How should Puppy be developed when Barry steps down?

[ttuuxxx]

So let's return to the aim of this thread: how should Puppy be developed in future.

Can you please address that question Ttuuxxx?

Sure no problem, Well I often felt that the 3 series what cut short, It had way more potential then what it was given. The 4 series came to quick. But I guess theres no looking back now. Move forward and onward.

With series 4 someone should keep track of the bugs and dislikes.

I named a few which really bugged me, LOL
Simple yet frustrating, Like the copy and paste not working like it did, I used be able to run the pet package manager, highlite the missing libs and do a search for them, all with the mouse, On 4.0 I can't if I even run Glipper on 1/2 the time it works, the other half its pasting the wrong stuff, or I have to click the Glipper icon and select the right lib/file to paste another add pain, I really hated that in 4.0. To the point where I wasn't going to use it and stick with 3 series and hope that sooner or later it would be fixed, even if you added the cutsel package it still didn't work.
Part of Pfind didn't work because of a missing package, and I had to help around 30 users install the dotpup handler. Really pups still work fine, I've never had a problem with them. Some developers like them better than pets, because you can configure them better, like opening up a folder straight after they install a package with a bin file and icon side by side so you can drag the bin to the desktop and set the icon. How good is that???


How puppy should be developedr??
First find out who wants what?

Base Team (function)
Who wants to compile packages
who wants to compile the kernel
Who wants to build it
Who wants to alpha/beta test and report bugs

Designing Team (look and feel)
Who wants to seek new icons/permissions ex Tango/gnome/kde all have excellent icon resources. Puppy's default 16x16 icons are in need of a face lift. Must have permission's and should be submitted to the Base Team for approval.
Plus we can build/donate our own icons- maybe have a competition, great way of building extra icons.
Same with backgrounds etc

Default Applications
Big things like browsers and media players should maybe be decided by the public, public vote.
Smaller applications can be decided/voted by Base Team / Design Team together.

Public Relations - Press Releases
Someone who will stay in contact will all team members so that they would have a good inside view of the current model

Web Team
Update the website / Wiki


Hope that what your looking for ?
ttuuxxx

P.S Has anyone tried my latest Icewm window manager pet package, I change something like 80+ icons from 16x16 to 24x24 and larger text size in the menu and it looks really nice. Only a handfull of default Icons I kept, I used Gnome/KDE/Tango for it.

[raffy]

@ttuuxxx - I guess one eventually faces the challenge of keeping mum for a while after sharing his/her ideas and coming in again later only if really necessary. (No hard feelings, OK?)

About the Foundation - Forming one tends to formalize the structure of the group, and eventually dissatisfied people make the "Foundation" a target for some unusual agenda. This slows down development and also hurts the feelings of those involved. (At least, this was how it developed for Puppy Linux more than a year ago.)

So it could be more helpful to put people in a list, like "regulars" or "coders" or "testers" --->> mysticmarks is inviting people to enlist here. We can then try to see what to do with the list and where to put it (most likely in puppylinux.org where there is this list example).

puppylinux.org has a blog facility, so everyone's name can be linked to his/her respective blog.

[ttuuxxx]

@ttuuxxx - I guess one eventually faces the challenge of keeping mum for a while after sharing his/her ideas and coming in again later only if really necessary. (No hard feelings, OK?)

.

No hard feeling, We both just want to see puppy stay on the straight and narrow, We could only hope the best and accept nothing less.
ttuuxxx

[DigitalCrypto]

disciple - DC - you are quite wrong. Puppy does not NEED by definition to stay developer focussed.

It doesn't have to but most of us will walk on you before you get anything worth while out the door.

2.x CE is a prime example of killing a project off by telling devs to go take a flying flip.

But I guess that's your choice since you seem to think you don't need them anyway.

Good luck.

[alienjeff]

@ttuuxxx

I thought you were on vacation, ffs ...

[ttuuxxx]

disciple - DC - you are quite wrong. Puppy does not NEED by definition to stay developer focussed.

It doesn't have to but most of us will walk on you before you get anything worth while out the door.

2.x CE is a prime example of killing a project off by telling devs to go take a flying flip.

But I guess that's your choice since you seem to think you don't need them anyway.

Good luck.

DC puppy needs developers who are active more and more, If we had 10xzigbert or 10xMU or 10xDisciple etc Just about anyone one of them, we would be light years ahead of everyone else.
Look how much Pburn has evolved this year, what an excellent little application it is.
Plus having more developers around spreads more knowledge in the forums and some of us are just like sponges, just waiting and wanting to learn more.

