Oh, I'm sure it's difficult. None of that changes the fact that we're all perfectly capable of judging new language features with:
- a readable description that is NOT up to JLS standard. - a thorough pros and cons list - plenty of 'real life' examples, preferably not constructed just to highlight the feature, but created by taking existing code and refactoring it to the hypothetical new feature - (optional, at least at first) some analysis on how to feasibly extend the grammar to accommodate the feature without changing javac's parser architecture compared to: - a JLS patch - a prototype I'd argue the first list is actually quite a bit more useful. And far less effort to create. I'm also not hearing you say that sun engineers themselves write up JLS chapter and verse BEFORE doing any of the other, dare I say, far more interesting work (even if it is hard work). There's also got to be a certain amount of stewardship on sun's part - nobody outside of the core contributors are really capable of making the kinds of decisions you're talking about (withholding one bug fix to avoid breaking backwards compatibility). On Sep 15, 4:25 pm, Joshua Marinacci <jos...@marinacci.org> wrote: > > Where coin went a bit wrong, I think, is in how you required more work > > from the community than what you require internally. I presume when > > (outside of coin) sun employees decide on java features to add to the > > language, they first do some analysis of which ones are worth it, pick > > one, nail down with decent certainty that this feature WILL be in the > > next release, and only then do the work of writing up specs and > > prototypes. > > Oh, if only it were that easy! I once spent an entire *week* on a 3 > line *bug fix* for the Windows Look and Feel only to have to *remove > it* later due to backwards compatibility breakage. The work required > on language changes is far far greater, and the cost of compatibility > far more challenging. > > It's just not easy doing software engineering on something that must > be used, compatibly, on nearly a billion computers. --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "The Java Posse" group. To post to this group, send email to javaposse@googlegroups.com To unsubscribe from this group, send email to javaposse+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/javaposse?hl=en -~----------~----~----~----~------~----~------~--~---
