Hey, Thanks for your attempts to put together a coherent answer for what's happening based upon reading our public utterances.
Unfortunately, we don't always realize that out comments on the bug tracker, on the Rails core list, on Twitter, etc. cannot easily be put together to form a coherent understanding of the current status of things. That said, as you have shown, there is sufficient public information for an enterprising person to figure out what's happening. I say that purely because we don't always realize that our scattered communications can come across as silence, and this is something we should work on improving. Comments inline. On Jul 25, 6:23 pm, Oldroy <roy.tsc...@gmail.com> wrote: > My best guess and a heavy dose of speculation for rails 3 rc1 > 1. Bundler needs to go to rc1 - maybe in a few days with no bugs - > maybe in a couple weeks with lots of bugs. We're down to basically no bugs, but this analysis was basically correct. Keep an eye out for an announcement soon. > That may be all - but.....if that isn't the last issue here are next > couple of issues that seem important. > 2. ActiveRecord is about half as fast as it was in rails 2.3.5 - it > won't remain that way. There is a lot of stuff and junk that has to be > cleared out of it - it will probably be a couple of weeks before jose > valim and possibly wycats turn their attention to a major clean up of > active record to get old stuff out of the way. Aaron Patterson (of Nokogiri), José and to some degree I have been actively looking at the current state of performance in ActiveRecord. The important open ticket is at https://rails.lighthouseapp.com/projects/8994/tickets/5098-rails-3-beta-4-activerecord-5x-slower-than-rails-235 (which overstates the magnitude of the performance issue, but not the reality of it). I recently commented on the ticket (https:// rails.lighthouseapp.com/projects/8994/tickets/5098-rails-3-beta-4- activerecord-5x-slower-than-rails-235#ticket-5098-60). Aaron is spending some time with Arel, which seems to be the crux of the issue. > 3. Memory leaks......seems to me that you wouldn't wan't to deploy a > high traffic site with beta4 and ruby 1.9.2 rc..x... until the memory > issues are taken care of. That might be a ruby 1.9.2 issue...it might > be a rails issue, that is beyond me at the moment. Yes. The two important tickets are at https://rails.lighthouseapp.com/projects/8994/tickets/5042-memory-leak-with-ruby-192-rails-30-beta-4 and https://rails.lighthouseapp.com/projects/8994/tickets/4183-possible-memory-leak-in-rails-3. This is something we're actively looking into. This is likely a problem in Ruby 1.9, but it may already be resolved (http:// redmine.ruby-lang.org/issues/show/3466#note-3). We're working with Aman Gupta (memprof), and have so far been unable to reproduce this issue. We are actively trying though. > > If I were a betting man, and I'm not, I'd bet we will see rails 3 rc1 > within two weeks, and we may see it in the next 4 or 5 days (today > being July 25, 2010) > > This is a (somewhat) educated guess after following the "f*cking bug > tracker" (is there just a "bug tracker"?) and reading through what to > me would be the most important issues. (And if a bug is on the "bug > tracker" how does it get bumped up to the "f"cking bug tracer"??) > > My best guess at the silence about what is going on? I dunno.... Seems > that one blog post or twitter post confirming the above would be all > that it would take now and all that it would have taken a month ago > just to let people who are interested what is going on. Part of the > big push for rails 3 is to expand not only what experts can do with > the system, but what beginners and intermediates can do, and to expand > the total base of users. This has been the stated goal of the core > team over and over. Not communicating is not the way you expand the > total base of users. And while whining about "where is it?" doesn't > help, how exaclty does it hurt? To paraphrase New Jersey's > governor...."you must be some of the thinnest skinned people I've ever > known". Hehe. I think it's perfectly fine for people to be asking what we're up to. We may not always be able to give a satisfactory response (lately, it's been "as soon as we stop getting so many darn bugs"), but it's a perfectly valid question to ask as often as you'd like. > If you are on the core team, and you don't get why people are excited > and confused and a little bit irritated at the lack of > communication....let me restate that, I don't know how you could not > understand after most of you have given presentation after > presentation about what is coming. Yep. We get it loud and clear. We've spent a lot of time building something kick-ass, have talked a lot about it, and people want it. We understand that fully. Thanks to the efforts of the community, we've gotten a huge inflow of bugs, as well as feedback from plugin authors. This is why RSpec, DataMapper, Haml, etc. are working flawlessly on the Rails 3 betas. We've come a long way, and are closing in on the finish-line. > But this is an opportunity. You build a community by > communicating.......see that the root of both words is the same? You > build irritation