How should Puppy be developed when Barry steps down?

A home for all kinds of Puppy related projects
Message
Author
disciple
Posts: 6984
Joined: Sun 21 May 2006, 01:46
Location: Auckland, New Zealand

How should Puppy be developed when Barry steps down?

#1 Post by disciple »

Well everybody, it had to happen eventually. Barry is stepping down from developing Puppy.
Some of us think this is cause for panic, but we actually need to sit down and think, and have a rational discussion about how Puppy's development can continue.

Please, for the sake of everybody's sanity, refrain from insulting people and starting big irrelevant arguments as has happened elsewhere. Also, I don't think we need to discuss details like SVN and bug reporters at the moment. What is much more important is who will work together to develop Puppy, and how.

I think we particularly need a discussion between those of you who can take a major role in development, as we need you, or the project will die. If you are one of these people (We all know who you are), please let us know what kind of role you are willing to play, and what kind of system you can work in.
Do you know a good gtkdialog program? Please post a link here

Classic Puppy quotes

ROOT FOREVER
GTK2 FOREVER

disciple
Posts: 6984
Joined: Sun 21 May 2006, 01:46
Location: Auckland, New Zealand

My Opinion:

#2 Post by disciple »

This is my perspective:
- There is probably no one who could successfully take on Barry's "benevolent dictator" role.
- There is too much difference of opinion between Puppy users, and too few really useful developers, for a highly democratic "community" system of development to work.
I suspect many of our really useful developers would be like MU, who says when he is doing development as a volunteer, he will only do what he wants to do - he will not be told what to do by a project manager or by a community vote.
Also, I think as well as being very inefficient and slow, a highly democratic approach is likely to stifle innovation, and is likely to lack any consistent long-term vision.

Therefore, I think the only realistic way for Puppy to move ahead is for leadership and decision-making to belong to a small group of core developers. People like myself who can't play a significant role in development could still express opinions, but in the end Puppy's long-term development will depend on these developers, so they must be able to control the direction they head. How this group worked would be up to them, but I imagine they would work fairly loosely, and I think the group will necessarily be small because at the moment we don't have a lot of people who are qualified for this role.
Do you know a good gtkdialog program? Please post a link here

Classic Puppy quotes

ROOT FOREVER
GTK2 FOREVER

disciple
Posts: 6984
Joined: Sun 21 May 2006, 01:46
Location: Auckland, New Zealand

Ongoing development vs "Community Editions"

#3 Post by disciple »

I'd also like to point out that ongoing Puppy development will be very different to past "community editions". As I see it, these have been more or less customised "remasters" of what Barry has already produced, with extra stuff added, and maybe a few structural improvements. The task of developing Puppy indefinitely will be very different from producing these one-off versions.

--------------
Have fun!

PaulBx1
Posts: 2312
Joined: Sat 17 Jun 2006, 03:11
Location: Wyoming, USA

#4 Post by PaulBx1 »

It's strange; everything I can think of, other than Barry running the show, seems doomed to failure. :?

Here is a possibility:

1) Freeze standard Puppy at 4.1, doing only minor bug fixes: 4.1.1, 4.1.2, etc. No application or kernel updates.

2) Let development continue in the derivatives, especially the ones that actually have some commercial potential (and thus can return money for all the developer time that goes into them). Lots of cross-pollination is possible of course.

3) Standard Puppy 4.1 can point to the most prominent derivatives in the initial info screen, for new users who need more up-to-date development.

I (and many many users) can be happy with a bug-fixed 4.1 for a very long time, or jump to a derivative when it suits us. Developers can be happy still producing useful code. Owners of derivatives still control their stuff (they are already the "Barrys" of their own projects). And there is no potential for community meltdown over how Puppy should be run.

I just throw this out as a backup alternative if nothing else seems reasonable. I too am very doubtful about the democratic model.
Last edited by PaulBx1 on Sat 09 Aug 2008, 13:12, edited 1 time in total.

raffy
Posts: 4798
Joined: Wed 25 May 2005, 12:20
Location: Manila

second

#5 Post by raffy »

PaulBx1 wrote:I (and many many users) can be happy with a bug-fixed 4.1 for a very long time
I second the points made by Paul.

