The latest fashion in certain “elite” development circles is using (unfortunate name) software factories to automate parts of the development process.
It is questionable whether this will happen or not, but there is one other frequently mentioned idea that I do agree with – sometimes it is easier to solve a particular problem by designing domain specific language interpreter/compiler first, and then solving the problem in that language than developing a solution with a general purpose language (C++, C#, VB.NET).
Again, some will be intimidated by the “domain specific language” mambo-jumbo and call bullshit. To those I present one not-so-well-known yet concrete example of such an approach that actually works – crm114.
Crm114 (crazy name, but check out the site, the name has a funny history) is a variant of Bayesian text classifier (it’s just one of the back-ends, there are others) and is used to filter out spam email. The author, William Yerazunis, has first designed a language called crm114 and then proceeded implementing spam filter in that language. The results are pretty impressive, but the system is not user friendly and that’s why it is not in widespread use.
I am guessing that we don’t really have many examples of this kind of approach because writing compilers/interpreters is hard, even when you know how it is done.
Be the first to rate this post
- Currently 0/5 Stars.
- 1
- 2
- 3
- 4
- 5