Recently in Red Hat/Fedora Category

It's been a little while since my last report on how Ubuntu 8.04 LTS has been doing on the $0 Laptop.

In short, all continues to go very, very well. At this point I could see ratcheting down my use of Debian on this machine and pretty much devoting it to Ubuntu all the way.

Why? Everything in Ubuntu works with as little effort as possible.

I have made some strides in getting Debian Lenny working better on the Gateway Solo 1450. I got sound to return by installing the ALSA modules myself. I'm having a problem with the upper GNOME panel looking a bit funky at times, with graphical "ghosting" marring its appearance. It's not a deal-breaker, but it also doesn't happen in any other distro.

And again, Ubuntu just does what it's supposed to do.

I still haven't conquered suspend-resume in any other distro. In Ubuntu, that just worked.

If for some miraculous reason suspend/resume works in CentOS/RHEL 5.2, I'll re-evaluate things, but a test of 5.1 today confirmed that it does not work out of the box. And I tried to install 5.2 on a free partition with the super-small network installer, which hung up early in the process. I bailed out of it and figured I'd forget about the whole thing until the CentOS 5.2 live CD image is released.

CentOS 5.2 — the free version of the recently released Red Hat Enterprise Linux 5.2 — is here.

I saw it on the mirrors last night, but as with most things Linux, a Distrowatch item means that it's really ready.

Here are the release notes from the CentOS team.

There are DVDs, CDs and a 7.7MB netinstall image. No live CD yet, but that will be coming soon enough, I figure.

For the past few CentOS releases, I've been trying the live CD just to see what kind of hardware detection I can get on my various PCs. I'll be anxious to give 5.2 a spin because Red Hat is promising better support for laptops.

Already CentOS/Red Hat 5.0 has been pretty good on my Gateway Solo 1450 laptop. Not so good as to bump Ubuntu or Debian off of it, but good nonetheless.

And Fedora 9 didn't suspend/resume it. So it doesn't look good for CentOS/RHEL 5.2, but I will still give it a try.

One thing that's new about RHEL is that Red Hat has pledged four years of "intensive" support, up from three, followed by what appears to be three years of less-"intensive" support, but support nonetheless.

So you can count on seven years of security patches on any Red Hat Enterprise Linux release, and that means CentOS will do the same.

Previously in Click:

• Red Hat's desktop strategy: Can you figure it out?
• Why yes, you can use apt and Synaptic in Red Hat or CentOS
• CentOS 5.2 almost here
• Red Hat Enterprise Linux 5.2 — a way bigger deal than you might think

Red Hat has a Linux desktop plan. It's just a little difficult to figure out exactly what it is.

I think Red Hat knows this. And it's OK with it.

One day Red Hat bigwigs are saying that they are not interested in aggressively pursuing the Linux desktop market, that Ubuntu has much of it sewn up, and why do it anyway when all the money is in servers and the support Red Hat so richly provides to those who want it?

Good question.

But I see a strategy in there somewhere. Steven J. Vaughn-Nichols, late of Ziff Davis, now writing just about everywhere else, including his own Practical Technology, has met recently with a bunch of Red Hatters. In SJVN's recent post, the Red Hat people still push Fedora, the community distribution that serves as a testing ground for future Red Hat Enterprise Linux releases, but the company is sometimes not-so-quietly working on making its flagship RHEL product a better fit for the desktop — and laptops, too. And Red Hat does see a niche for RHEL apart from the server:

What Red Hat is working on is continuing to make RHEL (Red Hat Enterprise Linux) business desktop friendly. Whitehurst said many business customers want the Linux desktop. They don't want to move their desktops lock, stock, and barrel to RHEL, or any other Linux desktop. What Fortune 500 companies do want though is to start moving up to 25% of their desktops to Linux.

Why? Because they want the benefits of Linux. Besides the usual advantages of improved TCO (total cost of ownership) and improved security, Red Hat's corporate customers want a Linux desktop that can be carried as a virtual machine on a USB key and can be be managed by Red Hat's management tools. Is this for someone who wants a Windows XP Home replacement? No. It's not. It is, however, something that can catch the attention of CIOs who want a Windows XP Pro replacement.

And who can resist SJVN's money quote from Red Hat's Jim Whitehurst?:

"There are companies that sell hundreds of products for millions of dollars and there are companies that sell millions of products for hundreds of dollars. Guess which kind of company Red Hat is?"

It's a riddle, right?

OK, forget about all of that. Just read Red Hat's own press release for RHEL 5.2, which not only talks up all the work they're doing to make suspend/resume work but highlighting the inclusion of desktop applications that aren't a generation too old for office use. I'm talking about OpenOffice 2.3 and Firefox 3, the latter of which just had its final release this week.

