Phoenix CE Base Version for next Community Edition

[alienjeff]

Before I do anything 2.14 specific, I just want to check: Is it final that we're using 2.14R? If so, should we still be using 'TS3.03CE' as the tag?

I trust Dougal and Pakt have been approached and consulted with on this change of course for the CE, as it may have some affect on their current and future 2.14R work. Project managers from earlier CEs should be consulted, too, as they've been down this road before and could share their experience (and nightmares).

Have any specs been decided upon? IIRC, 2.15CE had a lofty original goal of being a sub-100M ISO, yet after a barrage of requests of favorite apps, features, et al, it ended up being 30% larger. Any thought to this?

Speaking of specs, how about a strict minimum system requirements statement? Does the CE need to be able to run on machines that still sport carburetors? It's a large beach, so where is the line to be drawn in the sand?

[JohnRoberts]

In a small, 8cm, multi-boot LiveDVD I always keep close there are:
> 2.03CE
> 2.15CE
> 2.14R
> 2.17.1

Crossing out the obvious (2.03CE and 2.15CE), my best picks are 2.14R and 2.17.1. They both look solid candidates for a new CE...

An important issue is OpenOffice. I have successfully used the OpenOffice sfs file on a USB stick. Trying to remaster this to a LiveCD ISO though, did not work properly. In the LiveCD with OpenOffice made this way, Calc crashes (the rest of the OpenOffice apps work Ok...). Bottom line is, that it will be important to be able to properly integrate OpenOffice in the new CE (i.e. a LiveCD) if it is not included a priori like 2.03CE...

[tempestuous]

I discovered this thread late, but I just wanted to add my vote for the idea of basing the next CE version on 214R.
214R has had its foundations seriously tweaked and improved by Dougal and pakt. I believe that 214R is developmentally the successor to Puppy 2.16 (2.17 has a different kernel) and is an excellent basis for the next Community Edition.

[WhoDo]

I believe that 214R is developmentally the successor to Puppy 2.16 (2.17 has a different kernel) and is an excellent basis for the next Community Edition.

Except there is already a 2.16CE in pre-release. That kind of leaves nowhere to go/grow in the 2.1x series. And then there's the 3.xx series with no Community Edition and heaps to recommend it as a base. Sure it's hard work, but the 3.xx series needs consolidation (usually a function of the CE version) and it sits apart from both the 2.1x series and 4.xx series (Dingo).

[mbutts]

WhoDo I think sums it up quit eloquently. Personally, if we have 2.14 and 2.16 CE editions, I think thats great! I also hope that 3.xx won't be abandoned in favor of Dingo and hope that a CE will be offered later on. I think Barry said somewhere in a early post that it was not the purpose of Dingo to replace anything going on in the 3.xx series but to be a bold attempt to try something new.

New official releases have slowed in recent months. I can remember when it seemed like every other day there was a new 2.xx Puppy . It was amazing how fast they were pumped out and how much development occurred in a short period of time. Several members took it upon themselves to develop unique derivatives with specific functions and appearance. It's great to see the hard work of the 2.xx and 3.xx being implemented for specific uses.

Barry is working hard to build the foundation for Dingo and this gives the rest of us time to polish what he and the other developers have given us in the 2.xx. and 3.xx series. Since the official releases have slowed, it has given the members time to pick a favorite Puppy to polish. I know I got to the point that I quit trying to customize a release of Puppy for myself because half way through any attempt I was greeted with a new Puppy and I would try it and see what it had to offer for new features. Then the 3.xx series came out. More cool stuff..... maybe I should wait and see what comes out next... I was like a kid in a candy store.... decisions, decisions.......

Yesterday I was working on a friends computer. I needed Puppy for something and I popped in a customized version of Puppy I made for them. I was amazed how cool it was and what I had incorporated into it. I thought it was either 2.16 or 2.17. When I looked at the version I was surprised that it was 2.14. Next time I am at their house, I'm making a copy of it for myself since I didn't save one for myself when I whipped it out for them.

When 2.14r came out, it was a great reminder that there is a lot of potential in old Puppy's for development. I try to download any members release and save it for future use here at my place. A good example is Watchdog. It's one I want to use someday for security here at the compound. I appreciate the fact that someone took the time to build something that I want to use someday!

Right now, with all the hard work of the members to customize Puppies, it's like having your cake and eating it too... then, have someone offer you a slice of pie, and then ask what kind ice cream you want on top of it too boot. It's a win-win situation. And just like with pie and ice cream, if you don't have room for it now, you can always wait and have it later.

