Hi Gianluca

Gianluca Turconi wrote:
> I've just read the article present on your news site:
http://technology.guardian.co.uk/weekly/story/0,16376,1660763,00.html?gusrc=rss
> 
> and I have to say it includes misleading and false assertions,
> which I consider harmful for both my professionalism and my efforts in
> favor of the OpenOffice.org Free Software Community.
> 
[ ... ] 
> In addition to this fact, I want to underline, as a personal correction
> to the quoted article, the large contribution made by amateurs to the
> OpenOffice.org Community, either with patches to the code (still
> included into the OpenOffice.org Issuezilla tool) either with free
> contribution to linguistic tools such as spellchecking dictionaries and
> thesauri that *are* code and involve tens of volunteers and thousands
> of work hours.

I understand your point and see it friendly. As others have indicated
before, the articles' assertions could have been less emphatic. But I do
agree with Andrew Brown's main point: the quality assurance protocol, the
bugfixing process for OpenOffice needs a dramatic improvement.

You can search my previous posts during the OOo 2.0 development process. I
have become very very upset by the way user demanded features, or simple
buxfixes are dealt with. There are screaming examples of bugfixes delayed
for ages without a clear reason. I am sure any users that have visited
isuezilla has his/her own list. And what about releasing 2.0 final with
significant regressions?, like broken numbering styles. We are mirroring
other's practices.

I have reported some bugs myself, and found many others that were already
reported. I do use issuezilla, browse it, read the buxfixes part of
"release notes" etc. And have bad feelings from that readings. What really
scares me (and I see this fear also in Brown's article) when browsing
isssuezilla is the impression I get from the comments of the project
members. In many cases it seems that those programmers have never used OOo
themselves for a productive work. They do not seem to realize what's the
use for the feature they are working in. There are many talks about
interfaces, framework etc. but not on usability. In addition, live
discussions are not frequent on issuezilla, I get the impression they are
not wanted. It seems that issuezilla is seen by developers as a point to
pick up clear and well-defined tasks. But it is impossible for end-users to
setup such clear-cut commands. Issuezilla is complex and scares users, but
more fundamentally, users do not know how to implement a feature, and
programmers seem to listen only to "code-jargon": the two communities speak
a different language.

My impression is that we, OOo developers and users, need a better
communication channel between these lists/newsgroups and core programmers. 
An intermediate layer between end-users and programmers, managed by
experienced users with a knowledge of OOo internals (people like Andrew
Brown himself and many others). Dedicating payed people to read these lists
and extract statistics of most demanded features, most frequent FAQs, most
frequent user confusion with tools etc. This would not be wasted money,
even if that mean two programmers less for the core. Relying only in
isuezilla votes has revealed non-practical. I see the need for some kind of
authority that sum up user input, define tasks and priorize them for
isuezilla. I have no doubt this process is done now, but on the programmers
end and hidden from users. I would like to see this decicsion-making
process open, transparent and web published.

Last but not least, I must agree that bugfixing depends on good QA team. I
do not know the internals of this process in OOo. I do not know if
currently QA members are Sun employers or volunteers. I do see a need for
full-time (payed) people. But I do see this as a very good niche for
volunteer work. And I would not dismiss the efforts dedicated by QA
volunteers as "not writing code". I read Andrew Brown's article as meaning
that any effort (even non-programming tasks) put into bugfixing and quality
assurance is essential for overall quality and success of the software.
Thus, volunteer work into QA is, to my eyes, as important as the
core-programmers team. An essential, and community-contributed, part of OOo
development. The article missed this point: programmers are not all.

Enrique


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to