Devel

Release date for Slitaz 5.0? (37 posts)

  • Profile picture of Aleksej Aleksej said 10 months, 3 weeks ago:

    SliTaz distro should continue as minimalist or should evolve into a more complex distro?

    LiveCD ISO Image to taste
    Just collect all the packages you need into one flavor, and share it using mailing list or forum.

    Another point of discussion is the one that has been here

    I plan to return to this issue in the near future. It will be a package that installing, TazPanel will run faster. Some functions (implemented using shell scripts) will be put into a separate file. It will be “all the same official version.” In a separate package must be the same functions, but implemented on a compiled language. Most likely, it will be a Nimrod.

    Ever since I started the above topic, after fierce opposition to, my enthusiasm in this direction almost dried up…

  • Profile picture of christophe christophe said 10 months, 3 weeks ago:

    Quite frankly I do not believe it makes sense to wish for Slitaz to be more complex.
    There are a lot of different linux flavors around, each with its own personnality.

    Slitaz personnality is mostly being small simple and easy to run and maintain. I do not need another ubuntu or pclinuxos or arch. The other main features which differentiate it are ability to run in Ram, and frugal, which ensure the system is not corrupted, but also allows for easy install on top of a windows system (no partitioning required).

    This being said, the two areas where this community could do a better job are:
    - better security policy. I could not care less for a 5.0 slitaz, if it is not kept secure. We need the security updates out (someone has started a post on that, recently).
    - better hw support. We already all know that. But that comes second, after security.

    (As third, I would say more available packages – and, of course, improved documentation).

  • Profile picture of Aleksej Aleksej said 10 months, 3 weeks ago:

    Slitaz personnality is mostly being small simple and easy to run and maintain

    christophe,

    My friend interested in Linux. Mostly, Puppy and PuppyRus. For him, shell scripts — this is something complicated and incomprehensible. SliTaz, mainly written in shell scripts. But this does not automatically mean that it has become easier from that.

    SliTaz should be easier for users, not for developers. Come on, we drop this hypothetical possibility that if SliTaz was written only on scripts, then “everyone can help in the development“. And, behold, if any part is written in Python, C + +, or Nimrod, that “I want to help, but do not understand your fucking code“. Why drop?

    How many people engaged in the development of SliTaz over the past year? Who is it? What is their level? In fact, most likely they know “little more” than a shell-scripts. Here html, and css, php, python, gtk-dialog, various techniques and methods of programming…

    Who really wants to help the project, but he knows only Bash, will do so on the Bash. Who knows C, can write super fast, compact useful program (yes, I’m talking about TazWeb).

    I do not want to SliTaz has been slow because of its simplicity.

    In general, SliTaz — it’s a little secret, is a puzzle. Analyzing with its settings, with its work (which is not much different from that of other Linux), you progressively figured out the one riddle after another. You increase your own level. How pleasant it is when you learn something new today! How nice to solve a very complex riddle!

  • Profile picture of Trixar_za Trixar_za said 10 months, 3 weeks ago:

    The part I enjoy about SliTaz is that it constantly forces me out of my comfort zone. I’ve said before that I’ve learned more from running SliTaz for a week than I did running Ubuntu for 2 years. So this opposition to using languages other than Bash is a little silly to me, because pushing us out of our comfort zones is one things SliTaz does VERY well. I’ve been getting into Vala/Genie recently and it’s awesome. It’s compiler is small (1Mb package) and it depends only on glib (needs glib-dev) and gcc for compiling. For extending it, you just need to include the dev package of the package your using. It’s so easy that I already rewrote TazWeb in it!

    As for a more complete release, why can’t we have one on the website for users to pick along with our release copy and cooking version? Anybody that wants to head that project can do so. Remix to your heart’s content!

    Ok Guys, should we start a new topic and try drawing in the devs so we can start working on a SliTaz 5 RC?

  • Profile picture of Darjeeling Darjeeling said 10 months, 3 weeks ago:

    > Ok Guys, should we start a new topic and try drawing
    > in the devs so we can start working on a SliTaz 5 RC?

