Hi, Has it been decided the dates relevant to the next version? I have only looked here (http://www.boost.org/development/index.html) and didn't see anything. I'm not trying to change / hurry or delay anything. I just want to start moving some changes from trunk to release, but I work slowly and want to start closer to the beginning of a release, not closer to the end of one. Regards, Andy Tompkins.
On 17 April 2011 18:40, Andy Tompkins <atompkins@fastmail.fm> wrote:
Hi,
Has it been decided the dates relevant to the next version?
No, it hasn't. This release has been disrupted by the point release from the last cycle and the standards meeting. IMO we shouldn't start the beta until after BoostCon. It's less than a month away, which I don't think is enough time to release from the state we're at.
On 17 April 2011 18:40, Andy Tompkins <atompkins@fastmail.fm> wrote:
Hi,
Has it been decided the dates relevant to the next version?
No, it hasn't. This release has been disrupted by the point release from the last cycle and the standards meeting. IMO we shouldn't start the beta until after BoostCon. It's less than a month away, which I don't think is enough time to release from the state we're at. What state are you referring to? What has changed since your last
On Sun, Apr 17, 2011 at 9:00 PM, Daniel James <dnljms@gmail.com> wrote: proposed schedule? More communication to the library developers would be appreciated! We worked hard to get Phoenix V3 ready for release within our proposed dates. There was neither conformal nor denial whether these dates were good. And suddenly the current state is considered not release worthy ... I am disappointed by this information policy. I would like to see at least a beta version of 1.47 before boostcon so people who are listening to talks can actually download and try the code we are presenting. Thomas
On 18 April 2011 14:56, Thomas Heller <thom.heller@googlemail.com> wrote:
More communication to the library developers would be appreciated!
I wrote an email suggesting that we should decide on a schedule quickly, I didn't receive a response for 5 days. That gives the opposite impression.
We worked hard to get Phoenix V3 ready for release within our proposed dates. There was neither conformal nor denial whether these dates were good.
I think I was pretty clear that nothing was happening. http://lists.boost.org/Archives/boost/2011/04/179794.php
And suddenly the current state is considered not release worthy ... I am disappointed by this information policy.
A release takes time to prepare. At least a month given the desire for longer beta periods. Since we haven't started, it's unlikely that we've got enough time for a release before boostcon.
I would like to see at least a beta version of 1.47 before boostcon so people who are listening to talks can actually download and try the code we are presenting.
This is the first time I've heard that, I was under the impression that it could wait for the next release cycle, which would be after boostcon.
On Mon, Apr 18, 2011 at 6:28 PM, Daniel James <dnljms@gmail.com> wrote:
On 18 April 2011 14:56, Thomas Heller <thom.heller@googlemail.com> wrote:
More communication to the library developers would be appreciated!
I wrote an email suggesting that we should decide on a schedule quickly, I didn't receive a response for 5 days. That gives the opposite impression.
We proposed an alternative schedule ...
We worked hard to get Phoenix V3 ready for release within our proposed dates. There was neither conformal nor denial whether these dates were good.
I think I was pretty clear that nothing was happening.
... which apparently has been misinterpreted. So it was a misunderstanding on both sides. In any case, we tried hard to keep the deadline we proposed. But you are right ... apart from our queries there wasn't anything else.
And suddenly the current state is considered not release worthy ... I am disappointed by this information policy.
A release takes time to prepare. At least a month given the desire for longer beta periods. Since we haven't started, it's unlikely that we've got enough time for a release before boostcon.
I would like to see at least a beta version of 1.47 before boostcon so people who are listening to talks can actually download and try the code we are presenting.
This is the first time I've heard that, I was under the impression that it could wait for the next release cycle, which would be after boostcon.
Right, its the first time you hear, because we waited (like you suggested) for the next officially proposed release schedule. Which didn't came until someone else asked. Additionaly, if you read our proposed schedule again, you might notice that it would have been just in time for boostcon. The part about having it in the release cycle would have been the back plan if you, the release managers would have said: "Ok let us release on time." So, if I had known that the schedule would have been postponed so long ... it would have saved me a lot of trouble. So, who is responsible for making the schedule?
(...) because we waited (like you suggested) for the next officially proposed release schedule. Which didn't came until someone else asked. (...) So, who is responsible for making the schedule?
