slackware becoming outdated
 

Anyone else feel that slackware is becoming outdated? Still on kernel 2.4 after 2.6 has been out for almost two years? And it seems like the distro remains pretty much stagnate when it comes to making impovements in the core system, i.e. package,etc.

Also, I wonder what will happen now that 64bit cpu's are taking over. We have a 64-bit windows, mac-osx, freebsd, and some other linux distro have gone 64 as well. But it doesn't seem like pat has any intention of doing this and will most likely pass it off to some else, slamd64 i think.

Overall, i think it is becoming more and more of a server only distro, and it's losing ground quickly when it comes to the desktop.

No :)

Current is updated, packages have been added/removed (firefox, gnome).

The decision to keep 2.4 kernel was discussed plenty of times, and I think it's a right choice for now. Maybe the latest releases of kernel and software are being more stable now, but not a few months ago.

Patrick's thing is stability over new stuff. The 2.4 kernel is tried and tested and works - your system won't fall over if you use it. He does, however, include a 2.6.x kernel in the /testing folder on the cds (and on the ftp sites) so you can install it if you wish. You can also upgrade every package manually if you wish.

Pfft!!!

Outdated?

Don't forget that Slackware was one of the first distros to support the 2.6 kernel (Slackware 9.1). Other distro took some major hacking just to get it booting.

And if you look in -current, I'd say for an OS devoted to stability it is quite up-to-date.

- kernel: even Slackware stable has both 2.4 and 2.6. I use 2.6.10 from stable on my desktop without any problems.
- package management: most Slackware users are convinced that it shouldn't evolve (i.e. no automatic dependencies checking), so it's a deliberate choice
- 64 bits: good question. Has Pat ever given any information on this subject?

Slackware aims to stability, and Linux Kernel 2.4.29 is a good choice.

There is an unofficial porting of Slackware on 64bit processors: http://www.slamd64.com/ I didn't try it (no 64bit on my machines!)

I don't think Pat wil support a 64bit version for a while. There is someone working on a Slackware based 64bit port. It's called Slamd64. I've been thinking about it, but waiting a few more months to see what happens with it. And I was wondering if I could install it in addition to Slackware without any problems. That would give me triple boot system, but I don't know if that is possible.

And I don't think Slackware is outdated at all. It doesn't have all the fancy graphical stuff the others do, but if you really want them, get the source code and install it yourself.

Okay, stability is good but lets not get carried away. Even if u use 2.6 kernel you don't get many of the advantage becuase everything else was compiled again 2.4 headers. I'm just saying that the slow progress to adapt new features is making it stable yes, but it is also making it more geared for servers. Not for desktop where people don't have to be parnoid about stability and want thing like 2.6 and everything linked agaist it. They also want it compiled with the most recent stable gcc and flags at least for a pentium 2 or better. If not 64 as many people already have an amd cpu. I've tried slamd64 and it flys.

Also for it to go 64 bit at this point is unlikely as slamd64 has pretty much owns that port. So it's future as a desktop OS seems rather bleek to me.


Also, i'm not saying it should do automatic dependency checking but maybe at least tell you want things are linked against would be nice. Or any innovation in it's core system would be nice.

It's all a matter of personal preference, I suppose, but if I were to release a Linux distro, I would make the same decision. I would include only the tried and true for stabilities sake. I would rather have an older kernel that it is solid, then end up with an install that had a bunch of unresolved quirks.

Note that I'm not saying that the 2.6 kernel is unstable, but the 2.4 kernel definately is NOT. And then when the 2.6 kernel has been thoroughly tested and used, and there's no doubt that it is stable, I would then upgrade. You know what I mean? I want a for-sure solid system.

I would also point out that Slackware is essentially a one man show (ie, Patrick). He has also had a rough time of things health-wise, which naturally would be a distraction. Personally, I think Slack is still the best distro out there - like everyone else though I'll be looking forward to the next release, as always -- J.W.

This is *not* meant to be rude:
there's plenty of bleeding-edge distros out there, ArchLinux or Gentoo for example. So what keeps you here in the Slackware camp?

I think Pat has his eye on the AMD64 but it is not the focus of his energy.
If you look at some of his recent build scripts there is provision for the AMD64

Recent build script for alsa-utils

I feel just the opposite -- I'm running 10.1 with kernel 2.6.11.10, kde 3.4, gnome 2.10, xfce 4.2.2, the latest xorg, the latest mozilla, the latest firefox ...

