On Thu, Mar 15, 2012 at 8:22 PM, Noah Slater <nsla...@tumbolia.org> wrote: > On Thu, Mar 15, 2012 at 6:33 PM, Benoit Chesneau <bchesn...@gmail.com>wrote: > >> On Thu, Mar 15, 2012 at 3:52 PM, Noah Slater <nsla...@tumbolia.org> wrote: >> >> > How long before we can land COUCHDB-1426? ;) >> >> Depends how much feedback and tests are given on that. Want to help? >> > > I time for CouchDB is saturated, but I can help you test for sure. > > Is there anyone on this list that can spare a few cycles to help Benoît > test? > > Benoît, how would you us to test? > > Is there a list of stuff you'd like people to try? > > >> Keeping pressure like this start to annoy me. >> > > Understood, and I am sorry that you feel annoyed.
Thanks. > > I know some people work well under pressure, and some people do not. If my > last email did not adequately express why I want to apply pressure, no > follow up will. But this is not how I expect us to operate under normal > circumstances, so there is no need to worry. But these are not normal > circumstances. > > As Jan keeps saying to me, we need some good news. > I completely share the need to have some good news but in my opinion it shouldn't take the lead over the release quality. I can understand we are under pressure (though not so much these days since people are starting to understand what make couchdb a good fit for them) and that's why I think we should take the time to make the installation working for the users and packages maintainer. Which is a good way, by the way, to tell to anyone that they don't need anymore a binary from a third party to enjoy the CouchDB experience. > >> Some bugs especially though based on multiple configurations not on >> the dev control could take time to resolve. And I completely share >> paul's opinion on that. We should take time when it's about to keep >> CouchDB robust and stable. >> > > Agreed. > > (Though I will note that core stability, and configure time stability are > two different beasts.) > > >> Anyway was off during last 16h due to flights and saw that @davisp >> posted a new version that should fix latest issues. Time to test. >> > > Again, anyone wanna help out here? > > Most of us only have on specific environment we can test on, but the more > the merrier I expect. > > Benoît, could you throw a list of checkpoints for people to run through and > report back with? The best way to test right now is cloning the COUCHDB-1426 branch and do the follwing tests : - with spidermonkey installed by your system packaging tool - with spidermonkey installed in a custom path using --with-js-lib & --with-js-include - mixing both, havin a spidermonkey installed globally on the system and try to use one installed in custom path. I hope I'm clear, if you have any question bug me on irc or better by mail. - benoit
