On 2009-08-04T14:36:48, Henne Vogelsang wrote:

...

My take on this is: openSUSE in general uses the technically best solution - and this contains both features and stability - and where that's not feasible, the most popular solution is used. If you want to make this decision making process based on objective criteria you have to tell us how you want to measure all this. Stating this is only 2% of the work :)

Voting, aka "wisdom of the crowds" aka decision markets? Further, this is about defaults or exclusive choices. It doesn't mean that alternatives cannot be included for most scenarios, and where the default is truly unclear, possibly even offering a choice, no? Regards, Lars -- Architect Storage/HA, OPS Engineering, Novell, Inc. SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nürnberg) "Experience is the name everyone gives to their mistakes." -- Oscar Wilde -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org

On Tuesday 04 August 2009 14:47:53 Henne Vogelsang wrote:

Hi,

on 08/04/2009 02:40 PM Lars Marowsky-Bree wrote:

...

On 2009-08-04T14:36:48, Henne Vogelsang wrote:

...

...

My take on this is: openSUSE in general uses the technically best solution - and this contains both features and stability - and where that's not feasible, the most popular solution is used.

If you want to make this decision making process based on objective criteria you have to tell us how you want to measure all this. Stating this is only 2% of the work :)

Voting, aka "wisdom of the crowds" aka decision markets?

For the popularity criteria voting is okay. You just have to figure out who is eligible for a vote and how to vote and stuff but voting for stability or features sounds wrong to me...

Now I'm confused. I never mentioned voting for stability. And voting for features is done in openFATE right now - and either I understand you that you want the voting removed (and then see lots of +1 in the comments ;-( ) or - and that's the purpose of my write up: What does voting for features mean? How to interpret it? Andreas -- Andreas Jaeger, aj@{novell.com,opensuse.org} SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nürnberg) Maxfeldstr. 5, 90409 Nürnberg, Germany GPG fingerprint = 93A3 365E CE47 B889 DF7F FED1 389A 563C C272 A126

Hi, on 08/04/2009 02:50 PM Andreas Jaeger wrote:

On Tuesday 04 August 2009 14:47:53 Henne Vogelsang wrote:

...

on 08/04/2009 02:40 PM Lars Marowsky-Bree wrote:

...

On 2009-08-04T14:36:48, Henne Vogelsang wrote:

...

...

My take on this is: openSUSE in general uses the technically best solution - and this contains both features and stability - and where that's not feasible, the most popular solution is used. If you want to make this decision making process based on objective criteria you have to tell us how you want to measure all this. Stating this is only 2% of the work :) Voting, aka "wisdom of the crowds" aka decision markets? For the popularity criteria voting is okay. You just have to figure out who is eligible for a vote and how to vote and stuff but voting for stability or features sounds wrong to me...

Now I'm confused. I never mentioned voting for stability. And voting for features is done in openFATE right now - and either I understand you that you want the voting removed (and then see lots of +1 in the comments ;-( ) or - and that's the purpose of my write up: What does voting for features mean? How to interpret it?

Okay lets start from the beginning. What are you talking about here? I understood that you are talking about the decision making process for the distribution. Given your subject and your explanation in your mail. Maybe i misunderstood. Whats this mail about then? :) Henne -- Henne Vogelsang, openSUSE. Everybody has a plan, until they get hit. - Mike Tyson -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org

Hi, on 08/04/2009 03:43 PM Andreas Jaeger wrote:

On Tuesday 04 August 2009 15:32:39 Henne Vogelsang wrote:

...

...

...

on 08/04/2009 02:40 PM Lars Marowsky-Bree wrote:

...

On 2009-08-04T14:36:48, Henne Vogelsang wrote:

...

> My take on this is: openSUSE in general uses the technically best > solution - and this contains both features and stability - and where > that's not feasible, the most popular solution is used. If you want to make this decision making process based on objective criteria you have to tell us how you want to measure all this. Stating this is only 2% of the work :) Voting, aka "wisdom of the crowds" aka decision markets? For the popularity criteria voting is okay. You just have to figure out who is eligible for a vote and how to vote and stuff but voting for stability or features sounds wrong to me... Now I'm confused. I never mentioned voting for stability. And voting for features is done in openFATE right now - and either I understand you

On Tuesday 04 August 2009 14:47:53 Henne Vogelsang wrote: that you want the voting removed (and then see lots of +1 in the comments ;-( ) or - and that's the purpose of my write up: What does voting for features mean? How to interpret it? Okay lets start from the beginning. What are you talking about here? I understood that you are talking about the decision making process for

on 08/04/2009 02:50 PM Andreas Jaeger wrote: the distribution. Given your subject and your explanation in your mail. Maybe i misunderstood. Whats this mail about then? :)

My mail was meant to start a discussion on how we make decisions about features in the openSUSE distribution and how openFATE voting will help. We have to clarify what voting means - and can do.