I use slack on my dell 700m laptop -- a new piece of equipment, not some old Pentium sever. I have suspend-to-ram working, CPU throttling, widescreen aspect ratio, internal wifi with wpa encryption, and just about everything else on the laptop works just fine.

I used to use Mandrake, Red Hat, and then various Debian distros. I tried slack about 3 years ago but kept bouncing around from distro to distro. I finally "got" slack within the past year and I'm totally addicted to it. I've learned more about linux and my system than I ever had in all prior years with those other distros combined. Everything on slack just works. I've compiled tons of packages from source, something that never quite worked for me on those other distros, especially the rpm ones. I also have finally gotten to understand how to compile my kernel, and how it all fits together.

For me, slack is perfect -- stable yet bleeding edge if you want it to be. I hope it never changes.

Let's not forget that not all distros race to bring out the latest and greatest. Debian, for example, on their release distro concentrates purely on stability, you have to get Sid if you want to be cutting edge. Look at the distros that do bring out the new stuff regularly and then look at the complaints from users about the stability and things breaking.

One thing I do like about Slack is that it is bug free - I then choose to break it by going to -current :)

Slackware-current is probably one of the most current binary distributions available. All equally up-to-date distros I know take much more effort to maintain or big iron.

Let's take the highly acclaimed Gentoo: It downloads the latest source of Gnome, compiles them, downloads further packages to resolve the dependencies, compiles them, and so on, and finally install everything. Thus it only takes a week or so on weather simulation class cluster to update Nautilus. Of course, it's worth the wait, as all the newly installed programs rund 2 percent faster as they are compiled on your system. ;-) (NO FLAMES, please, dear Gentoo users --- your distro *is great*, just not for me; if I wanted a source based distro, I'd go for ROCK Linux, anyway).

Let's look at Debian and RPM based distros: Either they aren't as current as Slackware-current, or they do a lot of vendor patching, introducing vendor dependencies. So, I don't see anything wrong or outdated with Slackware-current.

Personally I prefer to run a released version (10.1) on my desktops *and* servers, with a few packages taken from current and 3rd party sources. I feel quite up-to-date with that configuration. And it's rock-solid.

gargamel

I have a amd64 processor, and the latest Slackware runs fine.

You probably aren't using the 64-bit processor to its full capability, though, pm. Code compiled for a 32-bit processor will run on a 64-bit processor, as long as the manufacturer of the processor knows their arse from a hole in the ground.

I'm a bit skewed in that direction... I try to avoid Intel when I can. In fact, I only just ordered my first *new* Intel-based system since my '386 DX. AMD is generally cheaper, and does what I need it to do. (I've bought second-hand systems for use as servers, and am currently running a P-II-350/192MB for my main webserver. The new system is going to replace that, and is a P4-3.2/2GB that I'm getting for a song.)


Anyway, I don't think that slack is getting outdated at all. The 2.6 headers are on the CD, and can easily be installed. There's a 2.6 kernel on the 2nd CD as well. I would not be surprised to see Slack 11 as a 2.6-based system, and I would not be surprised to see Slack 11 before the end of the year. As everybody else has said, Pat's focusing on stability over being on the front lines. That's the main reason I keep coming back to Slack, and that's certainly the reason that I will be installing Slack on the new server. There's nothing to stop you from compiling your own (and in fact, I do for Sendmail, Apache and PHP) if you want to improve performance.

(As an aside, does the 2.4.29 kernel in Slack 10.1 support Hyperthreading? I'm planning on going 2.6 anyway, but it's a point of interest.)

First, that babble about 2.4 and 2.6 headers is just nonsense. I have no idea where you got your info.

Second, the newest gcc 4.0 isn't included because it's really not "stable". There are alot of things that still need to be fixed. I'd rather wait a little while then see Pat just throw packages into current even though there may be huge flaws in them. So what if the newest version of <insert name> isn't in current 5 seconds after the developers post an announcement to their mailing list. Its more current then most any distro. If you want bleeding edge gotta have it now then use gentoo.

Third, pat has enough already on his plate. He doesn't need a whole other architecture to maintain. That's just stupidty. He's one person, not 20. Thats why there is slamd64.

I also use slackware as my desktop and server. It is an excellent desktop distro. Why wouldn't anyone want stability on their desktop? You like having things crash on you and act unstable? Slackware is also very current. I'm not running it on a 486. I've got an AMD64, 1GB Ram, Nvidia 6600 GT, 200GB SATA HDD, 12x DVD/RW drive and am running dual 19" LCD panels. I also run slackware on my brand new Toshiba tecra m3 which has all brand new hardware. 95% of the hardware works fine, the rest doesn't because it's not supported in linux period.
So your view that slackware's future as a desktop OS as "bleek" is very misguided and uninformed. I'm not saying Slackware is going to take over the world but it is as always going to have its own user base.