I can confirm that the schedule is this time quite fuzzy, and there are AFAIK no definitive answers or information in the calendar (but Daniel always answers promptly, thanks for that). Some deadlines would be helpful. Regards, Barend
On Tue, Apr 19, 2011 at 2:09 PM, Barend Gehrels <barend@xs4all.nl> wrote:
(...) because we waited (like you suggested) for the next officially proposed release schedule. Which didn't came until someone else asked. (...) So, who is responsible for making the schedule?
I can confirm that the schedule is this time quite fuzzy, and there are AFAIK no definitive answers or information in the calendar (but Daniel always answers promptly, thanks for that).
Some deadlines would be helpful.
I think Daniel's suggestion to aim for a beta after BoostCon is a good one. --Beman
On 19 April 2011 18:44, Beman Dawes <bdawes@acm.org> wrote:
I think Daniel's suggestion to aim for a beta after BoostCon is a good one.
We could pick a nightly release and label it 'alpha' so that boostcon people would have something to download. With the caveat that some things might be a bit screwy, but if we choose a rough date in advance, people can be prepared.
On 19 April 2011 20:42, Daniel James <dnljms@gmail.com> wrote:
On 19 April 2011 18:44, Beman Dawes <bdawes@acm.org> wrote:
I think Daniel's suggestion to aim for a beta after BoostCon is a good one.
We could pick a nightly release and label it 'alpha' so that boostcon people would have something to download.
Does anyone want this?
Daniel James wrote:
On 19 April 2011 20:42, Daniel James <dnljms@gmail.com> wrote:
On 19 April 2011 18:44, Beman Dawes <bdawes@acm.org> wrote:
I think Daniel's suggestion to aim for a beta after BoostCon is a good one.
We could pick a nightly release and label it 'alpha' so that boostcon people would have something to download.
Does anyone want this?
How would this be any different than just hooking your local svn system to the (next) release and updating whenever you want to? This is what I do so that I'm always testing against the "next" release. it's cheap, easy, and almost guarentees that I'll be ready for release before the actual release date. Robert Ramey
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
On 23 April 2011 18:14, Robert Ramey <ramey@rrsd.com> wrote:
Daniel James wrote:
On 19 April 2011 20:42, Daniel James <dnljms@gmail.com> wrote:
We could pick a nightly release and label it 'alpha' so that boostcon people would have something to download.
Does anyone want this?
How would this be any different than just hooking your local svn system to the (next) release and updating whenever you want to?
This is what I do so that I'm always testing against the "next" release.
it's cheap, easy, and almost guarentees that I'll be ready for release before the actual release date.
This would be mainly for conference attendees who want to try the new libraries. Downloading a file would be faster than checking out (especially if several people do it at once). There's also the possibility of an unexpected subversion check-in breaking a demo.
On Sat, Apr 23, 2011 at 4:38 PM, Daniel James <dnljms@gmail.com> wrote:
On 19 April 2011 20:42, Daniel James <dnljms@gmail.com> wrote:
On 19 April 2011 18:44, Beman Dawes <bdawes@acm.org> wrote:
I think Daniel's suggestion to aim for a beta after BoostCon is a good one.
We could pick a nightly release and label it 'alpha' so that boostcon people would have something to download.
Does anyone want this?
Sounds like a good compromise to me!
Hi Beman, Daniel, List, On 19-4-2011 20:44, Beman Dawes wrote:
On Tue, Apr 19, 2011 at 2:09 PM, Barend Gehrels<barend@xs4all.nl> wrote:
(...) because we waited (like you suggested) for the next officially proposed release schedule. Which didn't came until someone else asked. (...) So, who is responsible for making the schedule?
I can confirm that the schedule is this time quite fuzzy, and there are AFAIK no definitive answers or information in the calendar (but Daniel always answers promptly, thanks for that).
Some deadlines would be helpful. I think Daniel's suggestion to aim for a beta after BoostCon is a good one.
That means after Friday May 20. Can we set a date? Old paste:
Something like:
4th April: Release branch closed for beta (bug fixes only). 11th April: Beta target date. 2nd May: Release target date.
Probably the Monday 23 after BoostCon is too early (?), so it might be, let's say, Monday May 30 for "Release branch closed for beta (bug fixes only)." ? Sorry to ask this every now and then but it is useful to know w.r.t. my planning and schedule. Thinking further, we are close to 2 months late then. So either we might expand this cycle to 6 months, or do something like I proposed above and have also the next cycle during longer (then 4 months), to get back into the normal schedule. Thanks, Barend
participants (6)
-
Andy Tompkins -
Barend Gehrels -
Beman Dawes -
Daniel James -
Robert Ramey -
Thomas Heller