Hopefully, during this time, new development will shape up.
Puppy user since Oct 2004. Want FreeOffice? [url=http://puppylinux.info/topic/freeoffice-2012-sfs]Get the sfs (English only)[/url].

disciple
Posts: 6984
Joined: Sun 21 May 2006, 01:46
Location: Auckland, New Zealand

#6 Post by disciple »

Yes, that is a very good point, and I imagine any development would start like that anyway. It would probably have to start like that to be really sustainable. But then the question is still there - who would create any such versions and how?
I would also be happy without major new development, for several years at least, although I think any releases should probably include application updates (for things like pburn, as well as major applications like seamonkey and Abiword) as well as just bug fixes.

User avatar
WhoDo
Posts: 4428
Joined: Wed 12 Jul 2006, 01:58
Location: Lake Macquarie NSW Australia

#7 Post by WhoDo »

Here is a place to start:

Puppy Linux Foundation & Puppy Foundation History

A fair bit of the information is outdated, some of the names are no longer involved, etc. That said, there are new names to replace them and new projects to be added as well.

There is no reason why the Foundation shouldn't develop into the guiding body for core Puppy direction. It doesn't need any more than agreement from those named to play a part - any part - in the process.

The biggest hurdle, IMHO, is having one or two key people who would take the trouble to build Puppy core from Unleashed, T2 source or whatever. That's not a committee-friendly job. The rest is pretty much already in place. At the very least it is an excellent place to start, provided we can keep the balance that Barry has fostered thus far.

Cheers
[i]Actions speak louder than words ... and they usually work when words don't![/i]
SIP:whodo@proxy01.sipphone.com; whodo@realsip.com

disciple
Posts: 6984
Joined: Sun 21 May 2006, 01:46
Location: Auckland, New Zealand

#8 Post by disciple »

Is the foundation really functioning currently?
It looks rather complicated and perhaps unweildy. But it seems to be irrelevant to the question - it doesn't appear to address HOW development should happen. Or did I miss something big?
I don't think it really matters whether Puppy is developed under the auspices of the foundation, or what.
The biggest hurdle, IMHO, is having one or two key people who would take the trouble to build Puppy core from Unleashed, T2 source or whatever.
That's right.
The rest is pretty much already in place. At the very least it is an excellent place to start
What exactly is the rest though?
Do you know a good gtkdialog program? Please post a link here

Classic Puppy quotes

ROOT FOREVER
GTK2 FOREVER

disciple
Posts: 6984
Joined: Sun 21 May 2006, 01:46
Location: Auckland, New Zealand

#9 Post by disciple »

Actually, its not clear whether you're saying, "We've already got this great decision-making body - we just need some developers to do the work", or that we need a small group of developers who would provide the real leadership and decision-making under the auspices of the foundation.

Anyway - I'm going now, as I'm not helping anybody here :)
Do you know a good gtkdialog program? Please post a link here

Classic Puppy quotes

ROOT FOREVER
GTK2 FOREVER

big_bass
Posts: 1740
Joined: Mon 13 Aug 2007, 12:21

Re: How should Puppy be developed when Barry steps down?

#10 Post by big_bass »

removed
Last edited by big_bass on Tue 19 Aug 2008, 01:11, edited 1 time in total.

User avatar
dogone
Posts: 202
Joined: Tue 22 Apr 2008, 02:53
Location: Arizona, USA

#11 Post by dogone »

Here are some thoughts.

First of all, I implore all Puppy people to take heart. Puppy Linux is not just "Barry" for one very important reason. Barry could not have created Puppy without the support of and feedback from it's followers. Puppy is the product of many hands and many minds. Each of us is a bigger piece of Puppy is than we realize. If you doubt this, ask Barry.

As to moving forward, the first thing we need is a firm platform from which to step. I believe 4.1 is that platform. No Puppy release has ever been as current or well featured. It will be stable, largely understood and an excellent base for New Puppy Phase I.

I propose that New Puppy Phase I concentrate on making 4.1 all it can be. This means bug fixing, but it also means assisting derivatives toward it's adoption. We can't afford to lose 4.1 in the noise of very many Puppies. We must depend on it and use it to get the New Puppy community established. The community must first prove to itself that it can do just this one. If we fail with 4.1, the future holds little prospect.

