoCAD, ChemOffice, Genstat, MathCAD, MatrixX, Minitab, O-Matrix, Origin, S Plus, Systat, Unistat

Rather disappointing!

On closer inspection, nearly all of the products that I looked at in the "neither" category, support only one platform - Windows 32 bit. Having had a strategy of focusing on one popular platform, it is not surprising that they will wait until 64 bit is not only mainstream but majority. But I suspect that because these companies have no experience in porting and have made no effort to write portable code or muti-platform build systems, it may also be harder for them not make the shift.

With the release of Vista at the end of the year, the 64/32 distinction will be blurred, allowing these companies to hide the fact that they do not have 64 bit support.

So when you do make the switch to 64 bit, remember to ask if it will make any difference for the software you use.

Site comment

2006-04-18T17:19:00.000Z

Apologies that, with the easter break, I have not been able to write an article for a few days.

So instead, I will fill this item with an appeal. If you find this site useful or interesting, please add a link from your own site, or mention it to some colleagues.

Also please notice the "Submit news" link on the left hand side. If you have suggestions for articles you would like researched, or interesting experience you want to share. Drop me a line.

This goes especially to the regular readers that I have from some of the software suppliers - yes, I see you in the site statistics! We all like leaks and insider gossip, so please share.

I will get an article together for later in the week.

Beamer support for Scientific Word

2006-04-13T15:46:00.000Z

MacKichan have added experimental support for the Beamer class of slide show transitions to Scientific Word

Software makers add features, it's what they do. Typically to take them into new markets. But I have always felt that the MacKichan have been trying to make Scientific Word into things that it just isn't.

Scientific Word is a perfectly nice TeX editor, and with the right knowledge makes very nice documents. But rather than develop that idea, they tried to make it into a computation tool by bolting on, first a Maple engine, and then a MuPAD engine. It was never a comfortable mix.

Now with the Beamer class support they are trying to turn it into some kind of PowerPoint clone with typesetting. If you are one of those people who likes PowerPoint, then why not just export your work into it? If you are not happy with being limited to slides, then why not show your work in a real working environment where you can show it live?

Perhaps I am not in supportive because PowerPoint always seems to be wielded by salesmen!

Matlab GOTO joke

2006-04-11T09:48:00.000Z

As an exercise in not taking oneself too seriously, I will admit that, probably for the first time in at least 25 years, I fell for an April Fools Day joke.

I wrote the following piece, and only as I wrote the last line, I realized that this proposal wasn't real. Nice one Loren.

Matlab developer Loren Shure writes on her blog this week about a proposed spec for a adding the GOTO command to Matlab. Now I was trained to believe that GOTO is a source of pure evil in programming. The core Matlab system design has never been very "modern", showing too clearly its FORTRAN background, but even so, this seems like an unnecessary backward step.

Loren writes that "The problem with GOTOs is not the GOTO itself, but with the label." but then goes on to describe a bizarre spec that includes fractional line numbers, complex line numbers and random GOTO targets. This would be guaranteed to make "spaghetti code".

She wrote this on April 1, and it looks more like an April fools joke than a serious proposal.

S-Plus 8 goes to Beta

2006-04-06T08:16:00.000Z

Unusually for this part of the software industry, Insightful are being very public about the timing of the beta program for S-Plus 8, which is planned to run from now until August. So one might expect a quarter-4 release of this popular statistics software.

The feature list which will include:
-A new package mechanism for the distribution of addons.
-A collection of new statistics functions.
-A debugger in the Eclipse based workbench.

The package mechanism turns out to be a port of existing functionality in the R language (the open source cousin of S-Plus), and seems to be a move to get closer to the large R community and existing R resources.

The Eclipse debugger is already widely used by developers in other languages and so adding support for S-Plus will be beneficial to many. The Eclipse project has a lot of momentum behind it, and for Insightful to extend their use of it seems sensible.

