Barry Kauler announces his retirement from Puppy

[James C]

I'll just concur with most everything in jamesbond's above post ....... I'm also certain that the control/responsibility of Puppy Linux will be passed on in an orderly transition.It's not like we haven't been through this before......

[tronkel]

One can only assume that he is "quitting" Puppy Linux because he feels the need to adapt to a changing technological world. Ten years is a lot of time for anyone to invest in anything to do with technology. If I were in his shoes, I would also probably get to the point where I had a gut feeling that a change in emphasis is long overdue. One can get burnout as well after such a long stint. It's no bad thing IMHO. Taking a break from the usual routine is often a necessity - that's simply the way of the world.

No need to get concerned that users will simply get abandoned just yet. As time passes, the PC will slowly fade away in any case - to be replaced in due course with far more advanced stuff. When that time finally comes, Puppy Linux - along with all the other OS's that we are familiar with today - will simply become irrelevant and be relegated to the history books. That's what is known as progress.

[Burn_IT]

Perhaps he decided to make the change while it was his own choice and he will still be around to guide the transition.

[sunburnt]

In reality the std. base Puppy isn`t changing all that much. It`s reached a fixed format.
Many variant builders have pushed the envelope much further and advanced Puppy.

I`ve suggested that folks of like thought and interests form groups to develop branches.
This applies to the base Puppy distro. also, but it should be under more scrutiny.

Building distros. is a huge job, and they must have apps. Average users don`t compile.

A group would be a few doing the Puppy variant, and many app. builders contributing.
The app. builders could contribute to the core and many of the different Puppy variants.
Variant groups can concentrate on their O.S.s. App builders specialize in app. types.

# Suggestion: A new forum post section for forming and organizing groups, "Want Ads".
Hopefully a few current variant developers will group and get app. builders to support.

This way Puppy can grow and improve reliably. And most important... Lots of apps.!!!
.

[Karl Godt]

As time passes, the PC will slowly fade away in any case - to be replaced in due course with far more advanced stuff. When that time finally comes, Puppy Linux - along with all the other OS's that we are familiar with today - will simply become irrelevant and be relegated to the history books. That's what is known as progress.

I don't think so .

Even if people will have screen wallpapers on their walls with 24/7 internet access , a Kernel and Apps are required to manage it .

I personally don't know about linux touchscreen capabilities , but know that android should work .

[darry1966]

In reality the std. base Puppy isn`t changing all that much. It`s reached a fixed format.
Many variant builders have pushed the envelope much further and advanced Puppy.

I`ve suggested that folks of like thought and interests form groups to develop branches.
This applies to the base Puppy distro. also, but it should be under more scrutiny.

Building distros. is a huge job, and they must have apps. Average users don`t compile.

A group would be a few doing the Puppy variant, and many app. builders contributing.
The app. builders could contribute to the core and many of the different Puppy variants.
Variant groups can concentrate on their O.S.s. App builders specialize in app. types.

# Suggestion: A new forum post section for forming and organizing groups, "Want Ads".
Hopefully a few current variant developers will group and get app. builders to support.

This way Puppy can grow and improve reliably. And most important... Lots of apps.!!!
.

Yes this should be about community fits with an open model sharing the load, even average users can be testers though reporting back about software even testing betas of pre-releases anyone can contribute. The load must be shared.

[koulaxizis]

In reality the std. base Puppy isn`t changing all that much. It`s reached a fixed format.
Many variant builders have pushed the envelope much further and advanced Puppy.

I`ve suggested that folks of like thought and interests form groups to develop branches.
This applies to the base Puppy distro. also, but it should be under more scrutiny.

Building distros. is a huge job, and they must have apps. Average users don`t compile.

A group would be a few doing the Puppy variant, and many app. builders contributing.
The app. builders could contribute to the core and many of the different Puppy variants.
Variant groups can concentrate on their O.S.s. App builders specialize in app. types.

# Suggestion: A new forum post section for forming and organizing groups, "Want Ads".
Hopefully a few current variant developers will group and get app. builders to support.

This way Puppy can grow and improve reliably. And most important... Lots of apps.!!!
.

Yes this should be about community fits with an open model sharing the load, even average users can be testers though reporting back about software even testing betas of pre-releases anyone can contribute. The load must be shared.

+1

[koulaxizis]

Linux Mint has done a great work on community feedback, maybe we can do something similar... http://community.linuxmint.com/

[Iguleder]

In reality the std. base Puppy isn`t changing all that much. It`s reached a fixed format.
Many variant builders have pushed the envelope much further and advanced Puppy.

I`ve suggested that folks of like thought and interests form groups to develop branches.
This applies to the base Puppy distro. also, but it should be under more scrutiny.

Building distros. is a huge job, and they must have apps. Average users don`t compile.

A group would be a few doing the Puppy variant, and many app. builders contributing.
The app. builders could contribute to the core and many of the different Puppy variants.
Variant groups can concentrate on their O.S.s. App builders specialize in app. types.

# Suggestion: A new forum post section for forming and organizing groups, "Want Ads".
Hopefully a few current variant developers will group and get app. builders to support.