The "light" work of Phase I will permit our developers to sort out the "who" and "how" of New Puppy before getting involved in the "what". 4.1 will buy everyone time to get their footing. As confidence in 4.1 grows, it will move into derivatives and gradually become the gold standard. Once this happens, we'll have our collective footing and can move into Phase II.

New Puppy Phase II will involve moving Puppy from 4.1 to whatever comes next. The mission will be less one of fixing and more one of improving and enhancing. "4.2" will prove to the community that is capable of taking Puppy beyond what Barry himself envisioned. Anything will be possible thereafter.

As to the people behind New Puppy, amazingly little has really to change. Two things are obvious though. We need to develop a way of making the kinds of decisions Barry has made for us these past years, and we must find the people technically capable of implementing them. Replacing Barry the programmer will be a big challenge, but there are a lot of very talented people amongst us. Decision making will be far more challenging.

Puppy is a very developer-centric distribution (Barry's legacy) so developers will always be at it's core. I suggest that we establish a term-limited group (size ?) consisting of our best developers and our most productive thinkers. A term limit will permit us to tap more community brains over time while discouraging non-productive behavior. Guidance of New Puppy will probably have to be somewhat less democratic at the start so "the group" can get focused and find direction. All input will be welcome but the process must be orderly - more like Barry's blog and less like the forum. In time, perhaps more will be decided in the forum and less in the blog.

We must also plan strategic alliances with other distros and development efforts. Puppy has been too isolated in this regard. Partnering is a powerful tool and ideas must be traded like beads.

Well, I've said far more than I intended. I do feel very confident that the New Puppy community can accomplish virtually anything, once it finds it's footing.

User avatar
john biles
Posts: 1458
Joined: Sun 17 Sep 2006, 14:05
Location: Australia
Contact:

#12 Post by john biles »

Hello Everyone,
First let me say this "I'm no programmer couldn't write a script even if I had a gun against my head". What I am is a experimenter.
I take something I like from Debian or Mandriva or Slackware and on and on and see if I can get it to run in TEENpup. If one version of an app doesn't work, I try an earlier version or try one from another distro. Sometimes I have no luck at all.
Some forum Members have called TEENpup a puplet! Damn right! that's all it is. 99% Barry K 1% TEENpup.
I bring this up for 2 reasons: TEENpup has been based on Puppy 2.14 which came out early 2007 for a long time and still finds some happy and not so happy users. Basically Barry K could have left Puppy after 2.14 was released and it would have made no different to me.

So what I'm trying to say is that Puppy will live on in one form or another for years to come.

The Second reason is more important. At some stage the base LIB's eg glibc will go out of date so to speak. One day Flash, Firefox and other plugins/applications won't run on Puppy as they are updated and require a newer base to function. I found this out time and time again experimenting with TEENpup and its glibc 2.3.5 base LIB's.
I know it's only months away or if lucky a year where TEENpup 2008 user's won't be able to watch youtube etc as it will require a newer version of flash that just won't run.

So as much as I love the Puppy 2 series, I have to move to the Puppy 4 series, I have no choice and sooner or later either will other Puppy user's.

At the least someone has to kept the core dependencies up to date.

I have to thank Barry K for Puppy 4.0.5 IDE which has a glibc 2.6.1 base.
This will allow me and others a long compatability with all the plugins and apps that Linux has to offer.

Here's a really dumb idea......Maybe??
Let those who want to move Puppy forward create their puplets and let the end users decide which is best.

Again I have no proper Programming skills at all. Sometimes I'm amazed that TEENpup works at all. Most true programmers would be horrified to know whats gone into TEENpup and how I got some stuff to work.

This truely is the Magic of Barry K - one robust magically OS
Legacy OS 2017 has been released.

oli
Posts: 194
Joined: Wed 30 Aug 2006, 09:04
Location: Germany, Frankfurt

#13 Post by oli »

My suggestions:

First of all we have to find out who is able and has time to be a developer. If we can't find a core team of developers there will be no future for Puppy.