You want to see what a program is linked against?

root@Venus:~# which xmms
/usr/bin/xmms
root@Venus:~# ldd /usr/bin/xmms
/lib/libsafe.so.2 (0xa7fe5000)
linux-gate.so.1 => (0xffffe000)
libSM.so.6 => /usr/X11R6/lib/libSM.so.6 (0xa7fc5000)
libICE.so.6 => /usr/X11R6/lib/libICE.so.6 (0xa7fad000)
libXxf86vm.so.1 => /usr/X11R6/lib/libXxf86vm.so.1 (0xa7fa8000)
libxmms.so.1 => /usr/lib/libxmms.so.1 (0xa7f9b000)
libgtk-1.2.so.0 => /usr/lib/libgtk-1.2.so.0 (0xa7e7d000)
libgdk-1.2.so.0 => /usr/lib/libgdk-1.2.so.0 (0xa7e4b000)
libgmodule-1.2.so.0 => /usr/lib/libgmodule-1.2.so.0 (0xa7e48000)
libgthread-1.2.so.0 => /usr/lib/libgthread-1.2.so.0 (0xa7e45000)
libglib-1.2.so.0 => /usr/lib/libglib-1.2.so.0 (0xa7e24000)
libpthread.so.0 => /lib/tls/libpthread.so.0 (0xa7e12000)
libdl.so.2 => /lib/tls/libdl.so.2 (0xa7e0e000)
libXext.so.6 => /usr/X11R6/lib/libXext.so.6 (0xa7dff000)
libX11.so.6 => /usr/X11R6/lib/libX11.so.6 (0xa7d35000)
libm.so.6 => /lib/tls/libm.so.6 (0xa7d12000)
libc.so.6 => /lib/tls/libc.so.6 (0xa7bf6000)
/lib/ld-linux.so.2 (0xa7feb000)
root@Venus:~#

I'm not exactly sure what you mean about "innovation in its core system". Alot of innovation comes from software. Pat just maintains the distro not creating the software it uses.

From your remarks I get the impression that you just want the newest whiz bang thing even though you have no clue about it and what it is or does. You want all the newest packages the second they are available and you think anything less is "stagnation". I used to be like that too, then I realized I didn't care. My systems run without a hitch, I update them with swaret when I get an email from my server saying that slackware has been updated. I run 2.6.11 kernel on my servers and laptop and 2.6.10 on my desktop because it's so stable there is no need to upgrade my desktop. What does it matter if you have version 1.1.5 and 1.1.6 is released of whatever program. A rule of thumb is that minor version numbers are just bug fix releases so unless you're affected by it in some way it doesn't really matter. Besides Pat is always quite diligent with releases anyways. They may not be in current the next day but when they do I *know* that nothing will go wrong when I install it. But I'm crazy, I like a stable system that I can have installed for months and I will never notice a slowdown. I never to have to reinstall because my system seems to be getting bogged down for some unknown reason. It always performs just like it did the day I installed it even 6 months later.

Sorry to maybe come off quite harsh but you really don't seem to be informed at all and coming in here with an uninformed opinion saying slackware is irrelevant kind of pisses me off.

Umm ... do you have the slightest clue what you're
talking about?

Your average app will be compiled against glibc, and
not against "the kernel headers" ... the apps on your system
most likely don't give an owls hoot over what kernel was or
is on the system...



Cheers,
Tink

The 2.6 kernel will probably? appear in the next Slackware 11.0? release. When 10.1 came out (and PK was rushed with that one as he was sick) 2.6 was still evolving rapidly and definitely buggy so sticking with 2.4 was the right move. Just be patient

Heh ... what do you mean WAS?

How many successive 2.6.11s did we have in which
time frame? Current is 2.6.11.11, right? ;}


Cheers,
Tink

Mmmm - seems a few people are stuck in the old "paradigm".

2.4 was the stable stream, 2.5 was development.
2.6 is both - very unlikely there will be a 2.7 in the same way as there was a "testing stream" with 2.5.
I take it this was mandated by the man himself.

2.6 is changing a *lot* - Andrew (Morton) seems to think he and the product maintainers can keep it under control.
Seems to have worked so far.
From my perspective, it is spectacularly *not* buggy.

