When Architects Go Nuts

Here is a gem I came across today.

I was more amused by a comment on this image.

"Inscrutably brilliant excercise in seizmic design!! Any earthquake will think it's already been there so it'll leave the building alone."

Fireworks Display On WikiScanner

Franco Folini has this excellent post which gives us a good idea of the marketing strategies of a few "respectable" companies. A 24 year old hacker called Virgil Griffith has created WikiScanner with a very interesting goad in mind. In his words, "to create a fireworks display of public relations disasters in which everyone brings their own fireworks, and enjoys."

Not surprisingly, this guy has been sued before and obviously Wikipedia itself has an article about him.

3D Mojo Misses The Point

In a post titled "CAD file conversion misses the bus", Alex Neihaus says, "Here at Seemage, we marvel at the amount of time and energy people spend thinking about file format conversions. In a nutshell, we believe this is a solved problem...that people who want or need to convert file formats can and do, easily and accurately."

Alex seems to be living in a world wherein there is only one CAD software and that software can do everything: 2d drafting, 3d modeling, rendering, animation, FEA analysis, CAM, mould design, reverse engineering... I think you get the picture. Everyone in his world use that software only and they live happily ever after because their CAD file conversion problem has been solved.

However, in the world that we live in there are many CAD programs each designed to do a specific task. They come with proprietary file formats designed to store their application specific data. When users use neutral file formats to exchange data between these programs a lot of data is "lost in translation" which can include even geometric data. This is the CAD file conversion problem and in our world it is far from solved.

In the response to Scott Shepard, Alex says, "the manic focus on converting things obscures the need to do something with the info". I could not agree more. I have a small question though. How the hell are you going to do something worthwhile with the info if you do not have the correct info to begin with?

Free Magazines Anyone?

I read CAD magazines for two main reasons: to get news and to read product reviews. I am sure there are other uses for magazines but these two are what I use them for. I do not pay for any of the magazines I receive, even though I am in India. The advertisers bear that cross.

In recent years, magazines seem to be losing their punch. With news feeds I get the news as it happens, not when my postman decides to give it to me. With blogs I get product reviews and comments which seem to be more authentic than the ones we read in magazines, to put it mildly. I have much better things to do with my time than read a news item or a product review in a magazine, when people have already read, analyzed, dissected and commented about it much earlier in the online world.

Also, I find it interesting how the CAD media questions the numbers that the CAD vendors occasionally release showing their install base. Publications flaunt their readership numbers to attract advertisers. If the readers were made to pay for the magazines (instead of the advertisers) we would get a true number for readership, that is, the number of people who are actually willing to spend their time and money to read the magazines and hence may end up reading the advertisements in them.

Just because someone downloads a trial or is given a free educational version of a software, it doesn’t mean that he is going to use it. Similarly, just because someone is shipped a free magazine, it doesn’t mean he is going to read it.

I recently came across a term called "shelfware". I guess you could use the same logic for free magazines as well.

Just my $0.02

Solid Modeling For $195

If all goes according to plan, by mid-October, Michael Gibson will start selling Moment Of Inspiration (MoI) for $195 a license. He has finished writing the software and is busy wrapping up the documentation.

MoI is a 3D NURBS solid modeling program with an unbelievably small download file size of 4.7 MB only. It is powered by the SOLIDS++ modeling kernel by IntegrityWare, which has companies like Bentley, McNeel and Alias (now Autodesk) on it's customer list. So I guess MoI should be able to do most of the things other solid modelers do. The snapshot below should give you an idea.

I am curious to see how Michael markets MoI. His web site title reads "MoI, 3D Modeling for Designers and Artists", which, in my opinion, is not the best way forward. He started developing MoI as a CAD software for the Tablet PC. But now he admits, "There are actually currently a lot more non-tablet users of MoI than tablet users". I guess Michael can take a page out of SpaceClaim's marketing strategy. In my opinion MoI would be a perfect candidate to target the "extended development team" which includes people involved in conceptual design, engineering and manufacturing. Probably not in it's current form, but it's got what it takes, the main thing going for it being its price. You just cannot beat $195. I can see top management willing to dish out $195 a license to get the extended development team in the loop. After all, the idea is to get the extended development team to contribute to the design, not actually design. Therefore spending thousands of dollars on the extended development team doesn't make much financial sense, especially of MoI can do what is required of it.

