Addition of HBCI support, Maturity of 1.7-branch, next stable release time frame?

Derek Atkins warlord@MIT.EDU
04 Apr 2002 12:10:10 -0500


I would suggest working from the 1.7/CVS tree.

. The Postgres code in the CVS tree is more stable than the 1.6.
. The modularity is pretty cool, and seems to work well.
. There are some new features in the CVS mainline that I think
  should get into the next release.

I'm honestly not sure what particular features are in the works on
that branch.  I know that Scheduled Transactions are only
half-working, and the business code is mostly working but "disabled".
I don't know what other stuff people are working on.

I think it would be less work in the long run to get HBCI into the CVS
code and make a 1.8 release than it would be to get HBCI into the 1.6
tree, make an interim release, and then up-porting into 1.7.  But that
is up to you.


Christian Stimming <> writes:

> Hi all,
> In a message to this ML yesterday it was pointed out that there is an
> Open Source API for the German Online Banking standard HBCI available
> now. I had been looking forward to this since last summer, and
> eventually it seems like this API will actually be usable from within
> Gnucash.
> A note about the point of HBCI support: Roughly half of the banks in
> Germany use the HBCI standard as their online banking technology, so
> there is quite some demand for HBCI client software. There still
> doesn't exist ANY open source HBCI-capable software (Linux and/or
> Windows) out there in Germany, and there exists exactly ONE
> HBCI-capable proprietary software on Linux (Moneyplex by Matrica
> Price: EUR 59 and up). On
> Windows, there are a few more (of course), but they still are only a
> few. This means that IF we add HBCI-support in Gnucash, we would
> actually be the FIRST ONE to do this. Within all humility, this would
> probably give us press reactions up to the leading-edge computing
> journals ("First open source HBCI client ever available") and it might
> boost our user base in Germany by maybe a factor of 10.
> However, when I think about implementing HBCI support in Gnucash, I
> started to wonder what development branch of Gnucash would be suitable
> to do this. Of course, one would say, this is a new feature and thus
> it should be developed on the 1.7 branch, since this is per definition
> the development branch. Additionally, quite a bunch of Gnucash
> internals have changed now in the 1.7 branch, so the development in
> 1.6 and 1.7 is quite different and preferrably we should only work
> within one branch of the two. On the other hand, the time frame of
> implementing HBCI support *and* getting enough users to test it would
> be to do that during the next ~3-4 months. AND once the HBCI support
> has reached a beta stage, I would definitely want to make a *new
> stable release* with precisely the HBCI support added (and giving us
> the press etc. echo mentioned above).
> So I would like to ask the whole community:
> * What do you think about the maturity of the 1.7 branch?
> * What would need to be done before 1.7 reaches a beta-stable stage?
> This doesn't necessarily mean that all of the new user-visible
> features currently in 1.7 need to be stable and/or complete; we could
> very well postpone the completion of some of the user-visible features
> to a later point in time. But for the HBCI feature, we would need to
> have the internal new features of Gnucash to be definitely stabilized.
> * What would be needed to make a new stable release from the 1.7
> branch within the next 4-6 months?
> Cheers,
> Christian Stimming
> _______________________________________________
> gnucash-devel mailing list

       Derek Atkins, SB '93 MIT EE, SM '95 MIT Media Laboratory
       Member, MIT Student Information Processing Board  (SIPB)
       URL:    PP-ASEL-IA     N1NWH
       warlord@MIT.EDU                        PGP key available