(general comments above, not Slack specific).

I am not a Slackware user, but being a Debian user, I can understand the philosophy of not wanting to stay bleeding edge all the time. It's a bit risky to run mission-critical systems on the latest software.

This is especially true of the kernel. If 2.4 works for you, then I don't see any point in using a 2.6 kernel (unless there is some hardware-specific support in 2.6 which is not there in 2.4). Also once you install a fully working kernel, I don't see a situation where you have to upgrade it for a long, long time.

The kernel is the base of the system. I don't see many people (except Linux enthusiasts) upgrading their kernel every few weeks.

Well I do upgrade my kernel often on my laptop but even then it's not that often. I would only see upgrading a kernel if it benefits you somehow. I stopped upgrading my kernel after 2.6.10-cko3 on my desktop mostly because everything broke in 2.6.11. ACPI wasn't working right, it would just hang at shutdown and restart. There were lots of other little intricacies that made the upgrade a pain in the ass. So I just leave my kernel at 2.6.10-cko3 on my desktop because even though they might not get mentioned largely there are things that are changed in each version and they can brake a lot of other things. It's like throwing a rock in a pool of water. I got tired of screwing with newer kernels when I knew 2.6.10-cko3 worked for my system and there was really no reason to use 2.6.11. There was no gaining feature for me, no new support that I needed. On my laptop I'm running 2.6.11.10 just because I got it working on it without hassle so It wasn't a problem to use. With my servers they still had 2.4.29 on them so I wanted to get a 2.6 kernel on them and didn't mind screwing with them since I don't need a lot of hardware support other then the mobo, nics, and HDD controllers which work just fine. When 2.6.12 comes out I might try it out on my desktop just because it's been a long f'in time since they've released a new 2.6.x kernel. I don't mind the whole 2.6.x.y version stuff but it's breaking patches every other release and is getting damned annoying.

And as I recall, Pat did not move from 2.2 to 2.4 as the default kernel until 2.5 was opened. This showed the kernel maintainers thought 2.4 was sufficiently stable to start a testing branch. Let's face it, the 2.6 situation is a confusing mess right now. There was even some talk for awhile that the kernel people would basically always be testing and they would leave it up to the distro people to package a stable kernel. This goes against what Pat has always tried to package, a vanilla non-distro specific kernel. But one thing is clear, the 2.6 kernel is still in a state of flux and is not likely to be ready for prime time any time soon. Obviously Pat isn't as conservative as Debian stable but he does want his product to be safe to run a server on.

Mmmmh ... seems a few people caught the "johny come lately"
virus ...


Well - good for you is all I can say :)

From my personal experience I can only say the I
still strongly dislike 2.6, find it unpredictable at
times, and can't see the supposed performance
benefits. I'm forced to using it on one machine
because of the missing back-port of the cx2388
drivers in v4l, and the bloody thing is an adventure
with every upgrade it attempt.

2.4 hasn't given me trouble since what, 2.4.7? If it
wasn't for security fixes I'd still be on that old dog.

And I only moved away from 2.2 after 2.2.19 because
that's when I found 2.4 to be robust enough...



Cheers,
Tink

I don't generally find the 2.6 series to be too problematic. Upgrades from one revision to the next *are* flaky though. Things get changed so damn often that compiling a new kernel turns more into a crap shoot. One kernel version my harddrive is /dev/hda wham before I know it it's changed it to /dev/sda and I can't get a booting system. Took me 3 days to figure out why my laptop wouldn't boot and I only figured out cause I used SLAX which uses 2.6.11 kernel and automounts partitions in /mnt. Thats why I think I'm gonna just leave my desktop until there's some feature I can't live without that they add to it. Maybe Reiser4 support in default kernel might do it. Probably not though, 2.6.10 with cko3 patch already adds it.

I will say that 2.6.11 in whole has been disappointing.

Now that's a tad harsh.
I was merely trying to point out that waiting for 2.6 to go "stable" ala 2.4/2.5 is a forlorn hope.
If and when 2.7 comes out it likely won't be the testing arm of 2.6, it'll be a new level.

Unless of course Linus decides to change things again.
I tend to agree; that has been a bit of a problem child.

Does anyone have any idea how things are likely to look if Pat V can't administer Slackware anymore?

I'm not sure exactly what would happen. Pat surely seems to be almost unreplaceable. If it was taken by the community everything would be crapped up by a bunch of idiots who want to turn it into every other distro. There could possibly be a successor but even still, Pat's a hard person to replace. His ideals that he puts into slackware are unique and his unwavering ability to do what no one else wants to but needs to be done is hard to find.

