[gentoo-project] Re: Plan, then communicate (no-list)

[gentoo-project] Re: Plan, then communicate (no-list)

[Alec Warner]

On 1/20/08, John Lawles <jl.050877@gmail.com> wrote:
> Alec,
>
> > ...a thread about communication problems...

I've moved this to project.  Hopefully it will not be as flamey as you
expect; but if it is I apologize in advance.

>
> Not exactly.  Just for clarification, I was suggesting:
>
>   (a). Decide what the big-picture problems are, if any.
>   (b). If any, develop a plan
>   (c). Communicate it to users
>
>      From what I, as a mere user, can tell, the contentious issue
> is not so much (c) as it is (a).  The issue seems to be that some
> developers regard (a) as a personal insult rather than a request
> for better organization and coordination among the developers.

(a) is a bit vague.  Am I supposed to address issues that Gentoo
developers have?  Gentoo users have?  The Gentoo community?

The community currently has no good means to rank problems in the view
of users other than the forums; which currently have their own issues.

User Coverage: Not everyone has a forums account.  Not everyone uses
their forums account.  We have no idea how many users we have
(ancidotal numbers suggest ~200000; see
http://dev.gentoo.org/~antarus/bouncer-stats.txt).  It is difficult to
know what percentage of users responded and thus becomes difficult to
judge how important something is (we have only the respondants data to
use).

Arguably you could say that anyone who didn't vote doesn't care; but
you have to factor in people who didn't learn of the vote during the
voting period.

User Education: This is that whole Cathedral thing.  Below I'll talk
about Daniel's goal of maximizing developer impact and this plays a
big part.  Many developers don't talk to users because its draining
and they want to work on projects that they have a high impact on.  I
could sit in #gentoo and field questions all day (I've done it before)
but I have things I could spend my time on that are more worthwhile to
the project (and we are lucky enough to have a crack team of awesome
contributors that staff that channel).

Talking to users is exhausting when the user really has a
misconception about a given problem, program, or feature.  It takes
time to educate people why something works the day it does and
documentation only helps so much.  Give bad service and the user is
off to the forums to complain about how he was mistreated by that
Antarus guy on #gentoo-portage and how much Gentoo sucks.

That being said; talking to users who know what they are doing (doubly
so when they know more than me) is a delight and I'm generally happy
to take the time to respond.  If there was some way to aggregate user
complaints into concrete problem sets I'm all ears.

User Validation: Most systems that users can use to respond on a large
scale don't have a means to validate whether they use your software or
not.  This is more of a trend game; needing to look at the aftermath
of any given aggregate data and look for areas where people may have
given feedback that we should throw out (like automated voting).  I
don't think this problem is necessarily solvable or that big a deal
but it is something to consider/

>
>      Drobbins has addressed (a) and (b) and (c).  My suggestion is
> that the-powers-that-be at Gentoo address them also, starting with
> (a) and produce, hopefully, a far better plan.

Drobbins has addressed very little in my eyes.  Sure we have
communication problems (pr was basically dead until this incident) and
we have leadership issues.  His plan is not well specified:

1.  Open the lines of communication.  How?  We have an influx of
people interested in helping out with GMN and PR which is good.  We
have a new PR lead.  I'm busy working on news items and learning XSL
to try and change the webpages a bit.  The foundation obviously failed
at providing data in the past and I hope to change that.  We have
tried to be as transparent as possible with posts to -nfp, posts to
-project, news items on the website, etc.

Are there other places where communication is lacking?  What kind of
information are the users looking for?

2.  Maximize developer impact per unit time.  How?
I'm uncertain where Daniel thinks developers are wasting time stuck in
process.  We could kill the 30 day stability guideline in an attempt
to get packages into stable quicker; but I'm unsure what that would do
to overall quality (which a subset of the userbase seems to think is
subpar at this time).  I'm also unsure how much time it costs someone
to become a fully-fledged developer; however I think we have a decent
set of options for individuals who wish to contribute without being a
full-time developer (sunrise, proxy-maint, arch tester, overlays).

Are there specific processes we have that you think hold developers back?

3.  The project has hit several scalability issues; but Daniel does
not specify what these issues are or his ideas on solving them.  It's
difficult to specify management roles in a volunteer organization but
obviously some are needed.  Most projects lack a strong project lead
and finding qualified motivated people who want to spend some time
dealing with people instead of technical problems is a difficult
process that we could possibly improve.

What exactly are these scalability limits?  It is difficult to say
much about this item since it was vague.

>
>      This is sent off-list because, as you point out, it does not
> belong on gentoo-nfp.

Feel free to punch me if you dislike me forwarding it to project; I
think it is a worthwhile conversation to have and I wish to have it in
public.

>
> Regards,
>
> John
>

Thanks,

-Alec
antarus@gentoo.org
-- 
gentoo-project@lists.gentoo.org mailing list

Re: [gentoo-project] Re: Plan, then communicate (no-list)

[Graham Murray]

"Alec Warner" <antarus@gentoo.org> writes:

> The community currently has no good means to rank problems in the view
> of users other than the forums; which currently have their own issues.

What about the voting facility in Bugzilla? I know that it is very
rare for anyone to vote for a bug, but the facility is there.
-- 
gentoo-project@lists.gentoo.org mailing list

Re: [gentoo-project] Re: Plan, then communicate (no-list)

[Marius Mauch]

On Mon, 21 Jan 2008 08:17:41 +0000
Graham Murray <graham@gmurray.org.uk> wrote:

> "Alec Warner" <antarus@gentoo.org> writes:
> 
> > The community currently has no good means to rank problems in the view
> > of users other than the forums; which currently have their own issues.
> 
> What about the voting facility in Bugzilla? I know that it is very
> rare for anyone to vote for a bug, but the facility is there.

Has the same issues as forum polls, also see 
http://forums.gentoo.org/viewtopic-p-4733957.html#4733957

Marius
-- 
gentoo-project@lists.gentoo.org mailing list

Re: [gentoo-project] Re: Plan, then communicate (no-list)

[Robin H. Johnson]

[-- Attachment #1: Type: text/plain, Size: 604 bytes --]

On Mon, Jan 21, 2008 at 08:17:41AM +0000, Graham Murray wrote:
> > The community currently has no good means to rank problems in the view
> > of users other than the forums; which currently have their own issues.
> What about the voting facility in Bugzilla? I know that it is very
> rare for anyone to vote for a bug, but the facility is there.
Bugzilla voting only allows voting FOR an issue, and provides no means
of voting AGAINST.

-- 
Robin Hugh Johnson
Gentoo Linux Developer & Infra Guy
E-Mail     : robbat2@gentoo.org
GnuPG FP   : 11AC BA4F 4778 E3F6 E4ED  F38E B27B 944E 3488 4E85

[-- Attachment #2: Type: application/pgp-signature, Size: 329 bytes --]

[gentoo-project] Re: Plan, then communicate (no-list)

[Steve Long]

Alec Warner wrote:
> The community currently has no good means to rank problems in the view
> of users other than the forums; which currently have their own issues.
> 
> User Coverage: Not everyone has a forums account.  Not everyone uses
> their forums account.  We have no idea how many users we have
> (ancidotal numbers suggest ~200000; see
> http://dev.gentoo.org/~antarus/bouncer-stats.txt).  It is difficult to
> know what percentage of users responded and thus becomes difficult to
> judge how important something is (we have only the respondants data to
> use).
> 
> Arguably you could say that anyone who didn't vote doesn't care; but
> you have to factor in people who didn't learn of the vote during the
> voting period.
>
I could and I would; forum users see their accounts similarly to how gentoo
devs view their @g.org badges. Spammers are soon dealt with, so that anyone
who has posted more than 50 posts (not counting OTW) and been a member for
more than 3 months is not a bot. Let's say there's 100,000 active users. In
a poll say 5% of votes are fraudulent. It can be factored into the
calculation of significance. And when 90% say they don't like the way they
get treated by Gentoo devs, there is a real issue. Call it communicaton,
call it what you want, it's a real and valid concern.

And IMO it holds Gentoo back.

> User Education: This is that whole Cathedral thing.  Below I'll talk
> about Daniel's goal of maximizing developer impact and this plays a
> big part.  Many developers don't talk to users because its draining
> and they want to work on projects that they have a high impact on.  I
> could sit in #gentoo and field questions all day (I've done it before)
> but I have things I could spend my time on that are more worthwhile to
> the project (and we are lucky enough to have a crack team of awesome
> contributors that staff that channel).
>
There's also developer education. A junior dev (aka code-monkey) who comes
in at the start of their career is not expected to show the same level of
maturity as a 40 year-old.

> Talking to users is exhausting when the user really has a
> misconception about a given problem, program, or feature.
Yeah it's called requirements analysis (whichever model you use.) That's why
it's such a source of problems.

> It takes 
> time to educate people why something works the day it does and
> documentation only helps so much.  Give bad service and the user is
> off to the forums to complain about how he was mistreated by that
> Antarus guy on #gentoo-portage and how much Gentoo sucks.
>
So allow more advanced users to help the less knowledgeable. All your doing
is formalising what happens on irc.

> That being said; talking to users who know what they are doing (doubly
> so when they know more than me) is a delight and I'm generally happy
> to take the time to respond.  If there was some way to aggregate user
> complaints into concrete problem sets I'm all ears.
>
Votes. Require 75% majority from users if you want.

> User Validation: Most systems that users can use to respond on a large
> scale don't have a means to validate whether they use your software or
> not.  This is more of a trend game; needing to look at the aftermath
> of any given aggregate data and look for areas where people may have
> given feedback that we should throw out (like automated voting).  I
> don't think this problem is necessarily solvable or that big a deal
> but it is something to consider/
>
Yeah see above about statistical significance. We're not looking for a 5%
end of the normal here (which is what could perhaps be used to identify a
minority.)


