Getting money for Gnucash development [was Re: Newbie migration issues]

[Neil Williams]

On Sunday 30 January 2005 7:16 am, TC wrote:
>  > I also have no idea how to counteract the FUD of moving away from
> > Quicken.
>
> Actually, when I think about it, it's not really FUD that's the issue.
> Quicken is simply a Very Good application.

It's years since I've used Quicken, but it never handled double-entry 
accounting when I used it. It takes a lot of time on these lists to correct 
some assumptions that are only encouraged under the Quicken system or 
categories. 

From the user perspective, Quicken is more friendly because it (when I last 
used it) was dumbing down the whole process. That, unfortunately, does NOT 
help the user understand the system any better, making migration more 
difficult. It's only a weak form of lock-in but it hinders the migration of a 
very large number of potential users.

> Mozilla, especially in the form of Firefox and Thunderbird, really does
> offer something (e.g. security) that is lacking in the equivalent MS
> offering.  But Gnucash doesn't, I think, have the same differentiator
> w.r.t. Quicken.
>
> What Gnucash *does* provide is a quality finance package for GNU/Linux.

Quicken might be popular on Windows but when I was using it, accountants 
certainly did NOT deem Quicken to be a *quality* finance package for MS. I 
think that is a larger distinction than you seem to express. There is a big 
gap between cheapo Quicken and mega-deal Sage (in the UK at least).

> Which also means that the lack of a Windows port is not relevant.

It would be good to have a quality package for Windows that is free software. 
Proprietary programs should always be opposed by providing free software 
alternatives so it is always relevant. That said, the ideal is just that - an 
ideal - and reality is some way behind. There is nothing stopping a port of 
the g2 branch to Windows - except developer time and inspiration. If anyone 
decides to take it on, they will find the existing GNU/Linux developers to be 
encouraging and supportive.

All I will say is that GnuPG (which does have a windows version) has a lot of 
queries on their mailing lists from Windows users who simply have a very poor 
understanding of the larger issues - in their case, security. It's not 
usually solely their fault - their OS encourages a lazy attitude to security 
and users follow. We could see the same thing here - Quicken users migrating 
from categories to double-entry on the same platform could significantly 
increase the list volume and we need to be ready for that increase. It's not 
unwelcome but it would need to be taken off Derek's shoulders. At present, 
the vast majority of replies on these lists are from Derek. It all comes back 
to helping the team. If more users get involved in bug triage, if more users 
get involved in answering queries on the list, if more users contribute to 
writing documentation, if more users consider actually making the transition 
into developers . . . . .

> But maybe all of this is conceding too much ground.  Perhaps the trick
> here is not to get money to allow the existing developers to go full
> time.  Instead, perhaps it is to figure out why there aren't more
> non-full-time developers.

Precisely. Things like a Windows port would probably not happen even if the 
entire current team of developers became salaried to the same level as their 
current non-GnuCash income. The g2 port would be finished a lot sooner and 
releases would speed up, but a Windows port just does not seem sufficiently 
attractive to anyone currently developing GnuCash to do it themselves.

> And I would have thought that if a young coder, with spare time on
> his/her hands, wanted to hack on some worthwhile and valued OSS project,
> a Quicken-for-Linux would be one of the top three projects (the other
> two being the kernel itself, and something like Chandler).
> So, how come it doesn't?

1. GnuCash currently needs a lot of development time on the existing 
platforms, so new developers may well be attracted (as I was) to work within 
the application on existing platforms.

2. New developers need time to understand the existing platform and code 
before a port becomes probable. Porting an application to another platform is 
difficult work and creating one set of code that can be compiled on both 
platforms requires some detailed knowledge of both platforms. Developers with 
that kind of dual knowledge are rare. It took me the best part of 4 months to 
get under the skin of the code, partly because GnuCash is my first GNU 
project in C and adjusting from writing C/C++ on Windows takes time. During 
that time, I asked Derek, Linas and Josh a lot of questions and was probably 
quite a nuisance. :-)

3. New developers have to have an interest in finance to even look at GnuCash 
- many work in KDE or contribute to a distribution like Debian, others will 
find their niche in other projects like Apache, MySQL, multimedia players, 
content management systems, (just thinking of those teams that exhibited at a 
recent Linux Expo .org village), there are just so many projects.

4. Even if GnuCash IS in the top three (I'd say possibly in the top 10) most 
attractive projects for new developers, the port to Windows is NOT the 
motivation. Most projects don't have a Windows version.

> [And it's no use me answering - I can't program for toffee;

You can write an email, you can help with the documentation.

You are on the list, you can help answer some queries to take some load off 
Derek. Don't say you can't be involved - everyone can be involved - so start 
at home!
:-)

> that's why 
> I'm not involved.  And by the looks of things, helping out in something
> like Gnucash requires serious ability, or time to acquire it.

Only to get into the source code itself - there are a huge number of tasks 
that ordinary users should take on and this mailing list is one. 
Documentation is another. Bug triage is a large area that can include 
ordinary users as well as those with a little knowledge of code to help 
isolate the problems.

> But what 
> about those who *can* already code?]

Documentation and more answers on the list from people other than Derek would 
help!

-- 

Neil Williams
=============
http://www.dcglug.org.uk/
http://www.nosoftwarepatents.com/
http://sourceforge.net/projects/isbnsearch/
http://www.neil.williamsleesmill.me.uk/
http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://lists.gnucash.org/pipermail/gnucash-user/attachments/20050130/23660d62/attachment.bin