Looking to the future? (Score:5, Interesting)
Most (all?) of the other "major" distributions have gone the way of commercial and public acceptance (meaning ease of installation, and ease of use). Slackware, on the other hand, is still very much geared towards the linux user that already knows what they're doing. Do you plan on making Slackware "pretty", like the others, or do you plan on honing it into a development environment for power users? Or perhaps something else entirely?
But Slackware *is* easy! Trae McCombs even said he was very impressed with the ease of installation. And I think we set up the desktop (at least on KDE) with the nicest set of defaults. But I understand what you're saying -- many of the other distributions now provide a fully graphical installation. But, then you end up raising the hardware requirements, so it's a trade off. I do think keeping things primarily text based is the most flexible, then you can do things like installing with a serial console, and maintaining the machine remotely is more straightforward. Of course, I've been accused of being one of those "command line" kind of people...
Slackware, Inc. (Score:5, Interesting)
Now that you are a separate company (spinning merrily off...), what will your distribution channel be? Will it still be handled by Walnut Creek? What about the Slackware-by-subscription option?
The retail distribution channel for Slackware will continue to be managed by the same people -- we're distributed through Ingram Micro and Frank Kasper and Associates (who were recently acquired by LinuxMall) so you should be able to find the retail version of Slackware at CompUSA, MicroCenter, Fry's, and a lot of other big computer stores.
We're definitely continuing with the discount subscription program. Our loyal subscribers have helped fund the project for many years now, and other than downloads that's where most of the copies go. For now the subscription signup remains on the Walnut Creek CDROM site, but we'll probably be moving that over to slackware.com fairly soon.
The name (Score:5, Funny)
How did you come to the name Slackware? DId it hit you during a long nights of smoking from the holy frop with bob? Did stang climb in your window and wisper it in your ear while you were asleep? Was it the Xists?
Yeah, ok, I'll admit that it was SubGenius inspired. In fact, back in the 2.0 through 3.0 days we used to print a dobbshead on each CD. The name didn't come from Stang... I ran into him last summer and he asked me if that Red Hat organization was my company. I still think it's a pretty good name. I've been trying to put an ease-of-use spin on it, but it doesn't quite work. I think I'll just start telling people all the good names were taken to get them off the subject.
upgradeability (Score:5, Interesting)
I've been a Slackware user since 3.4 and absolutly love it. I don't like most package management systems out there and am glad that Slack doesn't use one (well, if you don't count pkgtool). Unfortunately this seems to be a bit of a problem when it comes to upgrading, seeing as you usually have to just reinstall from scratch and hope you have a good backup of your config files. How do you plan on addressing the issue of upgradability in future releases of Slackware, and do you think a better solution can be achieved through the install scripts without having to revert to an rpm-style package management system.
Although pkgtool is mentioned most often, there are also installpkg, removepkg, makepkg, and upgradepkg tools. The current system does install, track, remove, and upgrade packages, but lacks many of the extra features found in other packaging systems. Still, it *is* a real packaging system. We have used cleanly installable and removable packages since the start.
We're looking at ways to improve upgradability. One thing I've been discussing with the core team is changing the package naming format to be packagename-version-build-arch.tgz, similar to other distributions. The 8.3 filename length policy probably doesn't need to continue anymore, and having that extra information right in the filename will help people (or scripts) spot upgrades. I quit restricting package name lengths in the /contrib directory quite a while back and haven't gotten any complaints, so perhaps the time is right for that change.
David Cantrell is working on an experimental packaging system that would have a lot of advantages over our current system (or any other). It would track the user's config files and make it easy to transfer packages and configuration between their own systems. But, we're still evaluating it. It looks like the LSB is going to want everyone to use RPM, so we have to take that into account.
Download/Sales? (Score:5, Interesting)
I've been using slackware for years now, it was my first distribution back in the 2.x era and other then my little stint with debian for about a month I've been running it ever since.
It's been my observation that slackware has been the most "download friendly" distribution, by that I mean it's segmented into disk sets and you only need to download the ones you want to install it. Other distributions seem to obfuscate this process (redhat complains during install if it cannot absolutely find every package, as do many others).
