If we manage to from now on not to make any backwards incompatibility changes (there's still something which should be deprecated/delete imho) then we're fine with semantic versioning, but as soon as we introduce some backwards incompatibility then we should increase the version to 2.0, which we wouldn't want if having watir-webdriver in mind :) But that is all about future.
Right now there are some bugfixes which have been show-stoppers to some of the users and that's why i recommended about new gem version release. Of course i'm gonna ping anyone involved and ask their opinion about new release, no worries. Okay, i'll check out soon the made things and see if changelog is in sync and then we'll decide when and how to release exactly. Any arguments against the release? I'm pretty sure that this release won't include any backwards incompatible changes, but i'll make sure. Jarmo On Sun, Dec 19, 2010 at 12:44 AM, Charley Baker <charley.ba...@gmail.com>wrote: > Hi all, > > I'd also suggest semantic versioning and have in the past. We > haven't discussed it per se, and should soon, but I'm thinking > watir-webdriver is Watir 2.0. The 1.x.x releases will all be ongoing > maintenance and potential compatibility releases to make sure that > migrating is not too challenging and current large scale Watir > codebases are supported. Watir spec is the best avenue to support some > sense of cohesion between the various versions right now. > > You've mentioned RSpec, which has taken a similar route as far as > their older 1.3.x releases and the newer 2.x releases. That's a great > example and something we should all take a closer look at for a host > of reasons. Rapid release schedule is only one piece of that, but > something I was pushing and would like to have. > > Rapid schedule doesn't necessarily obviate the need for a roadmap or > some sense of where we're going, it just means that we pick things up > when we can and release when it's time. And I think what we have > currently and the release schedule I was pushing towards means we > should release in the next couple of days, based on what we have and a > few "go" votes for releasing a new version. > > I've been busier than I thought I'd be, working on side projects and > spending time with my family, along with the holiday season always > being a bit chaotic. :) I should be getting back into more development > on Watir now as I find a few hours here and there. Jarmo, I've added > you to the gem owners on gemcutter. Warning: my kids are out of school > for the next two weeks, so it's still going to be a bit hectic. :) > > Any new release does need a quick poll on this list, RCs on the main > list for feedback, and then a push. Having walked through this process > a couple of times with you, I think we're in synch. > > As mentioned, I'll try to be more available on IRC as well as coding > and integrating fixes, hopefully for both Watir 1.x and 2.x. :) > > I'm excited to meet up in person for whoever will be at the Selenium > Conference. There's still a lot to work out in the meantime, but that > should also be a great place to talk shop and figure out some issues. > > Hopefully that clarifies the current state of the union in my opinion. > > > Charley Baker > Lead Developer, Watir, http://watir.com > > On Fri, Dec 17, 2010 at 4:24 PM, Jarmo <jarm...@gmail.com> wrote: > > About the versioning. I'd suggest to use semantic versioning > > (http://semver.org) from now on. It just makes sense and would be nice > to > > adhere to some rules and let the world know that we are adhering to them. > > Also, wouldn't that be a nightmare if we'd support and develop multiple > > versions of Watir (e.g. fixing only bugs in 1.6)? It would make sense if, > > let's say Jari's Watir-WebDriver would be the next *official* version of > > Watir. In that case there would be 2 completely different codebases which > > should be developed separately, e.g. fix only bugs in the *old* one and > add > > new functionality to the new one. Why would you want to make 1.6 as so > > special? > > Makes sense? > > Why not be more agile and release when it seems that we've fixed some > > critical bugs and/or added valuable new features? I don't see how > "release > > plans", which should be filled, help us to get into rapid release cycle. > > Check out RSpec for example, where releases happen almost every week or > even > > few days in a row. Why not take that route also if it's possible? > > Of course plans for future development would make sense, e.g. what should > be > > done, what should not be done and so on, but specific plans about "does > this > > belong to the next release or not" doesn't make much of a sense in my > > opinion. > > Jarmo > > > > On Sat, Dec 18, 2010 at 1:05 AM, Bret Pettichord <b...@pettichord.com> > > wrote: > >> > >> I would like us to put the 1.6.x branch into maintenance,. which means > we > >> would only fix bugs in it. > >> > >> The new work would be in a 1.7.x branch. > >> > >> I've been in touch with Charley just a bit recently. Personally I was > >> thinking we should have a release manager for each release to help > >> coordinate what was and wasn't in scope for the release (and would > release > >> the gems when the release was complete.) It would probably be two > different > >> people. > >> > >> I've been collecting notes for a while on what I thought would be in > each, > >> and was planning on putting these on the wiki so we could discuss. I > hope to > >> get this done next week; if not then, then next week. I am off from work > >> until Jan 3, so have some free time to work on this. > >> > >> If you want to get the ball rolling, you could start by making a wiki > page > >> with a release contents proposal. > >> > >> Thanks for taking the initiative to bring this up and offer to be the > gem > >> owner. > >> > >> I also have several Jira tickets than I've been meaning to follow up on. > >> Hopefully will find the time over the holiday break. > >> > >> Bret > >> > >> On Fri, Dec 17, 2010 at 12:57 PM, Jarmo <jarm...@gmail.com> wrote: > >>> > >>> Hi! > >>> I've been trying to contact Charley on the IRC now for about a week. I > >>> wanted to ask his thoughts about releasing new version of Watir, which > >>> includes some quite critical bugfixes and wanted to ask how did his > venture > >>> with removing activesupport dependenfy from firewatir go? But Charley > hasn't > >>> been available on IRC for some reason. I'm not saying that he has to > be, but > >>> this is life and these things are expected. > >>> That's when i remembered that here was some time ago sent e-mails about > >>> who should the gem owners be and such. I'd like to get owner rights so > i > >>> could make the gem pushes myself if it's okay with everyone else, of > course. > >>> So, what do you think about releasing new version of Watir and granting > >>> me access to make gem pushes if the time is right? > >>> Jarmo > >>> _______________________________________________ > >>> Wtr-development mailing list > >>> Wtr-development@rubyforge.org > >>> http://rubyforge.org/mailman/listinfo/wtr-development > >> > >> > >> > >> -- > >> Bret Pettichord > >> Lead Developer, Watir, www.watir.com > >> > >> Blog, www.io.com/~wazmo/blog > >> Twitter, www.twitter.com/bpettichord > >> > >> > >> _______________________________________________ > >> Wtr-development mailing list > >> Wtr-development@rubyforge.org > >> http://rubyforge.org/mailman/listinfo/wtr-development > > > > > > _______________________________________________ > > Wtr-development mailing list > > Wtr-development@rubyforge.org > > http://rubyforge.org/mailman/listinfo/wtr-development > > > _______________________________________________ > Wtr-development mailing list > Wtr-development@rubyforge.org > http://rubyforge.org/mailman/listinfo/wtr-development >
_______________________________________________ Wtr-development mailing list Wtr-development@rubyforge.org http://rubyforge.org/mailman/listinfo/wtr-development