Christian Stimming <[EMAIL PROTECTED]> writes:
> The point of the backport is that I and the openhbci guys like to see a
> openhbci2-ported gnucash to be available sometime over the next 3 months. How
> long does it take until 1.10 or 2.0 is going to be released? Maybe we should
> think about planning some gnucash release schedule here.
The only problem with trying to plan this kind of release is that
we don't have dedicated developer resources, which makes it hard
to plan for how much you can get done in a period of time.
> If a 1.10/2.0 takes longer than 3-4 months (which is how it looks like ATM),
> then I would like to backport the openhbci2 code into the 1.8-branch in the
> meantime. But if we can come up with some release schedule that heads for a
> 1.10 in that interval (2.0 will probably take longer than that), then I will
> happily leave the 1-8-branch alone...
I guess it depends what features we "require" for a 1.10 or 2.0
release. For example I'm willing to forego the SQLite work in a 1.10
release; I am not so willing in 2.0. Also, the g2 branch is mostly
working (although there are a few major missing pieces) -- if we could
get some solid programmer time to complete that work we can move to
gnome2 quicker (which I think EVERYONE would like). I've been
spending most of my time in the g2 branch, as has hampton, but there
is still much to do. Moreover some work we've been "waiting on" for a
very, very, VERY long time, like the SQL re-work, etc (Matthew: hint hint).
Also, I don't think that our dependencies should dictate our release
schedule. If that were the case we should have had a G2 release about
18 months ago. Not that I'm adverse to a 1.8.9 release -- I'm just
wary that such a major change to the stable source tree may
destabilize the sources, requiring a 1.8.10 and 1.8.11 to fix the bugs
that cropped in during the backport.
>> > This would be a good thing because otherwise the openhbci guys have to
>> > continue supporting two versions in parallel, which is always a bad
>> > thing, and also openhbci2 is a necessary step for adding even more cool
>> > online banking features in Germany.
>>
>> Hmm.... It's unfortunate that they had to change the code so much that
>> the API changed... :(
>
> Well, yes, but that's what programming is like. :-\
Not always... Generally you can add new features without modifying
the API. Indeed, you can even do with without modifying the ABI if
you are careful enough and plan it out. It's just that the vast
majority of OSS programmers are not rigourous enough to think like
that. My guess is that it takes more time to think that way, and it's
not as glamourous to think that way, so they spend their time getting
features done rather than thinking about how it all fits together at
the API layer.
> Christian
-derek
--
Derek Atkins, SB '93 MIT EE, SM '95 MIT Media Laboratory
Member, MIT Student Information Processing Board (SIPB)
URL: http://web.mit.edu/warlord/ PP-ASEL-IA N1NWH
[EMAIL PROTECTED] PGP key available
_______________________________________________
gnucash-devel mailing list
[EMAIL PROTECTED]
http://www.gnucash.org/cgi-bin/mailman/listinfo/gnucash-devel