Than we can think about a "council" of some engaged Puppy users that brings new ideas to the developers and discusses what is added or changed at the next release.

In the past Barry developed Puppy too fast. There were bugs that had to be fixed after the release was published. Therefore we should have a "stable" release that lasts for about one year. In the meantime we enhance the stable version, called unstable. This version should be developed for about 8 months and than there should be 4 months of testing. Than it is changed into the new stable version.

User avatar
ttuuxxx
Posts: 11171
Joined: Sat 05 May 2007, 10:00
Location: Ontario Canada,Sydney Australia
Contact:

#14 Post by ttuuxxx »

dogone wrote: Puppy is a very developer-centric distribution (Barry's legacy) so developers will always be at it's core. I suggest that we establish a term-limited group (size ?) consisting of our best developers and our most productive thinkers. A term limit will permit us to tap more community brains over time while discouraging non-productive behavior. Guidance of New Puppy will probably have to be somewhat less democratic at the start so "the group" can get focused and find direction. All input will be welcome but the process must be orderly - more like Barry's blog and less like the forum. In time, perhaps more will be decided in the forum and less in the blog.

We must also plan strategic alliances with other distros and development efforts. Puppy has been too isolated in this regard. Partnering is a powerful tool and ideas must be traded like beads.
Well I read all you had to say and I must admit Your very "developer" orientated, Thats not the case for puppy anymore. Originally with Barry when he first started puppy 5yrs ago, yes it was, During puppy 1,2 yes it was, During 3 & 4 things changed. I was around for 1 (Just tried it out for a few weeks when it was released) Then the 2 series came out much better new programs etc. But then 3.0 came out which was based on Slackware sources plus puppy's added flare as we know. Then 4 came out. With 3 &4 No real new applications made by "Developers" Just updated fixes, that happen from start to finish of a series, Actually puppy 4 biggest noticeable change was the gtk2 theme and the update puppy manager. Oh and the Qt4 change. Same basic applications, well Seamonkey mail was dropped Sylpheed was added, but still not a developer thing. I compiled QT4 last week myself and released for puppy, I've made more complicated themes, I probably could of hacked the puppy package manager also and made it look like that or even added graphics to it. BUT guess what ??? I'm Not a developer and would be left out of your loop, So would WhoDo, Lobster, Sir Duncan, John Biles and many, many others etc. I think we really need to get off this developer Idea way of thinking. You'll be closing a lot of doors with that attitude.

Truly a developer is someone you can go to and say " I need an application built which can do this .... Show me a prototype in a few days" Puppy has what 4 maybe 5 who could do that,
But ask yourself
Do you have to be a developer to build puppy??? NO

Do you have to a developer update puppy NO

Do you need to be a developer to Change the way puppy runs? example, window manager replacement ? NO

Do need to be a developer to build graphics ? NO

Do you need to be a developer to install and remove applications? NO

Do you need to be a developer to compile packages? NO

Do you need to be a developer to find Bugs and Fixes? NO

Do you need to be a developer to get Gnome working on puppy by compiling sources? NO

Do you need to be a developer to get KDE working on puppy by compiling sources? NO

Do you need to be a developer to build and update websites and manage networks? NO

Do you need to be a developer to build pet and sfs packages? NO

Now lets look at the developer side
They make and maintain applications (my applications are pitiful compared to a real developers, Like Zigbert, He does excellent work)
They compile kernels and they do the above also, and in Barry's case he managed also and built a reusable framework and do most of it:)

The only reason why I broke the list down that way is that I was told I wasn't a developer this week in the forum, Tell you the truth, I never figured I was in the first place, but ALL those NO's you read up there I can do and a lot of non Developers like myself can do also.

Barry made a reusable base for puppy, by what you said, no new kernels, That just means from the developers side of things, fixes and patches, that happens all the time durring a regular cycle. but from the non-developers side of things, probably most of those "NO's" will be used in future 4.0 releases.

The cold hard fact is, right now you have developers and almost developers or junior developers, what ever you call it. You shut the doors on their faces and well I just don't know, your going to hurt a lot of feelings and lose a lot of talent if you think "DEVELOPERS" are the only ones to make decisions, also come to think of it developers aren't professional graphic designers, web developers, Networking managers. etc Theres a lot more to puppy then just developers its called a community, Lets keep it that way.

