Draft, State of the GnuCash project, part II: What can you do to help?

Wed Nov 5 15:37:31 CST 2003

On Mon, Nov 03, 2003 at 01:27:01PM -0500, Benoit Gr?goire was heard to remark:

> THIS IS A DRAFT!
> 
> 1  Introduction
> 
> The goal of this document is to guide potential contributors to where their
> contribution is most useful.
> 
> response, and will avoid current developers spending time on superfluous=20
> emails:

My overall impression is that much of what you are dealing with (if not all of
it) are 'project management' issues that a 'project manager' would deal with.  

I suspect that my definition of 'project manager' is much more expansive than
yours.  However, this document that you just prepared is the kind of document
that only a project manager would care about, read, or write ... 

> 2.1.1  Projects tied to other open source projects

This is an example of the kind of list that a project manager would keep.

> 2.1.1.1  Help with other projects on which GnuCash depend for many features

Can you add the following:  
-- a new/improved gtk2 tree implementation? The current gtk2 tree is complex,
hard to use, and fails to provide many important features.  The old ctree 
was much nicer ....

> 2.1.1.2  Improve graphing support
> GnuCash currently uses the Guppi graphing engine=20

Add gtkplot (from the gtkextra package) as a candidate alternate technology.

I've used gtkplot, I found it easier to understand/use/extend than guppi.
I don't know if gtkplot has been ported to gtk2 yet.

> 2.1.2  Small projects for new developers to learn the codebase.
> 
> 2.1.2.1  Adopting a bug
> Not all bugs are hard to find or fix.  Some didn't get fixed simply because=

We need a "screen team" that can review bugs, and sort them into
duplicates/fixed, and easy/hard.  So that the appropirate people can
work on them ...

> 2.1.2.2  Eliminating deprecated functions(code refactoring in the engine)

 There is a wording problem here.  It should say something like 
"change/update code that uses deprecated engine functions so that it
no longer uses deprecated functions". 

The 'refactoring' (if any) occurs outside the engine.

Most of these changes do not require any refactoring at all.

Refactoring should only be done by more experienced people who already
know how a module works, and not by beginners.  

Note that some of the old file-import/backwards-compat code uses
the deprecated functions, and that is OK, that code doesn't need 
to be changed.

Note also: I'd like to discourage these kind of changes until the gnome2
port is complete, and/or that these changes be done in the gnome2 branch only.
Otherwise, it will become very hard for the gnome2 port to move forward,
and merge back into the main tree.

> 2.1.2.6  Fix profiling support=20
> Skills: Strong C skills
> Current leader:=20
> Primary contact: Derek Atkins?

You may want to add the fields "important" and/or "urgent".  Liek this one is
important to me.

Also, you may want to add a 'sizing' field. Is this days, weeks, months or
effort?

> 2.1.3  Larger coding projects (longer commitment):
Oh, ok, well, never mind the last comment ...

> 2.1.3.2  Finish Scheduled Transaction implementation
> There is an extenseve todo list available at=20
> http://www.gnucash.org/lxr/gnucash/source/src/doc/TODO-schedxactions.=20

this list is hard to understand ... it would be nice if jsled explained it a
bit more ... 

> 2.2.0.2  Bug management (Tier 2 Technical support)

ok, this was the 'screen team' I mentioned above.
part of the task would be to figure out who should be fixing the bug,
and if its a coder, maybe fixing a few too.

> 2.2.0.4  Distribution maintainer
> =2D Debian (Need help fixing compile problems on some architecture, contact=

James Traecy ?

> 2.2.0.9  Wiki team leader and editor

I'm trying to add some additional security to the gnucash server, which
if all goes well, I would then be willing to run a wiki on it.  If you are
interested.   At the moment, I'm paranoid about being rooted.

> 2.2.0.11  Translation
> Current leader:=20
> Primary contact:=20

cstim -- Christian stimming

> 2.4  Evangelism and marketing:
> 
> 2.4.0.1  Improve feature list
> Compile a full list of GnuCash features, sorted in an intelligent manner to=
> =20
> keep it readable.  This would replace=20
> http://www.gnucash.org/en/features_5.phtml.

Actually, should replace features_1 through 4 as well, that whole thing is
hokey.  Also, change it so that it doesn't use the english alnguage in the 
graphics, since it cannot be translated.

> 2.4.0.2  Write a rebuttal to the GnuCash perception problem
> Why can't GnuCash gain widespread acceptance?  There are many factors, real=

Actually, I'd like to design a survey centered around this.  What are the 
problems, real or imagined?  

part of this should be going out to other mailing lists, and asking there,
'why don't you use gnucash'? and getting them to take the mini survey for
why not.

Note that surveys are useless if you don't ask the right questions.  So we
should take some time to design the right set of questions.  So for example:

"Answer the following questions only if you considered using gnucash 
but then decided not to":

1) Did you decide not to use gnucash because (check all that apply):
   -- too hard to install
   -- too hard to understand
   -- not web enabled
   -- I just needed a much simpler checkbook balancer
   -- I needed more business features
   -- I couldn't figure out how to import my msmoney data ...
   -- gnucash doesn't work with KDE (yes, some people actually think this...)

I think we can brainstorm some more of these ...

2) if you checked 'yes' to 'missing features', then which features was it
   missing:
   -- budgeting
   -- import from 

> Benoit Gr=E9goire, http://step.polymtl.ca/~bock/
> Pr=E9sident, Comit=E9 Jeunesse de l'Association des Dipl=F4m=E9s de Polytec=
> hnique.

and, appearently, the acting gnucash project manager.

You might discover, for the better or the worse, that being a project manager
can be a lonely job.   The developers never tell you what they're doing, you
have to ask them.  When you put together a list, such as the one above, it
can seem sometimes that no one really cares... people probably won't provide
feedback ... 

---

Anyway if/when you get an html'ed version of this, I'd like to post it on the 
website.  In particular, I want to get rid of/replace the current totally
obsolete link to the gnucash projects list. 

I'd also like to change the web link "how you can help" to look like a somewhat
modified version of this doc.

--linas

-- 
pub  1024D/01045933 2001-02-01 Linas Vepstas (Labas!) <linas at linas.org>
PGP Key fingerprint = 8305 2521 6000 0B5E 8984  3F54 64A9 9A82 0104 5933