Features meaning openFATE features i take it. Understood. You really need to think about your choice for a subject line then next time. Because openFATe features are a minor subset of "decision making for the distribution" :) Henne -- Henne Vogelsang, openSUSE. Everybody has a plan, until they get hit. - Mike Tyson -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org

Michael, On Tue, Aug 4, 2009 at 7:48 AM, Michael Loeffler wrote:

For openFATE the voting is a strong tool to speak up but we shouldn't base a decsion just on a vote. There are more things to involve, eg. survey results, mainainers opinion, quality aspects etc.)

I couldn't agree more! The challenge is, afaict, there's been a tradition of those additional considerations being included -- but, that process _seems_ to be waning. As a small-business/server user -- almost by definition "in the minority" -- although one that spends a lot of money every year on openSUSE(and SuSE)-related people, products & services, I'm, frankly, very concerned by the possibility that the distro evolves to a managed-by-huge-committtee, majority-rules product. Does this project really want to become another Fedora or Ubuntu? There are countless instances of low-level bugs, or fundamentally important design issues, that would never even register on an openfate popularity-poll, or may well rankle somone's "feelings", but -- imho -- should nonetheless be decided/directed by a much smaller group with the appropriate business, functional or technical competencies. Personally, I do NOT always agree with the decisions of those "smaller groups" either -- but that mecahnism has led to a distro that, imho, remains uniquely functional and valuable to my businesses' interests. -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org

Hi, on 08/04/2009 06:34 PM Alberto Passalacqua wrote:

Il giorno mar, 04/08/2009 alle 16.48 +0200, Michael Loeffler ha scritto:

...

For openFATE the voting is a strong tool to speak up but we shouldn't base a decsion just on a vote. There are more things to involve, eg. survey results, mainainers opinion, quality aspects etc.)

You should *remove* survey results from the factors, as long as the surveys are done without any criterion to at least try to make them representative.

Just a note. The past surveys where done by someone with a background in social science and statistics. So they are not as useless as you make them. Henne -- Henne Vogelsang, openSUSE. Everybody has a plan, until they get hit. - Mike Tyson -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org

On Tuesday 04 August 2009 14:36:48 Henne Vogelsang wrote:

Hi,

on 08/04/2009 02:12 PM Andreas Jaeger wrote:

...

My take on this is: openSUSE in general uses the technically best solution - and this contains both features and stability - and where that's not feasible, the most popular solution is used.

If you want to make this decision making process based on objective criteria you have to tell us how you want to measure all this. Stating this is only 2% of the work :)

Do you have a better proposal on what we shall do in similar situations that will come up? People expect that voting works - and I think we have to state where voting will work and where not, Andreas -- Andreas Jaeger, aj@{novell.com,opensuse.org} SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nürnberg) Maxfeldstr. 5, 90409 Nürnberg, Germany GPG fingerprint = 93A3 365E CE47 B889 DF7F FED1 389A 563C C272 A126

Le mardi 04 août 2009, à 09:58 -0300, Druid a écrit :

On Tue, Aug 4, 2009 at 9:45 AM, Andreas Jaeger wrote:

...

On Tuesday 04 August 2009 14:36:48 Henne Vogelsang wrote:

...

Hi,

on 08/04/2009 02:12 PM Andreas Jaeger wrote:

...

My take on this is: openSUSE in general uses the technically best solution - and this contains both features and stability - and where that's not feasible, the most popular solution is used.

If you want to make this decision making process based on objective criteria you have to tell us how you want to measure all this. Stating this is only 2% of the work :)

Do you have a better proposal on what we shall do in similar situations that will come up?  People expect that voting works - and I think we have to state where voting will work and where not,

I dont think voting works. Maybe the decision should be made by a technical commitee, or the release manager, or the board, I dont know. But wide vote I dont believe it works at all.

