Re: [libreoffice-website] [Proposal] A central employment-office-like web structure for LibO volunteers

In data 27 maggio 2011 alle ore 23:40:11, Andrea Pescetti  
<pescetti@openoffice.org> ha scritto:

> On 23/05/2011 Gianluca Turconi wrote:
> > since few days I'm discussing a proposal about the creation of a
> > employment-office-like collaboration tool for LibO volunteers. Here is  
> > my
> > original proposal:
> > http://wiki.documentfoundation.org/Website/LibreOfficeWiki/Proposed#Central_Employment-office-like_Web_Structure_for_LibreOffice_Volunteers
>
> It's a very interesting proposal, but I find it still too unspecified to
> start assessing tools (nobody wants lengthy discussions) or building
> broken-but-you-get-the-idea prototypes and castrating the project due to
> the constraints of a semi-random tool choice.

Absolutely.

As I've already explained in the Italian discuss list some days ago, the  
first test implementation of this tool has been severely limited to a  
specific CMS because of the as much limited available tech manpower. BTW,  
thanks to Daniele Pinna for his tech help.

> 1) The site must be fun to use. Not easy to use: FUN to use. It's a site
> for hundreds of people, not for the few that are highly committed and
> would contribute even without it. This must be priority #1 in design.

+1

> 2) Success should not depend on developers using this site. LibreOffice
> developers, rightfully, oppose everything that could hinder their
> productivity, and reserve to choose their tools. There's plenty of
> non-developers tasks and the site could now really focus on these only.

I strongly disagree on this point.

Developers are several grade of magnitude more important than any other  
contributor for a *software* project like LibreOffice.

Since "expert developers don't grew on trees", there must be a mean,  
whatever mean, in order to increase occasional developing contributions  
and to transform those one-time contributions into a more reliable and  
expert contribution.

I know developers have their bugzilla and their mailing lists, but when an  
independent, new developer approaches for the first time the project, he  
doesn't know anything about mailing lists (Libreoffice vs freedesktop.org  
ones?!), other developers' roles, tasks which he can complete. In a  
sentence: he doesn't know anything about how the dev part of the project  
works.

I don't want to force developers to abandon bugzilla, it's impossible, but  
rather to provide a more friendly entry level for those new developers who  
approach the project, a mean that can be used to spread expertise about  
sinple and more complex dev tasks and to lessen the now very steep  
learning curve about how the LibO code works.

> 3) The site must provide Faceted search for tasks: one can progressively
> narrow down the results to find all tasks needing knowledge of Italian
> and less than a day and easy-medium difficulty, by clicking suitable
> values on the "Language", "Duration" and "Difficulty" panels. In images:
> http://www.lucidimagination.com/files/image/articles/faceting/CNET_faceted_search.jpg
> (just found with a search engine, I'm not affiliated with that site).

+ 1

> 4) It shouldn't be necessary to register; optimally, the site would just
> use cookies like bit.ly does, at least for the first session.

I also think that registration should not be necessary, but the  
confirmation of starting a task is absolutely needed, otherwise in  
single-action tasks, more people may duplicate their work and get so a lot  
of frustration.

> 5) Cooperating should be rewarding; not (or not only) money as specified
> in the proposal, but bonus points that allow to identify the best
> contributors (may conflict with 4, but there could be some hybrid way,
> i.e., asking for an e-mail account and "registering" users this way, by
> sending them an e-mail in background).

A "stars" system?

You get some points for each contribution and 1 star every X points. When  
you achieve the "5 stars contributor" status, you can be "universally"  
considered an important LibO contributor.

> 6) The site must support free tagging: if the Italian community wants to
> post a series of tasks tagged "libo34it", they should be able to do it.

Uhm... -1

I think freedom of tagging may create a uncontrolled proliferation of tags  
and further linguistic fragmentation (besides needed translations),  
exactly what we want to avoid with this tool.

> 7) There must be an interface to clone an existing task. Optimally, to
> clone a series of existing tasks. I envision Italo having to clone all
> the "Translate LibreOffice 3.3.3 Release Notes to Italian/French/German"
> tasks to the corresponding 3.3.4 ones with one click.

That would spare a *lot* of time.

> 8) Links should be easy to refer and use, like
> http://SITENAME.libreoffice.org/task/308
> http://SITENAME.libreoffice.org/tasks/308+313+318+323
> http://SITENAME.libreoffice.org/tag/libo34it
> so that people could refer to "task 308" and link to it easily.

+1

> 9) Upon marking a task completed, the system should select the closest
> ones and propose them as further work. Like YouTube, if you wish.

And it should be suggested according to deadlines and priorities of the  
project too.

> 10) The site must be based on Free Software infrastructure, and this is
> not really a limit since there are plenty of systems allowing your
> requirements and my requirements.

Well, I suppose nobody *really* thought to use closed software for the  
whole tool.

> Sorry if I made a step backwards and brought the discussion back to
> specifications

Indeed, I thank you for doing so, because I asked for such discussion  
several times. :)

Thanks,

Gianluca
--
Lettura gratuita o acquisto di libri e racconti di fantascienza, fantasy,  
horror, noir, narrativa fantastica e tradizionale:  
http://www.letturefantastiche.com/

--
Unsubscribe instructions: E-mail to website+help@libreoffice.org
Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette
List archive: http://listarchives.libreoffice.org/www/website/
All messages sent to this list will be publicly archived and cannot be deleted