SlackwareThis Forum is for the discussion of Slackware Linux.
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
On one hand, the reason I love Slackware is its stability, and Slackware is as stable as it is because it has relatively few "official" packages, which Patrick + friends test fairly exhaustively.
The work involved in exhaustively testing a set of packages for mutual incompatibility is approximately proportionate to the *square* of the number of packages tested (actually less, because some packages are obviously not dependent, like xpdf and jove). (Three packages, A, B and C, require testing for interactions AB, BC, and AC, while four packages W, X, Y, and Z require testing for interactions WX, WY, WZ, XY, XZ, and YZ -- twice as many potential sources of incompatability.)
I actually wouldn't mind if Slackware releases got *more* QA testing, because problems do occasionally creep in there (like CUPS's problems in 13.1, or the installer's USB woes in 13.37), so keeping the number of packages small is a good thing.
The Solution is, as others have already pointed out, for users to contribute to Slackbuilds. Slackware could perhaps mention the existence of Slackbuilds in the installation tool, so more users could take advantage of it, and I wouldn't mind sbopkg getting included in /extras so pulling in everything else is trivial, but aside from that PV has quite enough on his plate.
On the other hand, I've often wished Slackware had better support for "enterprisey" type environments, like Turbo Linux did in its prime. The text installer is wonderful, and I miss it when having to deal with other distributions' installers, but it doesn't really cut it when one needs to install hundreds of servers at a time.
A project I've been sizing up for a while is writing a "cluster linux" slackbuild which converted a standard Slackware install into the "master" node of a TurboLinux-style cluster. It would set up zeroconf, ftp, and dhcpd for PXE-install, and create a local Slackware and Slackbuilds repository for the new nodes.
Additional nodes would PXE-install from the master node, and that install would include the "cluster linux" package, and some additional packages for compute and storage clusters: some additional perl and python modules, zeromq, openstack, hadoop, pig, sawzall, mffirewall, nagios/nagiosgraph (and appropriate plugins for monitoring server health: temperature, disk errors, load avg, etc), mrtg, rrdtool, puppet, and squid.
The onus would be on the package developer(s) to make sure these Slackbuild packages were all trouble-free, of course, leaving it to PV to focus on Slackware's base packages and making those bulletproof for all Slackware users (which he does magnificantly).
Every time I start this project, however, my other projects-in-progress grow little appendages and point at me, laughing. :-) Someday!
Koffice is now Calligra. I must admit I don't use it myself but people who do have told me it has started to become really nice. So personally I would prefer not drop it. Though I am also in favour of adding LibreOffice.
Slackware is fine as it is. Honestly, the only things that SHOULD ever be added are those things necessary for the function of existing programs already in the tree, like llvm was added to support the Gallium3D drivers in Mesa. No other extras are needed unless those extras are for expanded minimals to existing tools like Bash Completion is to Bash.
Also, functionality of packages in extras needs to be taken into accountability as well. Chromium, as someone suggested is a nice idea, but Chromium doesn't have many of the same features as Google Chrome does. To take out Chrome and replace it with it's open source sister project sounds nice, but you have to think about what is useful in a mainstream setting, not just private settings. Other projects from SlackBuilds that are problematic with compiles are the problems of the SlackBuild maintainers, not Slackware itself.
As far as LibreOffice goes, I'd actually say that I'd feel better with LibreOffice added, but ONLY if we could remove something like KOffice (or KDE and move it into SlackBuilds like Gnome was and use XFCE as the default UI of Slackware).
Slackware is a growing and ever improving project but it's already reaching the size of Dual-Layer DVDs. It really shouldn't be this big.
You know, when GNOME was dropped, I was highly disappointed, so much so that I temporarily quit Slackware. I say temporary because I was back with Slackware in less that six months. I got to like KDE (using AlienBob's Ktown right now in fact), KDE is really coming around and getting better with each update. That is my humble a??'d opinion anyway. I like XFCE too, except for one small detail, which has finally be fixed after all these years. So I may give it a shot, but alas I have grown quite used to KDE now so perhaps not.
My point is the desktop is not what makes Slackware, it's what is under the hood that makes Slackware. The desktop is just icing on the Slackware cake.