[nic2109]

It's true that Puppy 3 has stalled somewhat which is a shame. But there's going to be plenty of time to build a CE based on it. Right now the sap is rising for 2.14R so let's stick with it.

It seems to me that the thing we should be aiming for is stability and supportability as well as a wise selection of apps. Next; it MUST all work straight from the box and not be too scary for Windoze refugees/visitors. And lastly; it needs to look good too. Those first impressions are so important. The default theme etc needs to look like Vista in my opinion if we want to win over sceptical hearts and minds.

Where things get subjective is where we'll differ, but ecomoney will need the casting vote. It's a bit like informed consent in a medical context. The decision needs to take into account both the specialist's recommendation and the reason(s) for it, and the patient's preference and the reason(s) for it. Then the patient decides - and ecomoney is the patient in this analogy.

On the choice of apps: I have surprised myself to realize that I hardly use Open Office, but find Abiword or Gnumeric quick and easy. So my vote would be for the light-weight apps with clear and easy instructions for using the .sfs package. Office is typical bloatware with 80% of the features unused, so only include it as a user-requested option. I like the look of sylheed email, and the skipstone browser is also pretty impressive.

With judicious choices Puppy214R_CE (or whatever we need to rename it as) can still be suitable for old PCs rescued from skips AND look up-to-date.

[MU]

When I presented Build-a-Muppy, it was my intention to offer a system, that allows to build customized versions of Puppy, including fixes and enhancements we achieved in the last months.
This includes patches for initrd.gz by Pizzasgood and lots more.
It also has some of Puppy scripts localized.
I'd be glad if people would use it as a base for a CE version, and contribute further updates.

For example since 4 days I fight with Cups - the version I added does not work.
I made some progress, but it will take until weekend, until I can upload a fixed version.
Such annoying stuff keeps me off from trying other things, like adding compbiz.

Barry pointed out clearly, that he will concentrate on a complete rewrite of Puppy (Dingo), so don't expect a 3.02.
The 3-series with best slackware compatibility certainly already is a comunity-project, and will be continued like this.

ttuuxx has created brilliant addons to his firehydrant, NOP has added compbiz, Muppy is focussing on localization, and WOW has compiled the Gnome-libs and Gdeskletts (just to mention some, but far not all working actively on 301).

So instead of waiting for a 3.02 we now should try to combine the efforts that already have been achieved by the 3.01 based puppletts.

Mark

[Lobster]

It makes sense to me to do as Mark (MU) and Warren (Whodo) have suggested.

The reason for the project leader is to make decisions and implement them. Talking Stick was conceived at a point when 3.02 was expected in a short time. It was conceved as a 3 based Puppy.

The bulk of the work of putting an Alpha of Talking Stick together will fall on Robert. So it his decision and he has to be happy with it..

If he has the capacity (this has been done previously) two experiments for comparison can be released.

One of the original ideas of Talking Stick was to base it on an existing Puplet.

[WhoDo]

The bulk of the work of putting an Alpha of Talking Stick together will fall on Robert. So it his decision and he has to be happy with it

I understand perfectly well where Robert is coming from. I understand his focus. It is a laudible objective he is aiming for, but it is not what a CE edition is about IMHO.

The first half of my suggestion is that Robert does what he wants, and uses 2.14R as a base for a revised and updated 2.15CE - that would meet his objectives while still keeping the CE lineage intact.

OTOH, the project Robert was originally asked, and agreed, to lead was a Series 3 Community Edition. By no stretch of anyone's imagination can that therefore be based on a Series 2 Puppy!

The second half of my suggestion is that those in the community who wish to do so, work on a bug fix consolidation of 3.01 that will become 3.02CE when we are comfortable with that. It surely wouldn't take too big an effort to achieve that. It's happening already. Just ask MU, Pizzasgood, Tronkel, et al.

There is no reason why both projects cannot co-exist, or share technological advances and script changes/fixes for that matter. Develop it once and use it twice, if you like. There are already some great things that need to find their way into 3.02CE from 2.14R, and from the work of contributors like MU, Wow, Tronkel, Pizzasgood, etc.

I hope that will help put the "baby" to bed, instead of throwing it out with the bath water!

[tronkel]

WhoDo wrote:

I understand perfectly well where Robert is coming from. I understand his focus. It is a laudible objective he is aiming for, but it is not what a CE edition is about IMHO.

