Re: [gnome-db] GRAND MASTER PLAN

[email protected] (Linas Vepstas)
Newsgroups gmane.comp.gnome.office.general,gmane.comp.gnome.db,gmane.comp.gnome.apps.gnucash.devel
Message-ID <[email protected]>
On Wed, Mar 10, 2004 at 04:48:15PM -0700, Derek Neighbors was heard to remark:
> So much of what is discussed is already do able in GNU Enterprise.  I have
> not entered into the discussion as I know Linas, Dru and Rodrigo are all
> familiar with GNUe so I suspect they have their reasons.

I'll profer a couple of reasons (and these are *not* criticisms of
gnue, just general comments):

-- working out the c-to-python-and-back-again bindings are scary,
   what not with the gtk and scheme event loops already vying for
   dominence.

-- Install; repeatable, servicable user-interface experience. 
   Gnucash has already taken a pounding for being too hard to 
   install.  Most of our bugs are cross-interface bugs, where
   things work only if version x of this is installed, and 
   the obvious work arounds fail due to version conflicts.

   We're ciritcally dependent on obsolete interfaces, e.g. guppi.
   Another example, gconfd-1 vs. gconfd-2 is biting us these days. 
   No fault of our own; we don't use gconfd. Its simply that 
   gtkhtml uses gconf and lord knows why.  And scheme/slib/g-wrap 
   have changed in incompatible ways for reasons unbeknownst to us; 
   these changes only serve to make gnucash unstable, unusable.

   I'm scared of gnue mostly because I perceive it to be a moving 
   target, with its own set of dependencies and infrastructre.

In the same way, I suspect enterprise java beans have a lot of 
these functions as well.  But I figure that ejb is a platform 
that will never become stable, debuggable, supportable.

(crap the only java app I use on a daily basis cannot even communicate 
with the window manager correctly; it creates insane window sizes.
wtf I've complained about this, but that bug will never be fixed.
This is not a criticism of gnue, but is an anti-illustration of what 
I mean by "stable debuggable supportable".  And yes, its probably Sun's
fault.)

--linas

-- 
pub  1024D/01045933 2001-02-01 Linas Vepstas (Labas!) <[email protected]>
PGP Key fingerprint = 8305 2521 6000 0B5E 8984  3F54 64A9 9A82 0104 5933
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.