In my opinion, sooner than later, Michael will have to give up the Tablet PC friendly interface for a more normal interface. I would recommend going ahead with the Microsoft Office ribbon style interface, especially since SolidWorks and SpaceClaim have already joined the bandwagon and more are expected to follow. Normal PC users will be more willing to accept MoI if it has an interface they they are familiar with, instead of making them use their computer as they would use a Tablet PC. For example, right clicking on an object in MoI does not pop up a context menu. Instead, it repeats the last command. Being a programmer, I fully understand that Michael has taken pains to go down this Tablet PC road and back tracking is not going to be easy.

Currently MoI can read and write IGES files, which should be enough for it to fit into any company's workflow which contains other more capable and more expensive solid modeling software. STEP and SAT support would certainly be welcome since IGES comes with its own set of problems. CAD software vendors have abused the IGES standard by making their own "flavors" of the IGES file format to suit their particular needs, thereby creating many variants of the IGES file format. This can be seen from the fact that Rhinoceros can write 60 types of IGES files.

This first version of MoI will not have support for plug-ins (which is bad news for me) but Michael tells me that he would provide an SDK for third party developers in later versions. That is good news but brings up an important question. With MoI being priced at $195, how much money will I be able to make from a MoI plug-in? I am already having a problem with the pricing of my IntelliCAD plug-ins.

In my opinion, the key to MoI's success or failure will not lie in it's abilities but rather in the way it is marketed.

CADCAMNET.tv

CCNtv has its own domain name. Starting with the 27th August broadcast CCNtv will be available at http://www.cadcamnet.tv instead of http://www.engineeringautomation.net/CCNTV.html, where it is currently housed.

It's good to know that somebody is reading this blog.

Autodesk's View On PDF

Many blogs are reporting the NOVEDGE interview with Mary Hope McQuiston, Head of Autodesk Extended Design Group. When asked about the difference between PDF and DWF she quoted a customer by saying, "With a PDF or GIF it’s like you’re getting a brochure of the design. DWF, on the other hand, is the design.”

Nothing is further from the truth and the Head of Autodesk Extended Design Group ought to know, unless she intends to blame the ignorant customer for her statement. A PDF file can contain a precise NURBS model of the design, and that too at a fraction of the original file size. In fact this is one of the hallmarks of PDF. And putting PDF in the league of GIF is plain and simple dumb, if not malicious.

I could have understood if this was a statement made by an external PR firm who sometimes have no clue of what they are talking about. Edelman recently summed up the history of AutoCAD into one sentence and in the bargain rewrote it completely.

I wonder what Adobe has to say about this.

Real Virtuality

I came across this image today.

This is not a computer graphics image. It is real. See http://homepage.mac.com/helipilot/PhotoAlbum20.html

EULA Paranoia

R. Paul Waddington, the person who frequently comments on many CAD blogs, including this one, finally has a blog of his own. It is called "Caveat emptor" and, not suprisingly, his first post is an open letter to Autodesk, asking them to allay his fears on the audit clause in their End User License Agreement. He believes that the audit clause gives Autodesk complete access to his computer and premises and will compromise his IP and that of others. He suggests that the printed output from Autodesk Product Manager is more than required to determine a licensee's compliance to the terms of the license agreement.

This reminds me of a clause in my car loan agreement, which states that I am bound to bring my car for "inspection" to the office of the finance company whenever the company so demands, whenever that may be and wherever my car may be, failing which there would be a litany of legal consequences. I guess a paranoid person would be justified in calling this clause an excuse to walk over his privacy, disrupt his life, cause mental torture and a dozen other things. Probably a person who is paranoid to a higher degree may be justified in screaming that his insurance company may not pay for any accident while his car is in the custody of the finance company. Unfortunately there is no limit to paranoia, and I am beginning to wonder if there even is a cure.