Review of Maple comparison against Mathematica, MathCAD and Matlab

2006-04-04T11:11:00.000Z

I got sent an interesting document which appears to have been produced by Maplesoft (it uses their graphic design and asserts Maplesoft copyright). It is a bullet point list of Maple 10 compared to Mathematica 5.2, MathCAD 12 and Matlab R14.

Such documents are always interesting, not because of the conclusion, which of course is pre-determined, but because it reveals what the author considers important and where the author believes his product has an advantage.

The document contains 46 items across the following catagories: Ease of use, Interactive Document, Mathematical Power, Connectivity, Testing and Assessment, Add-on Products and Platform support.

My first impression was that its emphasis was centered far away from the core features of the products. The largest section, over a quarter, was on connectivity (but with no reference to many important areas of connectivity: code linking, SQL, web services, I/O board reading or import of any file formats. Their concept of connectivity was just file export and code generation).

Conversely, important areas were reduced to one bullet: "Full programming language", "2D plotting with interactive scaling", "Symbolic and Algebraic solving". Odd given the title of the document "Mathematical analysis software comparison chart".

There were several rather trivial items "Handwriting recognition", "Graphing Calculator Interface" which, I suspect are not supported by the other suppliers by choice.

There were several instances of misleadingly carefully chosen words. e.g Under Platforms Linux gets listed as 32bit and 64bit (Maple supports both), but Windows does not get split into 32 and 64bit support (Maple supports only 32bit, while Matlab and Mathematica support both). [[Correction 20-Apr-06: Matlab was not Windows 64 compatible when this comparison was written, but is now]] "Integrated math dictionary" scores a point for Maple and a half point for Mathematica. mathworld.wolfram.com provides orders of magnitude more than the Maple dictionary, but it is not "integrated". Likewise "2D plotting with interactive scaling" scores a point for Maple and none for Mathematica, MathCAD or Matlab, as they lack the "interactive scaling".

But the most unreasonable part of this document are the inaccuracies. As far as I can tell they are all related to Mathematica. "Support of units and conversions", "High performance industry standard libraries (Lapack, Atlas, Blas)", are given as "no" and "Non-linear optimization" as a chargeable extra. I wasn't sure if this was ignorance or mis-representation, but I did find that entering into google "Mathematica unit conversion" or "Mathematica Lapack", and "Mathematica nonlinear optimization" each gave links revealing the fact that Mathematica supported these features within the first 4 hits. Hardly taxing research.

Overall a rather unconvincing and slightly sleazy document.

A view on the new Origin Viewer

2006-03-31T13:29:00.000Z

Perhaps I am not very sharp today, but sometimes the purpose of a product escapes me.

Origin have just released a free tool to view Origin Project Files.

Origin Project Files, are a file format for storing origin data, results, graphs and layouts in a single file. So kind of like a .zip or archive. The format itself doesn't provide any additional features as far as I can tell.

Now the reader, lets you browse the directory structure and get data or graphics out of it. Well, again, like you can with a .zip file.

I can't see why anyone would use this, except to get at a piece of data or output from an Origin users project. If you actually cared about the way that different parts or the project related, you would already be an Origin user.

If Origin projects used standard formats in the first place, then you would be able to browse and retrieve parts already.

Its seems like a solution to a needless problem.

KNITRO for Mathematica

2006-03-28T09:39:00.000Z

Yet another new product launch from Wolfram Research, the 5th in slightly over a month. Again it is a third party developed add-on to Mathematica. Wolfram is being very successful in attracting developers to the Mathematica platform.

This one is a little more confusing than the others. The press release says it "Provides state-of-the-art solver for large-scale nonlinear optimization". What is odd is that Mathematica already has good built in nonlinear optimization methods which scale very well.

So the new active-set and interior-point algorithms are only going to be of interest to a small set of people with quite specialist needs, which are beyond the built in functionality.

The full info at http://www.wolfram.com/news/knitro.html

Open source scientific computing - follow up comments

2006-03-24T12:11:00.000Z

My comments on open source software in science generated more feedback last month than any other article.