    … yesss!!!

    But if producing a RC and eventually a SliTaz 5 release later this year is the plan, then the priorities should be on fixing/stabilizing what is there and not on fancy new stuff like Lua/Nimrod/Vala rewrites of tried & tested applications and scripts that work perfectly. Things are complicated enough as they are. Without identifying and focussing (!) on issues that need fixing or improving the plan to produce SliTaz 5 might fail.

    So, what’s needed? From my amateur(ish) perspective there are a few points …

    Harmony!
    I was shocked to see what’s been going on regarding the SOURCE variable in the receipts. One guy removes the variable without much asking, another guy reverses the removal without much explaining … really, guys, you should communicate before doing things that affect the work of others! In a project like this, asking questions, giving answers, discussing and explaining things are as important as hacking code.

    Boot Screen
    I don’t know why the boot procedure is in such a messy and convoluted state. I guess it’s a “Baustelle” and someone’s working on it? I hope that eventually it will be a single screen with options to select the system language, the keyboard layout, and a field for entering kernel parameters (as it was in SliTaz 4). The boot screen should also be completely removed from ISOs that have been created with TazLito or TazUSB.

    Xorg / Hardware
    With my old-fashioned Nvidia 7300 SE graphics card all the recent Rolling releases booted straight into the desktop using the proper resolution. Great! One tiny fly in the soup is this …

    http://www.dumpt.com/img/viewer.php?file=xr0hubhllg1eu7dqlizl.png

    … something I have never encountered in previous SliTaz releases, whether I was using the proprietary Nvidia driver or Nouveau. I’m not even sure the fragments of selected desktop areas have anything to do with Xorg. Maybe it’s Openbox?

    SpaceFM
    To me the biggest issue in the SliTaz Rolling releases is SpaceFM, which has multiple severe problems and is in its current state of development not very comfortable to use …

    - no “Open in Terminal”

    - setting internal drives to “visible” is constantly forgotten or reset to “invisible”

    - some global preferences can only be set by right-clicking on folders in the side-pane

    - internal drives can only be mounted as “root” which, strangely, makes the whole drive inaccessible for ordinary users

    - selecting files and folders doesn’t automatically de-select previous files and folders, which is irritating and dangerous

    One serious issue I came across some days ago was this: in a Live session I selected and copied a file in SpaceFM, then went to a mounted internal drive, selected a folder and pasted the file into that folder. There was no error message, but the file didn’t arrive in the target folder. I repeated the procedure and was then asked if I wanted to overwrite the file. When I accepted the overwrite the file was still not to be found in that folder. But on both occasions I could clearly hear the write access on my 20 year old Seagate drive. So, I looked around and eventually found the file one hierarchy level up in another folder. What a horror!

    I think that SpaceFM isn’t quite ‘there’, yet. For the time being it is probably best to stick with PCManFM, which also has its flaws but still works a lot better than SpaceFM. Just make sure not to use the dumbed-down and crippled version 1.1.0 currently in Cooking, which hasn’t even got a tree view in the side-pane. The original 0.5.x or PCManFMmod, the de-bugged version the Ignorant Guru made a while back, would be ideal until SpaceFM works reliably.

    Quality Of Packages
    Obviously, I cannot say anything about the vast majority of packages in Cooking. But there are some odd things I’ve come across: There seem to be bits missing in Midori 0.5.0 and moc (music on console) has an unnecessarily long rat-tail of dependencies including perl (!) …

    ——————
    moc deps Slitaz 3

    libid3tag

    flac

    curl
    —dep: openssl

    libsamplerate
    —dep: libsndfile

    libtool

    ffmpeg
    —dep: imlib
    ——dep: giflib
    —dep: jack-audio-connection-kit

    moc

    —————–
    moc deps Slitaz 5

    libid3tag

    curl

    libsamplerate
    —dep: libsndfile

    libtool
    —dep: libltdl
    —dep: automake
    ——dep: autoconf
    ———dep: m4
    ———dep: perl
    ————dep: libdb
    ————dep: gdbm

    ffmpeg

    moc

  • Profile picture of Aleksej Aleksej said 10 months, 3 weeks ago:

    2Darjeeling:

    I was shocked
    . . .
    One guy removes the variable without much asking
    . . .