There was this proposal at some point that many people liked: http://en.opensuse.org/User:Jproseve/CoreMaintainerProposal I still think it's the right way forward. Vincent -- Les gens heureux ne sont pas pressés. -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org

Le mardi 04 août 2009, à 15:11 +0200, Andreas Jaeger a écrit :

On Tuesday 04 August 2009 15:08:54 Vincent Untz wrote:

...

[...] There was this proposal at some point that many people liked: http://en.opensuse.org/User:Jproseve/CoreMaintainerProposal

I still think it's the right way forward.

What hinders us to implement something like that? This could need an update after the opening of factory where teams have responsibility for an area...

Oh, nothing hinders us :-) And we actually started "implementing" this in GNOME with a core team there. So, let's just approve it and do it? :-) Vincent -- Les gens heureux ne sont pas pressés. -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org

On Tuesday 04 August 2009 16:04:02 Henne Vogelsang wrote:

Hi,

on 08/04/2009 03:11 PM Andreas Jaeger wrote:

...

On Tuesday 04 August 2009 15:08:54 Vincent Untz wrote:

...

[...] There was this proposal at some point that many people liked: http://en.opensuse.org/User:Jproseve/CoreMaintainerProposal

I still think it's the right way forward.

What hinders us to implement something like that?

We have to get rid of Coolo 8)

He could be part of the team - consider cloning Coolo ;) Andreas -- Andreas Jaeger, aj@{novell.com,opensuse.org} SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nürnberg) Maxfeldstr. 5, 90409 Nürnberg, Germany GPG fingerprint = 93A3 365E CE47 B889 DF7F FED1 389A 563C C272 A126

On Tuesday 04 August 2009 16:04:02 Henne Vogelsang wrote:

...

Hi,

on 08/04/2009 03:11 PM Andreas Jaeger wrote:

...

On Tuesday 04 August 2009 15:08:54 Vincent Untz wrote:

...

[...] There was this proposal at some point that many people liked: http://en.opensuse.org/User:Jproseve/CoreMaintainerProposal

I still think it's the right way forward.

What hinders us to implement something like that?

We have to get rid of Coolo 8)

He could be part of the team - consider cloning Coolo ;) You know what? This is history repeating. KDE tried to get rid of Coolo as only decision maker and elected a "Technical Work Group" and we found that it's impractical to have a formal body to do technical decisions and now KDE has a mailing list where people doing the work decide (that Coolo is no longer

Am Dienstag 04 August 2009 schrieb Andreas Jaeger: part of that is bound to family not to KDE :) Greetings, Stephan -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org

On 2009-08-04T09:58:37, Druid wrote:

I dont think voting works. Maybe the decision should be made by a technical commitee, or the release manager, or the board, I dont know. But wide vote I dont believe it works at all.

This is quite opposite to the literature, by the way. Decision-markets, through polling a large audience and aggregating the results, deliver excellent and empirically sound feedback on the usefulness of features. They're an excellent and open way of gathering "market research" data. Compared to focus groups or expert panels, they have a pretty good accuracy, are cheaper to setup, and by definition "more open". Of course, you must describe them properly (ie, use case/benefit, not the detailed technical specification) - users are the wrong group to vote on the _implementation_, but the only group that can reliably vote on the usefulness. Now it's true that some aspects of voting in bugzilla aren't appropriate for this - for example, decision markets try to avoid feedback loops, so the votes from other users shouldn't necessarily be exposed, but those are more or less implementation details. I can dig out some references in case someone is really interested. But "expert judgment", in particular if it is only one expert, has - both for market research and effort estimation, by the way - an extremely bad track record, yet it often is the best (-> only) tool we have, which is somewhat of a pity. Regards, Lars -- Architect Storage/HA, OPS Engineering, Novell, Inc. SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nürnberg) "Experience is the name everyone gives to their mistakes." -- Oscar Wilde -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org

It usually leads to endless discussions without a clear decision, and forces to compromises which are often not in the interest of the distribution. You don't have to take "open decisions" in an open project. You need to take decisions in the interest of the project, the fact that they are open or not does not matter much.

Well it sure does matter to me, I don't expect a vote for everything opensuse related, but I prefer knowing how a decision has been made and that the decision has been evaluated by more then one person. I know the teams have regular meetings and anything important gets discussed there and decided upon, not everytime by popular vote but in an open manner... Karsten -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-project+help@opensuse.org