-- 
gentoo-project@lists.gentoo.org mailing list

Re: [gentoo-project] Re: Plan, then communicate (no-list)

[Donnie Berkholz]

On 18:19 Mon 21 Jan     , Steve Long wrote:
> I could and I would; forum users see their accounts similarly to how gentoo
> devs view their @g.org badges. Spammers are soon dealt with, so that anyone
> who has posted more than 50 posts (not counting OTW) and been a member for
> more than 3 months is not a bot. Let's say there's 100,000 active users. In
> a poll say 5% of votes are fraudulent. It can be factored into the
> calculation of significance. And when 90% say they don't like the way they
> get treated by Gentoo devs, there is a real issue. Call it communicaton,
> call it what you want, it's a real and valid concern.
> 
> And IMO it holds Gentoo back.

I think it could hold Gentoo's userbase from growing at the same rate, 
and I think it has little effect on Gentoo's progress in the technical 
sense (development, etc.).

Thanks,
Donnie
-- 
gentoo-project@lists.gentoo.org mailing list

Re: [gentoo-project] Re: Plan, then communicate (no-list)

[Richard Freeman]

Steve Long wrote:
> Alec Warner wrote:
>> Talking to users is exhausting when the user really has a
>> misconception about a given problem, program, or feature.
> Yeah it's called requirements analysis (whichever model you use.) That's why
> it's such a source of problems.
> 

True.  I think one of the underlying issues in this mess though is "who 
is the customer?"

If I'm assigned to a project at work I typically have a customer that 
I'm aiming to make happy.  If the customer ends up happy, I get 
rewarded, and if the customer is unhappy I get punished.  If somebody 
other than the customer gives me some requirements they're usually only 
taken into consideration to the degree that they can be handled without 
significant additional cost, or interference with the customer's needs.

It is the same in the open source world.  The customer is whoever 
rewards or punishes you.  Most often the customer is a peer of some sort 
- you take care of their php headaches and they take care of your 
baselayout headaches or whatever.  The customer is often yourself - you 
want to use some package not in portage - just add it.  Rarely is the 
customer an end user - the users aren't really in a position to reward 
you except in the most general sense - and if you take care of everybody 
else chances are you'll make a bunch of users happy anyway.

I'm not saying this is how it ought to be - but it is how it tends to be 
in the open source world.  How does IBM get its fancy hardware supported 
in linux?  They do a few things - one is that they write the drivers 
themselves, and two is that they give away all kinds of fancy code away 
so that the kernel maintainers are inclined to commit those drivers and 
take them forward.  They don't just stand up and complain that there 
aren't enough kernel devs buying mainframes and writing code for them.

You can always ask nicely - but one person asking nicely doesn't 
obligate anybody else to respond.  If what they're asking for entails a 
lot of work or a long-term maintenance commitment, they might need to do 
it themselves or find somebody else willing to do it.  However, asking 
nicely will generally get you further than asking rudely.

More than anything else gentoo needs people to step up and DO things. 
If your favorite package is stale volunteer to maintain it.  Or at least 
volunteer to proxy-maintain it - convincing a developer you're willing 
to take the responsibility seriously so that they can do commits for 
you.  What you can't just do is throw an ebuild into bugzilla and 
complain when it doesn't get committed - a dev would be reluctant to 
just commit an ebuild without personally making sure it works well - 
since they get to deal with all the complaints when 5,000 servers stop 
working.

I'm not too worried - people are stepping up and the current situation 
will pass.  Considering that even welfare recipients find something 
about their free money to complain about in the US I'm not surprised 
that users who are offered a free operating system can find time to make 
demands of the people who gave it to them.  That's just human nature.  I 
think that most users will just take all of this in stride.  I'd love to 
see more done to get users involved and make them feel like they have a 
voice, but users do need to understand that while they have a right to 
speak up nobody has an obligation to take orders...
-- 
gentoo-project@lists.gentoo.org mailing list

Re: [gentoo-project] Re: Plan, then communicate (no-list)

[Wulf C. Krueger]

[-- Attachment #1: Type: text/plain, Size: 967 bytes --]

On Monday, 21. January 2008 19:19:12 Steve Long wrote:
> And when 90% say they don't like the way they get treated by Gentoo
> devs, there is a real issue. 

"The best argument against democracy is a five minute conversation with 
the average voter." - Winston Churchill

s/voter/user/

> > That being said; talking to users who know what they are doing
> > (doubly so when they know more than me) is a delight and I'm
> > generally happy to take the time to respond.  If there was some way
> > to aggregate user complaints into concrete problem sets I'm all ears.
> Votes. Require 75% majority from users if you want.

No, most of those users do exactly that: They *use* what *we*, the Gentoo 
developers, provide. They can become developers easily enough and then 
they may vote all they want.

And they can always vote with their feet - on their way out. Gentoo is 
about choice and one such choice can be to leave.

-- 
Best regards, Wulf

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

[gentoo-project] Re: Plan, then communicate

[John Lawles]

Alec,

     I love Gentoo and I am very grateful for all the very
impressive that Gentoo developers have done.  As a mere end-user, I
make no claim to have expertise or answers, only questions that
other end-users likely have.

     As background, Gentoo users, like me, are happy spending a lot
of time configuring/upgrading if the end-result is a system
configured exactly as we want it.  Thus, for example, unlike users
of other distributions, we are happy configuring/compiling our own
kernels.

     The central issue that Gentoo end-users have, as I gather it,
is that running Gentoo takes a lot of time that does not seem to
fall into the above category.  For instance, every portage tree
seems to have various issues like blockers and faulty or circular
dependencies.  The work-arounds get hashed out over time in the
forums.  It is productive for the first, say, ten people to find
the problem and test a work-around.  It is a waste to have a
hundred or a thousand people repeat that process starting from
scratch.

     A possible solution is Daniel Robbins' proposal to have
separate "developer-facing" and "user-facing" portage trees.
Developers working on the developer-facing side would be freed to
experiment, hopefully improving their productivity.  Only the
tested and successful ideas would be ported to the user-facing
tree, improving end-user satisfaction.

     Is that a good idea?  Do developers here have still better
ideas?  I don't know but Daniel has (a) recognized a key problem as
users see it, (b) proposed a solution, and (c) communicated a plan
to users.  That, rightly or wrongly, gives end-users confidence in
him.

Regards,

John

P.S. I love Gentoo and I am very grateful for all the very
impressive that Gentoo developers have done.
-- 
gentoo-project@lists.gentoo.org mailing list

Re: [gentoo-project] Re: Plan, then communicate (no-list)

[George Prowse]

Alec Warner wrote:
> The community currently has no good means to rank problems in the view
> of users other than the forums; which currently have their own issues.

Have you thought about the use of some other forum, a web page for
example where you specifically as for user interaction, one with
questions, answer boxes and tick boxes. The page and subject could be
announced on all the current media (forums, lists, irc) and answers
could be automatically converted into percentages for the thought of the
community as a whole.
> 
> User Coverage: Not everyone has a forums account.  Not everyone uses
> their forums account.  We have no idea how many users we have
> (ancidotal numbers suggest ~200000; see
> http://dev.gentoo.org/~antarus/bouncer-stats.txt).  It is difficult to
> know what percentage of users responded and thus becomes difficult to
> judge how important something is (we have only the respondants data to
> use).
> 
> Arguably you could say that anyone who didn't vote doesn't care; but
> you have to factor in people who didn't learn of the vote during the
> voting period.
> 
> User Education: This is that whole Cathedral thing.  Below I'll talk
> about Daniel's goal of maximizing developer impact and this plays a
> big part.  Many developers don't talk to users because its draining
> and they want to work on projects that they have a high impact on.  I
> could sit in #gentoo and field questions all day (I've done it before)
> but I have things I could spend my time on that are more worthwhile to
> the project (and we are lucky enough to have a crack team of awesome
> contributors that staff that channel).

Maybe communication and people skills should be part of the job
description (language permitting of course)? If not then there should be
developers whose job it is to relay the comments of those who do not or
cannot want to communicate.
> 
> Talking to users is exhausting when the user really has a
> misconception about a given problem, program, or feature.  It takes
> time to educate people why something works the day it does and
> documentation only helps so much.  Give bad service and the user is
> off to the forums to complain about how he was mistreated by that
> Antarus guy on #gentoo-portage and how much Gentoo sucks.

Here is where the forums could help out, if a question or rfc was put
into a thread then any questions about the question could be answered by
those users in the know.
> 
> User Validation: Most systems that users can use to respond on a large
> scale don't have a means to validate whether they use your software or
> not.  This is more of a trend game; needing to look at the aftermath
> of any given aggregate data and look for areas where people may have
> given feedback that we should throw out (like automated voting).  I
> don't think this problem is necessarily solvable or that big a deal
> but it is something to consider/

Ask for a uname -a (not useful for me at the moment because I am stuck
on windows) or something that might give you a more precise answer like
requesting users set up an account (which could also give you more
useful information like platform, stable/unstable and demographics.
> 
>>      Drobbins has addressed (a) and (b) and (c).  My suggestion is
>> that the-powers-that-be at Gentoo address them also, starting with
>> (a) and produce, hopefully, a far better plan.
> 
> Drobbins has addressed very little in my eyes.  Sure we have
> communication problems (pr was basically dead until this incident) and
> we have leadership issues.  His plan is not well specified:
> 
> 1.  Open the lines of communication.  How?  We have an influx of
> people interested in helping out with GMN and PR which is good.  We
> have a new PR lead.  I'm busy working on news items and learning XSL
> to try and change the webpages a bit.  The foundation obviously failed
> at providing data in the past and I hope to change that.  We have
> tried to be as transparent as possible with posts to -nfp, posts to
> -project, news items on the website, etc.
> 
> Are there other places where communication is lacking?  What kind of
> information are the users looking for?