This way Puppy can grow and improve reliably. And most important... Lots of apps.!!!
.

Yes this should be about community fits with an open model sharing the load, even average users can be testers though reporting back about software even testing betas of pre-releases anyone can contribute. The load must be shared.

+1

+0.75

Until now, Barry and project leaders took care of that "standard base", while individuals contributed packages and new applications. I'm against branches - I like the terminology of "teams" better: "kernel team", "UX team", "packaging team", etc'. Everybody works on one, unified project (i.e not separate branches), but each team strives to achieve all goals of the next version, in one aspect.

I think we should create some sticky topic with contact details (i.e e-mail) of all forum members who want to be recognized as developers and have both the time and willingness to contribute. Such a list should contain a list of all areas each developer wants to work on (i.e kernel, automation, package management, packaging, translation, testing, etc').

That could be nice if we want to do some evaluation of our available human resources - that's the biggest problem of community distros, after all

By the way - I think we should set up trac or something similar, so we have a roadmap for each version. Each developer proposes a feature he or she is going to implement with help from his or her team, then we measure progress using some web application all developers have access to.

It's nice for small teams, since it makes it easier for coordinators to do that "time vesus man power" calculation and answer the eternal "postpone to next release, delay this one or drop that feature" question when time's up.

[tronkel]

Iguleder wrote:

I like the terminology of "teams" better:

Terminology is only terminology - a linguistic concept.

Barry's impending retirement has implications for the future of Puppy Linux that raises some questions that are even more fundamental than the management style of the project. Sure, there will have to be some sort of team/community structure in place in order to create an ISO called Puppy Reborn or some such other similar title.

As far as can be established, there is no Barry Kauler V2 walking around on planet Earth. Realistically there is probably then no chance that anyone would be able to step into the shoes of Barry V1 and carry on as before.

Would it not be better to simply accept this reality and start a completely new base for Puppy Linux from scratch? Any prospective Barry 2 would need to be a developer who has previously created a Linux distro from a clean sheet - just as Barry V1 did with the original Puppy. The only thing that would remain of Puppy original, would be its philosophy of "keep it minimalist, smart and full-featured" - in other words, a completely reborn Puppy Linux that Barry 2 plus his project members would be comfortable to work on and develop further..

The reason I say this, is that Barry V1 is really the only person alive on the planet who, is in the position of being able maintain the base of the present Puppy and also carry on with the truly creative ideas that would be required to take this distro into the foreseeable future.

Anyone whose CV fits the job description of Barry V2 please step up and make yourself known!

[Karl Godt]

Fact is : There are not many people here to develop the Puppy core : Kernel , Toolchain and compile core programs .

Me myself is not so much interested in using other distribution's binaries .

To me the puppy-4 core is good enough , does not need newest GLIBC !
I even start to use gcc version 3.4.6 on it
So far have upgraded GTK-2.0 to version 2.19.7 now .
Still using the 2.6.30 series kernel , want to upgrade it to 2.6.31 because that would show my USB-3G-Modem's Sim card and integrated micro-SD-Card slot.

If anybody is interested in Up/Downgrading Puppy-4.3.x , send me a private message !

[darry1966]

Iguleder wrote:

I like the terminology of "teams" better:

Terminology is only terminology - a linguistic concept.

Barry's impending retirement has implications for the future of Puppy Linux that raises some questions that are even more fundamental than the management style of the project. Sure, there will have to be some sort of team/community structure in place in order to create an ISO called Puppy Reborn or some such other similar title.

As far as can be established, there is no Barry Kauler V2 walking around on planet Earth. Realistically there is probably then no chance that anyone would be able to step into the shoes of Barry V1 and carry on as before.

Would it not be better to simply accept this reality and start a completely new base for Puppy Linux from scratch? Any prospective Barry 2 would need to be a developer who has previously created a Linux distro from a clean sheet - just as Barry V1 did with the original Puppy. The only thing that would remain of Puppy original, would be its philosophy of "keep it minimalist, smart and full-featured" - in other words, a completely reborn Puppy Linux that Barry 2 plus his project members would be comfortable to work on and develop further..

The reason I say this, is that Barry V1 is really the only person alive on the planet who, is in the position of being able maintain the base of the present Puppy and also carry on with the truly creative ideas that would be required to take this distro into the foreseeable future.

Anyone whose CV fits the job description of Barry V2 please step up and make yourself known!

I very much admire the work that BK did but he was not tbe Messiah I believe that a more community focused model can be used to get things going and this ridiculous who is going to be the next Barry is just that ridiculous it shold not be on the shoulders of one person but be a shared responsibiity even if your just a tester so please can we get away from the BK was God kind thinking that is harmful to the future of Puppy if he had suddenly been run over by a bus where would we be then.

So lets learn a lesson from this BK is a genius no doubt and I for one admire his brilliance but relying on one person is plain silly and that is an important lesson to be learnt.

Whether or not it is teams or whatever if Puppy is say to use an analogy a body anybody can be a valuable part of that body - last thing we need is people suffering burn out.