As well as a thread on sci.math.symbolic, there was an interesting response from Tim Daly, lead developer on the Axiom project.

There were three basic themes:

There was some heated debate on the best form of open source licensing. I have no particular expertise here, so I shall just duck that question.

There were several points that I had argued were a problem to open source, that people rightly questioned "so why is that more of a problem for open source than professional development?". Let me run through those now.

1) "The knowledge required for the topic and the interdependent parts of the system reduces the pool of possible contributors." It is quite true that the same issue applies to commercial development. But what I should have pointed out, is that the difference is that commercial development has an answer- it can pay more. Following the usual supply and demand laws of capitalism, the lack of supply of appropriate programmers forces salaries up until it attracts enough and equilibrium is reached. Or, as Richard Fateman pointed out in the case of Macsyma Inc, a failing company cannot attract the appropriate team and the failure is accelerated. In the contributed open source project, there is no salary or similar compensation mechanism. Indeed economics works in the opposite direction. Because the pool of contributors that you want to attract are particularly skilled, their time is more valuable, as they probably command high salaries in their day jobs, so it requires greater generosity to give their time for free.

The counter point to this are PhD students, who are typically highly skilled and expect to work for free, in pursuit of the qualification.

2) Tim Daly questioned why the complexity was a greater problem for part time contributors. Another point that I didn't make clearly enough. When I am deeply involved 8 hours a day in a programming project, I know every line of code that I am working on, the data structures, the function names and argument orders are all in my head. My 8 hours are all productive (apart from my human weaknesses of coffee breaks, day-dreaming etc). When I come back to that same code weeks or months later to fix a bug that might otherwise have taken 5 minutes, I spend the first two hours familiarizing myself with the flow of the code, that I have long since forgotten. If you take this over simplified example, and extrapolate. One full time programmer, achieves the same output as 24 programmers who work for one hour. Or another way, a full time programmer achieves the same in 2 hours, that a part time programmer manages in a year of working for an hour a week. Of course, the reality depends greatly on the amount you forget per break for given complexity of code, but I am sure you get the point.

As was pointed out there are many projects that are low on complexity and require widely available skills eg documentation translation, but these are pointless unless the central features are delivered.

The third strand, was from Tim Daly, and is the most interesting, being from a very different point of view (and well worth reading). If I can oversimplify his arguments- Axiom's purpose is not competition or dominance but in the act of donation where your work is available to others. He talks of a "30 year" time frame over which companies may well have gone bust, but free information lives on.

I was very much drawn to his view, but my cynical free-market thinking took over again. You can't separate yourself from the market, just because you are non-commercial. Even ideas need marketing, this is the purpose of publishing in journals, and presenting at conferences. If it was just to make the information public, it would be sufficient to put a copy of your research in the library and let it sit there waiting to be discovered. There is a wealth of research out there that is doing no good to anyone because it has long been forgotten or because competing ideas have been presented in a more compelling way. So Axiom must compete. Not for revenue perhaps, but at least for some level of mind share, and users, to keep attracting contributors.

The central question that I came back to was "If I produce some original piece of work that I want to share, should I publish it as a contribution to Axiom, or as a free piece of Mathematica or Matlab code and submit it to their share libraries?". Just like publishing a paper, I want it in the most prestigious place which gives it the widest exposure. Axiom must try to be that place, and I don't think it can take 30 years to get there.

Personal supercomputing news

2006-03-21T17:05:00.000Z

I commented last year on the embryonic market for personal supercomputing software.

Today, I happened to hear two related bits of news: A nice new piece of kit from Tyan (16 CPU computer with 64Gb of RAM). And the apparent demise of Orion multisystems.

As consumer PCs continue down in price, and approach the point that they deliver all the power that their users need, scientists and other users who's demand for CPU power is unlimited, and budgets are a little larger, are once again an appealing market.

But business can be a matter of timing. Orion, it seems, was just a bit too early to try and break through from specialist to large scale use of HPC and have run out of money before the market was ready. Maybe Tyan are here at a better time, or perhaps it is tomorrow's startup that will capture the prize.