Is a version of Puppy that is customised for an internet cafe the same thing as a community edition? I'm not sure what the answer to that is. Maybe what Robert is doing ought to be called 2.14IC (IC= Internet Cafe , lol.....)

Seriously though, there could well be a serious demand for internet cafe customised Puppy versions.

Great idea about a 2.15CE using 2.14R as the base - that's a terrific idea.

[Lobster]

The first half of my suggestion is that Robert does what he wants, and uses 2.14R as a base for a revised and updated 2.15CE - that would meet his objectives while still keeping the CE lineage intact.

That seems about right and a good place to start . . . if Robert agrees?

If anyone wants to be project leader for a 3.01 based community edition, please step forward.

Two community editions would then be possible . . .

[tronkel]

How about this for a plan of action?

In the time interval between now and the appearance of Puppy 4 Final:
1) Complete 2.14R
2)Apply this as 2.15CE on steriods by WhoDo
3)ttuxxx plus others who are interested takes on a face-lifted 3 series similar to Fire Hydrant.

All of the above are very worthwhile.

This clears the decks for Ecomoney to project-lead a Dingo 4 CE with all the help he needs.

[ecomoney]

First off, 2.14r is essentially all of the features of 3.xx with the older kernel, and therefore more hardware support. It has also been extensively bugfixed, and more useability features added. There nothing old about it apart from the number...2.14. It has more akin with 3.01 retro than 2.14, or 2.16 which has a different kernel. The only thing it lacks in terms of programs from 3.01 retro is the updated pdf reader which I have asked to be included, and the lite browser (Im in discussion with the developer of skipstone to see if he will make some useability changes).

Secondly, I work with puppy all day every day. I run a testbed cybercafe at a local cybercafe, which also takes in old computers, reconditions them and gives them out to the local community, and are used by AVERAGE USERS. We have nearly a hundred computers running puppy in our local community, including homes and businesses. The project is about far more than the cybercafe! I believe this, together with my formal university training and experience as a professional programmer in industry and local government, gives me a good insight into what most people require of their operating system.

Ive not been idle over xmas, I have been testing 3.01 (buggy and unstable), 3.01 retro (stable, but still buggy), and 2.14 ce (refined, tested, with plenty of support, will run on older computers). 2.14r is head and shoulders above the others in terms of what people want, based on my own experience. If people want to develop operating systems for themselves that is fine. My understanding of what a COMMUNITY is however, means users, as well as developers.

Lobster, I believe my background in the community is the reason why you asked me to head up the community edition in the first place. You obviously had faith that I would make the correct decisions at the time based on that experience, or you would have created the community edition yourself. Why are you now changing your mind about this just because you disagree with a decision you made me responsible for making?

Whodo you told me when I asked you for guidence that at the end of the day, I was the one making the final decisions (after consultation) because I am the one most qualified to do so. I was asked to create a community edition, which version upon which it was based was not stated, so I am deciding based not only on the opinion of many of the forum members, but also upon the opinion of the community of users around me IRL.

Its been my experience as a developer myself, that to create a good computer program, I must put aside my own personal wishes, and create something that fits the requirements of the people that will eventually use the program. The reason that most computer software projects fail is because this doesnt happen.

The best community edition possible, would be one based on 2.14r, with the features identified through user testing as part of the talking stick project. When I was asked to do this job, I took it as a serious responsibility, and I want to create the best possible result I can. As for the name or the number given to it, it really doesnt matter to me.

[Leachim]

I do not recommend any base edition here. I only want to line out the major difference (as to my knowledge) between Puppy 2.x and Puppy 3.x: glibc!

glibc is the most important library of an Unix system. Puppy 2.x contains a really old version of that library. So many old code is willing to run with Puppy 2.x, but lots of newer code rejects cooperation. On the other side nearly all new code should be able to link with Puppy 3.x' glibc but maybe not with the old one of Puppy 2.x.

I currently try to work around this problem. I use a Puppy 2.15CE based system and want to integrate new software that cannot be run with the old glibc. So I want to install two glibc-versions - one for old code, one for new code. Currently I try to figure out the necessary compilation options.

[Pizzasgood]

I don't really care what version gets used myself. Quite frankly, I won't be using it regardless of which is used. Not because of any important reason or anything. I just prefer Pizzapup. It's starting to get a little aged, but when I finally get caught up on all my commitments I'll be taking a break from using Puppy anyways, to try out some other distros (particularly Gentoo and Gobo). By the time I get done with that, 4.xx ought to be rolling along nicely, so I'll probably start fresh with that. Then when I get the time I'll start on the next Pizzapup. That won't happen until at least summer though, possibly later.