Finally please don't take this posting as direspect for BK because I really admire him but as I have said an important lesson here is development needn't falter because the father as people put it has left, it can be a team/commitee/whatever term effort.

[darry1966]

Fact is : There are not many people here to develop the Puppy core : Kernel , Toolchain and compile core programs .

Me myself is not so much interested in using other distribution's binaries .

To me the puppy-4 core is good enough , does not need newest GLIBC !
I even start to use gcc version 3.4.6 on it
So far have upgraded GTK-2.0 to version 2.19.7 now .
Still using the 2.6.30 series kernel , want to upgrade it to 2.6.31 because that would show my USB-3G-Modem's Sim card and integrated micro-SD-Card slot.

If anybody is interested in Up/Downgrading Puppy-4.3.x , send me a private message !

I agree Karl 4 Series could be still a good base for development with more modern kernels.

Need to be aware of bad code in some of the "Buntu" releases being inherited.

[sunburnt]

I`m not aware of any team that works close with Barry, but if there is, then these are the guys.
Other than that there`s a number of guys who have been variant builders for a very long time.

I say branches because there`s other ways of doing the O.S. thing than Puppy does it.
# Example: I`ve said for a long time that a union is an unnecessary complication.
And there`s still no PXE boot setup in the standard Puppy. Many improvements to be made.

# The part I left out above is: A dev. community is how Debian and Ubuntu does it so well.
.

[darry1966]

I hope people were not offended by my words but I care passionately about Puppy Linux and I want to see it suceed. It is so unique in its coding which makes so much more adaptable than anything else around.

[darkcity]

So lets learn a lesson from this BK is a genius no doubt and I for one admire his brilliance but relying on one person is plain silly and that is an important lesson to be learnt.

+1, baring in mind that some teams can be less effective than particular individuals

# Example: I`ve said for a long time that a union is an unnecessary complication.
And there`s still no PXE boot setup in the standard Puppy. Many improvements to be made.

I have mixed feelings about Union/AUFS. It certainly confuses newcomers, however it makes Puppy flexible and is one of the unique things about it.

[Iguleder]

Fact is : There are not many people here to develop the Puppy core : Kernel , Toolchain and compile core programs .

Me myself is not so much interested in using other distribution's binaries .

To me the puppy-4 core is good enough , does not need newest GLIBC !
I even start to use gcc version 3.4.6 on it
So far have upgraded GTK-2.0 to version 2.19.7 now .
Still using the 2.6.30 series kernel , want to upgrade it to 2.6.31 because that would show my USB-3G-Modem's Sim card and integrated micro-SD-Card slot.

If anybody is interested in Up/Downgrading Puppy-4.3.x , send me a private message !

I agree Karl 4 Series could be still a good base for development with more modern kernels.

Need to be aware of bad code in some of the "Buntu" releases being inherited.

Puppy 4.3 doesn't support most GPUs from 2010 and on, UEFI, Wayland, GTK 3, Qt 5 and much more.

It's old! It's great for the hardware it was designed to work on (PCs made until 2009). What you offer sounds to me like "let's use Windows 98, it is good".

By the way - we should replace Aufs with overlayfs.

[darry1966]

Ok I get the fact its old. However I would like to politely ask you as a Dev why it can't be upgraded to support these new GPU's and have a newer a kernel - is it a limitation of T2???

I am not talking about leaving it as is I'm saying as a base to use as for something new if it can be done then it will support older hardware and new. Afterall other issues like GTK etc can be easily compiled and updated so could it have an updated Xorg etc or is it totally impossible?????

Again I'm asking politely and would like to know not just that it is old.

[Iguleder]

Ok I get the fact its old. However I would like to politely ask you as a Dev why it can't be upgraded to support these new GPU's and have a newer a kernel - is it a limitation of T2???

It consists of many old packages. The kernel is responsible for its bad hardware support, but the whole graphics stack (all the way from libdrm, Mesa and DRI2 drivers to GTK) is the source of its poor GPU support.

Upgrading all these packages is a complex task - so complex it isn't worth it. You're better do a fresh T2 build with recent packages.

I am not talking about leaving it as is I'm saying as a base to use as for something new if it can be done then it will support older hardware and new. Afterall other issues like GTK etc can be easily compiled and updated so could it have an updated Xorg etc or is it totally impossible?????

Supporting both old and new hardware is pretty much impossible. Drivers for old hardware become deprecated, abandoned or just get dropped, while recent applications depend on a recent graphics stack.

This is true for both the kernel (i.e old PCI cards) and the graphics stack (i.e Nvidia drivers for RIVA TNT2), so supporting both means one operating system with two, completely different kernels and graphics stacks. Because things like recent versions of GTK won't work against an old X server (like that of Puppy 4.3), it means you'll have to replace the applications with old (e.g GTK 2) ones, too ... so you get two operating systems.

To sum it up: if you want a Puppy for old hardware, use an old version of Puppy (or a recent one, with horribly outdated packages). Otherwise, use the latest.

[darry1966]

Thank you for your reply and taking the time to answer it is appreciated.