To the Developers, I'm not trying to step on your toes, But I'm tired of everyone else thinking they can step on mine. Developers are a huge importance in any Linux distro, Without your previous work and current work we wouldn't be here chatting in this forum. I have all the respect in the world for developers, all my applications I use everyday hardware software etc is all developers and I'm very grateful of it. But sometimes the overlooked has to be looked at when making decisions. Together united we can make a better puppy.
ttuuxxx
http://audio.online-convert.com/ <-- excellent site
http://samples.mplayerhq.hu/A-codecs/ <-- Codec Test Files
http://html5games.com/ <-- excellent HTML5 games :)

DigitalCrypto
Posts: 12
Joined: Sun 26 Aug 2007, 19:19

#15 Post by DigitalCrypto »

Yes but you aren't changing anything.

http://puppylinux.com/development/project-statement.htm

Has been and always will be that way. You are better off starting from scratch and going forward or just forking the code base where ever you please and going from there.

If you've never been inside the code of Puppy you probably don't understand the crap Barry had to go through to get things to work.

I invite you to look. At basic you should be able to write shell scripts.

Welcome back from vacation Ttuuxxx. :D

oli
Posts: 194
Joined: Wed 30 Aug 2006, 09:04
Location: Germany, Frankfurt

#16 Post by oli »

@ttuuxxx: I think you are wrong. You need developers at least when you change the basesystem of a distro. You can't stay many years with the same basesystem because hardware and software change too much. And then you need developers. It is one thing to add aplications or change the look and feel. But it is another thing to build up a distro from scratch.

But perhaps this discussion is too sophisticated. So let's return to the aim of this thread: how should Puppy be developed in future.

disciple
Posts: 6984
Joined: Sun 21 May 2006, 01:46
Location: Auckland, New Zealand

#17 Post by disciple »

:(
DC - you are quite wrong. Puppy does not NEED by definition to stay developer focussed.
I personally don't think it could remain successful in the medium-term after Barry steps down unless it remains developer focussed. But that is only my opinion, and it is the whole point of this discussion.
I definitely think to successfully transition to a democratic, user focussed model, we would need some competent developer-type people to take over while a democratic system was implemented.

BTW ttuuxxx - personally, when I say "developer", I do not mean somebody who is writing a whole lot of code. In fact I wouldn't expect someone to have time to do that and to actually put Puppy together.
And if you have compiled a working KDE and Gnome from source, maybe many of us have underestimated your abilities (I don't know, as I don't know how hard that is to do these days).
Do you know a good gtkdialog program? Please post a link here

Classic Puppy quotes

ROOT FOREVER
GTK2 FOREVER

disciple
Posts: 6984
Joined: Sun 21 May 2006, 01:46
Location: Auckland, New Zealand

#18 Post by disciple »

So let's return to the aim of this thread: how should Puppy be developed in future.
Can you please address that question Ttuuxxx?
I'm not sure what you actually want... do you want to implement some sort of democratic system right away? Who can vote? Who decides what to vote on? Who will actually build Puppy and keep him alive and well?
Do you know a good gtkdialog program? Please post a link here

Classic Puppy quotes

ROOT FOREVER
GTK2 FOREVER

User avatar
ttuuxxx
Posts: 11171
Joined: Sat 05 May 2007, 10:00
Location: Ontario Canada,Sydney Australia
Contact:

#19 Post by ttuuxxx »

oli wrote:@ttuuxxx: I think you are wrong. You need developers at least when you change the basesystem of a distro. You can't stay many years with the same basesystem because hardware and software change too much. And then you need developers. It is one thing to add aplications or change the look and feel. But it is another thing to build up a distro from scratch.

But perhaps this discussion is too sophisticated. So let's return to the aim of this thread: how should Puppy be developed in future.
woo woo Noo noo. We need developers big time, I'm not trying to say that at all. I'm just saying developers can't be in 50 places at once and still try to concentrate on their task at hand, Plus they aren't Pros at every field of computers, Well unless your Bill Gates and you just steal everything from your competition, thank god they didn't have websites back at the beginning of Microsoft, It would of been hard for Bill to steal Xerox's website out from underneath them , LOL :). I was going off what the other forum member said about keeping the base for sometime, and just patching it and bringing it up a final release like 3 series did. That could be another year.