[Lobster]

Lobster, I believe my background in the community is the reason why you asked me to head up the community edition in the first place. You obviously had faith that I would make the correct decisions at the time based on that experience, or you would have created the community edition yourself. Why are you now changing your mind about this just because you disagree with a decision you made me responsible for making?

I fully support your decision to make CE for us and to make it reliable and to base it on what is the best available base.
Now it needs to come together. We have a base 2.14r. We have made that decision. We have a browser, Seamonkey. We have made that decision. Initial programs are available.

We now have to start building it. Telling us what we are doing. When an Alpha is available for feedback and so on

If you click on your first thread in this post, you can change the name of the thread to TalkingStickCE (or something more appropriate.)

We are also at the stage were a project leader for a community edition based on Dingo is also required. So I am interested in who is up for that . . .

[Pizzasgood]

Progress report: I have the technical stuff for syslinux/isolinux done in 3.xx now. The actual images and such will need tweaking, but it all works and I have the install script configured to use it (thought haven't had a chance to test that yet).

I'm done for tonight. Tomorrow I'll go back and work on compiling Pebble statically and maybe using uclib. Then I'll document all this stuff and upload it. Then depending on the time, I might start porting it all to 2.14R. I might have to wait until the weekend for that.

[WhoDo]

...[snip]...takes in old computers, reconditions them and gives them out to the local community, and are used by AVERAGE USERS. We have nearly a hundred computers running puppy in our local community, including homes and businesses....[snip]...My understanding of what a COMMUNITY is however, means users, as well as developers.

Where I think you are missing the point, Robert, is that the "Community" in "Community Edition" is not the user base of Puppy, or any other Linux. It is not your local community of Puppy users either. It is the community of Puppy users who frequent the forums, websites, etc and CONTRIBUTE to the Puppy development cycle. That means by testing, posting bugs, fixing the wiki pages, answering newbie questions, hosting downloads and torrents, etc. etc. THAT is the "Community" of which we speak.

You have broadened the definition to ALL Puppy users because that suits what YOU want to build. And because you are willing to do that, no-one is likely to disagree with you in any determined way ... except someone like me.

That's ok. Disagreement breeds discussion, reflection and hopefully consensus.

Lobster, I believe my background in the community is the reason why you asked me to head up the community edition in the first place. You obviously had faith that I would make the correct decisions at the time based on that experience, or you would have created the community edition yourself. Why are you now changing your mind about this just because you disagree with a decision you made me responsible for making?

I don't believe this is about Robert "ecomoney" Simpson, or at least it shouldn't be. If you were to take a more objective view of the issue, you might be able to see the problems. By all means develop your stable, reliable, bullet-proof, RetroPuppy. Use all of the resources the community is willing to offer. Call it 2.15CE-Revisited if you like. The point is that is not what the original CE charter that you accepted was about. Is it hard to go bugfixing in 3.01? Sure. I know how hard it was in 2.15CE, so I'm not minimising the issue.

Whodo you told me when I asked you for guidence that at the end of the day, I was the one making the final decisions (after consultation) because I am the one most qualified to do so.

No, because you are the leader of the project AS IT WAS OUTLINED. There is more than a semantic difference there. Let's draw an analogy you will recognise. A client asks you to develop a program to do a certain job. You accept the commission but part way through you see how big a job it really is and you get jelly-legs over it. So you decide to change the goal post and hope you can convince the customer what you want to do is better than what he asked you to do. Good luck. Sometimes you'll succeed but most times you'll be sacked, and probably not paid for what you've done so far. Is that a fair analogy of what's happening here?

I was asked to create a community edition, which version upon which it was based was not stated, so I am deciding based not only on the opinion of many of the forum members, but also upon the opinion of the community of users around me IRL.

Umm...sorry but that's not true. Have you visited the CE wiki page at all? Does it not say a "Series 3 CE Edition"? Did you not agree to that and start examining the approach to TS3.03CE? Heck, I'm darn sure I didn't imagine all of that.

Ok, so you decided a 3.xxCE is too hard for you. Fine. Do what you want WITH the community's open and transparent approval. Hold a proper poll. Don't just base your decision on your own predisposition and the plaudits of one or two vocal members. Gee, I sat through hours and hours of IRC chat sessions, emails, PM's, forum threads, etc before I made the final choices. You tried 3.xx on your cafe machines and decided it's too "buggy". Well, duh! That's what is your project leadership role to coordinate the fixing thereof ("Ending a sentence with a preposition is something up with which I will not put" - Winston Churchill ).