Why ever they told the developers in 2.15ce to take a hike, well that was then. This isn't going to be like that, at least I hope not
Developers are a hard thing to come by, Usually even harder to keep. LOL
Anyways if you want to be a team player, Join in and have some fun and share some productive thoughts.
Whats used is used and what isn't, isn't. Just the way it goes.
ttuuxxx

[DigitalCrypto]

Base Team (function)
Who wants to compile packages
who wants to compile the kernel
Who wants to build it
Who wants to alpha

In T2 dev this is one and the same person.

How do you expect to glue multiple T2 builds together and make sure everyone has the same thing?

You are going to CVS/SVN the SVN of T2? Sounds like a nightmare.

[ttuuxxx]

@ttuuxxx

I thought you were on vacation, ffs ...

I am, I'm working on a couple of non-puppy related projects. But I'm still staying up on top of the politics, But not helping in the forums, So I can stay focused.
ttuuxxx

[alienjeff]

@ttuuxxx

Just out of curiosity: what is your native language?

[tronkel]

Disciple wrote:

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.

I more or less agree with everything Disciple has said in this thread.

Ask yourself this question:

If Puppy Linux had never existed, could I have built this operating system?

Unless you can answer yes to that question, you could not be the leader of Puppy Linux. The truth is, that no-one in the Community can answer yes to that one. Even the best of the programmers/developers do not have the innovative and implementation skills and genius that Barry has. Makes no difference whether you are a "developer/applications-programmer" or simply a technician who can do things like graphics.

In order to try to find such genius, we are forced IMHO to look outwith the Community. Would anyone from another respected distro in the same (or similar) distro sector as Puppy be prepared to take Puppy under its wing? Can you think of such a white knight?

[cb88]

puppy will require a real developer to continue i believe we have a few around

if your not a developer then the best you could ever do is some form of remaster. Note: i am mostly still at this level myself but working toward being a developer

The svn i have setup on sourceforge is not just for puppy but for the scripts ie grafburn pfind etc... it would be very helpful to have those all in one place instead of haveing to sift through the forum for new versions i don't think that the developers would even have to upload them themselves but anyone with svn access could (i'll have to add you)

note: anyone can download from svn

ttuuxxx has mentioned that T2 7.0 has problems ie broken links in the source downloading ... to fix that we could use the SVN or wait untill 8.0 is released supposedly in the late summer/fall then download all the source and host it ourselves which is ideal anyway since we could all be working from the exact same source

iirc you can just manually extract source into T2 and it will find it but im not sure about that and im not in a position to be able to check

personnally i am very pro T2 since i can't really see ever automating a gnomeslack build of puppy....but it might be possible

[dogone]

As a politician would say, "I both agree and disagree a little about a lot of what you said."

Every one of the preceding posts in this thread contains some wisdom. That's the way it works here @ Puppy. Everyone contributes something, whether they be a gardener, music teacher, engineer or couch potato. Puppy has survived by virtue of Barry's leadership and wonderfully active and open communications.

In my book, people are not "developers", they just do "developing". And anyone who contributes their time, ideas or experiences to this forum is helping to "develop" Puppy Linux. So don't start looking around for developers to save Puppy. They're already here and they are for the most part, "us" - you and me.

Puppies don't have rank, they don't wear badges or power ties. Puppies play, and they play hard. We are all here because we love the experience, the people and the OS. Puppy Linux is meant for us to enjoy and learn from and we can do that with any version. Even if there is never a version 4.2, Puppy will live on for years. We have time to reflect and time to plan.

Reading the previous posts, I'm starting to hear people talk about which aspects of the Puppy project have worked well and which have not. What are the community's strengths and what are its weaknesses? What do we want to do with Puppy? I think these are the questions we must first ask ourselves.

[ttuuxxx]

I Have often wondered if we could split puppy into 2 distros, That it could actually be smaller and work better.
With T2 if that stays the norm, we should be able to compile all base packages at one including the kernel, Thats how its done now with the latest version that I'm working on. With some adjustments we could have a Smaller Xvesa model and a larger Xorg model, The Xvesa model could be for really older computers and hardware support, i386,i486 and the newer one i586,i686.
Another thing that has always given puppy a black-eye is logins, So many post on the forum about being root 100% of the time. I don't have any problem with being in root, actually it would be one extra step every time I boot up puppy that I don't have to do now. But for the users to feel secure would be nice, Also it would make a puppy server much more secure.
We have had attempts in the past at building servers, Maybe that could be part of puppy's future also, If we get the login/permissions implemented and if the team is large enough, It would only seam natural to also build a server edition. That would be a developer intensive release. Maybe post on the main homepage "Looking For Developers" I've seen other distro's doing this, actually arch Linux has a post on the homepage for developers. Not for this sort project.
These are just Ideas I'm tossing around.
ttuuxxx

[WhoDo]

2.x CE is a prime example of killing a project off by telling devs to go take a flying flip.