Transparency is great and some of the comments from the short-lived
userreps project was that most things seemed to be done in private, the
first any of the users knew about anything was an announcement.

Attitude is another. For instance, I made a comment on -dev (I think)
that without users Gentoo would be nothing and the response I got was
basically "p**s off, we develop because we want to".
> 
> 2.  Maximize developer impact per unit time.  How?
> I'm uncertain where Daniel thinks developers are wasting time stuck in
> process.  We could kill the 30 day stability guideline in an attempt
> to get packages into stable quicker; but I'm unsure what that would do
> to overall quality (which a subset of the userbase seems to think is
> subpar at this time).  I'm also unsure how much time it costs someone
> to become a fully-fledged developer; however I think we have a decent
> set of options for individuals who wish to contribute without being a
> full-time developer (sunrise, proxy-maint, arch tester, overlays).

Killing the 30 guideline would be a big mistake IMO. Quality is far more
important.

As for developers, Gentoo continually says it is under-staffed and one
of the things I suggested when userrel started up was a developer
fast-track, where current and former developers of other projects could
demonstrate they had the skills and could be moved into their areas of
expertise far quicker.
> 
> Are there specific processes we have that you think hold developers back?

Yes, the continual arguments over GLEPs. PMS is a good example - people
trying to add their political piece of the puzzle in and everyone
disagreeing.

I honestly believe that when you have something as serious as a PMS then
the constant bickering and the camps at gentoo have rendered it too
incompetent to get it done itself, you need outside assistance who can
listen to the arguments and make a decision. It would be great if an
agreement could be struck up with another distro where assistance was
given both ways. A request for discussion about the viability would only
take an email to someone like Mark Shuttleworth and what do you have to
lose? The worst that could happen is that someone says "we dont have the
time, sorry"

George

-- 
gentoo-project@lists.gentoo.org mailing list

Re: [gentoo-project] Re: Plan, then communicate

[Richard Freeman]

John Lawles wrote:
> 
>      A possible solution is Daniel Robbins' proposal to have
> separate "developer-facing" and "user-facing" portage trees.
> Developers working on the developer-facing side would be freed to
> experiment, hopefully improving their productivity.  Only the
> tested and successful ideas would be ported to the user-facing
> tree, improving end-user satisfaction.
> 

In theory we already have things like that.  We have overlays, masking, 
etc - all designed to keep messes out of the sight of users.  The main 
problem is that not everybody installs everything so it is hard to get a 
full test.

Arch teams are supposed to test all packages on a stable-keyworded 
system before stabilizing them.  Repoman also has automated checks to 
prevent in theory the worst offenses from happening.  Sure, things do 
get through, but it usually isn't all that bad.  When I get blocks 
typically I just need to unmerge one package and then do my emerge world 
to fix things - rarely do I get major circular dependency issues.

Out of curiosity, are you running stable?  If you have ~x86/~amd64 
accepted in your keywords then you are running what would otherwise be 
the developer-facing portage tree, and you'd tend to get breakage.

What I recommend is running the stable tree in general, and then 
accepting only individual packages in ~arch (usually tied to a version 
number) for the cases where you want cutting-edge.  You'll get a lot 
farther that way - you can still have the latest and greatest amarok or 
whatever, but not have to take an experimental glibc or kdelibs or 
anything like that.  Most users don't need EVERYTHING to be 
bleeding-edge, and you can benefit from just targeting this behavior for 
the packages you need this from the most.

Getting back on-topic - how will a separate portage tree for devs be any 
different from what we already have (masking and keywording), and how 
will we prevent users getting hit with even worse bugs because nobody is 
bothering to test anything in the environment that end-users actually use?
-- 
gentoo-project@lists.gentoo.org mailing list

Re: [gentoo-project] Re: Plan, then communicate

[John Lawles]

On Mon, Jan 21, 2008 at 09:42:21PM -0500, Richard Freeman wrote:
> how will a separate portage tree for devs be any different from
> what we already have (masking and keywording),

     Those don't help when it is the 'stable' part of the tree that
breaks.  The most exasperating problems that I can remember
involved basic libraries that were changed (upgraded or removed) in
the "stable" tree and that change, in turn, exposed broken
dependencies or some such elsewhere.  IIRC, libexpat was an example
of that and you can see the level of end-user frustration in a
thread such as:
http://forums.gentoo.org/viewtopic-t-575655-highlight-libexpat.html

> and how will we prevent users getting hit with even worse bugs
> because nobody is bothering to test anything in the environment
> that end-users actually use?

     The concept is that there would be a separate team maintaining
the stable tree.  A precedent for this would be the Linux kernel's
two-track development.

     I use my own mini-two-tree system.  I emerge --sync and then
wait a week.  During that week, many users will encounter bugs and
work-arounds start appearing in the forums.  After that week, I
upgrade in relative confidence that, while my system may break,
there is a good chance that there will be a known procedure for
fixing it.

     As a mere end-user, I am not recommending any particular
solution.  I am suggesting that more effort be devoted to
(a) recognizing user frustrations, (b) proposing good solutions,
and (c) letting us know that you see the problems and that
solutions are being worked on.

Regards,

John

P.S.
> What I recommend is running the stable tree in general, and then accepting 
> only individual packages in ~arch

Yes, that is what I do.  I would run completely stable but many
system essentials (such as, at my last upgrade, the X11 fonts) are
masked.

P.P.S. Again, my thanks to all developers for their monumental
efforts in making the distribution that I want to use.

-- 
gentoo-project@lists.gentoo.org mailing list

Re: [gentoo-project] Re: Plan, then communicate

[Richard Freeman]

John Lawles wrote:
> 
>      The concept is that there would be a separate team maintaining
> the stable tree.  A precedent for this would be the Linux kernel's
> two-track development.
> 

Actually, in theory that is what already happens with stable keywording 
- or is supposed to happen.  It is a different group of developers who 
maintain the stable tree (well, there is overlap, but some people wear 
multiple hats).

Expat was a real mess - mainly because they break ABI and it doesn't use 
slotting.  I'm not sure why slotting couldn't have been used in this 
case (others might be able to comment on this).

What we need is a better mechanism of warning users that they're about 
to do something that could cause them major headaches.  ELOG is useless 
when you find out after the fact.

revdep-rebuild working better would also be nice.  If it won't work the 
ebuild should abort prior to install with a link to a document 
indicating what will need to be done, and then users can read and 
understand it before they break things.

>      As a mere end-user, I am not recommending any particular
> solution.  I am suggesting that more effort be devoted to
> (a) recognizing user frustrations, (b) proposing good solutions,
> and (c) letting us know that you see the problems and that
> solutions are being worked on.
> 

I agree that this is something that does need better attention in 
general.  I find it a bit frustrating to have to carefully check my 
emerge -auD lists to ensure that the whole system won't blow up after 
the install - and prior to last Aug I'd never even heard of expat...
-- 
gentoo-project@lists.gentoo.org mailing list

Re: [gentoo-project] Re: Plan, then communicate

[Petteri Räty]

[-- Attachment #1: Type: text/plain, Size: 642 bytes --]

Richard Freeman kirjoitti:
> 
> Expat was a real mess - mainly because they break ABI and it doesn't use 
> slotting.  I'm not sure why slotting couldn't have been used in this 
> case (others might be able to comment on this).
> 

Expat saw extensive discussions in gentoo-dev so just search the archives.

> What we need is a better mechanism of warning users that they're about 
> to do something that could cause them major headaches.  ELOG is useless 
> when you find out after the fact.
> 

emerge --news
Just needs someone to champion it into a Portage release I think but not 
an expert there.

Regards,
Petteri


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 252 bytes --]

[gentoo-project] Re: Plan, then communicate

[John Lawles]

Alec,

On Sun, Jan 20, 2008 at 07:32:55PM -0800, Alec Warner wrote:
> I could sit in #gentoo and field questions all day (I've done it
> before) but I have things I could spend my time on that are more
> worthwhile to the project ....

To find the big-payoff items, I suggest first doing triage on the
threads:

(1) Ignore, at least temporarily, noob-does-something-stupid
    threads.

(2) Ignore threads about unstable packages because unstable
    packages are supposed to have problems.

What is left are (3) normal users having problems with stable
parts of Gentoo.  Then ask: what architectural changes to Gentoo
would have eliminated that problem, or better yet, that general
type of problem?

     Note that you can do this without participating in any
thread.  All that is needed is to scan the thread and classify it.

     For instance, in my observation, there are many problems that
are, at base, caused by portage leaking files and those leaked
files then cause something else to break, leading to lots of
confusion and misdirection.  That suggests that a little effort at
solving leakage might have a big pay-off in user satisfaction.

     For all I know, as I am just an uninformed end-user, leakage
has already been solved.  I mention it only as an example of using
user-frustrations to find the larger problems that are worthy of
developer time.

Regards,

John

-- 
gentoo-project@lists.gentoo.org mailing list

Re: [gentoo-project] Re: Plan, then communicate (no-list)

[Roy Bamford]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 2008.01.21 18:19, Steve Long wrote:
[snip]
> And when 90% say they don't like the way they
> get treated by Gentoo devs, there is a real issue. Call it
> communicaton, call it what you want, it's a real and valid concern.
> 
> And IMO it holds Gentoo back.
> 
[snip]

Steve,

For the furtherance of this discussion and simplification of the 
problem space, lets assume that we have already identified that Gentoo 
is split into two mutually exclusive groups which I shall call "them" 
and "us". I do not admit that it is really like this but it serves as a 
reference frame for further discussion.

How do you go about turning "them" and "us" into "we" ?
Discuss.

One further constraint to the discussion, you must not use the word 
"they", as in "they should ..."

The idea is not to point fingers but plan how to move from the 
situation described above to a different (better?) situation.