Its been my experience as a developer myself, that to create a good computer program, I must put aside my own personal wishes, and create something that fits the requirements of the people that will eventually use the program. The reason that most computer software projects fail is because this doesnt happen.

...but, Robert, the CE edition isn't for the "people" in general; it's for the "community" in particular. That's my point.

The best community edition possible, would be one based on 2.14r, with the features identified through user testing as part of the talking stick project. When I was asked to do this job, I took it as a serious responsibility, and I want to create the best possible result I can. As for the name or the number given to it, it really doesnt matter to me.

Well "best" is a subjective term, Robert. If you really don't care about the name and number then I suggest you finish the job and call it 2.15CE Revisited! That way you get your cake and those who want a 3.xx series CE won't have to eat it! Someone else with more, time, energy, determination or whatever will come along to take on 3.02CE, I'm sure. It needs to be fixed; especially since Barry has clearly stated that 3.xx Puppy WILL coexist with 4.xx Dingo in the development cycle, while we all KNOW that the 2.xx series is at an end, developmentally speaking; there will be NO 2.18 version from Barry.

Look, I am truly sorry this has blown up like this. I would have preferred if you had responded to my last PM on the subject. I believe you didn't because you didn't want to "hear" what I "said". It was too far removed from where your mind was already set at the time. That's a pity. I'll say no more about your decision unless you persist in treating it as the original project you were invited to lead. It isn't. That's ok. It's a different project. No problem. Let's leave it at that, ok?

[Lobster]

Let me put this in another perspective
Not many were willing to take on or had the skills to create an ISO
in the early days . . .

The first community edition 1.09CE, was very much based on Nathan's efforts but with help and feedback from others
He went to create Grafpup

By the time 2.03CE was created, Hacao just went away
translated his version into English and that was the CE

This was the '2.03CE charter'
http://puppylinux.org/wikka/MeatyInfo
What we got and people were pleased with, was a stable Puppy with OpenOffice. There was practically no interaction from the community.
We were a fledgling Puppy.

You kindly stepped forward to help create 2.15CE
You were not well known to the community,
which was a risk that proved worth taking
The first Alphas were massive but the final version was a triumph

I do not expect Robert to manage the project
the way I always like or to make the decisions you would
prefer.

Ecomoney (Robert) deserves a degree of freedom and support.
Just as Barry chose to use JWM through thick and thin
(and it is now a proven window manager)

However I do support the decisions he makes and has made
on our behalf.
That is to use a stable version of Puppy + Seamonkey.

If you decide you would like to be project manager
for DingoCE, I will support you (in my own style)

I would suggest (thinking back) that once decisions start to
be implemented, we need to be with a project . . .

So . . .
Which version of Ezpup should we use?
or should we use the new themes in Dingo JWM as default?

[tronkel]

All this who-does-what, to Puppy Linux version x.xx (CE or base) is a side issue IMHO.

Lets look again at the technical situation and where it stands at the moment.

1) 2 and 3 series Puppies are 2 different animals. The most important difference between then is that 3 series has Slackware compatibility. Nothing can be done to a 2 series Puppy version that will turn it into a 3 series version without building it again from scratch.

2) This puts Puppy 3 series in a special position. Even when Puppy 4 becomes a reality, there will be many users who will be happy that in case of necessity, they can always return to 3 series if need be. Same for 2-series afficionados.

3) Given this situation both 2 and 3 series projects should be left an a "stable" state so that any user at any time can go back and select any of these versions as suits their needs at the time. Therefore, both 2 as well as 2 series projects need to be done-and-dusted and signed off, ready to clear the decks for Puppy 4 and its inevitable CE version. The time between now and the arrival of Puppy 4 is the window of opportunity for this.

4) The question of who leads the CE project version of whatever base version, is really determined by whoever has the time, willingness and expertise to undertake the task. It does not have to be a single person who does this. This work could be divided up 2 or even 3 project managers to lighten the load on any one person. What the version no designation of any of these terminating projects is, "ist egal" IMHO

5) I think also, the community as referred to in Puppy CE really refers to the Puppy community. This is a sub-set of the wider community at large. Puppy Linux will never be able to be all thing to all men in the wider community.