That brings me hopefully to end, But first Probably the best people who should decide how puppy should be develop is basically the people who know Linux, Like the internals, what makes it tick, A few of us "Non-developers" have an excellent idea, Just because we haven't compiled a kernel doesn't mean we aren't educated enough from reading kernel reviews, Trying other Linux's and having a round enough opinion what works. Only 1 developer will be compiling the kernel anyways and building the base, but people like myself and many others could be using that base to compile the 200 default or so packages for puppy.
well I hope that helps
Developers are needed most, but others are needed more in other fields, hence we all should be equals.
ttuuxxx
http://audio.online-convert.com/ <-- excellent site
http://samples.mplayerhq.hu/A-codecs/ <-- Codec Test Files
http://html5games.com/ <-- excellent HTML5 games :)

User avatar
WhoDo
Posts: 4428
Joined: Wed 12 Jul 2006, 01:58
Location: Lake Macquarie NSW Australia

#20 Post by WhoDo »

disciple wrote:Is the foundation really functioning currently?
No. It isn't.
disciple wrote:It looks rather complicated and perhaps unweildy.
In its current, very dated format, it is! It was also way too formal at the time. It could do with less formality and more freedom. I still think its a good place to start though.
disciple wrote:But it seems to be irrelevant to the question - it doesn't appear to address HOW development should happen. Or did I miss something big?
For HOW Puppy is developed, you need look no further than Barry's own motherhood statement for the Puppy Project. I don't see why that should change. Perhaps the only thing we really needed was provided by cb88 (I think) - a CVS/SVN repo at Sourceforge.
disciple wrote:I don't think it really matters whether Puppy is developed under the auspices of the foundation, or what.
Now we come to the question of camels being horses designed by committee. The fact is that there has to be some general, loose, intelligent, informed oversight of future Puppy directions. Core Puppy has to stay true to Barry's ideals to survive. Otherwise it's just another also-ran distro like thousands of others out there. Core Puppy should stay sub-100Mb. That's its niche. It is NOT a direct competitor for sub-50Mb distros like DSL and Slitaz. It is NOT a competitor for plus-250Mb distros like PCLinuxOS, Ubuntu, SuSe, et al. The Foundations job is to protect Puppy's ethos, established by Barry, and keep it within its niche. Nothing more. Within that the development team, whatever that is from time-to-time, should have complete freedom to innovate, renovate and rejuvenate!
disciple wrote:
The biggest hurdle, IMHO, is having one or two key people who would take the trouble to build Puppy core from Unleashed, T2 source or whatever.
That's right.
Glad you agree. Barry will still be part of the team but, if I read him correctly, recompiling kernels from source and building Puppy core from Unleashed is just too much work to allow him to pursue the other directions he wants to follow.
disciple wrote:
The rest is pretty much already in place. At the very least it is an excellent place to start
What exactly is the rest though?
"The rest" is all the supporting infrastructure that means people get told what's going on (website, wiki, forum) and have the chance to feed back their own ideas in the appropriate way. If the Foundation simply formalises what is already in place, so be it. At least people will know. For example, nobody in their right mind would suggest that tombh and HairyWill aren't doing a great job with the web site. Only a lunatic would suggest that Lobster isn't the best promoter and cheerleader a distro could have, and he's handy with a wiki too! Who would argue with the likes of tempestuous for providing valuable driver support in a timely fashion? Certainly not me. The Foundation is the means for everyone else to know those things and have a focus for their ideas. That's all.

I certainly don't want to see a bunch of legal claptrap bogging Puppy down. That's not the way we do things, thanks to Barry. OTOH, I would like people to have a sense that there is a core group of committed individuals donating their time and expertise to keep Puppy going. THAT is where the Foundation comes in, IMHO.
[i]Actions speak louder than words ... and they usually work when words don't![/i]
SIP:whodo@proxy01.sipphone.com; whodo@realsip.com

Post Reply