(No points for spotting the 6sigma game)

- -- 
Regards,

Roy Bamford
(NeddySeagoon) a member of
gentoo-ops
forum-mods
treecleaners

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.7 (GNU/Linux)

iD8DBQFHll+jTE4/y7nJvasRAqwnAJ9B+/ODtYPMaRpux809y4tBrIilGgCfa33E
6VACx4CSZHjSr4ML24pUDNQ=
=/bT2
-----END PGP SIGNATURE-----

--
gentoo-project@lists.gentoo.org mailing list

Re: [gentoo-project] Re: Plan, then communicate

[Roy Bamford]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 2008.01.22 21:18, John Lawles wrote:
> Alec,
> 
> On Sun, Jan 20, 2008 at 07:32:55PM -0800, Alec Warner wrote:
> > I could sit in #gentoo and field questions all day (I've done it
> > before) but I have things I could spend my time on that are more
> > worthwhile to the project ....
> 
> To find the big-payoff items, I suggest first doing triage on the
> threads:
> 
> (1) Ignore, at least temporarily, noob-does-something-stupid
>     threads.
> 
> (2) Ignore threads about unstable packages because unstable
>     packages are supposed to have problems.
> 
> What is left are (3) normal users having problems with stable
> parts of Gentoo.  Then ask: what architectural changes to Gentoo
> would have eliminated that problem, or better yet, that general
> type of problem?
> 
[snip]

> Regards,
> 
> John
> 
> -- 
> gentoo-project@lists.gentoo.org mailing list
> 
John,

There is a third class of problem - experienced user does something for 
the first time and makes an error. There is a lot of that

I do pretty much what you suggest in my forums post. Users making a 
mistake, I try to show them where and what they did wrong without 
telling them the answer. The give a man a fish or teach him to fish 
idea.

When there is a regular issue, I write a NeddySeagoons' Guide on the 
forums that I can link to. It makes it easier for me and provides a 
tested path for others to follow. There are only about 7 or 8. I can't 
think of any that were Gentoo specific. Google Neddyseagoons +Guide 
site:forums.gentoo.org to find them.

I would say that the biggest issue is that Gentoo users actually build 
their own distro, using the tools that gentoo provides. There is no 
such thing as the Gentoo distro. Every install is different. Users 
therefore go though all the pain that binary distro builders go 
through,

that binary distro users never see.

Its this incremental distro update that causes problems from time to 
time. In general gentoo packages what ${UPSTREAM} provides. When that 
gets changed in incompatible ways, modular Xorg, libexpat ... its 
necessary to rip out the old put in the new and then pick up the 
pieces. That's never going to go away - its the price of being a source 
based distro.

Gentoo can (and does) provide the tools to ease the pain as much as 
possible. Can the tools be improved to make things easier ?
Of course thay can, and I've seen the way portage and gentoolkit has 
improved over the years.
Can the tools eliminate issues like libexpat - no. Whenever you have to 
take a jigsaw apart to put it together differently to fit in a piece 
you just found there is risk.

- -- 
Regards,

Roy Bamford
(NeddySeagoon) a member of
gentoo-ops
forum-mods
treecleaners

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.7 (GNU/Linux)

iD8DBQFHloM+TE4/y7nJvasRAukOAKC0CyYZRqS0wcksvrilosI0Yv9lIACgrk54
pg3m0YqMfqyrNntSAsnothQ=
=Y76D
-----END PGP SIGNATURE-----

--
gentoo-project@lists.gentoo.org mailing list

[gentoo-project] Re: Re: Plan, then communicate

[Steve Long]

Richard Freeman wrote:

> John Lawles wrote:
>> 
>>      The concept is that there would be a separate team maintaining
>> the stable tree.  A precedent for this would be the Linux kernel's
>> two-track development.
>> 
> 
> Actually, in theory that is what already happens with stable keywording
> - or is supposed to happen.  It is a different group of developers who
> maintain the stable tree (well, there is overlap, but some people wear
> multiple hats).
>
I agree that what with unstable, stable and herds working in overlays, we
have enough structure in place. Indeed some users complain that use of
overlays means they have to add stuff to layman to get bleeding-edge like
they "used to." :roll: IMO the balance is right. For anyone who hasn't
tried it yet, I recommend autounmask -n.

> Expat was a real mess - mainly because they break ABI and it doesn't use
> slotting.  I'm not sure why slotting couldn't have been used in this
> case (others might be able to comment on this).
> 
> What we need is a better mechanism of warning users that they're about
> to do something that could cause them major headaches.  ELOG is useless
> when you find out after the fact.
>
Well, we added /etc/warning to update[1] after the expat thing so we could
automatically protect users from ABI breakages. (That's why it took 2 or 3
months to bring out the new version; testing in chroot was a pita til we
got binhosts working nicely over the web.) It also picks up on any
elog/warn/info that tells the user to revdep on a lib; from a fresh 2007.0
install when it does the expat thing it also picks up on libintl.so.7
(iirc) which isn't in the warning file.

That's basically a mechanism for us to do what John was discussing: pick up
on breakages that hit us in #gentoo or -chat, and if they're more than a
simple revdep, codify a workaround. It's a generic mechanism, to do with
packages that must be built before the package (none atm) straight after
(eg gettext XML-Parser and libtool for expat), things we know will break
and so should be rebuilt before the revdep (eg dbus) and the actual lib/s
to revdep on. For all of these, they're only rebuilt if installed and
slotting is checked.

It's far easier doing that (and a whole load of other stuff) in a script
imo, since it's easy for users to change without requiring dev time. It
also keeps a logical separation between the front-end UI and the actual
package manager, which is rightly more concerned about maintaining your
system's packages in a coherent state, and building literally any type of
software.

> revdep-rebuild working better would also be nice.  If it won't work the
> ebuild should abort prior to install with a link to a document
> indicating what will need to be done, and then users can read and
> understand it before they break things.
>
Yeah that was the real problem with expat: revdep was going through a
rewrite at the time. The change so that the new version would install the
best-visible by default (as opposed to exact same versions) was fine (it's
what update does by default, since you sometimes need a package where the
version installed has gone from the tree) but it affected the apache2
upgrade since apxs is slotted. (Sorting that out added another week or
so ;)

[1] YAF update script: http://forums.gentoo.org/viewtopic-t-546828.html


-- 
gentoo-project@lists.gentoo.org mailing list

[gentoo-project] Re: Re: Plan, then communicate (no-list)

[Steve Long]

Richard Freeman wrote:

> Steve Long wrote:
>> Alec Warner wrote:
>>> Talking to users is exhausting when the user really has a
>>> misconception about a given problem, program, or feature.
>> Yeah it's called requirements analysis (whichever model you use.) That's
>> why it's such a source of problems.
>> 
> 
> True.  I think one of the underlying issues in this mess though is "who
> is the customer?"
>
I accept your points, but a distro's customers are its end-users. Simple.
Doesn't matter if some of them happen to be devs or power-users or
developers from other projects or a total newb. They are Gentoo's
customers.

The same applies to any software-project. If you don't look after your
users, you don't get paid (in the real world.) Without users Gentoo will
wither eventually. No real glory in working on a project no-one uses (even
if you and your mates think it's great and continue to use it; where will
you get new devs from when the others get a real job?)

> If I'm assigned to a project at work I typically have a customer that
> I'm aiming to make happy.  If the customer ends up happy, I get
> rewarded, and if the customer is unhappy I get punished.  If somebody
> other than the customer gives me some requirements they're usually only
> taken into consideration to the degree that they can be handled without
> significant additional cost, or interference with the customer's needs.
> 
> It is the same in the open source world.  The customer is whoever
> rewards or punishes you.  Most often the customer is a peer of some sort
> - you take care of their php headaches and they take care of your
> baselayout headaches or whatever.  The customer is often yourself - you
> want to use some package not in portage - just add it.  Rarely is the
> customer an end user - the users aren't really in a position to reward
> you except in the most general sense - and if you take care of everybody
> else chances are you'll make a bunch of users happy anyway.
>
IDK; I thought the whole furore over the Foundation was quite stressful. And
I can't imagine any dev worth the name liking users' consistent and
bitterly eloquent complaints about the behaviour of their peers. It makes
me ashamed on IRC when they tell me about and I'm just a user; I simply
can't understand how some of your peers can be so unpleasant and no-one
says a word about it.

You represent Gentoo: that should be in every dev's head whenever they post
using a gentoo.org address. Yes your opinions are your own etc, but if you
behave like a stroppy teenager it reflects badly on everyone else. Doesn't
matter what your position is, whether you're on the Council or leading a
herd; if you behave badly someone should pick you up on it, *publically*
imo. You behave badly in public, you get picked up on it in the same
medium. That's what would make your "justice" seen to be done and give
people confidence that Gentoo as a whole will not tolerate nastiness in its
ranks.

ATM the very opposite seems to be true (and no it doesn't need a BDFL to
sort that out, nor would a BDFL be /able/ to sort that out imo: it has to
be a collective thing.)

> I'm not saying this is how it ought to be - but it is how it tends to be
> in the open source world.  How does IBM get its fancy hardware supported
> in linux?  They do a few things - one is that they write the drivers
> themselves, and two is that they give away all kinds of fancy code away
> so that the kernel maintainers are inclined to commit those drivers and
> take them forward.  They don't just stand up and complain that there
> aren't enough kernel devs buying mainframes and writing code for them.
>
Er yeah, but when users do try to get involved, believe me they get put off.
I'm not saying you do it (you're engaging in this discussion) but there is
a prevalent attitude towards "lusers." The funny thing is others use that
to mean all Linux users, specifically including Gentoo devs in some cases.

IBM invests in Linux because there is a return on that investment. Users
invest a lot, in time, and also emotion - defending Gentoo to detractors.
Personally I find it amazing that devs are so short-sighted: if you don't
want to interact, simply don't. Don't be nasty, don't sneer at users, just
stfu and let someone else handle it; or maybe, you know, honour the
agreement you made when you signed on, and start representing Gentoo, not
your own ego.