Maplesoft goes for Control with Blockbuilder

2006-03-16T09:56:00.000Z

Back in February, I speculated over the contents of the rumoured plans for the Maple Blockbuilder for Simulink. The product is now out, and it looks like that analysis wasn't far off the mark.

The product appears to be an S Function extension to the code generator but also contains some control theory functions like transfer functions, state space and zero pole gain models etc, apparently like the Wolfram Research application Control System Professional 2. Whether these functions have similar range or depth is impossible to tell, as no documentation or serious examples are provided.

What is most interesting though is not the product, but the change to the company's presentation of itself that the release has triggered. What other company has someone else's product name (Simulink) larger than its own principle product on the front page of its website?
[Update March 22: Within six days, the page has been re-designed again, to return Maple to a prominant position, so I was probably not the only person to think this odd!]

There have been a couple of other small products for control engineering coming out of Maplesoft - ICP for system identification, Dynaflex for rigid body control. These had seemed rather fringe to a company that makes its money out of teaching calculus to students. Now with this web redesign, where they have been elevated to the front page and grouped together, I am starting to suspect that Maplesoft is trying to reposition itself as a provider of addons to Simulink and Matlab.

Matlab 2006a released

2006-03-14T17:42:00.000Z

Mathworks have released a new version of the Matlab product family.

The release itself is not very exciting, Mathworks has not even listed it in their "news" pages. The top Matlab new features are Windows 64 support and a new differential equation solving method.

What is very impressive is the meticulous documentation of the minor enhancements. While this is good marketing, as it makes the release look more significant, it is to be applauded as an irrelevant change for one person, is a compelling feature for another.

An interesting strategic change, is the shift to a planned two releases per year. Mathworks are showing some confidence in the plan by suggesting which months one should expect the release - and software releases are notoriously hard to predict.

While the virtues for the user in predictable release cycles are clear - user planning for deployment, shorter time to getting bug fixes - the benefits to Mathworks are also clear - a more compelling reason to pay for service rather than upgrades, and less "intellectual stock" tied up in development rather than in the market being sold.

This is the first release on this schedule, so they may not turn out to be as predictable as they hope. But while that might be embarrassing, it loses nothing. Greater pressure to release on time will be a positive drive on R& D but if too much importance is given to the deadline, then there will be pressure to cut corners - release unfinished features or buggy code.

But my biggest fear for this approach, is it makes big features- long term projects - harder to implement when you have to maintain several branches of code for different releases at the same time. We may see many more of these small incremental releases, and fewer "big new technologies" in the core products.

Did scientific computing just get a lot more boring?

New Mathematica applications: GeometricalGeodesy, LensLab, Rayica

2006-03-10T15:14:00.000Z

A flurry of new software releases since I last wrote about Wolfram Research: GeometricalGeodesy, LensLab and Rayica.

GeometricalGeodesy, which probably falls in the class of "if you don't know what it means, your probably don't need it", concerns itself with geodesy tasks like distances between points on the earth, different mapping coordinate systems and map projections.

Rayica is a ray tracing and optical design program, and the better named product, LensLab, is basically a "lite" version of Rayica.

What these all have in common, and with the last months announcement of Statistical inferencing, is that they are all, so called, "Third party" addon products. These are extensions to Wolfram Research's Mathematica product, marketed by Wolfram, but are written by other companies.

This is a growing trend for new technical software- to write as an extension to an existing scientific packages. While there are some benefits in access to the existing base of users and support from the manufacturer of the package you are writing for, I believe the central reason is reduced cost of authoring.

Many years ago, a programmer had to write every part of a program. Then came operating systems, and programmers no longer had to worry about disk operations, or whether you had an EGA or VGA monitor. Then operating systems built in all kinds of useful libraries like graphics libraries, interface building libraries, networking etc. All you had to write were the parts of the software that were unique to you.