Here are a few quotes from the RHEL 5.2 press release:

"We took part in the beta program of Red Hat Enterprise Linux 5.2," said William Cattey, Linux Platform Coodinator, MIT Information Services & Technology. "Re-basing the Red Hat Enterprise Linux desktop to have the latest Firefox, OpenOffice and Adobe Reader is very important to us because it gives our users the same key applications available on other platforms."

"LVM is very satisfied with our experience using Red Hat Enterprise Linux Desktop on the certified Lenovo T61 and X61 laptops," said Werner Schmidt, LVM's CIO. "We have deployed over 2,000 Lenovo laptops running Red Hat Enterprise Linux Desktop and plan to roll out several thousand more over the next several months."

The key in all of this is the corporate/enterprise connection, the idea not of wholly changing desktop platforms but bringing needed diversity to the desktop with Linux where appropriate, and leveraging the whole Red Hat relationship with server customers to solve problems on the desktop while adding incremental revenue and giving those customers even more reasons to stick with — and continue paying for — Red Hat.

And all those management tools, most of which I know nothing about, that Red Hat offers to keep servers in line and up to date — all that stuff can also make desktop management a more orderly procedure than the absolute mess that's going on now with Joe Worker's desktop PC.

Not that Ubuntu isn't also working on corporate, managed solutions for desktop PC management, but when it comes to paying for support, Ubuntu doesn't seem to be offering any deep discounts over what Red Hat is charging. And if a huge enterprise already has a lot of Red Hat on the premises, a little more doesn't hurt, right?

And there's another side to this valuable coin: While Ubuntu is mainly thought of as a desktop system, it's no secret at all that parent company Canonical is making a huge push into servers, with certifications coming for use on hardware from any number of vendors, commitments of long-term support and the same kind of sysadmin-helping tools that help leverage things for Red Hat.

So if Ubuntu is leveraging its desktop success to build a potentially lucrative server business, Red Hat needs to expand its own desktop commitment to keep and grow the already lucrative server market it currently dominates.

Who wins?

Damn near everybody, I figure. More competition means better products, most of which can be had for free. Remember, if you don't want to pay for Red Hat, there's always Fedora, or the RHEL clones put together by CentOS and Scientific Linux. And if you're deploying Ubuntu in an enterprise situation, you can pay Canonical, or leverage the substantial Ubuntu community to solve problems.

And while some of us can't imagine paying thousands of dollars a year for support on a server, that kind of thing starts to make sense in the enterprise when you weigh it with your own labor costs.

It's an equation that has worked in Red Hat's favor for a long time. And a few extra variables in said equation are just part of the game.

Today's rumor, which suggests that Oracle may buy Red Hat, or something along those lines, comes from Matt Asay.

I had no idea that the Debian-derived apt and Synaptic are viable choices for package management in Red Hat Enterprise Linux and the free RHEL-like CentOS. Not that I have anything against RPM and Yum, but it's nice to have choices.

Dag Wieers shows you how on his blog, which I found via Planet CentOS. (Have you noticed that Planet CentOS is a great place to find out stuff?)

It's all courtesy of a project called APT-RPM.

The best way to follow CentOS news is at Planet CentOS, which is just like Planet Debian and Planet Ubuntu, only more succinct.

All three of these blog-aggregator sites, which collect posts from developers, package maintainers and others involved in their respective Linux projects are very much worth reading on a regular basis.

But the reason for this post is that CentOS 5.2 — the free version of Red Hat Enterprise Linux 5.2 assembled by the CentOS team from the source code of RHEL — is just about ready for release, according to Tim Verhoeven:

We are currently in the progress of doing QA testing. All packages have been build. The current plan is to be able to finish all QA test this week so we might be able to release 5.2 next weekend or in the days after it.

While Fedora 9 didn't properly suspend/resume my Gateway Solo 1450 laptop, I'm still holding out hope that RHEL/CentOS 5.2 will, since greater laptop compatibility is one of the selling points of this significant new RHEL release.

I call it significant because it is bringing some new, very-much-up-to-date versions of popular applications to RHEL/CentOS. Until now, I think that desktop users of RHEL/CentOS have had to be content with Firefox 1.5 and OpenOffice 2.0.

Among the big changes: Firefox 3, which hasn't even had its final release yet, and Open Office 2.3.

So while the people at Red Hat may be downplaying any aspirations they have on the desktop, this new release, even though it's 5.2 and not 6, shows that they aren't relying on Fedora 100 percent for desktop users, many of whom are not anxious to do a major upgrade every six months.

Another thing about CentOS: Lately CentOS has been releasing a live CD and a small network installer image in addition to the full set of CDs and DVD.

I plan to grab the live CD as soon as it's available to see how the Gateway likes it.

But what about my VIA C3 Samuel test box? It runs CentOS 3.9 and won't boot anything after that ...