Why ever they told the developers in 2.15ce to take a hike, well that was then.

Wrong!!!! At no stage were any developers of 2.15CE told to "take a flying flip" or "take a hike" or anything of the sort!

Perhaps you are thinking of the Talking Stick CE project - based on 2.14R and coordinated by ecomoney (Robert S). I don't know what the devs were told on that project; I wasn't involved.

2.15CE remains arguably the most successful community project - some might vote for 1.09CE, 2.03CE or even 2.16CE, but my view is they were basically puplet projects following community ideas rather than a true community development project. I coordinated 2.15CE and I can assure you, I spent most of my time encouraging, assisting and supporting the developers who brilliantly donated their time and energy to the project.

[ttuuxxx]

2.x CE is a prime example of killing a project off by telling devs to go take a flying flip.

Why ever they told the developers in 2.15ce to take a hike, well that was then.

Wrong!!!! At no stage were any developers of 2.15CE told to "take a flying flip" or "take a hike" or anything of the sort!

Perhaps you are thinking of the Talking Stick CE project - based on 2.14R and coordinated by ecomoney (Robert S). I don't know what the devs were told on that project; I wasn't involved.

2.15CE remains arguably the most successful community project - some might vote for 1.09CE, 2.03CE or even 2.16CE, but my view is they were basically puplet projects following community ideas rather than a true community development project. I coordinated 2.15CE and I can assure you, I spent most of my time encouraging, assisting and supporting the developers who brilliantly donated their time and energy to the project.

I'd have to agree with WhoDo on that one, he was one of the key figures for that release, Plus its been awhile since I've booted it, But it had developer written all over it, like the blinky script was rewritten for icewm by I think, MU, the menus were rewritten by another developer, plus other advancements like the "flash patch" etc. I do think a lot of ground breaking development was build by developers and others for that excellent release and if we look at that as a project model for future Puppy releases we would be doing an excellent job.
ttuuxxx

[Auda]

Way back when, I used to loging here every day and read what was happening, not any more its all rush rush to get the next version out bugs and all.
Puppy 109ce was great I still use and install it on mates mates computers. To be honest I havent tried 2.15ce but most of the other versions have seemed to be unfinished and buggy many little things like not samba printing and cut and paste either not working or not well.
Hopefuly without Barry in the lead things will change and puppy can be seen as a stable distro that "just works" rather than, just about works.
Auda

[puppyluvr]

Hello,

@Ttuuxxx, I agree completely.. 215ce is a shining example of a community effort, doing the best with what was available. I cant wait for the 3.xxCE.....

To the point at hand...My question is...Who runs puppy now??? In almost a year here, Ive never seen Barry say "No, you cant do that, do it My WAY.." Barry wasnt a "ruler" he was/is a developer. He will continue to develop, no-one that brilliant just stops....
Puppy is a community, and should be governed as such...

If Barry had said, instead, "Im taking an extended vacation, but Ill be back", would it have created so much drama, NO.. The finality of "retire" sent everyone off. If we thought we had to come up with an "Interim" management system, rather than a permanent one, I believe we would have pulled toghther, not divided.

Puppy doesnt have to change at all. We may not have a developer of Barrys status, IDK, but if we Pull Together, not Apart, we can manage. And its not like Barry is dying, he`ll still be around. For Gods sake people, a great man has given us a great task, to pull together and carry on with his "Baby", the best damn distro around. That speaks of a lot of trust. Will we let him down???

Im no developer, (Duh) but Ill do whatever I can..And, I will learn. I am devoted to Puppy, So are Many others. I dont think Barry intended Puppy to be the "End All" Linux.
( Ironically, I believe the innovations Barry buitl in to Puppy may make it that anyhow!!)

We shouldnt be trying to guess what Puppy will be, We should continue to develop what Puppy is, to refine it, to fix all the things mentioned in this thread. The future will be there when we are ready. WE have Puppy now, Puppy is what it is, and what we make of it.
Puppy can, and will, go many different ways. Nothing new. I trust the "community" with the "official releases". The new kernels will come, lets make the most of the ones we heve first, and move on at a reasonable pace, to allow getting the most from what we have. I thought 3.01 had a long life ahead of it, but.....

