Tomas's plans about Slax project

[Tonio]

@fanthom

great but pls think for a while about fresh members coming from Windows

Yeap Sorry for being The modules have to work on the fly and mimick what *.exe's do for in windows Similar concept to the module is the *.pbi that PCBSD uses. I prefer pristine FreeBSD, where I learn the system rather than a system is smarter and will set things up for you But for your windows users friend(s) there needs to be a smooth transition and I get it

Second case is better cause you will be able to avoid duplicated deps in your system.

I did not know that the dependencies could be placed in different locations and shared instead of an all in one folder/module and avoid deps

module can't be inserted to union while being placed inside union (aufs), that's why it gets moved to /mnt/live/memory/modules/ first which is a real fs (user is using changes= cheatcode) or tmpfs (user is not saving changes). if you activate a module directly from usb/cd/hd it wont be moved anywhere.
second thing - module is mounted to /mnt/live/memory/images/ and not extracted so it does not consume memory.

I am quite intrigued by this concept. Why? As I have studied module creation & insertion on the fly, I noticed many things like the modules *.lzm & now *.xzm living in two places, ie., in images folder and in /mnt/live/memory/changes/ or /mnt/live/memory/modules. If I build the module from source & insert it on the fly, if I don't build it and insert it. There are differences. I had no idea that there was more to it Thanks for sharing this. I have more to learn.

Gentoo ebouild system is almost the same as ports in FREEBSD.

If you were to change to Gentoo build, one could use this idea Some folks have encouraged me to try out gentoo, but they tell me that there are stages, that one has to build the system and it could take a while to intall it? I use SystemRescueCD which is Gentoo based, but I just use it to make partitions & blank passwords for windows users

I am happy & thankful for porteus and I am surprised that there are few questions regarding modems(winmodems) & dialup users. Ahau has written some very nice howto's regarding pbuilds for some of these modems. This is where I did more work, but now I don't use dialup anymore and somehow I can't test the modules as I would like to.

[pizzar0]

previously by fanthom

'm wonder which desktop will be chosen for Slax-7. my wishes goes to Gnome-3 or Xfce/Enlightenment as KDE-4/Trinity/LXDE are covered up here.

Your wishes shall come true! Gnome-3 it is. (Just guessing, of course

[wread]

Thomas wrote, referring to the trailer video of Slax 7:

"Bear in mind that what you see is just a preview of raw KDE 4.7 desktop on 64bit Slax. All bundled in 180MB. There are TONS of things to be improved, though."

Not bad!

[ralcocer]

Seems everything related to Tomas M is out of the Internet

[francois]

It is back now. These were server problems.

[francois]

The first beta kde slax 7 should be coming out within hours:
http://www.tomas-m.com/blog/

[Hamza]

http://www.tomas-m.com/blog/18877-Slax- ... eview.html

8)

[francois]

Thomas will be at Linuxdays a Tchec conference today. He intends to talk about slax 7 that is in form of beta kde4 based, available for 32 and 64 bit. It is interesting to see the product evolve. Slax 7 will have to solve a lot of issues that porteus has already find some ways to manage. Thanks to our main conceptors and all the porteus community. A story to be followed:
http://www.tomas-m.com/blog/18897-KDE-i ... l#comments

[wread]

I am looking forward to it; Thomas will for sure have found things that I still haven't.

His beta kde4 (176 MB) is contained in our 000-kernel (13.5 MB), 001-core (46.7 MB), 002-xorg (71.1 MB) and 004-kde4 (my latest is 121 MB in 32bits). If we add the NetManager modules -ours is about 7 MB- to the 176 of Thomas, we are still 252 against 183....

Ok, Thomas put our kernel and core together in 48.8 MB (-11.4 MB), has xorg with 11.8 MB (-59.3 MB!) and for kde4 he has 2 modules (79.6 MB + 29.6 MB) that add to 109.2 MB (-12 MB).

Theoretically, I can put some expanded kde4 programs in rootcopy in slax to test it. I will try it!

[Thor]

Yesterday I got an 8 GB sandisk usb in the post. I paid $5 for it... A pint at a pub here, costs me $12-15...I seriously think time should be spent on other issues than reducing the size a 50-100 mb. Anyone owning a system that can run kde4 ok, can afford an 8gb usb...
If 50 mb makes it run noticeable faster than maybe yes, unless those 50 mb also take away a lot functionality.. If the first is the motivation, I find it weird that slax kept it's compiler optimalization at 486 for all those years.

[brokenman]

I seriously think time should be spent on other issues than reducing the size a 50-100 mb.

The more important aspect of a lightweight distro should be the memory footprint and not the storage size. Modular distros use compression to store everything in the ISO but these must be 'unpacked' before booting and it is at this stage that the memory footprint is made. Having a smaller memory footprint has advantages as of 2012.

In my opinion the only advantage of a small storage size is a faster download. I am more likely to try a distro that takes 2 minutes to download rather than 40 minutes. All my 256/512 flash storage devices died, were lost or the physical format changed long ago.

[Ahau]

yet another benefit of a small .iso size is for use in copy2ram. A lot of folks have older hardware, and copy2ram can make them run a lot faster, so long as all of the RAM isn't used holding the modules. If you have 1GB of RAM and you are using a 250 MB vs a 500 MB distro, you effectively have 50% more RAM available for use before you even get around to using the software in the modules.

I think it's a bit of a balancing act. There is definitely enough on the positive side for a "live" distro to go through the efforts to keep its iso size small. On the other hand, if you take this to the extreme, you wind up cutting lots of features and usefeul components (think tiny core and slitaz, for example). I think we want to keep a modern and user-friendly look and feel, while staying small enough to remain "light" and "fast".

[francois]

I agree with Thor that space is no more an issue.

In addition, building modules for big packages, even with porteus package manager (PPM) devised by the genius of brokenman, takes a big toll on the functioning of the linux box. But from what I see, we have a problem in terms of server space from which to download modules or iso, as well explained by Hamza:

Some are right invoking the main chararcteristics : simple and fast are two important features of porteus. Efficiency is also one of its characteristics. Lets think about all the work that has been done in terms of adapting porteus to graphic cards and to wifi cards. Tomas has not even begun to address that type of issue.

Modularity is another special characteristic of porteus. Modularity has been addressed further by the porteus team compared to slax. It seems to be a very promising tool compared to package fectching and installation on other standard linux installations. However, there is the complication tied to the dependencies and to the size of some modules, even though we now have the wonderful PPM (Porteus Package Manager) built by our genius brokenman. Modularity opens on the building of a system that could become a lot bigger.

This will be the case with a save changes folder that will grow bigger and bigger. If you add to that the use different packages that come standard with porteus OS, and that could include for example gimp or virtual box, you will quite rapidly fill your 8 gig usb key or a porteus frugal partition on the hdd. And many of us like to use porteus as a frugal system, as it is very efficient. Though, I sometimes go to see if there is not another more efficient system out there, most of the time I come back to porteus.

Should we offer also a bigger edition of porteus? Do we have server space problem as alluded by hamza in another thread? Is there simple and temporary solutions to these problems?

[Hamza]

@francois,

We haven't space problems on download server as so far I think. I can't provides any others info as i haven't any other.

We should open a new thread to talk about this because this thread is going so far from it's initial subject. Here, Tomas M's project about Slax distro.

[francois]

Sorry if I give some misleading information here. I will review the source. Evidently, I am the one who misunderstood you.

I imagine that you want us to continue on a thread called Tomas' project: slax 7 or Tomas' slax 7 and its impact on porteus ?