> Just like in Joel's post, where Netscape took years to rewrite its already working code, rewriting code because Coldfusion was “dead” or “inferior” would have

> Just like in Joel's post, where Netscape took years to rewrite its already working code, rewriting code because Coldfusion was “dead” or “inferior” would have served little business purpose. I understand there are plenty of counter-arguments to be made here. First and foremost, it gets harder every day to find a Coldfusion developer. Those who know the platform are most likely looking to give a facelift to their skillset, rather than prolonging the inevitable drying-up of the proverbial job pool. More, it may be harder to implement functionality in Coldfusion, particularly the kind we were dealing with, than, say, in .Net, Ruby on Rails, or even Go. Even so, there are other, better ways to move toward new technology without rewriting an entire codebase. Interesting that the author doesn't provide any real suggestions to this specific case, only the other examples, which are not similar. The facts are the following: 1. The Coldfusion community is dying. 2. Adobe is actively making it more difficult to commercially pay for the software. They are now seeking compensation for use of CF as part of "SaaS Revenue". You have to submit a form that says what your revenues are. I kid you not, thats what the sales people are saying to customers. These two situations have serious economical ramifications on a company running CF.