Salam,
Arabeyes has come a long way in the past two years. I would like to take
this opportunity to bring forth some suggestions on how to bring Arabeyes
closer to attain its long-term goals.
One can read through the Arabeyes Manifesto and notice that our goals have
been as follows:
- generate corporate interest
- motivate programmers-to-be to develop native Arabic software
- retain knowledge-base
- spread free software awareness in the Arab world
- money making opportunities via distributions, support, etc.
I think we have achieved most of the above stated goals and are well on our
way of achieving the rest. Today Arabeyes stands at a point where Arabic
support on Linux and Unix systems is a reality and not just a dream. The
major difficulties and issues have been resolved to a great extent. There
remains some issues that require great attention, but for the most part,
none of it should take more than a year to fully address.
So one must ask, what is next for Arabeyes? A comparison to our setup and
that of sourceforge.net has been made on several occasions. Are we to turn
to an Arab-centric sf.net? I think one of the main things that differentiate
us from sf.net is the fact that we are a community. We are also the Arab
world's first view of Open Source. In other words, we are the ones who set
the bar and standard for Arab open source development. This is a great role
that requires careful planning and organization.
I see Arabeyes goal more than just to add bidi and shaping letters. I see
Arabeyes as the cornerstone of Arab developer communities. A place where
the needs of the Arab user are answered by Arab developers. Today, the
needs are basic Arabic support. Tomorrow, they are different. In fact,
a good example of such a move is the addition of the Quran project. Judging
from sheer observation from the kind of interest it generated, one can tell
that the current needs are for things that are specific to the Arab world
(even though it serves the Muslim world at a large, the vast majority of
Arabs are Muslims). Duali (although yet incomplete) has also generated
a substantial amount of interest for users (but little from developers).
The point is, there are needs that are not being addressed by the generic
stock software.
One way to motivate programmers-to-be is by the Mentor Program that I am
about to propose. In short, it is a program by which veteran developers
take it upon themselves to mentor 1 or 2 apprentices and teach them what
it takes to develop open source software.
We can also gain more interest by shy developers by holding seminars with
specific subjects on IRC.
+ Mentor Program
I. Veteran Developer:
In order to make the Mentor Program efficient there are some pre-requisites
on part of the veteran developers. Veteran developers are defined as current
maintainers of Arabeyes projects.
+ Organize TODO list to note difficulty level (difficult/moderate/easy)
+ Comment code thoroughly
- Use Doxygen (www.doxygen.org) for consistency throughout Arabeyes
projects (w/ the exception of external projects it is optional)
- Note TODO items clearly on code to mark "entry points" for anyone wishing
to contribute
This is expected to be done by March 1, 2003 at the latest. Once that is done,
we can thn move on to the settling in. This can be done via IRC sessions and
will constitute of the following:
+ Ensure apprentice has all the necessary development tools
- compilter, debugger, libraries, autotools, etc.
+ Ensure apprentice is comfortable with using those tools together to
compile the project maintained by the veteran developer.
- this is done by allowing them to make changes in the code, re-compile,
test, etc.
- followed by a short session with the debugger
If IRC is not an option this can be done via email.
Once the apprentice is ready to contribute small fixes, the vetern developer
assigns tasks from the TODO list accordingly.
II. Apprentice:
It would be naive to assume that all apprentices (newbies) will continue
to contribute to their mentor's project(s) or any other. In fact, we can
even make the assumption that the majority will not. So, what do we get
in return from this process?
We must give some responsibility to the apprentice which would involve the
following:
+ Record time spent with mentor on irc, as well as log
+ Take notes and record what they are doing and what they are learning
+ Perform a minimum of 1 task assigned by the mentor (from the TODO
list)
Those records are to be kept and published for others to learn from. Simply
put, as diaries for others to watch (and perhaps learn from).
+ IRC Seminars
This is not an entirely new concept, but it is designed to invite others
to learn about a specific project. It can be employed for existing projects
or items on the global TODO list that are still at hand.
For example, a projct maintainer gives a 30-minute lecture about his project
on irc. This would involve a detailed discussion of the source code
organization and how things are implemented. It should be followed by a
30 minute discussion and Q&A.
Of course this can also be done for items on the global TODO list, where
a volunteer would elect to take the lead in bringing an issue forth and
either announcing the intent to take the lead on it, or hope a discussion
would lead to someone with enough motivation to do it.
+ Translation and Quality Assurance
This is a major concern as we are the official translation team for both
GNOME and KDE interfaces. It is our responsibility to maintain a standard
that reflects the excellence Arabeyes thrives for. This can be achieved
by ensuring the competence of translators. A place to note is
www.ArabTranslators.Net. They offer certification for a fee of $25.
What we can request from them is that they wave this fee for translators
who would like to join the Arabeyes project. There is a good chance of
them accepting it, since the money primarily goes toward funding translation
projects. Perhaps they can discount it. This can serve as an incentive
for translators -- a certification.
As a matter of fact, we should request that they try to promote Arabeyes
on their websites and/or publications as to attract professional
translators who may have the love of sharing.
In a way, we try to reach out to professional associations to help
out in any capacity they can.
+ Publications
One way to ensure maintaining a knowledge-base is by publishing handbooks
and booklets containing subjects ranging from Arabic howto's, to introductions
to Open Source development, etc.
Although it requires funding, the idea here is to provide comprehensive guides
covering a wide range of computing issues, while maintaining the Open Source
philosophy to further embed it in future generations.
This can also generate a revenue that can be invested directly back in
translation and documentation work.
+ Legal Representations
With growing interest from sponsors in investing in the Arabeyes project,
we must accommodate by registering Arabeyes as a non-profit organization
for legal and financial paperwork. The most suitable place for such
a registration in my opinion is the UAE.
Unfortunately we do not have the funds to hire lawyers. Instead we should
contact Arab lawyers through such sites as: www.mohamoon.com -- which is
a good place to see a database of lawyers. Perhaps through a rigorous
campaign we can find at least one lawyer who is willing to represent
Arabeyes (knowing what most likely this will require minimum amount of
work from his/her part).
I think our legal questions should be interesting enough for a passionate
lawyer, who may want to contribute in some small way.
Let me know your thoughts.
--
-------------------------------------------------------
| Mohammed Elzubeir | Visit us at: |
| | http://www.arabeyes.org/ |
| Arabeyes Project | Homepage: |
| Unix the 'right' way | http://fakkir.net/~elzubeir/|
-------------------------------------------------------
---
Was I helpful? Let others know:
http://svcs.affero.net/rm.php?r=elzubeir
Attachment:
pgp00000.pgp
Description: PGP signature