Re: 10 Questions for All Candidates

[Telsa Gwynne <hobbit aloss ukuu org uk>]

On Mon, Nov 12, 2001 at 10:26:11PM -0600 or thereabouts, Dan Mueth wrote:
> I ask that each candidate please answer the following questions and reply
> to foundation-list gnome org to help the Membership know the candidate
> better and make well-informed decisions at the ballot.

Good idea.

> 1) Why are you running for Board of Directors?

People kept pestering me to. Seriously. If you think this is not the
best reason to stand, then you have my full agreement and should cast
your votes for other people. 

> 2) Do you have leadership and committee experience? If so, please explain.

Leadership? No. 
Committee? Some, in real life and on the net. I am usually the 
minority of one who gets overruled. I have a lot of experience of
that. 

> 3) How familiar are you with the day-to-day happenings of GNOME?  How much
> do you follow and participate in the main GNOME mailing lists?

Not so much with the Gnome office stuff, nor the deep hacking. I
think I'm pretty familiar with which distros ship which Gnome
versions, which is the bug-of-the-week, whom to ask about what,
and where to find things out. Depending on time and my procmailrc
I follow between a dozen to twenty gnome lists from gnome-hackers
to gnome-list to the docs and bugs ones.

Like several of the others, I use irc a lot, and that does keep
me up to date.

> 4) One of the primary tasks of the Board of Directors is to act as a
> liason between the GNOME Foundation and other organizations and companies
> to find out how the two groups can work together to their mutual benefit.
> Do you feel you would be good at understanding other people and companies
> and finding ways that GNOME can collaborate with other companies and
> organizations to benefit both groups and their users?

Watching Sun get involved has been an education. Other than that
I have no experience of big businesses. I have never been employed
by such things. I am not the one to explain to someone why Gnome
is to their corporate and strategic advantage. 

> 5) One of the responsibilities and powers of the Board of Directors is to
> identify organizational weaknesses and needs of GNOME and to create
> committees, appoint coordinators of these committess, and act as liasons
> with them.  What do you believe are the current weak points of GNOME as an
> organization, and if you were able to, how would you change the GNOME
> organization?

Mostly infrastructure. 

	* People don't know how to get CVS accounts. Still. (I wrote
a doc on this, but it doesn't seem to be in CVS).
	* FTP mirroring and getting stuff out for releases to the mirrors
before announcing it. Gnome 1.4 release seemed to have learned very 
little from Gnome 1.2's. We *have* to fix this in time for 2.0. 
	* Points of contact other than the board are just not
known. ("Oh no, problem with something on widget! Who has access to
fix it?" "I have this idea for publicity, whom should I talk to?")
	* There seems great confusion outside (and sometimes within)
GNOME about the differences between the Foundation and the Advisory
boards. 
	* We have no-one who is willing to say "This release is
*not* going out until the developer docs are done". Every time
we discuss "And it must have good docs", and every time there is
no way of ensuring it. And so of course it doesn't happen.
	* Too much is word of mouth and not written down. That's 
everything from "who runs which boxes/servics" to "how to run that 
box/service". When someone vanishes and they did that job, they
typically leave nothing behind to explain stuff. (Does gnome-de have
a new listowner yet, btw?) And as for contingency plans: well, if 
we have a "what to do if we suspect cvs.gnome was trojaned" policy, 
for example, it's a very well-kept secret. (Surely we should have
this, or know that someone has thought about it?)

The other big one is that we have a fair number of hackers who
see any attempt to create a structure or a process for *anything*
as a restriction on their freedom, and who simply won't bother
using them. Whether they're right or wrong is immaterial; the fact
is that they won't want to follow any such procedures. Particularly,
I suspect, if it was set down by a committee :) So either the other 
60(90?)% all have to not form such things, or they have to accept 
that there will always be people outside them. The only thing I can 
think of to change this is to get a small working example of such a 
thing and say "See, it can work" and hope that people will go "Ooh. 
Perhaps" and give it a go. I don't think more committees would help 
this one. I think getting something working as an example would be
far better.

Finally, we don't seem to be able to make up our minds whether
GNOME is primarily a developer environment ("platform") or primarily
a end-user environment, or a 50:50 mix. Everyone has a different
opinion on this. 

> 6) The board meets for one hour every two weeks to discuss a handful of
> issues.  Thus, it is very important that the board can very quickly and
> concisely discuss each topic and come to concensus on each item for
> discussion. Are you good at working with others, who sometimes have very
> differing opinions than you do, to reach concensus and agree on actions?

Yeah. I just get overruled a lot. And then six months later people
start saying what I said then. (I have been burbling about X,
remote terminals, and so on for a while now. Now I see the "terminal
servers" thread and I'm thinking "Didn't I say this ages ago?" :)) 

> 7) Often Directors have to draft policies, form committees, find
> weaknesses or approaching problems of GNOME and work on solutions, and act
> as liason with various groups (both within and outside GNOME) and
> companies.  Please name three or more areas which you feel are important
> for the Board to address over the next year and which you would enjoy
> contributing some of your time to help get things started and possibly act
> as a liason between the Board and any other committees, groups, or
> companies if relevant.

* More bugs and testing stuff, more -- well, not formally-organised, per se. 
Just with more information for people and more "here's stuff to try" and
more ways to find out what current results are. In particular, more
testing on non-Linux-on-i386 boxes. Either we're portable and run on
UNIX and UNIX-likes, or we're not. 