Oh yea, SHAME, SHAME, SHAME on all those who made our forum negative....
I hope John deletes the whole thread.....makes Puppy look bad.

Solidarity, thats the answer...
United we stand, Divided we suck....
But what do I know...........

[ttuuxxx]

Hello,

Solidarity, thats the answer...
United we stand, Divided we suck.... <--- That made me laugh
But what do I know...........

[disciple]

But I guess that's your choice since you seem to think you don't need them anyway.

You could try reading DC. I've consistently said that I believe Puppy DOES need to be developer-focused to remain successful. You on the other hand were implying that because Barry's mission statement made Puppy developer focused when he was developing it, this isn't allowed to change when he stops. Of course it's allowed - if he allows it; even though I don't think it is wise.

Another thing that has always given puppy a black-eye is logins, So many post on the forum about being root 100% of the time.

Are you sure? I haven't noticed that. I guess in theory it would be good if it was easy to set up more users. BTW do you refer them to the excellent article Nathan put on his blog as an attempt to encapsulate why root is fine?
I actually think you mentioned the most likely reason for a fork before - Slackware compatibility.

I do wonder about all of you complaining about bugs and instability in recent releases though. There were significant bugs with CUPS and SAMBA in 3/3.01, but apart from that??? How many bugs does 4.0 really have? Taking dotpup handlers and stuff out isn't a bug. If lots of people need help figuring them out, it is a documentation problem, not a bug. But maybe the problem is really that people know Puppy "just works", so they don't read the release notes

OOPS - you've all dragged me completely off-topic

[technosaurus]

Coding/developing - One of the main issues that will have to be dealt with is code documentation which is written in BarryK - don't get me wrong, its all good code but the documentation was not intuitive to understand - it took me hours to partially understand the initrd coding- I learned a lot but afterwards I kept hearing the puppy login bark in my head - woof- woof There it goes again

packages - slackware compatibility was mentioned earlier but it could be any distro really as long as there are sufficient packages, maybe just use T2 and pacman, entropy or some other package manager that can compile from T2 (or whatever distro) source if a package is not available from puppy repos
also .pup, .pet or .sfs? or go to tgz, deb, rpm...? also kind of important that we choose a standard - preferably one that is easy to maintain so that we can increase our package selection to be somewhat competitive

direction - use the forum - ideas/suggestions and polls to achieve user input similar to brainstorm.ubuntu.com - granted we will still need capable developers to implement them but I know we have a lot of talent out there

More revenue streams (in addition to those previously mentioned) similar to damnsmalllinux.org, but use eeepc, ebox and DecTop systems (available from wdlsystems and dataevolution respectively) - both work with puppy, there is an eeepc and ebox version already and I will donate one of my spare DecTops to a developer who can make a DecTop optimized version compiled with gcc 4.3 (for geode optimization) and including the geode xorg driver, ltmodem, rtl8150, cs5535 etc... I tried - no luck so far

Developer rewards by forum nomination poll or as percentage of donation

more puppy versions for distro hoppers - more interest=more site visits = more potential revenue
<50MB(biz-card CD), <96MB(128MB flash), <192MB(256MB flash or mini-CD), <300MB (biz-card DVD), 700MB(CD), 4GB(DVD = all versions + puplets multiboot) - I won't delve into the opera-seamonkey-firefox debate etc.., but more release announcements typically brings more interest - release versions every 2 weeks = 12 weeks continuous on the distrowatch charts - 36 weeks if you have a free(seamonkey+gnash,486), non-free(opera+flash,586/geode) and firefox(686) version - released in order of size (basically just add packages) for easier development - add 3 alphas, 3 betas and 3 rc1 and 3 rc2 and you basically have a 1yr release schedule between major version changes

version the forums so that people discussing puppy 2, 3, 4 and later versions don't get confused by inapplicable info

wiki documentation team - great volunteer project for many puppy users that want to contribute

puppy promo team - to irritate Bob Barker and spread puppies everywhere - no spaying or neutering here

website maintenance - there are several puppy sites - many old ones that are no longer maintained or nonexistant and many dead links - we definitely need a lead maintainer and possibly a team of volunteer helpers

lead developer? or head developers? it would be an easier transition if someone who understood all of the puppy inner workings could take the lead since versioning could be tracked on their system - multiple contributing developers (more likely?) would require a tracking system - something to think about anyways if sticking with T2