> You can always ask nicely - but one person asking nicely doesn't
> obligate anybody else to respond.  If what they're asking for entails a
> lot of work or a long-term maintenance commitment, they might need to do
> it themselves or find somebody else willing to do it.  However, asking
> nicely will generally get you further than asking rudely.
>
Indeed: maybe devs should think about that when they complain they don't
have enough help. How on Earth do you (as a group) expect to get it when
a) you don't ask and b) you are contemptuous of anyone offering it?

> More than anything else gentoo needs people to step up and DO things.
> If your favorite package is stale volunteer to maintain it.  Or at least
> volunteer to proxy-maintain it - convincing a developer you're willing
> to take the responsibility seriously so that they can do commits for
> you.  What you can't just do is throw an ebuild into bugzilla and
> complain when it doesn't get committed - a dev would be reluctant to
> just commit an ebuild without personally making sure it works well -
> since they get to deal with all the complaints when 5,000 servers stop
> working.
>
Sure: I tell people in #-dev-help to submit and when it's maintainer-wanted,
to take it to #gentoo-sunrise. I find it suprising that they don't already
know about it tbh.
 
> I'm not too worried - people are stepping up and the current situation
> will pass.  Considering that even welfare recipients find something
> about their free money to complain about in the US I'm not surprised
> that users who are offered a free operating system can find time to make
> demands of the people who gave it to them.  That's just human nature.  I
> think that most users will just take all of this in stride.  I'd love to
> see more done to get users involved and make them feel like they have a
> voice, but users do need to understand that while they have a right to
> speak up nobody has an obligation to take orders...

Yeah take it as read that we accept that you are here voluntarily. (So are
we btw.) What specifically do you think could be done to help your users
feel like Gentoo cares about them, rather than views them as an annoyance?

Other projects (eg btrfs) *like* the attention they get from Gentoo users.
They don't balk at editing config files etc, and they don't mind being told
somthing's b0rked, and they need to start again. If it helps, think of your
users as your fans.

Yes they can be annoying: but every single one of them cares about Gentoo.

I sometimes wonder if some of the devs do the same; they seem more concerned
with their own glory and absolutely hate being told they messed up. So you
get bitchy little arguments on bugzilla denying a problem, when they should
just fix the mess and move on.

-- 
gentoo-project@lists.gentoo.org mailing list

[gentoo-project] Re: Re: Plan, then communicate (no-list)

[Steve Long]

Wulf C. Krueger wrote:

> On Monday, 21. January 2008 19:19:12 Steve Long wrote:
>> And when 90% say they don't like the way they get treated by Gentoo
>> devs, there is a real issue.
> 
> "The best argument against democracy is a five minute conversation with
> the average voter." - Winston Churchill
> 
> s/voter/user/
>
Churchill also said democracy is the worst form of government: apart from
all the others.

He also questioned why anyone should worry about Arabs being gassed in
Mesopotamia, so maybe you can just speak for yourself; would you prefer to
live under Saddam, Amin or Hitler than a democratic government?

>> > That being said; talking to users who know what they are doing
>> > (doubly so when they know more than me) is a delight and I'm
>> > generally happy to take the time to respond.  If there was some way
>> > to aggregate user complaints into concrete problem sets I'm all ears.
>> Votes. Require 75% majority from users if you want.
> 
> No, most of those users do exactly that: They *use* what *we*, the Gentoo
> developers, provide. They can become developers easily enough and then
> they may vote all they want.
>
Yeah but antarus was asking about "aggregating user complaints into concrete
problem sets." How would you do that?

> And they can always vote with their feet - on their way out. Gentoo is
> about choice and one such choice can be to leave.
> 
Yeah thanks for that contribution: really moved the debate forward. Don't
slam the door on your way out ;)


-- 
gentoo-project@lists.gentoo.org mailing list

Re: [gentoo-project] Re: Re: Plan, then communicate (no-list)

[Nirbheek Chauhan]

On Jan 23, 2008 7:23 PM, Steve Long <slong@rathaus.eclipse.co.uk> wrote:
> Wulf C. Krueger wrote:
>
> > On Monday, 21. January 2008 19:19:12 Steve Long wrote:
> >> And when 90% say they don't like the way they get treated by Gentoo
> >> devs, there is a real issue.
> >
> > "The best argument against democracy is a five minute conversation with
> > the average voter." - Winston Churchill
> >
> > s/voter/user/
> >
> Churchill also said democracy is the worst form of government: apart from
> all the others.
>
> He also questioned why anyone should worry about Arabs being gassed in
> Mesopotamia, so maybe you can just speak for yourself; would you prefer to
> live under Saddam, Amin or Hitler than a democratic government?

These statements on governance are completely irrelevant here. They
apply to large countries with populations in (at least) millions. For
small communities of several thousand (very small countries ie),
direct democracy works best. But even that is not relevant here, since
the responsibilities of a country's leaders are very different from
the "responsibilities" of a volunteer-run organisation.
`Responsibilities` is in quotes because you cannot really "police" an
organisation that works for free in it's free time.

Hence, a combination of meritocracy and democracy is the best way to
run these. The basic tenants of democracy of "equality" and "freedom"
have no meaning whatsoever in this context. In fact, I'm tempted to
invoke Godwin's Law at this point.

This is a *completely* different situation, and one simply cannot make
generalisations such as you have.


-- 
~Nirbheek Chauhan
-- 
gentoo-project@lists.gentoo.org mailing list

Re: [gentoo-project] Re: Re: Plan, then communicate (no-list)

[Wulf C. Krueger]

[-- Attachment #1: Type: text/plain, Size: 686 bytes --]

> Yeah but antarus was asking about "aggregating user complaints into concrete
> problem sets." How would you do that?

I'd wait for the bugs to come. If there are problems, we have Bugzilla  
to use for that.

And there's the user survey being assembled.

>> And they can always vote with their feet - on their way out. Gentoo is
>> about choice and one such choice can be to leave.
> Yeah thanks for that contribution: really moved the debate forward.

No, really. I don't want anyone to leave but if someone really thinks  
Gentoo is going down the drain (which it isn't), I can't do much about  
it.

> Don't slam the door on your way out ;)

I'll stay. :-)

-- 
Best regards, Wulf

[-- Attachment #2: PGP Digital Signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-project] Re: Re: Re: Plan, then communicate (no-list)

[Dale]

[-- Attachment #1: Type: text/plain, Size: 1945 bytes --]

Steve Long wrote:
> Nirbheek Chauhan wrote:
>
>   
>> On Jan 23, 2008 7:23 PM, Steve Long <slong@rathaus.eclipse.co.uk> wrote:
>>     
>>> Wulf C. Krueger wrote:
>>>
>>>       
>>>> On Monday, 21. January 2008 19:19:12 Steve Long wrote:
>>>>         
>>>>> And when 90% say they don't like the way they get treated by Gentoo
>>>>> devs, there is a real issue.
>>>>>           
>>>> "The best argument against democracy is a five minute conversation with
>>>> the average voter." - Winston Churchill
>>>>
>>>> s/voter/user/
>>>>
>>>>         
>>> Churchill also said democracy is the worst form of government: apart from
>>> all the others.
>>>
>>> He also questioned why anyone should worry about Arabs being gassed in
>>> Mesopotamia, so maybe you can just speak for yourself; would you prefer
>>> to live under Saddam, Amin or Hitler than a democratic government?
>>>       
>> These statements on governance are completely irrelevant here.
>>     
> Er I didn't bring them up. TBH it was only an aside to the main issue of how
> to involve users. The rest of the discussion has in fact been about
> governance, so I am puzzled that you think it "completely irrelevant."
>
>   
>> They 
>> apply to large countries with populations in (at least) millions. For
>> small communities of several thousand (very small countries ie),
>> direct democracy works best. But even that is not relevant here, since
>> the responsibilities of a country's leaders are very different from
>> the "responsibilities" of a volunteer-run organisation.
>>     
> In your opinion: others clearly feel it relevant.
>   

I agree.   +1

>   
>> `Responsibilities` is in quotes because you cannot really "police" an
>> organisation that works for free in it's free time.
>>
>>     
> Hmm and there's no such thing as forum moderators or irc ops?
>   

Any organization larger than a few must have rules and some "police"
enforcement. 


>  <  SNIP >


Dale

:-)  :-) 

[-- Attachment #2: Type: text/html, Size: 3050 bytes --]

[gentoo-project] Re: Re: Re: Plan, then communicate (no-list)

[Steve Long]

Nirbheek Chauhan wrote:

> On Jan 23, 2008 7:23 PM, Steve Long <slong@rathaus.eclipse.co.uk> wrote:
>> Wulf C. Krueger wrote:
>>
>> > On Monday, 21. January 2008 19:19:12 Steve Long wrote:
>> >> And when 90% say they don't like the way they get treated by Gentoo
>> >> devs, there is a real issue.
>> >
>> > "The best argument against democracy is a five minute conversation with
>> > the average voter." - Winston Churchill
>> >
>> > s/voter/user/
>> >
>> Churchill also said democracy is the worst form of government: apart from
>> all the others.
>>
>> He also questioned why anyone should worry about Arabs being gassed in
>> Mesopotamia, so maybe you can just speak for yourself; would you prefer
>> to live under Saddam, Amin or Hitler than a democratic government?
> 
> These statements on governance are completely irrelevant here.
Er I didn't bring them up. TBH it was only an aside to the main issue of how
to involve users. The rest of the discussion has in fact been about
governance, so I am puzzled that you think it "completely irrelevant."