by not communicating. Absolutely. Again, we've communicated in too scattered a fashion for people to put the pieces together reasonably. I'm going to try to resolve that in the weeks ahead. > And those of you who want the noobs to take a ticket and fix it and > send the patch......how long have you been working in rails...a couple > years?? As a member of the ruby and rails community you can't give > others a couple of years to catch up - and in some cases - surpass > your ruby and rails skill? Absolutely. While I think it's helpful for us to point people at the bug tracker if there's interested in contributing, saying "it'll be done when you start helping" is not actually that helpful. A lot of people on this list, and out there in the world, aren't necessarily capable of contributing, especially at this stage. The Bugmash project (http://wiki.railsbridge.org/projects/1/wiki/BugMash), which RailsBridge puts on periodically, is extremely helpful at getting new contributors past the initial hurdles. Their contribution guidelines are quite helpful as well. > And rails itself.....it's no accident that it took this long to re- > factor it. Rails 3 really is a lot of merb, isn't it? And only because > it was rebuilt with lots of new outside blood will it be as good as > it's going to be. All the more to look to the future of what will > rails 3 would have been without merb.....and what will rails 4 or 5 be > without @whineynoobgenius4 sticking around? Absolutely. Over the past year or so, we've added @tenderlove, @josevalim, and @spastorino and @xaviernoria to the group of people with commit access. Half a dozen or so more (Mikel Lindsaar, Neeraj Singh, Andrew White, Rohit Arondekar, and some others I'm forgetting atm) are actively participating in the day-to-day Rails 3 work. From my vantage point, we have far more activity and regular infusions of new blood today than we did when I started work on Rails 3 in January 2009. And I absolutely agree that continuing to support the participation of new, interested developers is critical to the progress of Rails in the years ahead. > So listen up.....@dhh, @wycats, @josevalim, @spastorino etc. You have > a giant opportunity to sooth the massive amount of people who have > become interested, well, largely by your own efforts to attract their > interest. Use that opportunity to communicate and set a foundation for > some of the outside blood that is coming in. For every noob that you > find irritating and will disappear in a couple of months because they > can't pay their dues, there are 5 others that are irritated and will > pay their dues and may be on the rails 4 core team, if you are will to > communicate. Just a blog post or twitter post a day from one of you > will more than suffice. I've noticed a couple of those things in the > last day or two and it really helps. My own personal position (can't speak for anyone else on Rails core) is that no noob is too irritating to spend time listening to. In fact, the perspective of new Rails developers is very important, since there are certainly irritations in the process that we can no longer see (as you have pointed out). I've said this before, and I'll say it again: if you are interested in participating in the development of Rails, feel free to email me directly. I spend a fair amount of my day-to-day work communicating with community members, and I don't mind helping people get up to speed. > And @jeremy/@bitsweet - good for you for speaking your mind, but the > FU stuff has a way of coming back to haunt you in things like client > negotiations. That post will be here for eons....and you have no idea > of all of the people who have seen it and taken note. If I had done > that it would keep me up at night. > > So yes....everybody remain calm. Our long community nightmare is > almost over. A few days - couple of weeks at the outside, and people > can finish their gems and plugins, and books, and ide's and projects, > and hosting platforms and.....well, you get the idea... Many of those who are working on gems, plugins, books, IDEs and hosting platforms have taken the drawn-out process of the Rails 3 beta period to provide extremely useful feedback about the current state of things. The feedback of David Chelimsky (rspec, on exposed testing APIs), Nathan Weisenbaum (haml, on templating APIs), several people at New Relic (on instrumentation), the folks at Engine Yard and Heroku (on bundler) and Sam Ruby (Agile Web Development on Rails, on general backwards compatibility) have been critical to ensuring that the final release of Rails 3 is up to the level of polish that you've come to expect from Rails. Thank you all. > > There I've said it. I'll go back to the shadows now.... Thank you for your lengthy and articulate response. You are correct: we need to communicate with the community in a single, public place. This will be resolved. -- You received this message because you are subscribed to the Google Groups "Ruby on Rails: Talk" group. To post to this group, send email to rubyonrails-t...@googlegroups.com. To unsubscribe from this group, send email to rubyonrails-talk+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/rubyonrails-talk?hl=en.
