Ideas for improvements in future releases

Forum Forums Official Releases antiX-19 “Marielle Franco, Hannie Schaft, Manolis Glezos” Ideas for improvements in future releases

Tagged: 

  • This topic has 161 replies, 22 voices, and was last updated Feb 5-2:12 pm by marcelocripe.
Viewing 15 posts - 16 through 30 (of 162 total)
  • Author
    Posts
  • #46301
    Member
    Avatarskidoo

    Due to many upstream changes in Debian, python2 and a lot of gtk2 will go.
    [..]
    We will keep to IceWM/Rox default

    Do you know of anyone who is rewriting a GTK+ 3 compatible version of rox-filer?
    If not, and you are averse to adding packaged gtk2 libs into antiX repos, we must wave byebye to rox

    apt depends rox-filer
    rox-filer
    Depends: libc6 (>= 2.15)
    Depends: libgdk-pixbuf2.0-0 (>= 2.22.0)
    Depends: libglib2.0-0 (>= 2.35.9)
    Depends: libgtk2.0-0 (>= 2.24.0)
    Depends: libice6 (>= 1:1.0.0)
    Depends: libpango-1.0-0 (>= 1.14.0)

    #46302
    Forum Admin
    anticapitalistaanticapitalista

    Do you know of anyone who is rewriting a GTK+ 3 compatible version of rox-filer?
    If not, and you are averse to adding packaged gtk2 libs into antiX repos, we must wave byebye to rox

    No I don’t.
    I’m not averse to keeping gtk2 libs in post-buster antiX repos if needed.
    At the moment, the libs you mention are still in Debian bullseye.

    If we know which gtk2 libs to keep, then I can package them for post-buster antiX repos (if required).

    • This reply was modified 5 months, 1 week ago by anticapitalista. Reason: added extra info

    Philosophers have interpreted the world in many ways; the point is to change it.

    antiX with runit - leaner and meaner.

    #46304
    Member
    manyroadsmanyroads

    I’d be happy to provide more help… if I could get things running on my gen 10+ Intel processors. Maybe after Bullseye. fwiw I have tried numerous times to build a version of antiX and install dwm (my wm of choice). But I can’t get a stable enough install to complete the tasks needed. No complaint, MX works just fine (ahs).

    Edit: I’ll give it another go to see what happens….

    • This reply was modified 5 months, 1 week ago by manyroads.

    Pax vobiscum,
    Mark Rabideau - http://many-roads.com
    "For every complex problem there is an answer that is clear, simple, and wrong." H. L. Mencken
    dwm ~Reg. Linux User #449130
    20 Jan 2021 ~ "End of an Error"

    #46311
    Member
    Avatarskidoo

    Does ceni’s “looks” put people off? Is there a way to dress it up with a yad-type “skin” — like live-remaster has?

    sudo updatedb
    locate Ceni
    ^—- capital C
    You discover 2 files, both are perl scripts. (plaintext, can view in text editor)
    Notice that one of ’em bears a filename “backend”. Yeah, the other is “the skin”, using ncurses.
    By my cursory reckoning, about 43 separately handled yad dialogboxes would be required to encompass the same functionality.
    .
    What do you anticipate would be the “value add” of some other skin? Superior info/help? If so, how about using a wrapper to launch Ceni ~~ the wrapper would launch a helpdoc (with embedded screenshots) into dillo immediately prior to presenting the Ceni UI.

    an option offering Devuan Ceres with just jwm (perhaps)

    JWM_Kit: A set of Graphical Apps to simplify use of JWM (Joe’s Window Manager)
    scroll down when visiting the linked page (to see the various beautiful screenshots)
    https://codeberg.org/JWMKit/JWM_Kit

    FYI, the JWMKit author has already (June 2020) released a “Devuan Ceres with just jwm” ISO
    (actually, I didn’t verify that it is based on Ceres)
    http://dev1galaxy.org/viewtopic.php?pid=25777

    ___________________________

    2 peeps in this topic mentioning, again, the prospect of merging with Devuan.
    Again, as-in, various prior topics I have belabored to explain why doing so is unappealing, IMO.