> They 
> apply to large countries with populations in (at least) millions. For
> small communities of several thousand (very small countries ie),
> direct democracy works best. But even that is not relevant here, since
> the responsibilities of a country's leaders are very different from
> the "responsibilities" of a volunteer-run organisation.
In your opinion: others clearly feel it relevant.

> `Responsibilities` is in quotes because you cannot really "police" an
> organisation that works for free in it's free time.
>
Hmm and there's no such thing as forum moderators or irc ops?
 
> Hence, a combination of meritocracy and democracy is the best way to
> run these.
I'm fine with both.

> The basic tenants of democracy of "equality" and "freedom" 
> have no meaning whatsoever in this context.
Again, in your opinion. How can the "basic tenants of democracy" have
nothing do with it if one is using "a combination of meritocracy and
democracy"?

> In fact, I'm tempted to invoke Godwin's Law at this point.
>
puhleez.

> This is a *completely* different situation, and one simply cannot make
> generalisations such as you have.
> 
I didn't; you inferred and generalised an awful lot all on your own ;)


-- 
gentoo-project@lists.gentoo.org mailing list

[gentoo-project] Re: Re: Re: Plan, then communicate (no-list)

[Steve Long]

Wulf C. Krueger wrote:

>> Yeah but antarus was asking about "aggregating user complaints into
>> concrete problem sets." How would you do that?
> 
> I'd wait for the bugs to come. If there are problems, we have Bugzilla
> to use for that.
>
That's fine for technical problems: it's a lot harder to use that to deal
with social problems. For a start, it would be seen as off-topic, and it
would take a very brave user indeed to file a {user,dev}rel bug. I guess
what I am getting at is that users need to be molly-coddled to a certain
extent: they're in a position of less power and not so comfortable with the
processes.

> And there's the user survey being assembled.
>
Yeah it looks great: what will be done with the results of it?

Also, will it be a recurring (eg annual) event?

>>> And they can always vote with their feet - on their way out. Gentoo is
>>> about choice and one such choice can be to leave.
>> Yeah thanks for that contribution: really moved the debate forward.
> 
> No, really. I don't want anyone to leave but if someone really thinks
> Gentoo is going down the drain (which it isn't), I can't do much about
> it.
>
Sure; but if they're going to leave that's up to them. Scaring em off (which
is how "if you don't like it, there's the door" sounds) isn't in Gentoo's
interest imo.

>> Don't slam the door on your way out ;)
> 
> I'll stay. :-)
> 
TF! (not that I had any doubt about it. ;-)


-- 
gentoo-project@lists.gentoo.org mailing list

Re: [gentoo-project] Re: Re: Re: Plan, then communicate (no-list)

[Wulf C. Krueger]

[-- Attachment #1: Type: text/plain, Size: 2958 bytes --]

>> I'd wait for the bugs to come. If there are problems, we have Bugzilla
>> to use for that.
> That's fine for technical problems: it's a lot harder to use that to deal
> with social problems.

I've written it before (in the context of some discussions on the -dev  
mailinglist): People need to grow a thicker skin. If people are scared  
by "fighting" among the developer's I can't really help it. We have  
quarrels just like anyone else. We will piss each other off sometimes  
and either work it out or agree that we consider each other idiots.  
That's fine, that's life.

Most users I've heard or seen so far aren't even really concerned with  
social stuff but (semi-) technical things - no 2007.1 release  
(technical issue), GWN (semi-technical; lack of manpower and user  
contributions), quality of the tree (technical) and the Foundation  
bullshit (non-technical and without any real relevance for our users).

If those are addressed (and all of them are), most of the discontent  
will either disappear or at least things will calm down again.

> would take a very brave user indeed to file a {user,dev}rel bug.

(That would be a UserRel bug.) Why? How many people have been eaten by  
UserRel so far? ;-)

Furthermore, seeing all those comments on Robbins' blog and from brief  
looks at the Forums, I'd say we have (more than ;->) enough brave users.

Yes, it will need some maturity but, yes, I *do* expect that from  
people who want to be taken seriously.

> what I am getting at is that users need to be molly-coddled to a certain
> extent:

We've had that. It was called User Representatives and it failed  
because the representatives decided not to represent anyone by either  
not showing up or simply doing next to nothing. I'm not into  
baby-sitting; I've three kids of my own. :-)

Nevertheless, whenever a user approaches me with a minimum (!) of  
politeness by mail, in the Forums or on IRC, I treat him/her like  
that, too.

> they're in a position of less power and not so comfortable with the
> processes.

Maybe but we need a certain degree of processes and people *will* have  
to follow them just like I have to as well.

>> And there's the user survey being assembled.
> Yeah it looks great: what will be done with the results of it?

I'm not sure yet.

> Also, will it be a recurring (eg annual) event?

Probably.

>> No, really. I don't want anyone to leave but if someone really thinks
>> Gentoo is going down the drain (which it isn't), I can't do much about
>> it.
> Sure; but if they're going to leave that's up to them. Scaring em off (which
> is how "if you don't like it, there's the door" sounds) isn't in Gentoo's
> interest imo.

Sometimes, that's the only reasonable answer, though.


>>> Don't slam the door on your way out ;)
>> I'll stay. :-)
> TF! (not that I had any doubt about it. ;-)

"TF"? :-)

-- 
Best regards, Wulf

[-- Attachment #2: PGP Digital Signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

[gentoo-project] Re: Re: Plan, then communicate (no-list)

[Steve Long]

Roy Bamford wrote:
> On 2008.01.21 18:19, Steve Long wrote:
>> And when 90% say they don't like the way they
>> get treated by Gentoo devs, there is a real issue. Call it
>> communicaton, call it what you want, it's a real and valid concern.
>> 
>> And IMO it holds Gentoo back.
>> 
> For the furtherance of this discussion and simplification of the
> problem space, lets assume that we have already identified that Gentoo
> is split into two mutually exclusive groups which I shall call "them"
> and "us". I do not admit that it is really like this but it serves as a
> reference frame for further discussion.
>
Fair enough, although I think it better to talk about devs and users, since
there is a distinction (a dev has gone through the recruitment process) and
there is overlap, and far more common ground than is realised when the
groups are arguing. It's like the People's Front of Gentoo vs the Gentoo
People's Army; we're losing sight of our commonality in factional fighting:
http://jerryleecooper.com/

> How do you go about turning "them" and "us" into "we" ?
> Discuss.
>
You just answered it imo: discussion, in a courteous framework. Having this
list makes a big difference, since we now have a forum wherein to discuss
non-technical issues without being flamed for being off-topic (if you're a
user.)

> One further constraint to the discussion, you must not use the word
> "they", as in "they should ..."
>
We should appoint you, NeddySeagoon El Presidente! ;p You've got far more
nous than most of us, and you have the maturity to see past your own ego.
Although TBH I don't personally believe we need a President or any other
changes beyond making sure the legal issues are appropriately handled and
addressing user concerns (specifically how they feel alienated by devs, in
contrast to other users). I'd like it if the culture changed so that user
concerns were seen as the priority, but I ain't holding my breath.. ;)

Presidents and premiers don't really work irl imo. Too much on one shoulders
only leads to collapse (usually evinced by delusions of grandeur, or a
conviction that complex problems can be solved with simple solutions) as
well as encouraging formation of a clique around the victi^W candidate.

People naturally form sub-groupings; if we ensure no one group can override
everyone else, that'll be good. A single leader works against that imo
since s/he will only ever be able to have effective communication with (or
trust in) a limited subset. Bingo, there's your clique, and everyone else
will feel excluded, or judged on how close they are to the "centre".

> The idea is not to point fingers but plan how to move from the
> situation described above to a different (better?) situation.
>
Agreed: whatever group you talk about there are always people one doesn't
like. Recriminations back and forth aren't going to help.
 
> (No points for spotting the 6sigma game)
>
Well I hope I don't lose any for having nfc what that meant? ;)


-- 
gentoo-project@lists.gentoo.org mailing list

Re: [gentoo-project] Re: Re: Re: Plan, then communicate (no-list)

[Nirbheek Chauhan]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Jan 23, 2008 10:07 PM, Steve Long  wrote:
> Nirbheek Chauhan wrote:
>
> > On Jan 23, 2008 7:23 PM, Steve Long  wrote:
> >> Wulf C. Krueger wrote:
> >> > "The best argument against democracy is a five minute conversation with
> >> > the average voter." - Winston Churchill
> >> >
> >> > s/voter/user/
> >> >
> >> Churchill also said democracy is the worst form of government: apart from
> >> all the others.
> >>
> >> He also questioned why anyone should worry about Arabs being gassed in
> >> Mesopotamia, so maybe you can just speak for yourself; would you prefer
> >> to live under Saddam, Amin or Hitler than a democratic government?
> >
> > These statements on governance are completely irrelevant here.
> Er I didn't bring them up. TBH it was only an aside to the main issue of how
> to involve users. The rest of the discussion has in fact been about
> governance, so I am puzzled that you think it "completely irrelevant."

The statements you made about Saddam, Amin and Hitler are irrelevant
here. The rest of the discussion is not.

> > `Responsibilities` is in quotes because you cannot really "police" an
> > organisation that works for free in it's free time.
> >
> Hmm and there's no such thing as forum moderators or irc ops?

You're mis-interpreting my statements. Perhaps I need to give a bit
more context to the words I use.

Policing in the sense of forcing devs to work on things they do not
want to work on.

When you're being paid to govern (as in the case of politicians) it's
their job, and they have to do it no matter how much they dislike it.
Not so in open source.
Of course, devs "behaving" is something that is a social necessity,
and is currently handled well in Gentoo.


>
> > The basic tenants of democracy of "equality" and "freedom"
> > have no meaning whatsoever in this context.
> Again, in your opinion. How can the "basic tenants of democracy" have
> nothing do with it if one is using "a combination of meritocracy and
> democracy"?

