Re: GnuCash page on GO site



Charles Goodwin <charlie xwt org> writes:

> On Fri, 2004-02-20 at 17:24, Derek Atkins wrote:
>> Oh, so this isn't even released yet??  UGGH.  Perhaps gnucash-2.2,
>> then.  I don't want co-dependencies when we release gnucash-2; I want
>> all the dependencies to be well-available.
>
>>From what you indicated, Gnucash-2 is months away at best.  libgoffice
> is progressing quickly and probably be released by the time gnucash-2 is
> ready.  Why waste your time messing about with guppi when you can help
> libgoffice evolve simply by using it?  It's not as if it would consume
> any resources.
>
>> GnuCash wont code to non-released dependencies
>
> Ordinarily this is a good approach.  But to go with guppi over
> libgoffice would be a really big mistake.

Note that I didn't say we'd go with Guppi.  Guppi is a non-starter
for gtk2, pretty much..

>> Basically, we choose a base system (let's say FC-1 -- it was RH9 (and
>> even RH8 for a while) but the g2 port is taking long enough that we
>> might change it)...  We want as many deps as possible to be available
>> on the base platform.  This means GTK-2.2.4 is our target.
>
> By coding custom widgets now that are in Gtk 2.4 you are just generating
> yourselves extra work when, by your own admission, you are short on man
> power.

Not at all -- we could just pull those widgets into our source tree...
Assuming what exists today meets our needs.  It may.  It may not...

Frankly the phrase wheel is a UI nicety. We don't NEED it for an
initial g2 port. We can easily live without it.  Until gc-2.2.

> You are also aiming at today's distros when, by your own admission,
> Gnucash-2 is not even alpha quality yet.  When you release you'll be six
> months or more out of sync and you'll be playing catch up again.

I'm looking at our target audience.  Our target audience is people who
have installed their machines on average 6-12 months prior to our
[even].0 release.  If we assume that the [even].0 release happens 3
months after our [odd].0 release, that means we've got 3-9 months of
deployment lag-time of the distros.  So, yes, we choose a distro
that's released 3-9 months before our planned [odd].0 release.

Right now that's looking like FC1.

> Your comments sound, to me, like you're willing to place limits to keep
> people happy.  You can't always appease everybody.  Distributions catch
> up incredibly quickly with things like Gtk.  And for everybody else,
> there's the Gnucash-1.x series.

Oh, we're definitely willing to place limits to keep people happy.
Our users are EVERYTHING to our project.  Without users, why are we
doing this?  Moreover, *I* want to be able to use it, and I don't
fully control the distro on my machine -- I certainly don't control
the version of Gnome that's installed.

I'm not sure what your definition of "quickly" is, but unless you mean
Debian/unstable, it can be a good 3-6 months between releases of a
Distro to "catch up".  That's WAY too long for us to wait for users to
easily build gnucash.

-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
       warlord MIT EDU                        PGP key available



[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]