So it is in science, where many of the technical software packages are pushing themselves as the equivelent of computational operating systems. You don't have to program numerical libraries, sorting, solving, optimizing, statistics, bignum or symbolic libraries. Just those parts of computation that are unique to your discipline.

You don't need a mass market or huge price tag, if you can write the code cheaply enough.

Tecplot to be split into two product lines

2006-03-07T17:14:00.000Z

Tecplot will shortly be reorganizing its product line up. The central product, Tecplot 10, will be split shortly to form two diverging offerings: "Tecplot 360" aimed at CFD post processing and visualization and "Tecplot Focus" aimed at "XY and 2D engineering plotting".

Upgrading customers will have to choose which branch to follow. Initially, it is claimed Tecplot Focus customers will notice little difference in the way of new features, but the company does not say whether the existing CFD visualization in Tecplot 10 will be removed from that product. If it has been, that is a tough upgrade to sell!

If it is true that their market is split between these two needs, then it is natural to target them seperately. If they are not, there are two possible interpretations:

1) Tecplot 360 will simply out develop Tecplot Focus which will become the "lite" version, at a lower price. A long established principle - see Photoshop vs Photoshop elements, Outlook vs Outlook Express, etc and in the Science arena Mathematica vs Mathematica CalcCenter or Mupad Pro vs Mupad.

2) By partitioning feature sets that do not have exclusive user groups, this is an attempt to force an existing user base to purchase two products where one used to do.

Details are expected by the end of March.

Mathsoft sales and finances news

2006-03-03T12:14:00.000Z

There is an interesting press release from Mathsoft today. On first inspection, it is a simple piece of cheerleading about how successful they have been but on closer inspection it is a little less clear.

First I wondered what the central premise meant: "[Mathsoft] has sold its 5,000th managed installation of Mathcad®, the world's most widely used engineering calculation tool. "Managed installation" refers to a team of users in industry, research or academia ranging from small workgroups to large engineering organizations with thousands of seats under management"

I searched their website for information on "Managed installation" and there is no reference to any product or scheme. It could mean "any quantity of licenses greater than one". Perhaps it refers to maintenance contracts, or volume schemes that need only 5 licenses. Lets suppose maintenance is 25% of a license fee (they don't list this), and volume discount means, say, 25%. Then a Managed installation, might be less than $1000 per year. That might mean this press release could represent as little as $5M per year. Not nearly enough to run a business on.

As I was musing on how this release was not as impressive as it first appeared, I noticed the most interesting fact. Towards the bottom was a reference to $3M in "follow on funding" from venture capitalists. So the funding that separated Mathsoft from Insightful five years ago is spent, and the company is still spending more than it is earning. This is normal for a start-up but for a mature company, this isn't healthy.

So how does a company close a deficit? Cut costs? No, Mathsoft is opening new offices not closing any. Sell more? Of course, if possible, though as we have seen from this release the number of customers isn't that large after 15+ years of trading. Make more out of existing customers? Here is where the excitement of having moved your revenues from 25% to 75% "managed licenses" becomes more clear. It looks like the strategy is based around growing the regular revenue from their regular customer base.

The only question left is- will this by getting them to use more of their products, or by charging them more for the ones that they have?

Security risks in web based scientific computing

2006-02-28T17:56:00.000Z

Pretty much every science software company has spun off a web version their products. But there is something odd about many of them - there is no sign of them on the makers websites. Sure there are pages of prose and pretty pictures telling you about them, but rarely any live demo. No chance to interact with an example deployment. It seems obvious that this would be the best way to showcase or trial such products.

And so it is with the Matlab Web Server that I was researching when this fact dawned on me. It's not that Mathworks doesn't consider its website to be an important part of its business and promotion. Then it occured to me that it might be because their website is important to them that they have no live demo.

The architecture of the system is: The web browser connects to a TCP/IP client "matweb", on the server that passes all requests to the matlabserver that connects to Matlab where the task is computed. So the matlab server is exposed to full scrutiny of the internet community, including the unfriendly ones.

