I agree with David. A release is a good idea even if it requires a "brown paper bag" release immediately when some show stopper pops up. We clearly need bug reports against the current codebase rather than the last released version. Given the extensive renovations since the last 0.11.x release, it doesn't seem like bug reports against 0.11.x are at all useful. That alone is a solid argument for a release ASAP.

While I'm not an actual FreeCol developer, I do have some experience maintaining a couple of projects and I have observed quite a few others. I don't think a "brown paper bag" release is something to be worried about. Especially since an immediate followup release can be made to minimize the damage.


Also, as a general player/user, I think it would be beneficial to have a current official release. Otherwise, players are going to continue tripping on the known problems with 0.11.x because that's all that's going to be packaged for most software distributions (Linux, etc.).

Of course, reasonable efforts to have a solid release are warranted. Things like giving the translators time to catch up, for instance.

Incidentally, I haven't played much lately. This has mostly been a time issue, but I have to confess that I haven't been as dilligent as I should have been about reporting bugs that pop up during such sessions. I've kind of been keeping an eye on the repo and waiting for some of the churn to subside.

On 2017-08-11 10:36 AM, David Lewis wrote:
/> All true.  However we have had a few "Oh shit" releases, where we put
out version N only to have to put out N+1 almost immediately to fix a
glaring bug. /

In an ideal situation, and I'd say once FreeCol is considered out of beta (stable), we'd have release candidates where this sort of thing would be found before creating a stable release. It's kind of expected with beta software in general. While we can try and do everything we can to avoid having this happen, we don't have enough eyes on trunk, and we need more eyes on it to catch these kinds of problems before they would hit a release. Aside from creating a release, I'm not sure how else to get those eyes on it. I'd love to do a pre-0.12 release to do just that... but since this is still beta, 0.12.0 /*is*/ the pre-release to 1.0.0 anyway. That's why my opinion/vote would be to just go for it -- creating the 0.12.0 release soonish, knowing there are some issues. I think the major thing we want to try and avoid is big blockers (P0). So far most have been fixed, and we're down to what I'd consider annoyances, and other minor issues that don't block gameplay as far as I've been able to find.

/> Agreed in general.  Pity this release required a major rewrite that took/
/way longer than it should have./

Such is the way of OSS that doesn't have enterprise backing, with a very small team and you doing 99% of the work. I am amazed and pleased that you were able to stick with it for so long! I don't know that I could have done that.

/> The question then is "how soon do you/
/want to do 0.12.1?":-)./

Shortest answer: As soon as necessary.
If it needs to be the following week, or 2 weeks later. I'm okay with that. I would like to automate the release process so that /_if we wanted to_/ we could release on a 2-week cadence. Not saying that we would (or wouldn't ;) ), just that we should be capable of it.

Again, if there are objections, I have no problem pushing out the release date further, plus if we find another P0 blocker, we would have to do that anyway.

I'd love to hear other folks on this mailing list chime in as well... what are your thoughts, are you playing the game? If not, could I ask that you play a bit and let us know how it looks now? :)

-- David

On Fri, Aug 11, 2017 at 1:40 AM Michael T. Pope <mp...@computer.org <mailto:mp...@computer.org>> wrote:

    On Thu, 10 Aug 2017 17:57:26 +0000
    David Lewis <highwayofl...@gmail.com
    <mailto:highwayofl...@gmail.com>> wrote:
     > I agree that 3083 should be fixed before release, that is
    currently the
     > biggest annoyance in the game.

    Its now top of the list.  However, once fixed, its going to need
    play testing as there are a lot of cases.

     > Part of the reason I'd like to get this release out sooner rather
    than
     > later (RERO: Release Early, Release Often)...

    All true.  However we have had a few "Oh shit" releases, where we put
    out version N only to have to put out N+1 almost immediately to fix a
    glaring bug.  Perhaps I should pay less attention to this as it
    would now
    be the new release manager's problem:-).  You could equally argue that
    because it is a while since the last release a bit more testing will not
    matter that much.  I certainly would prefer the bug flux settles
    down from
    current levels.  It has been a busy week on the bug tracker.

     > My background in working for Starbucks Technology, Disney
    Interactive, and
     > Marvel Games is that early and frequent releases created a
    tighter feedback
     > loop between developers and testers or users, and generally garnished
     > higher usage. I believe it would allow the game development to
    progress
     > faster, and since it puts the newer code in the hands of the users,
     > enabling them to help define what the game will become, and
    ultimately
     > result in higher quality software.

    Agreed in general.  Pity this release required a major rewrite that took
    way longer than it should have.

     > ...just stating why I'm hoping we can do a release soon.

    Understood and generally agreed.  The question then is "how soon do you
    want to do 0.12.1?":-).

    Cheers,
    Mike Pope



------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot



_______________________________________________
Freecol-developers mailing list
Freecol-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freecol-developers



------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Freecol-developers mailing list
Freecol-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freecol-developers

Reply via email to