    Do you think that I had to ask someone about it?

    Too many of my questions on the forum went unanswered. Do you familiar feeling when you ask something, when you want advice on how best to do: so or so, but no one, absolutely no one answers. After a few questions you understand this: do as you see fit.The desire to question disappears. You start to think how to do better. You start trying.

    Indeed, in those days it seemed that all was still. And then suddenly it turned out that there are many people who are interested in your (my) actions.

    the priorities should be on fixing/stabilizing what is there and not on fancy new stuff like Lua/Nimrod/Vala rewrites

    Please, wait me until I finish to puzzle out with Xorg, upgrade it.
    And note that “hypothetic” TazPanel addition is just addition, no more. Original TazPanel can work as good as with it, and without it.

  • Profile picture of Aleksej Aleksej said 10 months, 3 weeks ago:

    2Darjeeling:
    Strange, check this:

    moc depends for SliTaz 3;
    moc depends for SliTaz 4;
    moc depends for SliTaz Cooking.

    Do you see the difference between them?
    Do you see the differnce between your output and output from pkgs.slitaz.org?

    Anyway, I guess that “libtool” branch of depends is unnecessary. Can anyone to check it up?

  • Profile picture of Darjeeling Darjeeling said 10 months, 3 weeks ago:

    > … but no one, absolutely no one answers.

    That’s an experience I know only too well from this forum. I suspect it is also the reason why some people who register here to ask a question get frustrated and never return. But I had no idea things were the same “behind the scene” on IRC and on the mailing list. I had always thought that people with access to the servers talked with each other and that they actually did “collaborate” in the best sense of the word. Seeing that this is apparently not (or no longer) so I’m wondering how this project is going to continue and how it ever came off the ground in the first place …

    > Please, wait me until I finish to puzzle out with Xorg, upgrade it.

    … I will, and I have to say that I am mightily impressed by your enthusiasm and your willingness to invest time and work to realize your innovations. But, as I said above, I’m not so sure if such systemic changes are what’s needed right now, when there are problems all over the place that need sorting out if making a RC is the aim.

    > Anyway, I guess that “libtool” branch of depends is unnecessary.

    That would be great. Installing bulky Perl to get moc working defies the purpose of using a small console application.

  • Profile picture of Trixar_za Trixar_za said 10 months, 3 weeks ago:

    I wouldn’t say that fixing Xorg would be unneeded. It’s where most of the issues come from so it’s the #1 thing that needs fixing. Next is sound, udev support and sda naming conventions. We’re not going back to PCManFM for the simple reason that the old one depends on HAL (which would make you complain that SliTaz doesn’t support certain devices) and the new one depends on gvfs (which would make you complain that SliTaz is bloated).

    As for why we’re answering less and less: For me it’s that I’m tired of answering the same questions over and over. I even wrote an IRC bot to answer FAQs automatically because of that. The thing is, if people bothered to google or search the forum for the answer then 9/10 it pops up. My second problem is some users have a sense of entitlement or is just plain rude. I either react harshly to this or I just outright ignore it. Also idling on IRC for a minute after asking a question then quiting with a shitty remark does not constitute asking for help. And my third problem has little to do with SliTaz or it’s users. I just don’t have the time like I used to anymore.

  • Profile picture of gibor gibor said 10 months, 3 weeks ago:

    Senza offesa e con il massimo rispetto verso Pankso ed il forum vi dico questo.

    La mia opinione conta meno di nulla, ma tuttavia colgo l’occasione di dirla per l’argomento trattato.

    Le doti di slitaz è indiscutibile che sono tutte nel kernel, unico veloce e intrigante, con questo affascina tutti quelli che lo provano.
    Il merito di questo è tutto di Pankso che non finirò mai di ringraziarlo.