    Devuan uses “amprolla”, most of its packages are hosted on // served from Debian repositories.
    Devuan is not forthcoming, does not advertise this detail & many of the Devuan users are unaware.
    How many packages are the Devuan devs actually packaging and maintaining?
    (I knew the count as of about a year ago, researched while editing the nosystemd wiki)
    How many in-house software projects, aside from refractasnapshot and refractainstaller, are the Devuan devs actually maintaining?

    antiX kernel(s) include framebuffer module, and wide range of component firmware support.
    Devuan kernels?
    .
    antiX stable repos share buildtime//runtime compatibility with Debian stable and, more importantly, with MX Linux stable.
    A repeated topic of help requests at dev1galaxy forum: “Can I use MX repos?”
    The consistently repeated answer: “dunno, don’t care, you get to keep the pieces when it breaks, don’t make a Frankendebian”
    .
    antiX forum, the level (and quality) of support… vs the “in Devuan, no one can hear you scream” forum

    #46492
    Forum Admin
    AvatarSamK

    Any official position by the main devs? I think this is a fruitless discussion if we are going to get shut down.

    Apologies for the delay…

    I’m not too keen on the idea of a Welcome app.

    If instead of a Welcome-screen we develop a gui to manage desktop-session-antix options/configuration, and have it as the “STARTUP_DIALOG_CMD” that Dave mentioned? Would it work if the first tab was some kind of “Quick setup” introduction, as BobC suggests?

    I also find a welcome screen unappealing, particularly if it is shown by default. Almost no-one reads them when evaluating a distro. In general users tend to be interested in whether the distro works, then expect it to work exactly to meet their preconceptions, all this straight-out-of-the-box. Welcome Screen is a polite misnomer, in my view it is more accurately How to Fix or How to get Your System Going. It sends out the unconscious message the distro is problematical. Additionally over the last few decades there has been a obvious tendency for people to do little or no reading and researching. Instead there is a general expectation that others will always guide them, as can be seen from repeated forum questions.


    AntiX seems to be leaving the ‘basic’ distro idea behind these days, which is a shame, I came from #!, when it ceased, several years ago now, because it was a basic lightweight distro, but it has been growing ever since – are you trying to court ex Windows users(?) nowadays, rather than being the basic lightweight distro suitable for old machines that it originally was?

    Too many ‘improvements’ are spoiling the distro, I fear.

    A very good point that all should consider.
    antiX should not lose its identity in search of improved usability.
    One part of the antiX identity is the combination of Rox-filer and a window manager as default.
    That is not going to change (well until rox-filer is no longer usable that is).

    I agree entirely with those comments.

    Focusing on new users coming from Windows is in my view a fruitless exercise. The basic concept is flawed but is raised every few years. If it was going to work there would be a distro in existence that had already succeeded, to which all would go.

    Usability is not an absolute, it is a qualified concept. That means there can never be a universal agreement on what represents an improvement. Too often an idea represents what an individual user thinks is an improvement, i.e. something which that user or local group might desire. Sometimes it is proposed because something similar exists in another Linux distro or Windows.

    It must be remembered, antiX is not Windows (or any other distro) and will never try to be. From its inception antiX had, and continues to have, a goal which is discussed less often than being lean and mean. Encouraging the user to learn Linux and understand what they are doing rather than just blindly clicking on an icon is every bit an antiX tenet as being lean and mean.

    In releases since antiX-13, in my opinion, the main menu has become become overcrowded to the point of hampering finding what is being looked for. This complexity grew mainly from, in my view, a now obviously failed idea to improve discoverability of progs. It even duplicates some things that are more appropriately accessed via the Control Centre.

    We are now in danger of exacerbating that overloaded state. antiX has not been improved by this drive to put everything in the menu together with suggestions to ship with more items on the task bar. One reason antiX is so configurable, is to put the choices into the hands of the user. That fits with another goal of the distro, to enable the user to fine tune antiX to suit their personal choice. To this end, antiX should return to shipping a greatly simplified user interface, having far fewer items in the main menu and an almost bare task bar. Let the user produce the interface they want. This is can be supported by the now much more active user forum.

    We have already seen the adoption of some useful user scripts. In the main, they have extended the scope of antiX, rather than attempting some disputable improvement of an existing mechanism or a questionable easier way of working. Extending the scope is more likely to enhance individuality of antiX. The current work to restore the SIS graphics functionality and support of HD displays are encouraging examples of extending the scope of antiX.

    Much of the current discussion seems to encompass the idea of making things easier by emulating other distros. antiX has a strong individual identity. It stands out from the herd because it does things its own way and covers aspects eschewed by others. Making it look and behave like other distros moves it closer to the rest of the herd. It dilutes its individuality.

    The adoption of some user contributions seems to have encouraged a subliminal anticipation that others under discussion will be treated similarly. It is a better state that a user produces a script with the expectation that it will be sufficient they alone adopt it.

    anticapitalista has mentioned some areas where work is needed for the next antiX release. Perhaps some of the current user enthusiasm to help could be directed towards those.

    #46496
    Moderator
    christophechristophe

    +1

    #46497
    Moderator
    christophechristophe

    Any official position by the main devs? I think this is a fruitless discussion if we are going to get shut down.

    I think this is a very fruitful discussion and well worth continuing.

    How I see future antiX (ie the next stable ‘bullseye’ version).

    Due to many upstream changes in Debian, python2 and a lot of gtk2 will go.
    That includes a few apps that have been ‘standard’ on antiX eg disk-manager, lxkeymap, lxappearance(?), gtkdialog and others.

    Our in-house apps also have to be ported from python2/gtk2 as well.
    Dave has already started porting them (in testing/sid branch). Most seem to work ok, but there are bugs – testing will be needed.

    The live-system (including frugal, remaster, snapshot etc) seems to be working ok. However, the UEFI side probably needs to be improved to cater for more and more frugal/live users booting via UEFI since they have newer hardware.

    Documentation certainly needs updating.

    All the above will need to be done for the next stable release once Debian11 is out.

    Making antiX more usable:

    Certainly, IMO, including user written scripts have made antiX more user friendly.
    eg the IceWM toolbar script, antiX-wifi switcher, android-device-connect, fast_personal_menu_editor, icewm-menu-recent-files and others, though I personally don’t use any of them. The downside to including these useful scripts is that the menu could be considered to be ‘bloated’.

    Having some sort of script/app for hd screens is necessary.
    Updating the control centre from gtkdialog to using Xecure’s dialogbox version will probably be needed at some point.
    I’m not too keen on the idea of a Welcome app. However, that doesn’t mean that it will not be included if most of us here want it.

    What I don’t want is for antiX to be turned into something it is not meant to be.
    For example, some in the past basically wanted it to be like a Debian lubuntu. If people want lxde, Xfce etc and ‘bells n whistles’ out of the box, then they really should try another distro.
    We will always be systemd-free.
    We will keep to IceWM/Rox default
    We will ship with/promote our own (less known or even unknown) tools/apps over bigger and better known versions eg droopy, 1-1-voice etc.
    We will try to keep CPU and RAM usage at idle down to a minimum. We mainly achieve this through not running unnecessary services on boot.

    Of course, antiX is far from perfect and needs to adapt to changes posed upstream.
    We have already done so eg by no longer making the base iso fit on a cd.
    Other compromises will no doubt follow at some point in the future.

    #46503
    Moderator
    AvatarBobC

    The Internet connection detection and connect helper is working. It should be tested more and added ASAP, IMO. I don’t see why we should wait for a new major version on that?

    The HiDPI screen detection and adjuster script is close. It needs work on multiple display setups. I will try to see if I can work on it over the weekend as I have multiple 4K screens of different sizes connected now. Again, I think once its working it could be added to the repos and then added on the next minor release.

    Both of those would be run from ~/.desktop-session/startup.

    The last major issue we seem to hit a lot of is people not selecting the right language, so then they will likely have lots of problems if they can’t read the screens. They should have done that in the boot, but maybe need another chance as soon as they come in the first time.

    Maybe we should be saving the screen layout and language like we save the internet connection info, and using them when coming in if found. I still don’t think it would be bad to show them the links to the help, too at that point, if nothing else. Name it what you want, but if they don’t get those things right, they get messed up and potentially frustrated and upset, and you see the results here and on DW. These are small things that could be done to improve. The people with those older machines typically are running them because that’s what they have, and probably aren’t so technically savvy, so anything we can do to help them a bit would probably be appreciated.

    Just my 2 cents worth… BTW, did you know another distro copied and tweaked our IceWM setup? Imitation is flattery 🙂

    • This reply was modified 5 months ago by BobC.
    #46510
    Member
    Avatarolsztyn

    I also find a welcome screen unappealing, particularly if it is shown by default.

    I am just a user, so I do not mean to influence decisions and direction but let me just mention my perspective as user:
    – antiX should maintain being lean and fast as much as possible under circumstances. It should boot to user desktop without any intermediary such as ‘welcome screen’. Lean does not need to imply mean however. IceWM as default is usable enough and can be made aesthetic too, if you find a good theme out there… However it took me long time to tweak parameters to make it behave correctly, with the help in this forum. After benchmarks of various ‘DEs’ I came to conclusion that unfortunately none of the common ones is code efficient, being just bloat, so not worth wasting memory footprint. The only one that is lean and almost matches overall memory footprint of antiX/IceWM is DE called ‘Fly’ adopted by Linux Astra (Thanks again Dzhigit!). But it seems not open source, so unlikely to be added to antiX.
    – I agree with much of what SamK pointed out that default menu screen became way too crowded. I had to simplify by removing some entries, e.g. most IceWM configuration items and more. Once IceWM is configured they become just noise on the menu and prone to accidental changes if left there.
    – antiX should maintain Debian compatibility. This is as long as it does not defeat the identity of being lean. I do not know gtk2/gtk3 difference but it seems in general that Linux in general has lost its own identity by blindly trying to follow Windows bloat, while past efficiency could have been still maintained with old leaner programming.