There is a great power struggle going on out there between security companies and the hackers. With mighty companies like Microsoft and real experts like The Apache Foundation struggling to keep up with the latest tricks of the enemy. Would you trust the security of your web server made to a company who's expertise is not in security? Well it looks like neither would Mathworks!

Perhaps this is paranoid, but when you look at the documentation for the Matlab Web Server, there does not appear to be even a mention of security, let alone any serious instruction.

Since most scientific software are essentially programming languages, they are very dangerous to give unrestricted access to. You must block two levels of access:

First you must control the tasks that the user requests to get computed (Matlab, like most other languages, can be instructed to delete files, launch programs like telnet daemons, or copy and send files such as your password file).

Second you must make sure that you are opening a door only to the application that you intended.

One line of the documentation that interested me was:
"observe that the line
<input name="mlmfile" value="webmagic" type="hidden">
sets argument mlmfile to the value webmagic. The mlmfile argument contains the name of the MATLAB M-file to run."

So the choice of what program to run is visible in the page source! Sure enough, if you find some user's pages and do "Show source" you can see this "hidden" field in the HTML. This means that I can instruct that user's Matlab to run any .m file that I can predict to exist on his server, just by copying the HTML and editing that line and opening it in a browser. Now hands up who knows the matlab installation and most popular addons well enough to know which could be misused if malicious user could run them? Also, if I can place a file on the server by some other exploit, I now have a way to execute it. I'm sure a real hacker would have plenty of ideas.

If you are considering any other web based deployment system for a language, a good test might be to see if the supplier trusts it enough to run on their own websites.

Open source technical software - dead end?

2006-02-23T13:44:00.000Z

The open source movement has had a large impact on the software industry in recent years. I for one am a huge fan of Firefox and Thunderbird from the Mozilla project and a modestly satisfied user of OpenOffice. Can scientific software expect the same kind of open source benefits?

Firstly, it has to be said, on the smaller scale end of software, scientific software is dominated and always has been by open source code. When small specialist problems are solved in software, then there is often no commercial opportunity, so developers put code into the public domain rather than do nothing with it.

But the reality is very different with large software products. Complexity of software development rises disproportionately with the size of source code. This requires teams, which in turn require coordination and more people. It rapidly becomes expensive. And it becomes too involved to have developers spend just small amounts of their time on it.

When you look around at large scale open source projects they are nearly all Commercial software that has been released. e.g. Firefox was once Netscape. OpenOffice was once StarOffice.

On this basis one can probably assume that any future major scientific software open source project already exists. It is either one of the existing open source projects (Axiom, Maxima, Reduce, Scilab) or it is one of the commercial products if its owners fail to make it pay or fail as companies (Matlab, Mathematica, Maple, Mupad etc). Since these all these products are the central revenue of their companies, we can rule out the code being donated, like IBM released Eclipse. With the exception of Mupad, their companies look pretty healthy too.

So let us turn our attention to the three main large open source projects Axiom, Maxima and Scilab. Do any of them have the potential to become the killer free science app? Well no, and let me argue why:

1) The knowledge required to be a contriubutor is a level above that required to be a contributor to, say, Firefox. You need to be both a competant programmer AND a competant mathematician to be able to code complicated and debug algorithms. The world may be full of keen students who want the kudos of getting a few lines of code into Firefox, many fewer who are actually good enough programmers, and fewer still who might know the math to add to, say, Axiom. This means that these projects will never be able to keep up with the development rate of the commercial packages. They are behind already, and will only fall further back.

2) Large technical computing systems have a lot of internal dependancies. You may be able to take a task like "change the bookmark mechanism" in Firefox and be pretty confident that that team won't affect the way your browser renders pages. Ask someone to add features to, say, linear algebra, and you might affect equation solving, statistics, ODE solvers and many more areas that each affect other components. This again raises the technical requirement of your contributors, reducing the pool further AND adds a significant management overhead and system design overhead.