The "inspection" clause I mentioned above is there for a reason - to make the recovery of the car easy in the event that I default on my payments. The law in India forbids the use of force to recover a car (although force is often used, but that's a different story). Thanks to this "inspection" clause I am legally bound to get my car for "inspection" at which point the finance company will promptly seize the car if I have defaulted. While a paranoid person will read all sorts of nightmare scenerios in such clauses, the normal people will use a little trust and some common sense while reading and accepting such clauses.

Similarly the audit clause in Autodesk's EULA, and that of SYCODE's as well, is there for a reason - to make it easier, or even possible, to catch the bad guys. How are you going to prove in a court of law that someone is reverse engineering, decompiling or disassembling your software? How are you going to prove that someone is using a crack of your software? Can you accomplish this using a pretty tool like the Autodesk Product Manager? I don't think so. You need hard evidence and you need to get it the hard way.

I mentioned in an article earlier ("Intrusive License Agreements"), that the FBI could not nail the guy who tried to sell SolidWorks source code, even after a successfull sting operation. Such is the state of the legal system and pretty license agreements will not help the situation.

The audit clause should bother only those who intend to do something wrong, and of course, the paranoid as well.

My Affair With AutoCAD

I first met AutoCAD in the CAD/CAM lab of Goa Engineering College way back in 1993. She was in her 10th Release and looked gorgeous on the 486 computer on which she lay. She was a breath of fresh air compared to the drawing board of the drafting class. Since she was the only one of her kind in the lab, she was kept quite busy with the guys falling all over her. I remember asking for the lab keys at night since I could not get enough of her in the day. I really enjoyed the night sessions. There was nobody to disturb us, apart from a lone bandicoot which had made the lab it's permanent residence.

She looked simple on the outside, but was capable of doing a lot of things... really nice things. I realized how complex she was when working on my final year project. The project was an application of optimization techniques to a few mechanical design problems. I wrote all the programs in C but wasn't too happy with the output. Although accurate, the output was basically a bunch on numbers showing the details of the optimum design. I decided to draw the designs on the screen and then animate them to show how the optimum design was arrived at. And that's how I entered the world of graphics programming. I tried to ape a few things that AutoCAD did, but failed miserably. It was then that I realized that graphics programming was not as simple as I thought it was. All thoughout my project I kept wondering about AutoCAD and what she must be like underneath.

I got a chance to work closer with her at my first (and only) job at ACGL. By that time she was in her 12th Release and more mature. And so was I. At ACGL we built bus bodies. I joined a team of programmers (actually, a couple of them) who were entrusted with the job of customizing AutoCAD so that she fitted into the company's workflow. I helped in developing an in-house system wherein the assembly drawings of a bus body were automatically generated. The designers didn't have to draw anything. They first ran a FoxPro program that took litany of input parameters regarding the bus body design, such wheelbase, overhang, number/type of seats, etc. and stored these parameters in FoxPro DBF files. Therafter they would fire up AutoCAD and make her to load a master AutoLISP program which would call other AutoLISP programs depending on which assembly drawing the designer wanted to generate. These individual programs would then automatically create the assembly drawings by reading the parameters in the DBF files. By assembly I mean a giant structure of steel members and sheet metal welded together to form the side, roof, floor, etc. of a bus body. It was truly amazing that she could create such complex drawings without the designer having to move the mouse an inch. She was actually doing parametric design.


And it didn't end there. The AutoLISP programs added extended entity data to the objects in the drawing. Extended entity data is basically extra information, not related to geometry, such as material, finish, etc. We wrote AutoLISP programs to read this data and automatically create the BOM (Bill of Materials). The BOM was then imported into another system in Oracle which would create the required paperwork (work orders, indents, purchase orders, etc.) to get the actual parts fabricated or procured for assembly in the shop floor.

The drawings, data and all related information was then recorded back into a central database so that they could be reused for another bus body depending on which parameters matched. I believe people are now calling this process "Product Lifecycle Management" or something like that and think it is something great. My AutoCAD did this years ago. It was like an opera and she was playing the best part.

