It's strange that the HN crowd, consisting mostly of developers who work for companies that exist on average for 16 months, has the audacity of calling Wasabi, a tool that has been in use for close to 15 years, a failure.
Probably because it isn't some hipster badass ninja rockstar web-scale pipe-to-dev-null Wangular.js Ruby-on-Fails Web-2.0 SaaS app, and therefore automatically a failure.
What would you have done, given the starting point of a VBScript codebase, to bring it to Unix and later to .NET? Remember, it does no good to talk about what should have been, years before. You have to make the best decision with what you already have. In that light, writing a custom compiler for a variant of VBScript doesn't seem so bad.
If your point is that it was strategically necessary given the constraints at the time, that's fine with me. There are lots of terrible things that became strategically necessary due to the past decisions of business people.
I still wouldn't want to have to learn this language, would you?
I would. I might not like it, but I can't say for sure now, not having used it. And you never know, there could be some good ideas lurking deep inside.
I recently played around with Atari 2600 BASIC (which is quite possibly worse emulated than on a real system) but found that it had an incredible tracing mechanism I've never seen elsewhere. Also, you could control the speed of execution as the programs runs (imagine---running a program at full speed, then slowing down to watch exactly what happens). You won't know until you try.
The author of this post started at Fog Creek around a year after Wasabi was invented, if I'm getting my timeline right. He was also the one to finally kill Wasabi. Hardly sounds like jumping ship.