3) To support the costs, free software needs major financial contribution. OpenOffice gets the backing of Sun, who would like to sink Microsoft Office, Eclipse is backed by lots of big companies like IBM (who want it as a platform for their commercial tools). Only Scilab has any backers. The Axiom group recently commented that they had their first funded project. A $4500 grant from the Google "Summer of code" one time charity. And that 2 month project does not appear to have been delivered after 5 months!

4) The major selling point of free software is "it's free". But of course there are other costs to installing new software. Principally the training cost. It takes next to no time to learn how to use a word processor so OpenOffice is still almost free. But the learning time for scientific software can be major. So if it takes a month to learn then Axiom is now a $2000 product. If you add the same cost to, for example, Mathematica it is a $4000 product and arguably with the better support of the large user base, lots of materials, books training etc, that comes with a commercial system, perhaps it is only $3000. Free is still cheaper but the relative difference is much smaller than it is with simple consumer products.

5) Every successful open source project has established itself as THE alternative. Firefox is THE choice of browser after IE (who really has heard of Opera?) OpenOffice is THE office suite etc. This isn't true yet in the science software. Two must die to give the third any recognition, and if MuPAD were to join the group, the problem would get worse.

Maple handwriting recognition: useful or gimmick?

2006-02-21T18:18:00.000Z

One of the more original features of Maple 10 was its handwriting recognition tool. Quoting the PR:

" With over 1000 symbols, it could be a daunting task to find the symbol you want. But with the handwriting recognition built into Maple 10, it's easy!

Simply use the mouse to sketch the symbol in the Symbol Recognition palette, just like you would draw it on paper. The symbol recognizer will search through all the symbols and find the symbol you need."

The question is- "Is this a useful tool or just a gimmick". I set out to find out.

The first problem is a human failure. With only a mouse to do the handwriting, it is actually quite hard to draw the symbol you want unless you slow down and concentrate. Here are screenshots of my first attempts at drawing an Infinity symbol:

The list of characters at below the two buttons are Maple's suggestion for what I meant. With this writing, it should probably be forgiven for its suggestions of "M" and "&".

However, when you take more care, the suggestions don't get much better.

These seem like pretty good infinity symbols to me, but I never managed to get a match, even though the symbol is supported. "%" was the most common response and I can't see why it matched to "D" in this example.

The results varied depending on the symbol, arrows and comparisons frequently gave useful suggestions, but lots of the symbols were frustratingly hard to match or impossible. Following are some more screenshots of decent attempts to draw supported characters, with the failed matches:

The most surprising was Pi, since it was on the face of the button you press for a match and very likely to be needed. I never managed to get a match for Pi out of lots of attempts. Most often Maple suggested "H"

Where it did do well, with almost flawless matching was ASCII characters. Presumably because ASCII character recognition was already a solved problem. Of course, those are the characters where it is of no use in this context.

n reality, the need isn't really there anyway. 1000 characters may sound like a lot, but there are over 100 on your keyboard already, and once the remaining 900 have been broken down into 12 catagories, it is pretty easy to scan, say 70 arrow symbols, by eye, for the right one because, unlike Maple, your eyes are good at character recognition.

This feature was a lot of fun, seeing what the character lottery would throw up, but useful or gimmick? Strictly gimmick.

Mathematica released for Mac Intel

2006-02-17T10:30:00.000Z

It was not long ago that I wrote about the hints and statements coming out of the scientific software companies about support for the new Mac Intel platform.

http://scientificcomputing.blogspot.com/2006/01/mac-intel-race-is-on.html

Well the race has a winner, and it is Wolfram Research, who have now started shipping a native Mac Intel version of Mathematica. Wolfram even claims that this is the first professional application, not made by Apple, for Mac Intel.

http://www.wolfram.com/news/intelmac.html

Now we must wait to find out if it was a close run race or if the other horses are heading for the glue factory!

While Apple will be pleased that there is serious support for its new platform, this will mean little to Wolfram's sales in the short term. But it does tell us much about Wolfram's attitude to new technology and the efficiency of its development group.