Page 2 of 19

Re: The future of Porteus

Posted: 06 Nov 2015, 15:55
by francois
@aus9:
Good luck for all your decisions and feel free to ignore me

Don't be shy, we try to be a very inclusive community. This is a brain storming thread. You have your right to be part of this discussion as anyone else. This new but very enthusiastic look that you have since you came on porteus is important. Newcomers have a word to say, and specially those who take but also according to their means try to give back.

As bogomips commented in a private discussion that we had, many interesting and linux talented persons came across the porteus forum in the last few years and they finally went by. We have to find a way to keep such persons with us. Maybe if we we would have been able to keep some of these, we could have today the luxury to continue with both slackware and another distribution.

Also we have very faithful members, we want to keep them too.

Maybe some of us here would benefit to be able to have this ability to adopt a different perspective. This said I do not to insunuate that one perspective is necessarily better than another. These perspectives have to be debated each to their merit on the basis of facts more than on the basis of opinions. See systemd, a can of worms:
viewtopic.php?f=48&t=4384

Which includes a link to the position of Patrick Volkerding already in 2012 not only on systemd but also about google, apple and microsoft:
http://www.linuxquestions.org/questions ... re-949029/

Re: The future of Porteus

Posted: 06 Nov 2015, 15:56
by donald
Hello all
I do not intend to start a new (fruitless) debate on systemd because if you’re not somebody who likes to tweak your system you won’t be affected by systemd too much at all.

Also, If you installed Linux as a replacement for Windows and you use it the way
you used Windows, you won’t even notice it.

But I have a simple point of view -- Clarity is better than cleverness --.
Therefore I do prefer init concepts like "runit".

And a 2nd point is:
I fear that we will loose our freedom of choice, as soon as more and more programs will need systemd as a dependency.
That's all.

Re: The future of Porteus

Posted: 06 Nov 2015, 15:59
by Tonio
wread wrote:I am sorry, but Porteus will die, just as we are overheating the Earth and the planet will die again.....
It will probably not die, but just because of the change to systemd based pile of dung some people will stop using it :cry: :(
Not all people died, they just
left
. Planet may not die, just yet but sure many prophecies are coming of age. Hope 4 the best to all.

Re: The future of Porteus

Posted: 06 Nov 2015, 16:04
by Blaze
brokenman wrote:Will the Arch system be called Nemesis or Porteus Nemesis or Porteus II or Porteus 4.0?
Porteus 4.0
+1 for name (about Porteus 4.0). I think it's right way.

I have a suggestion - for each release Porteus to use the code name.
For example Android 6.0 Marshmallow or Porteus 4.0 Nemesis etc :)

brokenman, graphics for Nemesis is no longer needed?

Re: The future of Porteus

Posted: 06 Nov 2015, 16:16
by Tonio
Lastly, I am using modal language above because the decision has not yet been made. It depends on a few factors.
What are those factors? Can we at least know some of these?

BTW: Salix releases a slackware live kit, and before Porteus goes way to Arch, how about releasing a linux-live kit(script/?,) to make a Slackware installed system to a liveusb, livedvd, livecd from our installation. I see those are still hard like Tomas' live scripts. Something easy to just run ./createlive /dev/sdb2 or something and have our installation on a thumb drive before the transformation. Otherwise we will have to cling on to our Porteus 3.1 and use it as our last LLS(Linux Live System).

Re: The future of Porteus

Posted: 06 Nov 2015, 16:21
by francois
@wread:
For you and for most of us the perspective of the end of porteus is a rather depressing. But this not necessarily means that we are there yet. All depends on good will of the members of the forum and a common project that we could come with. We are a lot further than we were at the beginning of porteus, it would be really silly to spoil all that gray matter and knowledge.

@all:
Each of us have to be able to come up with the pros and cons ofthe proposition offerred by brokenman, but maybe more also with the pros and cons of not adopting his proposition. Brokenman is making a important point, how can you criticize any system, if you have not really worked with it in some sufficient depth. Personally systemd seems to be above my level of competencies, I am no IT. However, I have the impression that my learning experience of linux has been increased since I got in touch with archlinux (this is clearly an opinion).