Such as stopping releasing gnome. Thats something not a lot of people would have done. But Pat recognized that there are other projects to tackle that beast and that his time would be better spent elsewhere. Keeping slackware so true to its roots after all these years is something a lot of people would be hard pressed to do. I think replacing Pat would have to be a gradual process. Where he possibly finds a successor and "phases" them into position slowly. Starting with them helping out on package releases and such.

I just don't know though. Pat is very much a one in a million person.


EDIT:

Oh and btw I believe Linus said that there was not going to be a 2.7 branch split in the near future. Thats the reason for the whole 2.6.x.y numbering now. They added another minor number set to signify testing branch. The developement model changed to suit the changing needs of the kernel. People weren't wanting to feature freeze the 2.6 series just yet because it still has so much potential. I don't know what they'll do after the 2.6 series. Maybe we'll have a brief 2.7 release and hit 2.8. I don't know. It's up to Linus. For right now though things seem to be going to stay this way for a little while.

Outdated?

Only if I want it to be. I kind of see each release as a stable snapshot at that time and place. But it don't take long for me to start adding and updating to suit. Sometimes from current, sometimes from linuxpackages and half the time I compile from source.

I'm running slack 10.1 with 2.6.11.9 patched for cx88/with kde-3.4. Lots of things are non standard add-ons so how am I supposed to know whether or not I'm really up to date?

Maybe its the eye candy that counts or maybe it has to be just like one of those rpm based be all end all distros that is so new the patches are holding it together and if there ain't a shiney new wizard in charge, then it aint configurable.

I think I'll stick to a (old fashioned?) stable distro for a base and configuration with an editor. If I need something warm and red there's cvs, gcc and mcedit.....well vi is a bit long in the tooth........ I can put anything I want in there, I'll be able to see how it is behaving, I''ll be able to take it out if it misbehaves and I can do all that without having to do a complete reinstall.

Actually that sounds kind of advanced.

I gave up being up to date at about mandrake-8 when I found out there was a standard linux file system, a sensible (bsd style) init system and a compiler in every box.

Then it all began to make sense and I could be as up to date as I wanted to be.

Amen brother.

I see what chris318 is getting at, some of the significant apps and services like Apache, PHP and MySQL are staying one major revision behind the curve and that spells a *very* conservative distribution. But in each case upgrading to the current version has compatibility and configuration issues that not everyone who depends on them will want to deal with on a distributor's schedule, but on their own schedule. People who really want Apache 2, PHP 5 and MySQL 4.1 (much less MySQL 5!) can upgrade on their own.

"Very Conservative"? I really don't see how. What reason is there to use apache 2 right now? There actually are drawbacks to using apache 2 right now. What feature of the "new" MySQL cant you live without right now? PHP is about the only one that I could see someone really wanting because of its new OOP style. Like you said though, If you need it, you can install it.

Namaseit, your arguments are a prefect example of what I meant by "very conservative".

I guess I've just been spoiled by slackware these last 3 or 4 years. It just seems to me that Apache 1.x is stable, secure, reliable, and fast. If you run a webserver how many people are you going to server daily? 1 thousand? 5 thousand? 50 thousand? Something tells me probably not. I'm not going assume the same for every situation but is having the newest brightest thing really make it "cooler". Like I said, I can see something like PHP which in the newest version changes the whole programming style. Hell I'd like to have it, I might just install 5.0 to play with. But thats slackware.

It's easy for you to add what you want to if its not included. I really don't get this complaining about the packages included with slackware. If you want the newest of packages nanoseconds after their release(and this isn't pointed toward anyone in particular) use gentoo. I'm sure they could use another fanboy or two.

As far as I can tell 99.99% of slackware users are happy with its current state and how Pat is running his distro. Keyword *his*.

Pat V. personally stated when he visited ##slackware on irc.freenode.net that he's shooting to have the 2.6 kernel in Slackware 11.0 (and not in /testing). See http://wigglit.ath.cx/slackware_botl....log.14Jan2005

[13:40] <volkerdi> Anyway, I can see the Great Restructuring happening for Slackware 11, but first 10.1 (the last to support a 2.4.x kernel) needs to release so that NPTL and such can come along.

