I can sympathise with Kevin and find his experience similar to my own. Whether his recommendations make any sense ... I'm not so sure ...
I've been involved in programming and computers for way too long, but in an intermittent manner, with gaps of years doing something totally different followed by bouts of programming, each time in a different environment and with different languages. Having seen technologies come and go, I do not feel optimistic regarding the survivability of RoR.
During the '80s I was involved (in a very tiny way) with research into the theoretical underpinning of Ronald Reagans "Star Wars", the question being whether a given piece of code could be mathematically proven to be fail-save. The conclusion seemed to be a resounding "no". This was frustrating, perhaps one tended to view programming languages as being "mathematical languages" and felt that if properly written, code should follow some sort of mathematical laws.
The impression I get with RoR is that it is much closer to being an "organic language", where you can learn to speak it and become proficient but not without a significant effort. Being young is obviously a huge asset when learning languages and I suspect this to be the case here as well (as opposed to learning standard programming languages - I find Ruby easy to learn, much easier than I found learning Pascal was 30 years ago). RoR, as opposed to Ruby, is perhaps best seen as a step towards the development of truly intelligent programming languages, where the machine moves towards an understanding of human thought - not the other way round.
RoR is obviously a very temporary phenomenon and it will eventually be replaced by something totally different. It seems to have a very narrow application window which, combined with the effort of learning to "speak" RoR, sets it on the path to its own eventual demise. I'd estimate something in the region of 5 years, definitely not as much as 10.
So the question is really: Is documentation worth the effort? And the answer is probably: No. Attempting to do what Kevin suggests would probably kill RoR off much faster than it's "natural" lifespan would otherwise be.
Personally I am not willing to invest the time and effort needed to become proficient in something that, to me, seems a very temporary and fast-changing phenomenon. But I can see younger programmers benefiting hugely from their efforts in RoR, not least from being involved in development. RoR points to a future where programming will be quicker and easier, but also with a higher tolerance of individual instances of code failures. A more organic way of programming, one that moves closer to what our brains are built for - communicating, conceptualising - and away from what computers are good at - calculating and shuffling of minutiae.
Hoping to have complete documentation of future programming environments will be as futile as an American hoping to learn Japanese by reading a book on Japanese grammar.
Just my (not so humble) opinion.
Binni
-----Oprindelig meddelelse-----