You're not understanding how I meant this combination of meritocracy
and democracy works. Meritocracy to choose the "leaders" (devs), and
Democracy between them. The users don't come in the picture at all.
If the devs wish to listen to the users, well and good. If not, no
one's forcing the users to stay. It's nothing like a country where
you're tied to the place you're living, and you have no choice.

>
> > In fact, I'm tempted to invoke Godwin's Law at this point.
> >
> puhleez.

You showed as though the opposite of democracy is a gruesome
dictatorship led by people such as Saddam, Hitler, and Mussolini.
You're telling me that isn't an application of Godwin's Law?

Godwin's Law:
"As an online discussion grows longer, the probability of a comparison
involving Nazis or Hitler approaches one."


> > This is a *completely* different situation, and one simply cannot make
> > generalisations such as you have.
> >
> I didn't; you inferred and generalised an awful lot all on your own ;)

Perhaps the phrase "wrong place to use these examples" instead of
"generalisations" would've worked better?

- --
~Nirbheek Chauhan

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.7 (GNU/Linux)
Comment: http://firegpg.tuxfamily.org

iD8DBQFHl3scb1z91vbKYbYRAlt/AJ9t5ZN63XPyVkZ79bTbU1XOdiIgAwCggGY0
hhP7CbrCQ4Vz28qo+B4aoJU=
=Ae5w
-----END PGP SIGNATURE-----
-- 
gentoo-project@lists.gentoo.org mailing list

Re: [gentoo-project] Re: Plan, then communicate (no-list)

[Donnie Berkholz]

On 13:37 Mon 21 Jan     , Richard Freeman wrote:
> True.  I think one of the underlying issues in this mess though is "who is 
> the customer?"

Thanks for your well-put description of this.

Donnie
-- 
gentoo-project@lists.gentoo.org mailing list

Re: [gentoo-project] Re: Re: Plan, then communicate (no-list)

[Donnie Berkholz]

On 13:48 Wed 23 Jan     , Steve Long wrote:
> Richard Freeman wrote:
> > True.  I think one of the underlying issues in this mess though is "who
> > is the customer?"
> >
> I accept your points, but a distro's customers are its end-users. Simple.
> Doesn't matter if some of them happen to be devs or power-users or
> developers from other projects or a total newb. They are Gentoo's
> customers.
> 
> The same applies to any software-project. If you don't look after your
> users, you don't get paid (in the real world.) Without users Gentoo will
> wither eventually. No real glory in working on a project no-one uses (even
> if you and your mates think it's great and continue to use it; where will
> you get new devs from when the others get a real job?)

Users come naturally once you've got a great product, which comes 
naturally once you've got great developers. The motivation of developers 
to create such a product exists before there are users, and studies of 
OSS projects have shown that it's rarely that a developer's intrinsic 
motivation is to get users.

New users may be drawn by the userbase (the community aspect) in 
addition to the product. New developers are drawn by the product or by 
the existing developers. Since new developers generally come from the 
user base, there would be some decline, but I suspect smaller than you 
would think.

Thanks,
Donnie
-- 
gentoo-project@lists.gentoo.org mailing list

[gentoo-project] Re: Re: Re: Re: Plan, then communicate (no-list)

[Steve Long]

Wulf C. Krueger wrote:

>>> I'd wait for the bugs to come. If there are problems, we have Bugzilla
>>> to use for that.
>> That's fine for technical problems: it's a lot harder to use that to deal
>> with social problems.
> 
> I've written it before (in the context of some discussions on the -dev
> mailinglist): People need to grow a thicker skin. If people are scared
> by "fighting" among the developer's I can't really help it. We have
> quarrels just like anyone else. We will piss each other off sometimes
> and either work it out or agree that we consider each other idiots.
> That's fine, that's life.

Agreed; I don't have an issue with disagreement: I have just seen too much
*nastiness* (especially on the dev m-l.) And it really annoys me that it
goes unremarked from devs, since they have signed up to _represent_ Gentoo.
Go and have the flamewar off-list, or on IRC. If it's someone you can't
speak with, deal with it professionally on-list by sticking to the topic at
hand (as far as possible.) Yeah there'll be times where that doesn't work,
but they should be exceptional, not commonplace.
 
> Most users I've heard or seen so far aren't even really concerned with
> social stuff but (semi-) technical things - no 2007.1 release
> (technical issue), GWN (semi-technical; lack of manpower and user
> contributions), quality of the tree (technical) and the Foundation
> bullshit (non-technical and without any real relevance for our users).
> 
> If those are addressed (and all of them are), most of the discontent
> will either disappear or at least things will calm down again.
>
Yeah I don't think that addresses the issue of involvement though, and it is
a central problem for devs as well as users, since you are overstretched.
I've spoken to too many user who've complained, with good reason, to accept
that technical problems, especially lack of manpower, don't sometimes have
social causes.
 
>> would take a very brave user indeed to file a {user,dev}rel bug.
> 
> (That would be a UserRel bug.) Why? How many people have been eaten by
> UserRel so far? ;-)
> 
> Furthermore, seeing all those comments on Robbins' blog and from brief
> looks at the Forums, I'd say we have (more than ;->) enough brave users.
> 
Heh, maybe so; thing is if someone's nasty to you (wherever it is) and
no-one else comments, you just assume that's the culture. If you then act
similarly, you will be flamed. That just leads to confusion and
ill-feeling, for zero benefit to anyone. And often that person will just
walk away (no need to show them the door) and you've lost an advocate and
gained a detractor.

> Yes, it will need some maturity but, yes, I *do* expect that from
> people who want to be taken seriously.
>
Sure, but devs have to show just as much, if not more, maturity; or we won't
take your claims to be snowed-under seriously. "You haven't got enough
people? Hmm I wonder why.."

>> what I am getting at is that users need to be molly-coddled to a certain
>> extent:
> 
> We've had that. It was called User Representatives and it failed
> because the representatives decided not to represent anyone by either
> not showing up or simply doing next to nothing. I'm not into
> baby-sitting; I've three kids of my own. :-)
>
Hehe I know what you mean (not as many kids though.) I actually feel that
way about the devs (collectively.) I don't care how much you whinge about
being over-worked, you're doing it out of your own self-interest; you can
quit whenever you want, and some of you even think it acceptable to treat
Gentoo as your own "private playground." If you want more help, ask
*nicely* and *behave* nicely or no-one will want to play with you.

> Nevertheless, whenever a user approaches me with a minimum (!) of
> politeness by mail, in the Forums or on IRC, I treat him/her like
> that, too.
> 
>> they're in a position of less power and not so comfortable with the
>> processes.
> 
> Maybe but we need a certain degree of processes and people *will* have
> to follow them just like I have to as well.
>
Sure; but if you look at users as potential future devs, it's worthwhile
helping them when they are clearly out of their depth. A quiet bit of help
at the beginning makes the world of difference, and brings you a "convert."
There've been a few times in #friendly-coders when someone has asked a
really stupid question, and I really haven't been in the mood to deal with
them. The times when I've bit my tongue and just pointed them in the right
direction, have pretty much always ended up positively.

There's been the odd time that hasn't worked, but others have stepped in to
explain (usually after I've hit !diy, !ego and !igli ;) and only once have
we had to kb someone (after days of problems; he's come back a bit wiser,
finally.)

>>> No, really. I don't want anyone to leave but if someone really thinks
>>> Gentoo is going down the drain (which it isn't), I can't do much about
>>> it.
>> Sure; but if they're going to leave that's up to them. Scaring em off
>> (which is how "if you don't like it, there's the door" sounds) isn't in
>> Gentoo's interest imo.
> 
> Sometimes, that's the only reasonable answer, though.
> 
Well it's the answer I'd personally give to the drobbins ultimatum, but I'd
never say that to my users for any software I've ever written. I might
say "I don't want/have the time to add that, feel free to patch it, and
[maybe] I'll help you with it." If you don't see users as your client, you
won't get anywhere with software development ime. You might get paid, but
you won't be asked back; the best software is always the stuff that
end-users (be that admins, end-users or other coders) feel ownership of,
because it's been developed in response to their needs. No user buy-in, no
use of your code, as many large organisations have found (usually just
after they rolled out their new EIS that was a year late and /way/
over-budget, only for everyone to ignore it.)

Yeah devs are users too; but imo please the users and the devs will be happy
(if they're not they can change it) since they get more kudos as everyone
wants to use it. Please the devs only, and all you get is technically nice
software that no-one else knows how to use, and frankly you code yourself
into irrelevance since everyone's using the other software that makes them
*feel good* about wastin^W spending all their time on a computer.

Devs are just as susceptible to the "ooh, shiny" bit as well; it's human
nature. Many books have been written on HCI and they overlap with
psychology (although the resultant doctoral students annoy the hell out of
me, as they have nfc when it comes to implementation and they claim to know
CS.) It's not something a coder typically worries about, but we like the
same interfaces, especially for apps we don't know.

To be an effective coder, imo, you have be able to think like a user (which
is why it's fun: you see what happens in all the other industries close up
at a data-level.) Spending your days sneering at users is not a good career
move, and imo Gentoo should stamp on it-- hard.

NB: None of this means entertaining unreasonable demands or tolerating
discourteous behaviour.