By the time I left ACGL to start my business in 1999 my love for AutoCAD had only multiplied. I never liked coding in AutoLISP. Debugging was a nightmare. I am more of a C++ guy. After leaving ACGL, I was free to try my hand at the ObjectARX SDK, the C++ tookit used to build plug-ins for AutoCAD. By that time she was called AutoCAD 2000 and what a piece of work she was. I soon realized that ObjectARX could make her to move in unimaginable ways. As you can imagine, I had a ball of a time.

They say wine and women get better with age. And so did she. She is going to be 25 this year. Over these past years I have been closely watching her every move. And she gets more of them every year. At SYCODE we have made her do things for customers that have left them spellbound. And customers keep coming back for more.

People calling her a "dumb 2D system" don't know her the way I do. If we could make her do the things she did in ACGL on a DOS platform way back in 1999, you can only imagine what she is capable of doing now on Windows.

My affair with AutoCAD is far from over. In fact, we have just got started.

AutoCAD 25

Autodesk is celebrating 25 Years of Design Innovation and has an interesting timeline at www.autodesk.com/autocad25. According to the timeline, "1982: John Walker and 16 employees start Autodesk with the goal of creating a CAD application for under $1,000."

25 years on, I guess they are still trying to do the same thing. It's a good thing that $1,000 is not as much as it used to be back then.

New Feed Address

The feed address of this blog has been changed to http://feeds.feedburner.com/deelip

Please update the feed address for this blog in your feed reader.

Getting Out Of AutoCAD's Shadow

We recently received reports from users claiming that our IntelliCAD plug-ins were not loading in their versions of IntelliCAD. We had tested our plug-ins with a few IntelliCAD variants, not all of them. So this time around, I decided to get a hold of all existing versions and variants of IntelliCAD out there and test our plug-ins with them. This began the process of visiting the web sites of all the IntelliCAD Technology Consortium (ITC) members and downloading trial versions of their products.

While doing so I could not help but observe how almost all the ITC members were raving about how IntelliCAD "almost is" AutoCAD but is priced much leser. Some were talking more about AutoCAD than their own product. I was literally fed up of reading the word "alternative" over and over again. There were just a couple of members who seemed to get it right. They mentioned DWG compatibility as one of the many features and not the main reason to buy their software. I really liked the way they portrayed their software as a fully functional and extendible CAD system ready to do some serious work for the customer.

Let's face it. Low cost CAD software does not get noticed anymore. Almost everyone is offering a low cost (or even free) version of their software. What gets noticed is how a CAD software can make a difference in an end user's workflow. Most of the ITC members still seem to be stuck in the AutoCAD alternative rut.

I am eagerly waiting for "The New IntelliCAD" which is supposed to be released early 2008. I hope the "20 man years" invested in the "complete redesign" has not been wasted in merely aping the features of AutoCAD. In a press release on 27th February 2007, ITC President Arnold van der Weide said, "The mission of the ITC is to bring high-quality CAD technology to the industry." By high-quality CAD technology I hope he means new technology and not what Autodesk has already come up with in the year before.

Autodesk itself is trying hard to add new features to AutoCAD every year. And all this while the ITC has been spending its resources in reinventing the AutoCAD wheel. If they had spent it on innovation, they probably would have something different, which may or may not have been better than AutoCAD.

It looks as if Autodesk is beating a dead horse and the ITC is beating the people who are beating a dead horse.

Autodesk has often used FUD (Fear Uncertainity Doubt) to discredit IntelliCAD, the ITC and the OpenDesign Alliance. And why wouldn't they? How would you treat a person who keeps following you around and does everything you do? I guess, you would first try and shake him off. Then pretty soon you are going to start telling people that he is not mentally stable.

I believe that IntelliCAD has quite a few things going for itself. However, it badly needs to get out of AutoCAD's shadow. If you continue to portray yourself as someone's whore, nobody is going to treat you like a lady.