Re: Some Points, about Arabeyes [Part I]

[Arafat Medini <lumina at silverpen dot de>]

Disclaimer: speaking of the fact that ppl come and go to this project
and that our #working# contributors are very few, based on this fact I
am answering this mail.

Am Fr, den 04.06.2004 um 13:02 Uhr +0100 schrieb Muhammad Alkarouri:
> Salam Arabeyes people,
> 
> As we all know, Arabeyes has done a lot of contribution to the Arab open source
> and continues to do. In fact, it is now the flagship website in this area. As
> it comes of age, it needs various changes to adjust to its current size and
> position.
> 

(But flagship in a world based on some hundreds of Linux users...)

I think that the project is still too small to be too burocratic. I
think that YES some day in the future your (good) idea may be adopted
but today it's too much burocracy for a few ppl.

> I have some suggestions that I hope you may consider:
> 
> 1. Documentation on Arabeyes policies needs to be improved in two dimensions:
>    a. Historical records of valuable discussions need to exist

The problem is here, what is valuable?
In the case of very critical discussions and points of view we have the
WIKI which gained very much acceptance btw the users, maybe I am the
only one who doesn't know how to use the WIKI ;) (what a shame...)
But the wiki as a tool proved to be very good, accepted and indeed
helpful!

>    b. Arabeyes policies and general 'agreed-upon' things need to be widely
> known and easily accessible.

I am very unsure regarding this point. Does the majority of ppl who only
pass by Arabeyes have the right to know about our decisions?
hmm,well I don't really know. (really) 
Looking at other OSPs/FSPs like the kernel-project, freedesktop.org,
gnome, Apache... this is not really the case. There are announcements
made, yes, but the whole process of deciding, making roadmaps etc... is
taken internally.
AND whith the distinction that Arabeyes is a meta project, Arabeyes
itself has NOTHING to speak about, because it is nothing #real#!!!
Under it's umbrella decisions are made, ppl meet, etc...
but there isn't any Arabeyes roadmap cz arabeyes doesn't stand for a
special project.
Maybe some info about internal work as an rss feed, or as an
announcement on l4a or anything in that regard but nothing really
systematic...

> Both these points are now actually met by Arabeyes to some extent. Failures do
> occur. Specific examples that I know of:
> a. Due to lack of feedback, some projects do most of their feedback through irc
> and outside projects lists. A case is the Arabization support in Debian
> Installer[1]. This will lead to duplication of work or loss of valuable support
> opportunities.
> b. Recently, a row has happened about the Arab/Muslim duality which some assume
> to be from fully existing to non-existant. While there is a policy in Arabeyes,
> it is not documented in the About page, the manifesto or other similar pages.
> Result: everyone has to know about it on his/her first row.

The problems you spoke about accured, that's true but it's more the
preference of a hacker to have more personal contact with others that
makes him IMHO contact via irc.

For the moslem/arab problem I think it's more of a cultural problem that
even if we do a manifesto will hount us with many new members coming to
the project. cz it's more the culture itself that stands in the
foreground and not Arabeyes's manifesto IMHO.


> Suggestion: 
> a. Use a kind of Request for Comment (RFC) of Policy Enhancement Proposal
> (PEP).

I am not against this. But my Q would be then: how burocratic is this?
If it's not so burocratic then it's ok else it'll be good to find
someone he wants to make this on himself maybe you? ;)

> If any topic needs to be a policy, somebody can suggest it as a policy
> proposal. Discussion can be done about it and it may be changed to reflect the
> general consensus. In the end, the core team will decide the policy taking into
> account opinions and/or consensus. I am speaking with
> http://www.python.org/peps/pep-0001.html in mind.

Well that's what ppl are doing from time to time.

> b. All important correspondence should be through public e-mail, as it is
> already archived. In case there is some good reason to do otherwise in a
> particular project, a feedback channel should be established frequently enough
> to get any possible contributors, coordination between two projects, etc. This
> should be the maintainers responsibility.

Here I am more skeptical, if the goal is to tell every newbie what's the
way to contact, then I'll tell you that different ppl prefer different
ways; many use private mail as the first way of contacting cz they are
too shy or they don't know about ML, others use mailing lists, while
others use irc. in some cases wiki is used too!
It's not right to force a newbie to use a specific way of contacting us.
A newbie is a ptotential hacker/translator he's a king till he starts
work then he becomes our slave MUAHAHA ;D

For every other point I agree with you.

> c. Every document should be dated. Yes, that includes the charter and the
> manifesto.
> 

Very good idea

> 2. In all projects, enough research should be done in the start so as not to
> duplicate other work done in other places, or at least to give good reasons for
> doing so. This includes duplicating work in Arabeyes, but also includes
> attempts to translate every single technical word without referring to various
> Arabization projects (not web based) and authorities around the Arab world[2].
> 

I am for duplicating work, and I'll tell you why: I think a hacker has
the freedom to work on whatever he wants, if he can make his passion
come true we will have two projects competing with the same energy and
passion to make the best technology. Telling someone to stop and work on
another project mayb make him loose interest.
We are NOT a company, lately this company thinking is detroirating every
linux project out there.
Everyone does not want to duplicate work, what about KDE and gnome? BSD
and Linux, gnumeric, abiword And OO.o, etc...
As for existing translation projects, as they have their own agenda
they'll not let anyone do the same work anyway if someone has already
registered on working on them.

> Enough for a first part..
> 
Salam

> Salam,
> Muhammad Alkarouri
> 
> [1] From my point of view, what has happened is that Debian Installer needs
> Arabic support, both as translation and as BiDi supporting in text mode.
> Translation went along nicely. We needed BiDi and shaping in newt
> (http://lists.arabeyes.org/archives/developer/2004/January/msg00042.html). The
> Arabeyes community had no feedback about that, until it was working. Though it
> was mentioned before that proper support for Arabic in newt will benefit other
> distros.
> 
> [2] Many of the universities in the Arab world teach computer science in Arabic
> already. When Sudan tried to go that way, we found a large experience from
> Egypt, Syria and Iraq. No doubt there are other countries as advanced.
> 
> 
> 	
> 	
> 		
> ____________________________________________________________
> Yahoo! Messenger - Communicate instantly..."Ping" 
> your friends today! Download Messenger Now 
> http://uk.messenger.yahoo.com/download/index.html
> _______________________________________________
> General mailing list
> General at arabeyes dot org
> http://lists.arabeyes.org/mailman/listinfo/general
>