Attack of the Clones

In an earlier article ("The Future of AutoCAD Clones") I laid out a scenerio whereby companies adopting 3D could turn to AutoCAD clones to fulfill their 2D needs. This brings about an interesting question. The clones have been around for many years now and companies get into the mood of cutting costs from the time they are formed. So why are the clones still languishing where they are?

I believe there are many reasons for this. The main reason is the FUD (Fear Uncertainity Doubt) campaign unleashed by Autodesk aginst the enemy, namely the IntelliCAD Technology Consortium (ITC), the OpenDesign Alliance (ODA) and both their members. And you know what? If you look at it as closely as I have, you will begin to realise that Autodesk is not entirely wrong.

The IntelliCAD 6 platform is rather "weak" as compared to that of AutoCAD, even several versions earlier. The IntelliCAD 6 platform is built using SDS (Solutions Development System), a clone of Autodesk's ADS (AutoCAD Development System). ADS became history many years ago when Autodesk came up with ObjectARX, a far advanced platform which took AutoCAD to a whole new level. The AutoCAD verticals are actually a bunch of ObjectARX plug-ins working inside AutoCAD. Since ObjectARX applications share the same memory space as AutoCAD, they work much faster and better than those using ADS technology. This makes AutoCAD far more powerful and scalable, capable of handling large amounts of data very efficiently.

On the other hand, IntelliCAD is still crawling on the SDS platform. Crashes are not uncommon. Large drawings take forever to load and refresh. We have all been led to believe (by Autodesk) that the cause for this undesirable behaviour can be traced to the reverse engineered DWG format from the ODA. DWG is only a file format which stores data. It has very little to do with the internal architecture of IntelliCAD. Sure there may be cases when corrputed files crash IntelliCAD, but that happens with AutoCAD as well.

I believe the problem lies in the recipe, not just a particular ingredient. SDS has been written in C, a language which reigned during the times of the 486 computer. I am not aware of any modern day CAD system whose platform is written in C. The nature and volume of data that modern day CAD systems are expected to handle, and the speed at which they are expected to do so, is just not manageable by an antique platform such as SDS. Something is bound to misfire and it often does.

Earlier in May, I had written about the New IntelliCAD. Since then I have had the chance to work with the ODA's DRX, a clone of AutoCAD's ObjectARX. Just like how ObjectARX revolutionalized AutoCAD, DRX is all set to do the same to IntelliCAD. Whether it actually does is a different question. Using DRX, developers will be able to create verticals similar to those of AutoCAD. This opens up a whole new world of possibilities and opportunities.

Graebert has gone one step further and is working on an ARX of their own which is source code compatible with AutoCAD's ObjectARX. This has profound implications. Although DRX is a clone of ObjectARX, an AutoCAD developer still needs to port his ObjectARX code and make adjustments (sometimes major) so that it can compile with DRX. At SYCODE, we write plug-ins for AutoCAD as well as the new IntelliCAD. We need to maintain two sets of source code for the same kind of plug-in. But with Graebert's ARX, an AutoCAD developer can use his existing ObjectARX code as it is, compile it with their ARX libraries and presto, an AutoCAD plug-in becomes an IntelliCAD plug-in. Why is this important, you may ask. Very often an AutoCAD user is unable to dump AutoCAD for IntelliCAD because a third party plug-in that he relies on is not available for IntelliCAD. In such an event, Graebert's ARX SDK becomes the magic potion that allows the developer to convert an AutoCAD plug-in into an IntelliCAD plug-in in a minute, literally. Wilfried Graebert tells me that they are not quite there yet, but they eventually will.

But there appear to be a few issues the clones need to sort out before they launch their attack. While developing SYCODE's range of IntelliCAD plug-ins, I was in touch with a few of them, on a technical front as well as a business front. From what I gather they do not appear to be working along the same direction and some have no idea what the others are up to. That's understandable because they are not only fighting Autodesk, but competing among themselves as well. However, to take on the might of Autodesk, a little more organization and unity could come in handy.

Good technology can make good products, but only good marketing can sell them. I expect Autodesk to kick its FUD campaign into higher gear after the new IntelliCAD is released. I wouldn't be surprised if a couple of law suits spring up. The clones need to come out with a united strategy to counter this.

And most of all, the clones badly need to shake off their image of being an inferior and cheap (the other meaning) product. And this will be extremely difficult to do in the first release, when the dam of bugs and goof ups always breaks open.

I believe the attack will start at the start of 2008.