Anyways, I don't see the exclusion of a 2.6 kernel being an issue. I still run the 2.4 kernel on my home Slack machines and even had a 2.4-kernel'd Slack machine at work pulling Perl-crunching duty. I think the main thing Pat is focusing on is stability. The original poster may not like the fact that 2.6 isn't set up by default to be used by Slackware but there's nothing stopping him from grabbing the 2.6 kernel (whether it be from kernel.org or slackware.com as a slackpack) and compiling what he needs into it. As with most Linux software, the source is out there for the benefit of everyone...to complain about what isn't included in a stock distribution is pointless, IMO.

Hi Namaseit,

Just to be clear, nothing I'm saying should be considered a criticism of your points or Patrick's reasoning behind his distribution choices. Slack rocks my house, etc.

It's legitimate to ask "why upgrade?", but it's also legitimate to ask "why not?" Well, in the case of major revisions to major apps and services, I've previously given one reason "why not" and we seem to be in agreement. But there are other reasons to wonder why not.

You've said that there are drawbacks to Apache2. Well, Apache2 hit minor revision 54 a while ago. If my work depends on a webserver and revision 54 of my webserver still has some showstopping drawbacks, I might need to contemplate that and maybe make some hard decisions about my future with the product. The *entire* product line, since inevitably the day will come when Apache 1.1.3 will receive precisely as much attention and patching as Linux kernel 2.0.x.

MySQL is another example. (Note: I'm a PostgreSQL guy myself, so everything I'm about to say might be utterly wrong. Don't quote me on anything here! :-D) Slackware has MySQL 4.0.x but a stable 4.1.x seems to have been around for a while, while the scary things are being done in 5.x. MySQL has been steadily moving towards more support for SQL features and standards, and I'm sure that 4.0 -> 4.1 brought some improvements with it. So, "why not" upgrade? Or perhaps, "Is the guy I look to to design my distribution implying something about MySQL 4.1 that I should know about?"

In short, both "why" and "why not" are legitimate questions to ask.

Oh I completely understand. I'm not trying to attack anyone, in particular. I can see the point you try to make with the comparison of apache and the kernel but it's more or less wrong. The 2.0 kernel has much less hardware support and features then the 2.6 kernel. While apache 1.3 will still serve webpages just fine and still does for most anyones needs.

Sorry if I seemed hostile in anyway. But the original poster was making insane claims and trying to spread false truths. I don't mind a debate but the other person better at least have a clue about what they are talking about if they want to make any kind of point that they want to be taken seriously.

Forgetting about all the software version numbers, slacks future seems not well defined. What well pat do in a year are two when 64 is pretty much standard. After slamd64 spent all this time debugging and what not, will he just take those improvements and tell the slamd64 people sorry better luck next time. That wouldn't seem fair after they put so much work into the project. The server market may not upgrade as fast so that is fine for them. But desktop is a totally different story and this is my point. Slack's future as a desktop OS does't look good to me.

Hi Namaseit,

I admit that by comparing Apache 1.x to kernel 2.0, I was exaggerating for effect. But otherwise why is my example "more or less wrong"? The box that's my home server today is from 1999 and using all the same hardware that kernel 2.0 supported just fine, the only reason I'm running 2.4-2.6 is that it would be more trouble to downgrade! Meanwhile, Web technology keeps growing, along with users' expectations. If I were a mighty Web-geek, I wouldn't mind at all running Apache 2.0.54 on a kernel 2.0.whatever (assuming that's possible) on that box, and only upgrade the kernel for security issues. So what we care about upgrading depends on what we're upgrading and what we're doing with it.

Again you're making assumptions that Slackware is going to take over the server market and the world and that Pat is going to be all of a sudden releasing for x86 and x86_64. Why would he take the work of slamd64? why? why? why? It makes no sense. They are the maintainers of that port so why would Pat take it away from them? They can do whatever they want. Pat has no say in what they do. You are looking at things in a completely wrong context. And this assumption that everything will be running 64-bit is a wrong assumption as well. You nor I have any proof either way but as it is now I don't see that happening in the grand scale that you describe. Besides, what servers besides those that geeks setup will Slackware find its way onto. Company's want big support and so they go to Novell, RedHat, etc. Unless they actually trust their staff. But that is a small amount company's.

to entrumbo:
You are more or less wrong because there is HUGE difference between 2.0 and 2.4 and 2.6. Large memory support, SMP, Filesystem support, Pre-Emption, etc. Apache 1.x still servers webpages. It can still do the "flashy" sites. I personally like simple and elegant websites like the slackware mainpage. But thats just me. I like websites that give me all the info I need easily.

I can't help myself!

I looked at this thread again this morning and I thought that chris 318 might have a future with the ministry of misinformation or maybe the ministry of fud.

IMHO slackware is a pretty impressive outcome for a one man operation. It has outlasted every other commercial linux distro for the past (what is it?) 13 years. It will probably fold up because Pat decides to fold it up and not because it is out of date and no one is interested in it anymore.

I can't help myself:

"seems like the distro remains pretty much stagnate"

Have you bothered looking at the changlog? IMHO it don't look very stagnant, change occurs in a methodical well considered way with who knows how many man hours of compiling/recompiling/testing/retesting not to mention correspondence with developers of the said packages. Consider the glibc entry for 13 May. Glibc is right and it ain't going to fall over or it wouldn't be there.

"when it comes to making improvements in the core system, i.e. package,etc."

What has to be done to the core system? whatever you mean by core system, and do you mean packaging system? Whether any other (newer?) packaging system is any better would be a rather subjective point to make.

"Also, I wonder what will happen now that 64bit cpu's are taking over."

Taking over what? I don't see any groundswell of takeup, usage or indeed advertisments for 64 bit systems here in Oz. 64 bit is still in early adopterland. If someone has the time or fat why not roll out a 64 bit OS but how many people need or want one?

"We have a 64-bit windows, mac-osx, freebsd, and some other linux distro have gone 64 as well."
How about "There are a 64-bit.............." and "".........other linux distro offer/provide an alternative 64...."

"But it doesn't seem like Pat has any intention of doing this and will most likely pass it off to some else",

I don't think passes anything off to or onto anyone, but there might be people who offer Pat/Slackware their support and/or assistance.

"slamd64 i think."

You are correct.

"Overall, i think it is becoming more and more of a server only distro, and it's losing ground quickly when it comes to the desktop."

It always has been a server distro, but not just a server distro or maybe the inclusion of productivity and multimedia aps is an oversight. What do you mean "desktop" anyway? I'll venture a quess that you mean workstation and I think you'll find that most folks on this list would be quite satisfied with slack's suite of workstation productivity aps. Just like the backend software it is all solid and stable. If you want more or newer you can upgrade/install yourself.

IMHO, most of the productivity and multmedia applications have a much more rapid development cycle than the backend programs, but who wants to constantly track beta? and who wants to constantly upgrade the libraries to run beta and worry about security risks and instability for the sake of latest and greatest?

There seems to me to be an overabundance of 'help me' type posts (especially from newbies) to some lists I've been on every time a certain rpm based distro with a tendency towards bleeding edge comes out with a new release...and then everyone has to wait for help on a list, or a fix, or a patch or a stable version of the ap to replace the newest and greatest version, etc.

This is not the case with Slackware, never was, never will be.

Hi chris318,

I don't understand your objection to Patrick possibly building on [ not "taking"] slamd64's work. If slamd64 is based on Slackware, didn't they already "take" Patrick's work? That's the way Free/Open software is supposed to work! Some Germans wanted to apply RedHat's package manager to Slackware, the result grew up and became Suse. Someone wanted to compile RedHat with Pentium optimizations, and Mandrake was born. Don't even get me started on how many current distributions started by "taking" Debian.

Anyway, getting back to your point about Slackware on the desktop. There are alternatives, and in fact if I were really looking for the best desktop and the most current hardware support I'd be looking at the latest Suse. Slackware might be an uphill battle for you based on what you want, just as I was always fighting with RedHat a few years ago. Doesn't mean that Slackware is dying, just that you and it may have irreconcilable differences.

I don't see 64-bit becoming a standard anytime soon. While the hardware may be available, 64-bit hardware still costs more than 32-bit hardware. Also, there's the software to contend with. No one seems to want to redo current software into 64-bit so applications can take advantage of this new hardware. Until that happens, 64-bit isn't gonna be economical, at least on the desktop.

What Pat does is what Pat does. He OWNS Slackware (and BTW, from what I understand, he doesn't own Slamd64 and I doubt he'll 'take' its improvements without at least asking first). You'll probably get your questions answered if you email him and voice your opinions directly. Until that happens and when his response is posted here, I'm treating this topic as FUD. I've seen enough of FUD like this on the Slackware newsgroups and sites like Slashdot to know when threads such as this are pointless.

I'm not attempting to attack or demean you or your arguments, but you're making some HUGE assumptions, IMO. If you're not happy with the way Pat's doing things, you can always try a different distro and be happier. I don't think Pat's gonna be changing the way he does things because of posts such as this, as he's been doing what he does for a LONG time.

The incentive to upgrade from a pentium 2 to 3 or even 4 is not that great. But the benefits to upgrade to 64 bit when apps and OS's are moving in that direction is tremendous. Especially on a desktop. Server not that much becuase most stuff is not cpu intensive anyway and stability is the most parmount concern. With windows xp 64 out the apps will follow, a lot of engineering software like matlab and stuff already have 64 bit version. In the area of engineering, video encoding, gaming, etc... 64 bit computing offers a tremendous boost in performance.

So to upgrade from a pentium 2 or 3 to at athlone 64 w/ pci-express, Sata, gigbit lan, etc is well worth it on a desktop. With the christmas season coming up in 2005 and such people will be buying those 64 bit computers... more and more. Prices are really starting to come down for a base 64 bit box and apps and drivers are really starting to get moving in that direction as companies realize that is the future of computing.

Without slack going in that direction are as well, it will be dead on the desktop. And plus the slamd64 project has that already. Server it will do just fine for a long time to come.

With the removal of gnome, it tells me that direction it is headed. Not that I used gnome but a lot of people on the desktop do. With pat giving his permission to the slamd project to use the trademark tells me that is not a direction he what to take with slack. The incredible slow adoption of 2.6,etc for stabililty concerns.. I conclude that it is moving more and more to a server role.

Hi Namaseit,

----------------------------------
You are more or less wrong because there is HUGE difference between 2.0 and 2.4 and 2.6. Large memory support, SMP, Filesystem support, Pre-Emption, etc. Apache 1.x still servers webpages. It can still do the "flashy" sites. I personally like simple and elegant websites like the slackware mainpage. But thats just me. I like websites that give me all the info I need easily.
----------------------------------


What you and I like in websites is utterly beside my point. What you and I need in a kernel is (almost as much) beside the point. Each of us has certain priorities based on what we're doing and what we plan to be doing in the near future. It's presumptuous for us to dismiss someone's concerns about package and port currency, as it was presumptuous for him to question Slackware's future because of it.

As for my server, it likes the SMP improvements since 2.2.x. It couldn't care less about Large Memory support or 20 different filesystems available. But that's just me.

http://www.linuxquestions.org/questi...69#post1587469


Read that one ... you and chbin should get on quite well ;}

And btw, I conclude that you're just a troll who wants
to upset slackers :D



Cheers,
Tink

Why do I stick with Slackware --

1: Stability (has been said before here, I second, third twentieth that)
2: Upgrades, Ihave found this to be by far one of the easiest Distributions to upgrade, how hard is it to FTP the -current and then cd to each fdirectoryand tupe upgradepkg *.tgz, I mean really, how hard is that!!?
3: Time frames - Every distribution tries to push out the Most Current and Bleeding Edge packages included, and then what do you get, unstability. EG Debian 3.0rc4 of which onthe debian site is a whole pastle of things that were considered bugs, and now I see why I am not using that today. (Originally my first Choice of Linux Distributions, but could not get it to work out of the box.)
4: Support, there are so many venues of support for this Linux Distribution. Here, and evenothers that use it. Other forums have also got support, but from reading, this is Patrics "Baby" so to say.
5: since reading of his illness, I have come to find that only a few even acknowledge it. Get better Pat, you have the best here.
6: Ease of use, and the best most stable tools out of all of the Distributions.

Just my personal observation. Slackware 10.1 - current and using the RC May 2, 2005 and so far it has been the one that works best after the Feb 2005 release. the others seem to have 12MB too much on the CD ISO to let it be burned to CD, but found a cool way past that, and can even upgrade without having to burn a new CD.

What other Distribution has that kind of ability?

Not one!!

Ok I'm not going to argue moot points with you anymore. I have no idea where you are getting your information but it is utterly wrong.


to etrumbo:

I understand that those are besides the point. The point is that right now and for what I'm sure is the near future apache 1.x will server its purposes just fine in most situations if not almost all of them. Development on apache 1.3 is still going on. Here is an overview of new features in apache 2.0 link. On that list is not much that will make your site all of a sudden whiz bang full of features just for having 2.0. I'm sure theres alot more then just whats on that list but my point still stands. Just having 2.0 is not going make everyone's ugly sites pretty. From what I see most people use templates they find or whatever comes with whatever software they use like phpnuke or blogging software. They may change it slightly but usually not a hell of a lot. I bring this up because you said "users' expectations" are growing. I'm just unsure what you mean by expectations. As I said 2.0 won't make all sites pretty all of a sudden. Sorry if I am coming off harshly.