The reason behind this I think is that they want people to buy it, so they obfuscate and make it difficult to download the distribution.
Now with Slackware apparently getting spun off into a seperate company, will there be more pressure to sell more units, and will this "download friendliness" change?
If anything, you're going to see a much stronger commitment to download friendliness. After all, a lot of our users are more technical and prefer to download. Sure, we like to ship paid copies as this helps support the project, but our first priority has always been to make things as good as we can, and then give it away on the net. We will never try to sell more copies at the expense of our user community. We've even been known to give free support to people who download.
Some of you might have noticed that our home site moved from ftp.cdrom.com to ftp.freesoftware.com (also reachable as ftp.slackware.com). This machine just had an $80K overhaul, adding 500 GB of RAID storage and gigabit ethernet, and should make downloading Slackware faster and more reliable. This machine, BTW, is expected to handle more free software downloads than any other machine on the net. Now if I can only get them to switch it to Linux. ;^)
Boxers or briefs? (Score:5, Funny)
by Anonymous Coward
C'mon... lets see if moderation REALLY decides what gets asked of Patrick... or if Roblimo just does it himself...
Boxers. With Grateful Dead skeletons.
The Magic Wand questions (Score:5, Interesting)
by Anonymous Coward
If you could wave a magic wand and change any one technical aspect of Linux, with no negative side effects on the OS or its users, what would you change?
I wish that limit on kernel size would go away. Or, if I really wanted to go overboard on this one, I suppose it would be nice to be able to use hardware drivers designed for Windows or NT, even if it were unreliable. Just to be able to play around with it.
If you had one wave of the wand and could change only one thing about the Linux community (in the traditional and/or the new, more business-oriented sense), what would it be?
Suddenly they would all want to run Slackware. ;^)
No, I think I'd just like to see the community base its decisions on good technical reasons rather than profit motive, and for the people who are marketing and advertising Linux products to represent the products honestly. If they don't, then we all look bad.
How powerful is this magic wand? Maybe the IPO funds raised by all of the Linux companies should appear in a giant pile, and then K&R will decide how to divide them up among all the various Linux projects. I think they'd be fair about it.
Slackware's Direction as a Linux Distribution
I've always viewed Slackware as the all purpose workhorse of the Linux distributions. It's always done things better and faster in the server role. Now as everyone is pushing to get Linux on the desktop, I'd like to know what Slackware's Direction in this area. Will it remain focused on playing the server role, or will the distribution splinter into different job roles, or will it follow the crowd and push for the desktop?
While we're not really trying to be one-size-fits-all, it seems like by not focusing on some niche that we've ended up being the most flexible.
One of the fortune quotes I like is "A successful tool is one that was used to do something undreamed of by its author. -- S. C. Johnson"
Anyway, I think the more you try to second guess the user, the more you put up barriers. So we like to keep things uncomplicated as much as possible. I think this leads to a system that makes a good server (where you aren't even required to install X if you don't want it), or a good desktop workstation if you do a full installation with KDE or Gnome. It all depends on what you install, and we like to keep that open-ended, and give you the options for whatever you might have in mind at install time. We are talking about making a higher priced retail version that would bundle some commercial tools and office applications, since we're asked about that quite a lot. This will probably also take the swiss-army-knife approach, as some of the add-ons will probably be server oriented, and some will be for the desktop.
Installation options -- FTP install (Score:5,
I was wondering if Slackware would include an ftp install method in some future version similar to FreeBSD, NetBSD, Redhat. I realize ftp has some serious drawbacks compared to NFS or CD install but I found it quite handy when I wanted to give FreeBSD and NetBSD a try. If not ftp, what about the possibility of opening up a public NFS server that will export the latest stable version of Slackware since many of us may not have an extra machine to set up NFS on. It could just run off the same machine as the ftp server for Slackware, right? Just a couple of thoughts.
