Re[4]: Anyone has read the FSF Usenix speech ?
Hello!
On Thu, 12 Jul 2001, dim wrote:
> PC> 1. The best results are obtained by cooperation. In this case do not
> PC> compete against GNU/Linux, cooperate.
> This is in progress, but only on half:
> There is a porting Linux2Hurd which makes Hurd more usual to Linux
> users/developers. So, Hurd developers helps to Linuxoids.
> But how many Linux developers helps Hurd standing more perfect and
> stable? Most of them just don't know enough about Hurd to have a wish
> of helping.
Documentation is information, and we need to inform people about the Hurd.
We think, the Hurd is a great system and a great idea, and we think, if
people know about the Hurd, they would agree. If they only knew... I had
problems getting docs and info about the Hurd. Getting into FreeBSD was
much easier. The Hurd simply lacks of docs for users, and it's not because
the Hurd is not usable in the moment. It may not be a production system,
but it's stable enough to try it out and to get an idea about the "Hurd
Idea", e.g. the translators.
> PC> 2. Let people chose and make up their own minds.
> Yeah, of course. But I can't get my choice between what I know and
> what I didn't heard about. ;)
Exactly. People don't think about the Hurd, or the have a bad opinion
about it. That's bad, we need good PR. Has someone a good idea for PR?
> PC> 3. Make certain that the product or service you are offering is of good
> PC> quality.
> That's true for comercial products. But Hurd isn't one.
Even a document is a product which has its quality. If the Installation
Guides were not useful, noone would recomend using them.
> I suggest to welcome developers at now, not users!
We need both. A developer can be too much into the system to see problems.
"It's a known issue, but we can cope with it" is not acceptable for an
first time user or developer. And users are the "naive" people which would
say: (I don't know how to do it, but) foo would be very nice!"
> Hurd now is more a great idea of OS than a complete product that will
> be popular to all in nearest future. It's need to show this IDEA, not
> just things realized at now.
> I believe, many developers are ready to work for such IDEA. For their
> own future. For new features.
The challenge for developers is a not yet completed system with some new
concepts an a lot of space to make a name.
If they can't find info about the internals, they will turn away from the
Hurd and search projects with a better environment.
> Gabriel said:
> > Debian is (mainly) the stable distribution (OK, and the
> > community). Once the HURD is part of the stable distribution it will
> > be there (and it will be a very important advantage for Debian, having
> > multiple systems with the same administration).
>
> Yes, it will be important advantage for Debian in future.
> But RIGHT NOW Hurd needs Debian community.
Debian is part of the development environment for the Hurd, and it works.
The Hurd should work without Debian, but in fact the Hurd is useless
without the rest of GNU, and the easiest way to get it is Debian. But we
have to take care of "Distribution Wars". People could dislike the Hurd
beacause of "Debian".
People don't like barriers. If you can't find info about something, you
won't buy/like/use/develop it.
A new user/developer has to be guided to the info he need from the very
first sentence and image.
This means for users:
info where to get the Hurd and how easy it is to install it (you can tell
him anything, if you can move him to install, you've done your job) and
short overview about the benefits and features of the Hurd.
For developers: present him the benefits and features and tell him it has
a good, comfortable development environment. Do we? And a lot of projects
to join/start.
And you can't put the good hint: "search at Google for Hurd", we have to
collect or at least link information.
Do we have a complete searchable mailing archieve other than Geocrawler?
Do we have a recommendet development environment?
Do we have a guide to fix known frequent bugs/issues while porting?
Do we have some goals or basic principles for design? Is there any
definition if "Hurdishness"?
Generally: We need good documentation and high quality information. we
can't get users nor developers without these.
Patrick
----
Ingeneers motto: good fast cheap choose any two
Patrick Strasser: pstrasser at bigfoot dot de
Student of Telematik, Techn. University Graz, Austria
Reply to: