Originally published December 31st:

http://on-being-open.blogspot.com/2007/12/apps-and-platforms.html

"Anonymous" pointed out on my post called "Repairo" that one reason people still use MSWin is that running things like AutoCAD under WINE under Linux are fraught with problems. At best, it requires work and experience at this point in time.

As much as I am a fan of Linux and OS.X these days, I want to state that there is nothing magical about either platform. All computer operating systems are amazingly complex bits of code written by human beings, and at any point in time it is possible that one platform is better than another, and even the term "Better" would require one to state what they mean by "Better". Better at memory utilization? Easier to cluster? Less prone to crashes? More virus proof? Sells more hardware?

That there are currently more MSWin applications is also not magical. To repeat what I said in the comment to the "Repairo" post: AutoCAD started as a Mac app, and when Autodesk saw people were willing to pay for copies of it on MSWin they ported it. If Autodesk thought enough people wanted it on BSD or Linux or the iPhone they would make a version that worked there as fast as their fingers could code.

Phrased that way, the question would seem to be, do enough people want any given app on Linux to bring it there. But that is not the way Linux works either. Here is where companies get into trouble not just with Linux but with all Open Source.

The real question about any Open Source platform always is: Are there enough people that want any given type of application that they are willing to take the time and effort to create it. In the example of AutoCAD: Are there any technical people out there that want a CAD package on Linux bad enough that they will write it themselves?

To look very quickly into that, I googled up two search terms: CAD + Linux

http://www.tech-edv.co.at/lunix/CADlinks.html

Over 50 CAD packages for Linux, some commercial, many Open Source. And that was just one of many hits.

The problem for a commercial company like Autodesk is knowing when the market has moved enough to a new platform to make it worth their while to port to a new OS platform. That costs a fair amount of money, and if the product is not written in a portable fashion, then it costs even more money to either port it or better, redesign it to be portable.

The problem is worse than it seems though, because in this case the Open Source world will look, maybe ask once, get a response like " We are waiting to see which way the market goes before we decide if we are going to move to the new platform", and then the Open Source folks just move on without them. By the time they decide to enter the market, the field is crowded and they are no longer the number one choice. In fact, there is now probably a free, Open Source solution sitting there and now they have to convince people that it is worth money to pay for their newly ported version.

I pay for the best. I think most people are willing to. Since I don't use CAD, I'll switch to Office packages for a second: Even though I do not have to, I send money to OpenOffice because I use their product and I like it. I like it better than MS Office because it uses Open Standard file formats, and runs on whatever platform I happen to be on.

Another example: I just paid for a copy of Scrivener recently. I am working on writing a few books, and found OpenOffice to be unable to do some things that I wanted it to do (Chapterization and organization). Some research found Scrivener, and I am extremely happy with it... other than I wish there was a Linux version. It's OS.X only. Guess what happens next? If someone comes up with a decent manuscripting program that is Open Source and cross platform, then I'll be retiring Scrivener...

All software applications are like that. Companies need to know and understand this new Open Source dynamic of platform, or find themselves playing catch-up.