    We will always be systemd-free.
    We will keep to IceWM/Rox default
    We will ship with/promote our own (less known or even unknown) tools/apps over bigger

    – I like systemd-free architecture. But is it not going to become increasing difficult to maintain? This is just a question, not questioning…
    – IceWM default is fine with some tuning. However you not want to enhance Fluxbox with Tint2 or something, so as memory footprint is maintained low and usability is enhanced? Also Rox seems entailing additional complexity of maintaining older, unsupported by Debian libraries going forward, so why not SpaceFM, which seems much superior to Rox? Just a question…
    – Provided with antiX tools such as ISO-Snapshot, Live-USB-Maker and Remaster are a great value of antiX and should be maintained at all cost.

    he live-system (including frugal, remaster, snapshot etc) seems to be working ok. However, the UEFI side probably needs to be improved to cater for more and more frugal/live users booting via UEFI since they have newer hardware.

    – Indeed. Boot enhancements would be great, and to simplify boot process. I would suggest also bringing up in the boot menus the discovery of bootloaders and grubmenus, currently nested in deep submenus.

    Just my two cents as user…
    Thanks and Regards…

    #46525
    Moderator
    christophechristophe

    (Those quotes above attributed to me were actually my quoting anticapitalista, but.. it’s open source, right?)

    Just as an aside: I personally LOVE my tint2 setup on fluxbox. But I don’t expect others to want it on their computer. I really like the simple interface of antiX, and I like to tweak it myself. Just my preference. For example, I really don’t like IceWM. I know others DO like it. That’s part of the appeal of antiX, isn’t it? You can make it what you want.

    So I hope antiX stays light, fast, and configurable.

    #46529
    Member
    AvatarPPC

    – I like systemd-free architecture. But is it not going to become increasing difficult to maintain? This is just a question, not questioning…
    – IceWM default is fine with some tuning. However you not want to enhance Fluxbox with Tint2 or something, so as memory footprint is maintained low and usability is enhanced? Also Rox seems entailing additional complexity of maintaining older, unsupported by Debian libraries going forward, so why not SpaceFM, which seems much superior to Rox? Just a question…
    – Provided with antiX tools such as ISO-Snapshot, Live-USB-Maker and Remaster are a great value of antiX and should be maintained at all cost.

    – I’m system init agnostic- antiX without systemd flies on even my slowest computer- I’ll stick with it
    – I don’t like Rox, even after having learned to configure it to be a more “normal looking” file manager- but it uses less resources than SpaceFM and is almost as configurable as it… and anticapitalista already stated that he’ll keep the needed libraries to keep it working in future antiX versions. We already have spacefm out of the box, is just a matter of making it your default browser… If anything, I think spacefm only lacks, out of the box, a “recycle bin”/”trash”, but I have that covered…
    – antix tools are basically the biggest part of what makes this system so unique… so I’ll guess they’ll have to pry them from my cold dead hands… 🙂
    – I currently using min-fluxbox with a custom tint2 toolbar on all my 3 systems- it looks great, and runs light as a feather…

    – The only thing I miss in antiX? The fact hat no one found a way to make snaps work without systemd- there were some apps I wanted to try that way, other than that, it’s as perfect as it gets…

    P.

    #46540
    Member
    Avatarolsztyn

    : Those quotes above attributed to me re actually my quoting anticapitalista, but.. it’s open source, right?)

    I was aware this was anti’s quote. I just quoted that quote and the forum system attributed that to you, not recognizing it was referencing anti’s comment.

    Just as an aside: I personally LOVE my tint2 setup on fluxbox. But I don’t expect others to want it on their computer. I really like the simple interface of antiX, and I like to tweak it myself. Just my preference. For example, I really don’t like IceWM. I know others DO like it. That’s part of the appeal of antiX, isn’t it? You can make it what you want.

    By saying this my point was that if antiX provides Fluxbox along IceWM, I do not mind IceWM is default but if enhancement such as Tint2 can be made to Fluxbox without increasing memory footprint then would this not make sense to embrace such enhancement, if it is a low hanging fruit? This would contribute to overall value to antiX as enhanced second WM, not replacement of IceWM… Additional enhancements were done to IceWM, after all but it seems Fluxbox is out of favor… I am not questioning such decision, just an observation…

    – I don’t like Rox, even after having learned to configure it to be a more “normal looking” file manager- but it uses less resources than SpaceFM and is almost as configurable as it… and anticapitalista already stated that he’ll keep the needed libraries to keep it working in future antiX versions.

    I am using SpaceFM as File Manager only. No Space-IceWM, no Icons on desktop and by doing just that the memory use is significantly less than Space-IceWM. I may try Rox and compare if any memory savings at all, not to flip to RoxFM, but just for benchmark. If it is just a few Mb, then I do not understand all this trouble with separate libraries. If the difference in memory use is significant the it is a different story…

    – antix tools are basically the biggest part of what makes this system so unique… so I’ll guess they’ll have to pry them from my cold dead hands…
    – I currently using min-fluxbox with a custom tint2 toolbar on all my 3 systems- it looks great, and runs light as a feather…

    My coming originally from mainframe systems, where system integrity and recovery were of utmost importance, such antiX Live tools are what makes antiX a serious system with integrity and recovery infrastructure included. In comparison most other Linux systems are mostly for just playing around with no particular recovery needs.
    Just my two cents as a user, not a developer…

    • This reply was modified 5 months ago by Brian Masinick.
    • This reply was modified 5 months ago by Brian Masinick. Reason: Incorrect quote cited
    #46545
    Member
    Avatarolsztyn

    I am using SpaceFM as File Manager only. No Space-IceWM, no Icons on desktop and by doing just that the memory use is significantly less than Space-IceWM. I may try Rox and compare if any memory savings at all, not to flip to RoxFM, but just for benchmark.

    Just to update:
    I have done a rudimentary benchmark on freshly booted antiX with SpaceFm/IceWM and then RoxFM/IceWM and my memory reading was with my setup on Thinkpad X61/4Gb:
    – antiX with SpaceFM/IceWM with no icons on desktop, so just IceWM – 223Mb
    – antiX with Rox/IceWM – 229Mb

    I am not sure if it is comparing apples to apples or apples to oranges, as someone may point me to imprecision in such testing but with this comparison I did I have slightly better memory footprint with IceWM with SpaceFM (Not Space-IceWM) comparing to Rox-IceWM.
    So if if my testing is valid and not comparing apples to oranges then my conjecture would be that there is no particular evidence RoxFM saves memory footprint of antiX vs. SpaceFM.
    Please correct me where my testing is not precise and result should be different…
    Thanks and Regards…

    • This reply was modified 5 months ago by olsztyn.
    #46573
    Moderator
    AvatarBobC

    “antiX is a fast, lightweight and easy to install systemd-free linux live CD distribution based on Debian Stable for Intel-AMD x86 compatible systems. antiX offers users the “antiX Magic” in an environment suitable for old and new computers. So don’t throw away that old computer yet! The goal of antiX is to provide a light, but fully functional and flexible free operating system for both newcomers and experienced users of Linux. It should run on most computers…”

    I didn’t write that, but I agree with it. We should do what we can for people to realize that goal. MS has literally decided to toss all these people with older computers by the wayside. Maybe they don’t make enough money from them to bother. We don’t need to make a profit. Profits here could be measured in smiles.

    I have some ideas to nix the Welcome Screen concept itself, and replace it with something that will help the newcomers but without bothering experienced users, and not cost much in time, effort or space, but to be honest, I don’t want to suggest it if the goal above is no longer true.

    #46578
    Moderator
    AvatarModdIt

    On Distrowatch it looks this way:
    antiX is a fast, lightweight and easy-to-install Linux live CD distribution based on Debian’s “Stable” branch for x86 compatible systems. antiX offers users the “antiX Magic” in an environment suitable for old computers. The goal of antiX is to provide a light, but fully functional and flexible free operating system for both newcomers and experienced users of Linux. It should run on most computers, ranging from 256 MB old PIII systems with pre-configured swap to the latest powerful boxes. 256 MB RAM is recommended minimum for antiX. The installer needs minimum 2.7 GB hard disk size. antiX can also be used as a fast-booting rescue CD, or run “live” on a USB stick, with or without persistent file storage.

    Important is we let anticapitalista do the shaping, he is the head cook and already made a clear statement about the recipe.

Viewing 15 posts - 16 through 30 (of 162 total)
  • You must be logged in to reply to this topic.