At the risk of repeating myself, Ubuntu 8.04 LTS works great

When it comes to my main computer — a late-2002 Gateway Solo 1450 (1.3 GHz Celeron, 1GB RAM), Ubuntu 8.04 LTS is the best operating system I've ever run.

After pretty much a full year of Debian (first Etch, mostly Lenny), also great but not as great as this new version of Ubuntu, so many things are working so well that I'm reluctant to do anything but keep using this long-term support version of Ubuntu, which will have three years of updates and patches on the desktop.

I keep cranking live CDs of new Linux distributions into the laptop to see if they can do Suspend/Resume, how their desktop environments look and work, and basically whether or not they can do as well.

Fedora 9, Mandriva 2008, PCLinuxOS 2007, OpenSuse 10.3, nothing has been able to handle this particular collection of hardware better than Ubuntu 8.04.

I'm still waiting for CentOS to release its free version of Red Hat Enterprise Linux 5.2, which might offer greater hardware detection on the Gateway than Fedora, or might not.

And I'm open to any distribution that can meld as well with what I call the $0 Laptop.

But for now, I'm reluctant to mess with what, since its release in April, has been a very good thing.

I don't know why I'm compelled to continually report on how well Ubuntu 8.04 LTS is running on the $0 Laptop, but I keep doing it.

From the graphical polish to Suspend/Resume, Alps touchpad control and everything else I've done with it, this is the most impressive Linux distribution I've run thus far.

For use on this laptop -- a Gateway Solo 1450 -- it's better than Debian Lenny, my other go-to OS.

Today I tried a live CD of Fedora 9, since Red Hat Enterprise Linux 5.2 supposedly has beefed up its support of Suspend/Resume on laptops, I figured that maybe, just maybe, that functionality was present in Fedora 9.

It very well might be, but in the Fedora 9 live environment, Suspend/Resume doesn't work on this laptop.

Moving on to Debian, in the Lenny updates I installed today, there was a new kernel among them. I booted into it after the update, and the new 2.6.24 kernel still doesn't support the ESS 1988 Allegro sound chip on the Gateway.

In order to have working sound, I'm still using the original Lenny 2.6.22 kernel, which does support the chip. I do understand that I can manually add the module I need to support sound in the new kernel, but I'm waiting to see if and when Debian decides that it would like a certain number of its users to enjoy sound. Until then, I'll stick with 2.6.22.

In case you were wondering, and I know you were, Ubuntu 8.04 LTS supports sound just fine on the laptop, even with a 2.6.24 kernel. Score another one for Ubuntu. If the binary blob in the kernel for the ESS 1988 Allegro sound chip were the only such blob left in the kernel as configured by Debian, then I'd understand its sudden exclusion from the distribution, but I have a very good feeling that this is not the case.

I will consider adding the sound modules myself, as detailed in one of the relevant bug reports, but I'll more than likely turn to Ubuntu for the simple reason that it just runs better. And this is coming from a person who has championed Debian quite a bit. (As an aside, I love bug reports that give you a fix for the problem that often works but leaves the bug intact to a) annoy some users and b) drive others away.)

I've thought about it quite a bit. If you're running a standard desktop computer, it's easy to make just about any Linux distribution work well, if it will work at all. You're not often worrying about unsupported touchpads, uncontrolled CPU fans, flaky or nonexistant Suspend/Resume, other power-management issues and the like. I can run Debian Etch with carefree abandon on some of my desktop systems, but getting Etch to work well with an Alps Touchpad is just not in the cards ... or maybe it is, since I found some new suggestions for configuring xorg.conf to make the Alps perform better. But since I've made the move to Lenny, I'm probably not going back to Etch on that system, even though the sound-chip issue continues to piss me off.

I stumbled across this on Slashdot, which led me to Red Hat's own release on all the new things in Red Hat Enterprise Linux 5.2 (and eventually in the free CentOS clone of RHEL).

The most shocking: Firefox 3. The Red Hat people must have a lot of faith in Mozilla's latest browser.

When it comes to the up-to-date applications, RHEL purposefully stays behind the curve so as not to break anything, especially on servers. But for desktop users, having to run Firefox 1.5 for-freakin'-ever is a bit of a bummer. Same for OpenOffice; the version I last used (probably in CentOS 4) didn't even have ODF compatibility.

Users of RHEL 5.2 will enjoy the following newish applications:

• Evolution 2.12.3
• Firefox 3
• OpenOffice 2.3.0
• Thunderbird 2.0

This is one of the parts of the release that makes me eager to try RHEL 5.2:

We also significantly improved laptop support, with Suspend/Hibernate/Resume enhancements that allow us to certify more laptop systems.

Also, many graphics drivers where updated, including a backport of the "intel" graphics driver commonly used in Desktop and Laptops.

Bottom line: These improvements make RHEL/CentOS much more attractive on the desktop (and especially for laptop users).