But before: can we state on what really distinguishes porteus from other distributions in addition to being slackware based:
Pros:
- speed
- small size
- modularity
- simplicity: panel menus, intuitive, intelligent and functional organisation of the information
- relative transparency and access to all of the components of the linux box: root and normal user mode (users are responsible persons, they are not immature), the easiness with which one can access to every parts of the system (all devices are mounted)

- user level target: from complete newbie to advanced linux user
- educational approach adapted to the user level
- open, tolerant and warmful community, at least as much as possible
- ready to help community members, anti RTFM philosophy

- survival instinct (I hope)


Cons:
- smaller universe of packages
- difficulty to resolve some dependencies
- package manager limited by the small universe of binaries and of builds (thru sbo packages), which needs a lot of effort to maintain and to improve.
- relatively small number of additional modules xzm not integrated to the package manager (no critic here to our modules producers, they are doing a great job, but the size of the task is simply overwhelming)
- difficulty to produce a variety of modules at a large scale
- minor distribution based on a bigger distributions, with no pretention for now to be able to dictate the rules

- difficulty and almost impossibility to produce and maintain documentation because the lack of manpower
- lacking documentation resulting in the repetition of the same strategies on the forum (people do not read the documentation)
- very small though dedicated community

- user empowerment
- we do not take enough care of our developpers

Re: The future of Porteus

Posted: 06 Nov 2015, 16:34
by brokenman
they are almost dead by now... take a look at that
At least quote credible information and read it. The article says the creator of 'gnome' is tired of linux. Also, the word systemd is not even mentioned anywhere on the page!

Tonio I felt the same way at one stage, but after I took the time to learn about systemd I found tweaking it is quite easy. In any case, the idea is to have users barely notice systemd is running.

I have also used slackware for MANY years and I like it's principles. It is great when you have a full blown install and don't have to worry about dependencies. Variation is the beauty of linux but also the downside. Every distro has its own way and package structure. The winners are those that make life easy for the user. Systemd overcomes some severe limitations of other systems and is the beginning of unifying a small part of linux. Many will think this herecy, but I believe it is the only saviour. If systemd was really the problem, then slackware would have a larger user base since it is one of the last to resist the change (and relatively easy to use). Sorry, but I don't see the masses flocking to slackware or gentoo. I personally agree with the linux philosophy of 'do one thing and do it well', but I would that this also needs to kee up with the exponential advance of hardware.

When would we finalise on the base of Slackware or Arch Linux? How do we do this?
Finalisation of the base would be a few months away. I am still watching slackware's movements.

Is there any timeline for the alpha, beta, RC or final release date?
The decision to change has not yet even been made.

Blaze, for now I will continue developing nemesis, but revert back to the porteus graphics and name just to save some time should we change base.

Re: The future of Porteus

Posted: 06 Nov 2015, 16:42
by brokenman
What are those factors? Can we at least know some of these?
Sure. I have mentioned them before.

1) Workload - One person maintaining 4 desktops on 2 architectures (8 desktops) is not an option anymore. On arch, building a desktop is MUCH less work.
2) Workload - Maintaining the custom Porteus gtk apps is extra work.
3) User opinion - If I get enough good reasons NOT to switch I could be swayed to stay with slackware. "I hate systemd" is not a good enough reason in light of other factors.
Mostly these reasons would be assistance from the community to actively develop. Over the years people have come and gone. Finding people with ability to develop and ability to stay is not so easy.
4) Personal commitments - I have to choose an option that I can maintain for the foreseeable future without burning out. I have less time now than I did 2 years back. I would love to be able to develop Porteus full time, but this does not out food on the table.

That's some of the factors.

Re: The future of Porteus

Posted: 06 Nov 2015, 16:52
by beny
hi all i agree with you francois,for the necessity of release the scripts that make an aufs system,i have the whole slackware and arch too into the sd or other removable media, slackware too is at a crossroad, the release lag tell us this,systemd crap or not it is the future, wread you can istall plasma without any problems,if someone want something or have the skill or wait the work of other peoples,i have used slax 6.1.2 since the release and when i have trouble with my system is that i use,so don't have the necessity of the last release,brokenman have done a good job to give us an alternative.....

Re: The future of Porteus

Posted: 06 Nov 2015, 17:16
by Tonio
@Brokenman

Sadly some of the factors that you have listed are very important :( I agree that maintaining 8 desktops is not very ideal. But can this be fixed in some way?, i.e, some of us agree to build/maintain a desktop and upload these on your site. Only the base with core+devel should be provided in my opinion and then us should make our desktops like we want. We are burning some of our developers out, I understand this point.

I can argue about init scripts because I have played around some, I used red hat 8/red hat 9, mandrake, then fedora core and fedora. There were sysv init scripts and used a tool called chkconfig and there were run levels and I had the pleasure of playing with these and learning about them. Sysvinit vs BSD style init scripts. I used a winmodem called slmodemd and made it work in red hat, mandrake, then fedora. Chkconfig worked nicely and then I used Slax which is slackware based and I made the modem work in there and used the rc.d/ directory and made it work at the first stages of porteus we had a porteus script that created it a module to make this modem work in porteus. Try to do this with systemd, it is cryptic and yes, I have played and play with Fedora it has some good things, but also adds a great deal of packages. I have not done this because now winmodems are a thing of the past and with broadband and wireless widely available, but it does make things more complicated. It may have good things. What some of us users liked about Slax/Porteus is that it is small/modular and can be adapted to basically anything we want. I even succeded creating a livecd/LiveDVD/ liveusb with KerTeX (a new lightweight version of tex and latex) something I had not done before, but it seems that not too many people cared for it I stopped playing with it. I have it for myself and I have not shared the modules anymore or uploaded them anywhere. I see some of your frustrations and why you are taking this path. I have not used Arch nor know much about it, but many folks tell me that it was good, but then it went to systemd and just because of this it is not good. The systemd thing is one that apparently tries to take over the whole linux and it should just be an init system like openrc in Gentoo. I have not used Gentoo but have used FreeBSD and it got many of its ideas from there. it has *shisms to substitute software to make gnome and some other desktops work without systemd and I get your point that it gets frustrating and too hard a job to maintain this.
If I were you I would tell folks here is the base, you want this* or that*, here's the source, build it and use it yourself* . If your module works for you you can share it with others * otherwise build it/use it to your taste. If you want porteus like you are used to, you will only get what we can provide for you because it is not easy to maintain 8 desktops and two architectures. Now i486 is being dropped/replaced with i586 many packages are being rebuilt in slackware-current. what will happen? If you want porteus with this desktop and that desktop like you are used to, we will have to use Arch as base because maintaining them on a slackware base is overkill and can be done easier in Arch. This is what I mostly see, would that be a fair assumption?

Some folks use Crux how what work? They use bsd style init scripts also and how do running and installing packages work there if there are users who have used it/use it? Just in case another distro could be used that is not Slackware before going to the other one?

I do not mind systemd too much, but developers who develop programs and desktops do mind its intrusiveness. Here's a site you can read in case you want to find out more arguments against it:
http://www.textplain.net/blog/2015/pro ... bsd-init/

Re: The future of Porteus

Posted: 06 Nov 2015, 17:46
by roadie
The only constant is change, and it happens for lots of reasons. The issue of Porteus changing the base, or going to systemd, is kinda like a patch o' fluff in a sea of cotton.....it just ain't that big.

I say that coming from, as I said, 15 years of Slackware, first Peanut Linux, then I spotted the first version of Slax, and I haven't had an hd install since. I think we need to look at why the change is coming.

Porteus, as it stands now, is too much for one maintainer to handle, going to Arch would lessen the load....but also force users to do more of their own work in getting packages. I've found pacman to be no better than crap apt-get in what it will, or won't, do. This brings up another thing...would the new Porteus still have the same layout, as in saving changes and using modules?

Cutting down on the number of desktops will, I think, kill Porteus. The clamor for this or that DE would be endless. Going to just a base, core release will push away new users, new users know little to nothing about building their own stuff....they want someone to do it for them...for free.

What really needs to happen is for the moaning, bitching, and whining to stop, and someone step up and lend a hand in maintaining. I imagine the pay is the shits, but I'm sure the fringe benefits are amazing. Ignore the family harping about never seeing you. Really, we get this great distro for free and I may be naive, but I don't see it bringing in scads of money....massive prestige maybe, but no big bucks.

In any case, I too would like access to the live scripts if possible, if only to maintain my own builds....I'll stick to Slackware as long as I can, I don't see Arch as a viable alternative for me, haven't looked into systemd yet. I do want to play with a Porteus/Arch version though.

Re: The future of Porteus

Posted: 06 Nov 2015, 19:00
by francois
Roadie and others.

I would say yes, before criticising the project please give nemesis beta a try. It is exactly in its present unpolished form the best of porteus (simplicity, structure, easiness of use) as a wrapper over arch linux (its pacman package manager and pasture) to provide modules.

For example, even a moron or handicapped linux user like me can build a desktop environment like xfce or plasma5 over the actual system already built by brokenman. And with a little creativity nemesis could become even more.

When you look at all the work that arch linux has done to maintain such a huge documentation as provided by their wiki, its hard to criticize such a community. It should rather inspire some respect.

I do not have a cue about the work involved to keep porteus base module at level with slackware current version. However, there might be enough voices and then hopefully brains and fingers to replace fanthom and brokenman for maintaining that part: donald, knallkopf, wread, tonio, phhpro ...

Also, we should not forget fanthom. Very few current members have given a hand to provide support to porteus kiosk users. In the end, is there so much return for the developper, except for the development of a certain expertise?

Re: The future of Porteus

Posted: 06 Nov 2015, 20:07
by brokenman
Interesting read Tonio. Especially the problem with journald hijacking the 'debug' kernel command line argument. It seems some systemd developers are a little arrogant at times. Your assumption is one factor.

Maintenance under a slackware base is not so difficult since it doesn't move at a rapid pace. There comes a time after a new release when things settle in the slackware world and only security patches are the focus. That's where we are right now, waiting. This probably wouldn't happen on an arch base. On the other hand knocking up a desktop on arch is much simpler. Either way you look at it, I don't think I can realistically maintain 8 variations of a desktop with any sort of high quality. Help from community members will be very much welcomed, whichever way we go.

Re: The future of Porteus

Posted: 06 Nov 2015, 20:46
by roadie
@brokenman

From your perspective, maintaining Porteus would probably be much easier under an Arch system as far as providing different desktop versions. I'm assuming Porteus 4 would have a package browser for the new users, though PkgBrowser needs QT or python-pyqt so would affect size of iso. I'm also assuming there will be a module creation wrapper as in USM.

Arch pretty much forces the user into the world of the CLI and we both know that new users are not at all comfortable there....they want to click a button and see the results pop up. The Arch wikis are very good...trouble is most new users lack the patience to research problems....I envision a large increase in questions on the forum.

Another point is that a lot of new users seem to want Porteus installed to an hd as opposed to a frugal. I can envision a lot of borked systems happening with Arch, it happens fairly easily. Apparently, it's even possible to bork pacman itself, then you're really screwed with an hd install....you get to reinstall all over again. I don't see new users appreciating the "simplicity" of Arch at that point.

I've found over the past week that Arch is nothing like simple or easy but the issue will not be the underlying base of Porteus, the new user cares little, I think, as long as they can have what they want...easily.

And it's here that Porteus/Arch may fall down, Arch is, when all is said and done, a distro designed for power users.

As you say, Slackware doesn't require a lot of updating, where Arch is cutting edge and this will have an impact simply because everyone wants the newest, shiniest whatever...things will break.

Re: The future of Porteus

Posted: 06 Nov 2015, 21:27
by roadie
@brokenman

Maintaining 8 desktops regardless of the base is ludicrous. Either the community has to come forward, or a new way will have to be put forth. Possibly supply the base with X and a light desktop...xfce, lxde...and the user will have to get whatever else they want.