> "TF"? :-)
> 
/me looks innocent: Thank Freitag! ;p


-- 
gentoo-project@lists.gentoo.org mailing list

[gentoo-project] Re: Re: Re: Re: Plan, then communicate (no-list)

[Steve Long]

Nirbheek Chauhan wrote:
> On Jan 23, 2008 10:07 PM, Steve Long  wrote:
>> Nirbheek Chauhan wrote:
>>
>> > On Jan 23, 2008 7:23 PM, Steve Long  wrote:
>> >> Wulf C. Krueger wrote:
>> >> > "The best argument against democracy is a five minute conversation
>> >> > with the average voter." - Winston Churchill
>> >> >
>> >> > s/voter/user/
>> >> >
>> >> Churchill also said democracy is the worst form of government: apart
>> >> from all the others.
>> >>
>> >> He also questioned why anyone should worry about Arabs being gassed in
>> >> Mesopotamia, so maybe you can just speak for yourself; would you
>> >> prefer to live under Saddam, Amin or Hitler than a democratic
>> >> government?
>> >
>> > These statements on governance are completely irrelevant here.
>> Er I didn't bring them up. TBH it was only an aside to the main issue of
>> how to involve users. The rest of the discussion has in fact been about
>> governance, so I am puzzled that you think it "completely irrelevant."
> 
> The statements you made about Saddam, Amin and Hitler are irrelevant
> here. The rest of the discussion is not.
>
I asked *one* question wrt the preferred mode of governance (I did not make
a series of statements). Clearly everyone here prefers some form of
representative democracy (feel free to chime in if you don't.)

It wouldn't hurt to add the occasional "imo" btw. You seem to enjoy making
generalised pronouncements of your opinions as if they are accepted truths,
which believe me, they are not.

>> > `Responsibilities` is in quotes because you cannot really "police" an
>> > organisation that works for free in it's free time.
>> >
>> Hmm and there's no such thing as forum moderators or irc ops?
> 
> You're mis-interpreting my statements. Perhaps I need to give a bit
> more context to the words I use.
>
No I'm really not: if I'm misunderstanding your meaning, it's because you're
using the wrong words.

> Policing in the sense of forcing devs to work on things they do not
> want to work on.
>
Er that's not policing, it's coercion. It might be what drobbins proposed,
but it's not what I am arguing for.

Policing in the sense of enforcing acceptable behaviour guidelines when they
have *clearly* been breached is something that all of us would accept, I
hope, as the only way to ensure harmonious (or at worst civil) interaction.

> When you're being paid to govern (as in the case of politicians) it's
> their job, and they have to do it no matter how much they dislike it.

That's not true; they can quit too, and in fact usually have far more
options available to them than the average worker who has to take crap at
work or lose their wage.

> Not so in open source.
> Of course, devs "behaving" is something that is a social necessity,
> and is currently handled well in Gentoo.
> 
In your opinion; others feel the dev m-l has been out of hand (I personally
feel it's a lot better than it was, but there are still too many flames,
characterised by 2 or 3-line snipy posts that shed no light on anything)
and also that some devs are arrogant and supercilious, especially on
bugzilla.

I don't myself run into the latter much, since I know a few devs personally.
It's only happened once in the last 6 months or more, and that dev simply
won't be getting any more bug-fixes from me, however critical they might be
to his/our project. Such is life, I'm afraid: if you're nasty to people
they get turned off helping you; blather on about meritocracy all you want,
you've simply shown you're actually not that great a developer.

>>
>> > The basic tenants of democracy of "equality" and "freedom"
>> > have no meaning whatsoever in this context.
>> Again, in your opinion. How can the "basic tenants of democracy" have
>> nothing do with it if one is using "a combination of meritocracy and
>> democracy"?
> 
> You're not understanding how I meant this combination of meritocracy
> and democracy works. Meritocracy to choose the "leaders" (devs), and
> Democracy between them. The users don't come in the picture at all.
> If the devs wish to listen to the users, well and good. If not, no
> one's forcing the users to stay. It's nothing like a country where
> you're tied to the place you're living, and you have no choice.
>
Agreed that a distro is not a nation-state. So the tenets of democracy /do/
have some place in your vision? Good.

I totally disagree that users are irrelevant; Gentoo would not be running
without its users, some of whom provide the infrastructure it runs on. And
think about this for a second: why does every other distro apart from
debian actively chase users?

Users are *critical* to the survival of a distro. Or do you believe a couple
of hundred devs can possibly do all the bug-reporting and fixing that users
currently do? What about the support to other users so that they can keep
their machines running?

Please note: I am not advocating user votes to determine technical
direction. The question raised was "how do we aggregate user concerns that
we may address them?" The only answer that makes any sense is voting. If
that requires 2/3 or 3/4 majority for it to be seen as valid, then so be
it. But there really is no other method that I can think of. Can you?

>>
>> > In fact, I'm tempted to invoke Godwin's Law at this point.
>> >
>> puhleez.
> 
> You showed as though the opposite of democracy is a gruesome
> dictatorship led by people such as Saddam, Hitler, and Mussolini.
Hmm what form of governance would you call "the opposite of democracy"?

> You're telling me that isn't an application of Godwin's Law?
>
No, I don't think it's an application since Godwin's Law is more about
people using the "you're a fascist" argument when they have no rational
response. In this case, philanthrop specifically stated that he thought
democracy to be useless, and forms of governance were the topic.

> Godwin's Law:
> "As an online discussion grows longer, the probability of a comparison
> involving Nazis or Hitler approaches one."
> 
Yeah I know what it is, thanks; that's why I said "puhleez". It doesn't add
much of substance does it? (It is, after all, a joke.) Treating it like a
scientific law with an "application" is nonsensical imo, and frankly
foolish in this instance, since democracy vs some other form of governance
was the (sub-)topic.

>> > This is a *completely* different situation, and one simply cannot make
>> > generalisations such as you have.
>> >
>> I didn't; you inferred and generalised an awful lot all on your own ;)
> 
> Perhaps the phrase "wrong place to use these examples" instead of
> "generalisations" would've worked better?
> 
Not really: see above and recollect I didn't start the whole "democracy is
crap" argument. You inferred an awful lot about what I was saying, and
generalised it to match people who say "fascist!" to their parents.

Can we declare this sub-thread dead please? I feel like we're getting into a
cyclical argument about a minor aside atm; I'll answer any points you want
to talk about wrt user-involvement, but I'll be ignoring anything about
Godwin's Law as I think it's a non-starter and drifting off-topic. (Just a
heads-up.)


-- 
gentoo-project@lists.gentoo.org mailing list

Re: [gentoo-project] Re: Plan, then communicate

[Joanet]

A Dilluns 21 Gener 2008, John Lawles va escriure:
>      I love Gentoo and I am very grateful for all the very
> impressive that Gentoo developers have done.  As a mere end-user, I
> make no claim to have expertise or answers, only questions that
> other end-users likely have.
>      As background, Gentoo users, like me, are happy spending a lot
> of time configuring/upgrading if the end-result is a system
> configured exactly as we want it.  Thus, for example, unlike users
> of other distributions, we are happy configuring/compiling our own
> kernels.

I am also an end-user, and I have been in Gentoo for about a year. I must 
completely agree that I love Gentoo and I believe it is a great distribution 
(if not the best for people wanting to learn linux). 

>      The central issue that Gentoo end-users have, as I gather it,
> is that running Gentoo takes a lot of time that does not seem to
> fall into the above category.

This very much depends on how often do you upgrade. When I started I used to 
upgrade each week. Nowadays, I upgrade when I am on holidays ;-)

> For instance, every portage tree 
> seems to have various issues like blockers and faulty or circular
> dependencies.  The work-arounds get hashed out over time in the
> forums.  It is productive for the first, say, ten people to find
> the problem and test a work-around.  It is a waste to have a
> hundred or a thousand people repeat that process starting from
> scratch.

I do not have much knowledge of computing (I am myself a social psychology) 
and currently I am able to maintain 4 computers running gentoo with different 
architectures. In my case, being able to apply the solutions has given me a 
very good knowledge of linux. 

>      A possible solution is Daniel Robbins' proposal to have
> separate "developer-facing" and "user-facing" portage trees.
> Developers working on the developer-facing side would be freed to
> experiment, hopefully improving their productivity.  Only the
> tested and successful ideas would be ported to the user-facing
> tree, improving end-user satisfaction.

I must say I do not have much knowledge of the organisation of Gentoo and its 
history. All I know is from I have read lately in the forums and I very much 
appreciate Dominik Riva effort in gathering information all in one place. 
From this very naive perspective, I very much dislike the messianic aura 
given to Daniel Robbins in some of the posts, and I find that the discussion 
is sometimes a reaction to his proposals instead of coming from a community 
reflection.

With respect to the thread "Should the Gentoo trustees accept Daniel Robbins 
offer?", I have been unable to vote on that thread as it was closed when I 
noticed it. Nevertheless, I believe that 696 votes and 346 posts in three 
days speak more of the strength and involvement of the community than of a 
deep crisis in the gentoo project. If we look at the posts in the gentoo 
forums, there is plenty of activity and it does not look like we (users) are 
flying away.

I am also concerned with the stereotyping and division between devs/users that 
the discussion is somehow assuming. I believe that many people are involved 
in the Gentoo project in different ways, with more or less involvement, and 
with more or less responsibilities. Being a "community" does not mean that 
everybody has the same "rights and duties", and I am not sure how the 
discussion on "democracy" is helping here ("should I have a vote on the menu 
when I am having guests at dinner when I am the one who is cooking?"). At the 
same time, stressing the differences between users/devs leads to the 
attribution of "idealised" characteristics and interests to those groups 
like "devs are such and want such", "users are such and want such". An 
interesting example of this kind of social processes can be seen in the 
documentary "A class divided" 
(http://www.pbs.org/wgbh/pages/frontline/shows/divided/). I feel the 
discussion on "what the users want" and "what the devs want" promotes this 
differentiation, and we might end up with a book tilted somehow like "devs 
are from mars, users from venus". 

I believe there is space for improvement in the gentoo community (there is 
always room for improvement ;-)). Nevertheless, I do not see (as a gentoo 
user) any urge for drastic measures, and the present discussion does not seem 
to be the adequate atmosphere to make such moves. 

I would like to take to opportunity to enthusiastically thank to all the 
people involved in the gentoo community, and I am very much proud of being 
part of it. 

-- 
Joanet
--
gentoo-project@lists.gentoo.org mailing list