I recently posted the following reply to an article about how management failed when they made incompetent choices about an implementation technology simply because they wanted to use the popular tools instead of the right ones.
Back in the year 2000...
I worked for LingoMotors, a fantastic startup some years ago which had built an amazing AI system that would parse the English language. The whole thing was built using VisualWorks Smalltalk. We had more than a dozen doctorate level computational linguists, several of them famous in their community. The system was amazing, and it leveraged the strengths of Smalltalk in a great way. This was a way to tag information with real metadata about the meaning of the data, and our first customer was a large online bookseller.
When we needed another round of financing we kept on getting pressure from potential investors who agreed we had something great, but they wanted us to port the whole thing to Java before they would back us. They had no clue that doing this would have killed our competitive edge, and it was hard to secure more funding.
The choice of implementation technology had become an issue of political correctness, in a sense. A lot of really wonderful programming language products died when the Java juggernaut trampled the world with overhyped promises. Now it seems that Java is over the top of its S curve, and things are more relaxed. It could happen again.
2 comments:
Maybe Python will be the next language everyone needs to use. It reminds me just a little of this classic Dilbert.
So what happened? Did the project succeed or did it fail because of lack of funding or because of implementation problems due to not using the best technology?
Post a Comment