Languagesdontfailpeoplefail

Google is still insisting titles in this blog need to be in devanagri script. Since I didn't change any settings, they must be having a true Sanskrit-moment at headquarters. Weird. As long I add no spaces or other delimiters, the titles remain Engfish, but one wonders.

Eliotte now insists VRML isn't dead. It failed. Languages don't fail. People, companies, investments, initiatives all of these fail. Languages work or don't work. But if the metric is adoption, then it is just a 'size matters' metric. Here are some notable failures by that metric:

1. The Macintosh. All profits to the contrary, it is very much a minority platform. It is the classic loser of the desktop wars.

2. Netscape/Firefox/Mozilla. Huge embarassing failures. Despite the technical excellence of Firefox, it still can't chart better one in five, but like the dead newt, it's getting better.

3. SVG. It couldn't get a hearing in Congress with Jimmy Hoffa's body in its trunk.

4. XML. Where is that XML browser that was going to liberate us all from the tortures of programming in the DOM with innerHTML?

5. Ruby, Python, PHP, and a host of other open source minority languages. Javascript, on the other hand, claimed a place front and center and never gave it up.

It's an easy game to play but a game it is. Let's try looking at the more interesting points.

1. There has never been a time since VRML 1.0 that a viable VRML viewer wasn't readily available. There are more available now than when VRML was 'winning'.

2. The VRML is Dead or Failed meme originates in the States. Americans will declare 'mission accomplished' without ever checking to see if they really have and they will declare failure and loss as soon as the first trendy pundit tells tells them they are losing two points of market or mindshare. Webheads are lemmings.

In fact, the uptake and application of VRML and now X3D in European universities and markets have been wildly successful. That is why the Intergraph Corporation had to go to France to buy the French Homeland Security applications when Ingr had VRML experts on their own staff. They declared it dead, refused to support internal efforts and got caught with their pants down. Ooopsie.

3. The idea that all successful computer science originates in Silly Valley and that if the Americans don't use a language, it is a failure is the height of American zenophobia and arrogance not to mention a bit of ignorance.

4. Just a bit of Googling (if you can get it to quit transcoding) will show you some wildly successful applications of VRML. The same bit of Googling can show you some wildly impoverished American companies and professors. IBM among others are spending a very considerable amount of investor dollars to convince the world that we need open standards for 3D on the web when we already have them. The indirection is there to hide their own ambitions to obtain encumbered IP and licensing advantages through Second Life. I understand that. That's business. It's sleazy but it is altogether American business practice these days.

Languages work or don't work. People fail.

The latter case is possibly more disturbing because it means American professors in otherwise respectable universities are educating a generation of students in the politics of failure. Maybe that is not the right thing to do. Maybe a computer scientist should look as the applicability of a technology, become conversant in the means and uses of scene-graphs and leave the political analysis of failure to the marketing majors where learning to lie early and often is required if it isn't a natural talent.