Well, I don't think there's much possibility that ftp.freesoftware.com will run a public NFS server, and slackware.com doesn't really have the bandwidth for it. And sure, NFS is supposed to be secure now, but as recently as last fall there were holes fixed that you could get a root shell through, so I don't think I want to run NFS daemons that are exposed to the net at large. :)
An FTP install would be a good thing, though, and there has been a test version of color.gz with support for an FTP installation floating around among the Slackware developers. I'd like to see it added to the standard installation disk, but the rootdisks as they exist now are running out of extra space. So, to add FTP capability we will have to either have a dedicated FTP rootdisk, or the other option would be to add a second install disk. This would give us the extra room needed to merge together the existing color, text, and umsdos installation disks, and add FTP support as well. I'd rather not have to require 3 installation disks, but it's getting harder to fit all of the tools needed to install Linux onto a single floppy anyway, and at some point the change will have to happen.
Closed Development (Score:5, Interesting)
I have been a loyal user of Slackware for many years. I have always wondered why isn't the development process more open. For example, Debian has a very open process in which volunteers can contribute to the packaging of the distribution. Slackware does not seem to allow for that, that is, you seem to be in complete control of what goes out the door. Do you plan to allow for users to assist in development or do you wish for things to remain the same?
There's a trade-off there. Being pretty much the Slackware czar, I can make sure that there's a high level of quality and consistency. Keeping the source open is one thing, because then people will help find bugs and send in patches, but having people making changes directly is a lot harder to manage. That said, we have added more developers to the team over the past year, but so far everything still goes through me so I can check it over, and I still do most of the development myself. I'm hoping to start distributing tasks more once I'm satisfied that the quality will stay the same or improve.
Porting Slackware (Score:5, Interesting)
With the formation of your company, will this give you the resources to port Slackware to the PPC and Alpha? Are there any plans for this?
I can tell you that everyone working on Slackware is interested in other platforms. Besides a load of x86 machines, we have two Alphas, two G3s, and a couple of Sparcs floating around. I actually got Slackware running on my Alpha, but never got as far as building an installer for it. There is work going on for Alpha and Sparc, but I don't know when any of it will be ready for public consumption. David Cantrell just picked up a new Sparc today, in fact. Chris Lumens has been working with a dual processor Alpha loaned to us by Alpha Processor, and has a basic Slackware system up and running on it. And Logan Johnson just got a shiny new iMac DV the other day, but he told me not to tell you. :)
Corporate Structure? (Score:5, Insightful)
by Spud Zeppelin
Now that "Slackware Linux Inc." is being spun off, are there any plans to honor J. Robert "Bob" Dobbs by designating him Chairman Emeritus? What kind of poison-pill-defenses are going to be included in the corporate bylaws to prevent being taken over by X-ists, or for that matter, anyone from Cupermond or Redtino?
Sure, why not? I doubt he'll spend much time in the office, though. :)
On the serious end of the question, since we're not publically held and don't have outside investors we don't have to be concerned with a hostile takeover. Someone could offer to buy us, but if it's not in the best interests of Slackware's users we'll tell them to go away.
(Non) Participation in the LSB. (Score:5,
I understand that you have chosen not to participate in the LSB. The reasons mentioned were:
a) That you prefer the old "unix" way of doing things.
b) You feel that these ways should be THE standard.
There must be good technical and marketing reasons behind your preferences. Could you please elaborate on both? Thanks.
Did I say that? Well, I suppose I must have, but my main reason for not participating in the LSB was simply a lack of spare time. It wasn't like I was boycotting it or anything. That said, I do like the more traditional ways of doing things in most cases, and I'm reluctant to move things around for no reason. So, it'll have to go through a little bit of Darwinism first -- if it succeeds, and compatibility improves among LSB compliant distributions, then I'll definitely adopt it, at least to the degree needed to get those benefits. I am hoping that the LSB succeeds to that degree at least. Given the current state of divergence, I think Dan Quinlan has his work cut out for him. But, the Linux Filesystem Standard and the File Hierarchy Standard were both extremely beneficial to Linux (where would we be without those?), so I'm sure this will be another step in the right direction.