Re: Breaking www.gnome.org down a bit

[Claus Schwarm <c schwarm gmx net>]

On Thu, 16 Sep 2004 18:27:55 -0400
Curtis Hovey <sinzui cox net> wrote:

> On Thu, 2004-09-16 at 18:06 +0200, Claus Schwarm wrote:
> 
> > 
> > Thank you very much for the details. :-)
> > 
> > I understand these are not the last words, and will probably be
> > refined.
> > 
> > One question thought, if you don't mind: What software should be
> > listed there? This is the most important part, IMHO, to get an idea
> > on how the use cases could be reached.
> 
> I typed as fast as I could and still couldn't get a whole use case out
> in the last email.  I've been meaning to write some use cases for the
> software catalog, but I haven't found the time.  However you have done
> some work already, so why not put forward your use cases for us to
> review.  Now we need a wiki to put this up, critique it to come to a
> final list of cases.  We can make do with email for now I suppose.
> 
> > For example, you could mean software.gnome.org to be an index for
> > selected projects, maybe only those hosted by gnome. Or it could
> > lists all projects that "hook" themselves in, just like
> > planet.gnome.org
> 
> We would like to include all apps that use the GNOME APIs.  Our
> interest in DOAP is that it lets local and remote projects maintain a
> file that we can pickup and turn into a page.  Once a project is
> registered, we would check it on a timely basis and regenerate its
> page.  There is some hope that DOAP will be used by other sites,
> making their task of communicating information about their project
> easier.
> 
> > I ask because from your description it reads a little bit like the
> > "project detail page" should be a replacement for a project
> > homepage, and the apple software site also looks like this.
> 
> We want to give a consistent presentation for users and developers. 
> It will be easier to make comparisons between software.  Returning
> users should be able to apply the rules they learned during their last
> visit. Users will not get lost moving between sites to answer their
> first questions.  Maintainers do not need to make a pretty site to
> attract the users.  Maintainers do not need to redesign their site to
> accommodate the information we want.  We do not want to keep uses from
> visiting another site, but if users comes to the GNOME software
> catalog looking for software, they should not have to go offsite to
> answer simple questions.
> 

First of all, thanks for the answer.  :-)

Coming to your question, I didn't think of use cases when I started to
be absoluty honest.

I was just uncomfortable with a few things:

1.) The GNOME software map didn't look appealing, users were complaning
    about it including me, and I was told there's somebody needed to
    write it.
2.) The TROVE categories on freshmeat are confusing.
3.) The descriptions on freshmeat are (usually) confusing.
4.) There seems to be some good quality software out there that didn't 
    make the jump from from GTK1 to GTK2, and nobody bothers.

Item 2.) and 3.) are probably due to the fact that there are apps for
developers, server administrators, and desktop users alike.

So, the use cases developed were:

a.) A beginner should be able to find applications for his needs (browse
    the web, get Email, view video, listen to audio, etc.), and install 
    them afterwards.

b.) An advanced user should be able to get an overview of existing
    replacements for his current applications if he wants to do so, 
    and install them afterwards.

c.) An developer should be able to get an overview on all existing
    applications in a certain area before he decides to reinvent the 
    wheel.

d.) A developer interested in learning or hacking for a certain language
    or binding should be able to find existing projects to study the
    code.

e.) The GNOME community should be able to learn where more work is
    needed to get a complete set of applications for the most common
    needs.

I dropped another use case after I learned some more about Open source
development, that was:

f.) A developer good in implementing X should be able to find a project
    that is looking for somebody good in implementing X.

After the first discussion about the idea on IRC, another, ehm, 'use
case' was:

g.) GTK-only projects shouldn't be discriminated today, so they might
    get GNOME apps tomorrow.

I do not know if that was helpful but I hope so. Please excuse any
errors since english is not my native language and let me know if I
didn't make myself clear.


Claus