From: "M. Edward (Ed) Borasky" Date: 2009-02-12T01:12:38+09:00 Subject: [ruby-core:21992] Re: New hash : syntax for the 1.8 series? Yukihiro Matsumoto wrote: > Hi, > > In message "Re: [ruby-core:21977] Re: New hash : syntax for the 1.8 series?" > on Wed, 11 Feb 2009 10:33:56 +0900, Evan Phoenix writes: > > |This causes massive confusion and doubt about ruby. When a user > |upgrades to 1.8.7 and a significant amount of code begins to fail, > |they worry. When the upgrade to 1.8.8 and discover they've been using > |new syntax that their coworkers don't have, they get worried. > > As for 1.8.7, it was unfortunate to break Rails at the beginning, but > "significant amount" is overrated, as far as I understand. The > problems were Rails design was bit weak for name conflict, and more > importantly, NO ONE ever had tried to check Rails on 1.8.7 RC during > the release candidate time. I had talked with DHH and people behind > REE after that, to ask for help when 1.8.8 will out. We will see less > problems, I hope. > > matz. > > Programming is as much, if not more, about communications within a team of programmers as it is about communication between programmers and computers. This lesson is drilled into us at early stages in our careers -- at least it was in mine. For example, FORTRAN (I) was first created in 1957, and by 1966 there was a committee-defined language standard, FORTRAN 66. Roughly every 11 years thereafter a new standard appeared, although the growth of C in the application domain that used to be FORTRAN's exclusive niche has slowed the standardization process. FORTRAN 1995 seems to be about it. Ruby is older now than FORTRAN was when it was first standardized, and yet we don't have a formalized structure of users and implementers in place to *define* the language. We have the tools and test suites, we have the community, and we have, I think, enough "industrial support" to make an ISO standard possible now. I think it's time we got together and did this. We have two "de facto" standards now -- Ruby 1.8.6 as defined by the second edition of the Pickaxe, and Ruby 1.9.1 as released and documented in "Programming Ruby 1.9" and "The Well-Grounded Rubyist" and "The Ruby Programming Language". I don't have a huge installed base of Ruby code, so I am moving to 1.9.1 to take advantage of the YARV engine. I do not see the point in testing 1.8.7 or 1.8.8 when 1.8.6 works in the MRI implementation, JRuby, Ruby Enterprise Edition and perhaps others. -- M. Edward (Ed) Borasky I've never met a happy clam. In fact, most of them were pretty steamed.