Re: [RFC] Measuring skills of a Debian Developer
On Wed, Nov 15, 2000 at 12:49:35AM +0200, Eray Ozkural wrote:
> On Tue, Nov 14, 2000 at 03:02:47AM -0600, Manoj Srivastava wrote:
> > >>"Eray" == Eray Ozkural (exa) <erayo@cs.bilkent.edu.tr> writes:
> >
[snip]
>
> It's a good idea, because it has been effective in preventing duplication
> of effort and it has let others to watch progress.
>
mmh, have you ever happened to see real open source development?
If someone has the need for a piece of software he either finds
it, or he doesn't find it cause there is nothing with the
appropriate functionality. So what is he/she doing then ?
Probably he/she is going to write some program that solves *his/her*
problem. Then he puts it somewhere on the web, so others might
use it as well. When someone finds the software eventually it doesn't
fit his personal needs, so he/she might adopt it to solve his/her
problem.
* That's what I suppose the normal way of open source development *
If someone is interested he will probably join in on the project.
When he doesn't communicate to the other people that are related to
the project, then it's his own fault. But normally people tend to
communicate what they are doing.
So for me there is no obvious reason for such a database, but for the
sake to get spam from computer firms who are seeking employees.
When the idea of the database is a bit adopted it would make sense
for me. I mean creating a database of projects that need a helping
hand at ... is a nice idea. People who want to contribute by working
on it can contact the project leader who will in turn assign them
some work.
[snip]
>
> > Eray> The idea is making things in the open. Why is that? Because, as
> > Eray> you will have noticed we are interested in the work done rather
> > Eray> than authorship!
> >
> > So, in order to help in some project, we need a databse of
> > skillz to have it in the open? Without a databse things are closed?
>
> Exactly. The list is a digital heap of information, however unarranged
> and disorganized. People shouldn't have to follow this list in order
> to know what's going on. That way "outsiders" can have a look at what's
> been done. As a verification, consider the news on debian's home page.
> Why do you think only significant items are there? The list is not
> efficient enough. There are of course a *lot* of other reasons which
> I'm sure that you can figure out by yourself if you think about it.
> That's why I wrote briefly, I thought people here could evaluate the
> implications. If not clear, just ask.
>
do you think that when setting up such an information database that
people will have time to answere mails they get directly? Did the
thought that there are maintainers who are really busy by only
packageing ever occured to you? These maintainers may simply ignore
that mails if they get too many cause they *don't* have enough time.
And I suppose that there might be a bunch of mails going directly
to the maintainer that otherwise wouldn't.
> > Eray> If people have a way to electronically examine how projects are
> > Eray> going, then they can use their volunteer time more efficiently,
> >
> > How does a databse of skillz help us see how well projects are
> > going? And how exactly is knowing how well a project is going (and
> > knowing that such an estimate is not an absolute metric) help one
> > efficiently use ones time? Should one not helkp in a project going
> > well(Debian)? Should one not help on a project in its infancy? Should
> > one not help a project that is foundering?
>
> It enables a program to match skills with skill requirements of tasks.
> The "shoulds" you mention are irrelevant.
>
mmh, that would mean: He all newbies, go get away - YOU DON'T HAVE THE
SKILLS REQUIRED -
Do you think that will ever work? What do you suppose to gain from this?
Nobody will work for it? I don't think that this will work out.
> > Eray> and more people can work on the projects they want. That's one
> > Eray> aspect. Another aspect is about automation. Now, of course you
> >
> > A databse of skillz helps us automate projects?
> >
>
> Not the projects, but the work distribution and certain communications.
>
> It should be persistent (then a database) and formally organized
> so that a program can work on it.
>
what do you think? Cause I am capable of doing automake/autoconf and
some dejagnu things, you would suppose that I am going to assist
in perhaps a dozen projects simple cause I know how to use that
tools???????????????? To be true I won't, but I will help people who
have problems with these things when they get in contact with me. For
now only personal friends know about this and the load they create
is enough for me. When I service others too then perhaps I will
only be a knowledge base but nothing more cause then I won't have the
time to do anything else but answering emails. And this is definitly
not a desireable state.
[snip]
>
> More than that. It should be a system with plausible interfaces, but
> more than a database. There must be programs that process the data.
>
I never want to get automated mail saying: "Hello this is the
autoproject manager. Please drop by at http.... there is a project
that needs you to assist in autoconf/automake design."
[snip]
> > Eray> The thing is that the system should not care who it is dealing
> > Eray> with. So basically, there won't be "someone" assigning you "a
> > Eray> task". You'd just register if you're looking for some project
> > Eray> to work in your volunteer time. As I said, like the GNU task
> > Eray> list. You shouldn't have to register or it shouldn't be a
> > Eray> strict requirement to get things done. [***]
> >
> > Oh, I see, No oe assigns you to a task, you just register. So
> > this databse is an information sink? Who uses this information, i
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>
> ha ha ha!
>
A database of going on projects might be the right idea. Where people
can seek someone with a set of skills they need.
But a database where you register your own set of skills - I simply
don't think that this will work. The open source community has never
ever been working that way and I don't think that it ever would.
Perhaps the best idea to proove this false or true is to simply test
your thoughts and create the database. Then it will show us if there
is actually a need for it when people are going to use it.
But when people aren't using it (* the cause is irrelevant *) there
is simply no need for it.
[snip]
My intention is not to discurage you, but just giving some insight
on my thoughts and feelings about this kind of database and to be
true I don't think that I will be using it.
--
kind regards,
Michael Moerz
Reply to: