The future of Porteus

[francois]

Roadie, nemesis already has all you are asking for:
module creation wrapper as in USM
pman as devised by brokenman is a wrapper over pacman that strips the packages and build them into functional units that we like to call modules.

Arch pretty much forces the user into the world of the CLI
Porteus likes cli, but tries to provide gui wherever possible. Under arch there are some interesting gui interface, no need to build them.

The Arch wikis are very good... ...trouble is most new users lack the patience to research problems
This where the porteus and arch linux philosophy differ. We take the time to help each new member according to his level of linux knowledge. In fact, sometimes I have the impression that we do too much.

In addition, these wikis are readily available to people intervening with newcomers and could be cited or hyperlinked easily. The manual is there. The exception would be for the particularities of porteus-nemesis. Personnally, I have the impression that it will make my life a lot more easy. Those new members that will want to participate will more easily find the information to guide their less knowledgeable porteus forum mates.

new users seem to want Porteus installed to an hd as opposed to a frugal
This could be a difference between the porteus forum philosophy and arch linux and the other arch linux based distros: we are frugal and modular. They could be referred to the other distros for full install or invited to experiment at their own risks.

really screwed with an hd install
The advantage of porteus is right there in terms of repairing or reinstalling the system. With modules it could be darn fast. In addition, there is the save changes folder or persistence function. You get rid of it or of some part of it, and the system is as it was at the beginning. When you are happy with your settings you just turned theses changes into a module.

archlinux ... ... everyone wants the newest, shiniest whatever...things will break
And here maybe porteus could be based on the manjaro servers. I have asked on the manjaro forum the difference between them and arch linux in terms of stability:
Manjaro is a stable release, compared to arch linux rolling release: what is meant here? How does manjaro achieves its stable release if it is based on arch linux?
https://forum.manjaro.org/index.php?top ... ;topicseen

[sean]

@brokenman,

Yes, I did say "Bottom line is "Big Money" wins yet again".

That was certainly no admonishment to you personally! All here should appreciate my praise for you and the other developers and contributors, every one of them.

I'm extremely paranoid about microsoft and it's gigantic bag of tricks. I see every little deal a Linux distro, or Linux company makes with them, as being another microsoft hook into opensource, thus Linux.

Porteus is not mine, it is your's, and you are gracious in sharing it with others.

We do not agree on systemd, and again that's my paranoia. I believe I understand the Unix concept, and choose the original form of it. I'm probably the very least computer literate person on this forum. My resistance may come from age :-) My fear is that microsoft will gobble up and sue Linux out of existence in another 20 years or so, hopefully after my own expiration date :-)

Now, you said "There is a reason why most distros are adopting systemd, and (do your homework) it is not because they are forced to." I disagree, and my opinion is that they are absolutely being forced to. I also realize development is a very big picture to take in, with thousands of players, developers, applications, etc. included.

Either way I will not run away from Porteus. This must have been an anxious day for you with the commentary here. However, you did the right thing, you were honest, and you asked for it. And you did not lose anything.

I promise not to mention my 2 favorite bad words again in this thread.

Sean

[aus9]

as brokenman has discussed systemd here, so shall I.
Systemd

udevd
udev is a device manager for the Linux kernel, which handles the /dev directory and all user space actions when adding/removing devices, including firmware loading. In April 2012, the source tree for udev was merged into the systemd source tree

Count the number of distros that don't use it, I can name a few not already mentioned. Tinycore and AntiX but it will get harder due to udevd.

Altho slightly off topic, I use cars which don't reveal their source code, I use routers and not all routers reveal their source code. So some purists who get excited about being anti-sytemd are missing in action on some other matters.

Get over it. If the dev team need to change, thank them for their hard work in the past. If systemd is some evil thing move to a distro that doesn't have it.

Offtopic but kind of related.....Speaking of firmware, some sound devices need it. I don't see any flak about that to the same level of venom by anti-systemd users.

[normalGuy]

Hello everyone
(sorry about my english)
I being using porteus for a 1 year now, I am just a normal guy – well not an IT I mean - porteus was my first step into the world of linux. I learn something with it that help me with other distros, but in the end I always come back here.
From src2pkg, sbo, a few command lines, partitions, etc was a learning curve. A lot of games: pychess, minecraft, peg-e etc etc I even installed it in the hd and was always working: rock solid!!...but the deps was hell , I was tired...I`m using a arch base distro and it`s good but because it is a rolling distro there are some bugs: I had to change to the old kernel right now even.
With so many people doing small distros and It is so hard to join forces?
Looking forward for the new porteus.
DuckDuckGo porteus every day and thanks for all your hard work.

[brokenman]

Sean I did not take any personal offence from your comment. I know you mean well and your input here is valid. Interesting to note is that at compile time you can absolutely choose how much of systemd you want to use. The only way I see people being forced into systemd is the issue with udev. The choice will be either adopt systemd or maintain udev separately. Gentoo offers with or without systemd, but gentoo is not for the new user.

My intention is to bring 'ease of use' to arch linux where newcomers can stay away from the command line. We can put speed humps down for the rolling release so that we stay slighly behind the razors cutting edge. Breaking the system should not be too much of an issue (hopefully) and it would most certainly support frugal installs. I am using it frugally right now and it retains all of the benefits of regular Porteus in that it is modular, it took about 20 seconds to install and I can remove a module that breaks anything.

[brokenman]

We could use an internal update system that updates just the base modules (001,002,003). On top of this we could limit the date of repository using Arch Linux Archive. In this way we could keep the base modules in sync with the arch repositories.

Advanced users would be free to update to latest but would also be expected to fix their own problems if bugs originate upstream. This seems like the simplest native solution.

[aus9]

Umm, altho its purely selfish of me to say this, some people might be in the same situation as me?
I should mention that I have just found out that Arch is one the Linux distros mirrored by my ISP

That is important to me for 2 main reasons.
1 I live in Australia and our internet is about the 44th fastest in the world, but I am on copper wire over 3 km line of sight from exchange so maybe 3.5km in actual distance. So the size of the iso and size of updates and size of extra packages is relevant.

b) My ISP counts uploads with downloads and most Australians are on capped plans. Even those that boast they are on unlimiited plans should read the fine print as ISPs tend to have some kind of "fair go" policy. And money is tight.

2 Now altho I am a slow learner, it has finally dawned on me, the future of Porteus, for those who stick around, is repo ArchLinux.
So it look like I can forget trying to learn Slackware and start afresh. Maybe other new Porteus users might see some merit in that as well?

If above is too obtuse, its time to start testing Nemesis. But can't promise anything.

[zer0-G]

Hello,

I have been around the Linux live scene since the early days of Slax. While there are hundreds of great distro's out there I have always felt a live edition is great tool. Unfortunately, recent years slackware based distros cause too much frustration when a needed application is not supported or just too difficult to get running. So, I would like to support any effort to move Linux live to a modern OS base that hopefully has the following priorities:
1) Huge repository of all applications (Arch does have a large repo).
2) Focus on laptop use with excellent power saving control.
3) Minimal memory resources.
If a working beta can be released I think the community can get on board and help out in development efforts.
Exciting times ahead...

[brokenman]

It seems to me that the new Porteus in its basic package, unlikely to be more than 1Gb

I am aiming at less than 300MB

[tome]

Maybe Porteus could be based on Arch (x86_64, new hardware) and on Slackware (i486, support for old hardware), but probably it requires much more work. On forum I have found only one person that asked for Nemesis i686. Now it is hard to say what would be better. After, for example, a year of using Arch as x64 base we could see that it was correct (ahead) or wrong (back) step and do next one. I prefer Slackware feature and stability based release cycle over novelties, that are not worth attention they get ("Never change a running system" ).
People choose Porteus, because its target is to be simply (to install, understand and repair, modular), lightweight and clean. To have rich of packages is average important.
By the way why people choose something? One example of interesting, tendentious answer could be found here: https://www.youtube.com/watch?v=qp0HIF3SfI4 (translated subtitles available)

[Bogomips]

Friday not most appropriate time for bombshell. RL priorities. Would like to make contribution. Hope debate if it is as such not yet over. Would like to present different scenario. Still ingesting facts.

[KnallKopf]

This message was predictable when i see that you start this poll and this discusion.
But it is a shock too when it is reallity.

where is the result of the poll ?
I m shure that slackware was the winner.

I do not Understand the decision to leave the good working Slackware.
Because i can not believe that makes less work to change the underlying OS.

In the other hand i really thankfull for your work.
And when i see how much work it is, to get going small thinks, and when i see the things that should work but they doesn't.
I can understand that you want to reduce the work.

What is your propulsion ?

I will stay at your great work Porteus 3.1 and make careful look to the new OS.
Possible that will a better OS like Porteus 3.1 but systemd is crap.

[roadie]

This message was predictable when i see that you start this poll and this discusion.
But it is a shock too when it is reallity.

where is the result of the poll ?
I m shure that slackware was the winner.

I do not Understand the decision to leave the good working Slackware.
Because i can not believe that makes less work to change the underlying OS.

In the other hand i really thankfull for your work.
And when i see how much work it is, to get going small thinks, and when i see the things that should work but they doesn't.
I can understand that you want to reduce the work.

What is your propulsion ?

I will stay at your great work Porteus 3.1 and make careful look to the new OS.
Possible that will a better OS like Porteus 3.1 but systemd is crap.

Here is my take on the situation....not speaking for brokenman.
There is no doubt that Slackware was the winner of the poll and for good reason. The thing is, when the poll was done, there were 2 maintainers, now as far as I know, there is just one. It's too much for one maintainer, so I guess the thought is that by moving to an Arch base, the users will be responsible for getting extra packages apart from the base Porteus 4.

The bulk of the maintaining will shift to the users and Arch...I don't see that as a terrible thing as I don't believe in spoonfeeding users. Whether or not the users will agree remains to be seen. I'm sure brokeman is planning to release a very polished and smooth Porteus 4 with lots of apps for new users to get up and running. It's the aftermarket goodies that might be the problem.

My take on it.....I doubt very much that I'll be moving to Arch, my main system will stay Slackware. Then again, I'm not a power user, but I do like small and no Arch I've seen comes close to Porteus in size. I'll be somewhat surprised if brokenman brings it in at 300 megs, hell, a barebones net install of Manjaro is around 325 megs. Do these Arch guys know about compression?

[brokenman]

Yes slackware was the winner of the poll. Roadie hit the nail on the head. The poll was at a different time and under different circumstances.

One of (if not the main) common complaints of slackware based porteus is the lack of available applications. To the average user this is essentially what an operating system is about. It is a means to procure the applications they want to work on a day to day basis. The OS itself is largely irrelevant to the average user. In fact I would say the desktop environment is more important to most users. Power users are another breed. The OS matters here.

On size ... my base lxde system (arch based) is 176MB. I am not sure why people think that systemd is insanely huge. If compiled in full, it is 69 binaries at a combined weight of 5.2Mb. It also makes some other packages redundant.

[roadie]

True, Slackware can be frustrating as far as getting packages you want. I end up grabbing a deb or rpm and converting it, often it doesn't work because of mismatched libs or glibc...it's a pain in the ass, for sure. And it leads to a cluttered system which I hate.

I think I understand the reasons for a move to Arch, and under the current circumstances, I think the move is needed. But, I also think the move to Arch is the end of Porteus as it is now, in that the new users will need to be of a different mindset. They're probably gonna have to be willing to get there fingers dirty under the hood. I know some will disagree, but thats how I see it. To have it any other way will mean a high level of maintaining and support.

I'm waiting to hear Bogomips idea, and playing with a hacked Slitaz I cobbled up....kinda neat, it uses a Porteus kernel and saves changes Porteus style....just something different....while patiently awaiting a 32 bit Porteus/Arch.