* Developer docs. Especially with the change to 2.0 there's a *lot* 
for new hackers to take in. I can be of little help with this, because
I don't know what constitutes good docs for developers. 

* This bloody memory usage issue. Numbers bandied around as same minimum
defaults for using GNOME range from 64Mb RAM (g-h, last March) to 
256Mb (spotted on linuxchix mailing list very recently, from one user
to another). With that kind of range, I really don't see why so many
people are so keen on GNOME on handhelds (sorry, Jim!) unless we're 
all anticipating some really monster handhelds arriving next year. I
really really really want to see either an official acknowledgement 
that GNOME *does* require 128Mb of RAM (I still run it in less, but
I don't need a file manager and like text-based apps), or a lot of
CVS commits which are doing something about it. There *are* lots of
machines out there with 64Mb or less. The "memory is cheap" issue is
*not* an assertion you can make to all people: it's really only useful
to people with credit cards or a very nearby shop, who are prepared
to take their box apart and invalidate the warranty (I can just see
my parents' face at the suggestion they put it in themselves), and if
you live miles from said shop in a country which is on many internet
vendors' blacklists due to "too much fraud from there" (and this is 
not just south-east Asia: I saw a friend in .ch get caught by this),
"memory is cheap" is no help at all. And adding RAM to laptops is a
whole different game. How many people are really happy about taking 
their laptops apart to add more memory? Mess that up and you have 
acquired £1000's worth of doorstop. 

This issue really does bother me a lot (as you can tell from the length).
I will cheerfully report results from running GNOME on my 32Mb box
(it actually has more these days, but "linux mem=32" solves that,
and the chip is not even a pentium), but I would like to see a few
more people run on a 64Mb box for a week. Developers can do this, too:
just do the compiles on a different faster box. 

* Finally, I still consider that GNOME lacks some of the networking 
features that X is so good at. Before I got a keyboard/monitor/mouse
switcher, I had two monitors. One was a lot nicer than the other. Using
GNOME on that, I'd have a bigger resolution, more fancies in the panel,
and could get away with some smaller fonts. On the other monitor, normally
attached to other machine, my GNOME settings were much plainer. So far,
so good. But when I ran GNOME remotely, I had no way to tell it "I am
on crap monitor, use different settings". The assumption was: this account
on this machine has this monitor. There is one .gnome/ directory and no way 
to have separate "connected to crap monitor" or "running over slow connection 
to other machine" sort of settings. I don't understand X and I don't
understand networks, but I like messing around like this, and I like
seeing what breaks when you try this. (Another one: gnome-terminal ->
ssh to other box -> start applet. Watch it crash, because there's no
panel running on that box. Why should it crash? Why can't it connect
to the right panel? I have the orbitrc settings which should make this
happen. This should Just Work.) 

I'd love to see this kind of thing sorted. Telling Windows and Mac users
about "we have a GUI too" gets a complete lack of interest. "I have one 
anyway. And?" Showing them the "Okay, this is box A. And we have the
stable version of foo-app on it. That box over there has the new beta
version. But we don't have to walk over and login there to see it" and
starting beta version of foo-app running on box A is far far more 
impressive to them. The networking aspect *is* something outside their 
experience and which elicits much more impressed reactions. But it 
has to work right. 

I can't offer much in the way of helping here other than to keep
testing (when computers are back talking to each other).

None of those really require committees, btw. Mailing lists, perhaps,
but it's more a matter of getting people to think of it than having
a Committee For Running On Small Machines and so on.


> 8) Do you consider yourself diplomatic?  Would you make a good
> representative for the GNOME Foundation to the Membership, media, public,
> and organizations and corporations the GNOME Foundation works with?

Not really diplomatic. I try hard instead, to make up for it :)
I only rarely have sent emails that the next day I wish I could call
back and destroy. I use my "postpone" key and check the next day
when I'm not sure. (And have had occasion to be thankful for the
habit!) I am not good at dealing with the media: I have had very 
little experience of it and I always cringe at the reults. I don't 
have a smart appearance for talking to corporations. I loathe 
speaking in public to people I don't know. I forget everything :) 
This and the "point of contact" aspect of the foundation are the 
things I'd be worst at. I have the nasty feeling they're about 90% 
of the job, too. 

That said, I'm alright at representing GNOME at shows: I may
not know the technical answers, but I know how to clear the screen
down to just one thing for the person to concentrate on and to let
them have the mouse :)

> 9) Will you represent the interests of GNOME and the GNOME Foundation over
> all other personal or corporate interests you may represent?

To an extent. If the interests of GNOME involved 72 hours of staying
awake, as seems to happen before far too many releases, then I will
go to sleep without a qualm, and delete all complaints unread. 

I don't represent any corporate interests. 

> 10) Will you be willing and have the available time to take on and
> complete various tasks that the Board needs accomplished?

Willing, yes. Time, probably (although it's possible I'll have a month
without net access next year). Taking stuff on, I'm great at. Completing
them: um.

The only thing to add is that all the things I have mentioned I
can do, I can do without being on the board. If the point-of-contact
and talking-to-big-companies things are the bulk of the work, people
should really not be voting for me. 

Telsa