    In passato avevo già espresso un parere negativo sulla roadmap di slitaz, dove ha raggiunto secondo me il culmine fra la versione due e la tre, dalla cooking successiva e la quattro siamo usciti dalla corsia d’accelerazione.
    È più facile raggiungere il successo che mantenerlo si dice, e in questo si raccoglie anche slitaz.
    Riassumo tre punti fondamentali che non mi piacciono.
    Ho letto in giro già annunciato che la prossima versione sarà con spacefm, scelta da me assolutamente non condivisa, altro il bug oramai storico delle ultime versioni, la schermata grafica di scelta lingua e tastiera, un vero trabocchetto per tantissimi utenti.
    Ultima osservazione su tazpanel non so come fate a trovarlo interessante, tutto quello che macina fuori controllo nel pc mi irrita figurarsi se riesco ad apprezzarlo.
    Reputo addirittura terribile lo stile fuori catalogo di gnome e le icone di faenza.

    Qualora cambiaste modo di vedere le cose, sarò felice di contribuire nei limiti delle mie possibilità.

    Un cordiale saluto a tutti.

    English version

    No offense, and with the utmost respect and Pankso this forum tell you this.

    My opinion has less than nothing, but nevertheless I take this opportunity to say it to the subject matter.

    The qualities of slitaz is indisputable that they are all in the kernel, only faster and intriguing, with this fascinates all those who try it.
    The merit of this is all about Pankso that I will never cease to thank him.

    In the past I have expressed a negative opinion on the roadmap slitaz, where he reached the pinnacle in my opinion between version two and three from the cooking and the subsequent four of us left the acceleration lane.
    It is easier to achieve the success that they say keep it, and in this it also collects slitaz.
    I summarize three key points I do not like.
    I have read around already announced that the next version will be with spacefm, chosen by me absolutely not shared, the other historical bug now the latest versions, the graphical login screen to choose language and keyboard, a real pitfall for many users.
    Last observation of tazpanel do not know how do you find it interesting, everything grinds out of control in the pc irritates me alone if I can appreciate it.
    I think it is terrible even the style of gnome out of print and icons, faience.

    If you change your way of seeing things, I’ll be happy to contribute to the extent of my ability.

    A cordial greeting to all.

  • Profile picture of sidhu sidhu said 10 months, 2 weeks ago:

    Simplicity is power – I have a wiki running on a Neoware thin client and 8 MB or RAM. How cool is that! We have version 4 at a major hospital in Kenya doing daily duty and I had to revise and fine tune the thing for weeks before we settled on something which works for the locals. And I have not had to respond to a SINGLE support call for the last 8 months!

    Slitaz shines and I’d love to help out in getting a more stable version 5.

  • Profile picture of fatmac fatmac said 10 months, 1 week ago:

    I tried ‘rolling’ & like what I saw; however, there are two problems with the GUI installer, that I found.

    1) It doesn’t find a pendrives UUID automatically, & I could not write anything into that field.

    2) It can’t install grub because it says no disk is selected; even though I had selected sda2 as the root partition.

    I also tried to install using tazinst, but I could not get my pendrive’s UUID to enter into the script, & it wouldn’t work using /dev/sdb1

    I hope this info is helpful toward the release of version 5.

  • Profile picture of devl547 devl547 said 10 months ago:

    >I do not want to SliTaz has been slow because of its simplicity.

    Being fast, but too complicated to modify? No way!

  • Profile picture of Aleksej Aleksej said 10 months ago:

    Complexity — a relative term. What is complicated for you? What do you expect to modify in the future? What experience of modifications you already have?

    Is cook complex? Do you want to change it?
    Is tazpanel complex? Is tazweb complex?

    I say again that the shell has a certain “bottlenecks.”
    Do not want to rewrite everything. But, sometimes even in the fast C code are inserts in assembly language.

    So, let me rephrase you:

    Being fast, but too complicated to modify? No way!

    Being fast, and moderately complicated in the pair of functions specially written in C? Sounds interesting!

  • Profile picture of holkfoor holkfoor said 10 months ago:

    I am very happy with SliTaz. SliTaz 4.0 is great, fast, easy and very functional for me. Congratulations to the developers.