Could this mean a greater push from Red Hat on the desktop, even though the company has stated recently that it will not focus on that very market?

I say yes.

Red Hat 5.0 (OK, in my case the free CentOS 5.0) runs pretty damn well on my Gateway Solo 1450 (the $0 Laptop), except that Suspend/Resume doesn't work ... and if it did, I would be very happy about it.

The Red Hat release didn't mention the fact that RHEL didn't suffer from the same OpenSSH vulnerability that has affected Debian-derived Linux distros, but the CentOS team does point it out while also telling CentOS users to check suspect keys from users of Debian-based systems that have had SSH contact with your RHEL/CentOS box.

Note: I originally wrote this post on 2/15/08. Today is 4/24/08. Since that time, I've looked into updating in the BSDs a bit further. In FreeBSD, it's certainly possible to update both ports and packages.

Anyway, here is the "original" 2/15/08 entry:

While it's pretty easy to install software from precompiled packages or from ports in OpenBSD, FreeBSD and NetBSD, I've hit a bit of a wall when it comes to keeping any of these systems up to date with periodic security and bug patches.

I don't know if such updates are either not as necessary in the BSDs, even though my Linux boxes have a dozen or so of them every week, or that it's just to hard to do for the average BSD user.

I see plenty of Web help on how to upgrade from one version of a BSD to another, but I don't see anything that covers searching for periodically updated packages and updating an installation on, lets say, a weekly basis as security and bug problems arise and are presumably updated in the repositories of packages and ports.

O, BSD users, correct me if I'm wrong -- and I do hope that I am wrong. But with apt/Aptitude/Synaptic in Debian-based Linux distributions, rpm/Yum in Red Hat- and Suse-style systems, and upgradepkg (and slapt-get/Gslapt) in Slackware (with security announcements going to the mailing list and the www.slackware.com/security page) ... need I go on?

The point is that almost all Linux installations are easily upgraded with precompiled binary packages. Gentoo ... well, I won't go there because I know it has its own BSD-like ports system, but I've never used it and don't know how it works.

Again, the point is that all of these Linux distributions have me conditioned to expect -- and to install -- updates on a regular basis.

But what do I do with BSD? In OpenBSD, for instance, I've never even downloaded the ports tree. Everything I've installed has been a precompiled binary package for the i386 architecture. It's very slick, works perfectly ... but am I exposing myself to undue risk by running Firefox 2.0.0.6 instead of the newer 2.0.0.12? Is all that extra OpenBSD security for nought if I'm running applications rife with security holes?

I'm being completely serious. Is there something I'm missing here? Since OpenBSD, at least, updates the whole system every six months, am I OK to keep the same packages running until the next release? What does this say about BSD vs. Linux when it comes to security and bugs?

But wait. I did run DesktopBSD for awhile, and I remember that system having a GUI package manager that not only fetched new packages but upgraded those already installed.

So that's what Matt Olander was talking about when he said that PC-BSD and DesktopBSD were working together to share technology when it came to package management.

As far as I'm concerned, I don't need to do my updates in a GUI app. I'm perfectly OK with using the console. Just being able to do that updating is enough. That is, unless someone out there can convince me that Linux has conditioned me to think I need something that I really don't.

Those on all sides of this issue, please enlighten me -- and quickly.

Or should I say, "How Canonical wants to be more like Red Hat," because the profit-seeking company behind Ubuntu, which wants to compete not just on the desktop but in the server room as well, has a new product called Landscape, which for $150 per node (unsupported), will allow for the full administration of any number of remote Ubuntu-equipped boxes:

Landscape provides users with a hosted web interface on which all machines are registered. From this single interface, packages and security updates are deployed to the entire network of servers and/or desktops with a single click. Additionally a wealth of monitoring data is provided graphically to the administrator showing process and resource use as well as flagging any available security fixes for the system.

If "$150 per node" means $150 per box, that can add up pretty quick. But such remote management of a plethora of boxes is something that a lot of people might want, I figure.

By the way, how much does Canonical charge for support? A lot. It makes Red Hat look like a bargain.

I've heard of quite a few people still running Debian Sarge -- the stable version of Debian before Etch went stable in April 2007. As per Debian policy, support for what is referred to as "old stable," in this case Sarge, is slated to last for a year after the next Debian release is declared "stable" (Etch).

So now we're bumping up on March 31, 2008, and Debian is telling users about the end of updates for Sarge:

One year after the release of Debian GNU/Linux 4.0 alias 'etch' and nearly three years after the release of Debian GNU/Linux 3.1 alias 'sarge' the security support for the old distribution (3.1 alias 'sarge') is coming to an end next month. The Debian project is proud to be able to support its old distribution for such a long time and even for one year after a new version has been released.

The Debian project released Debian GNU/Linux 4.0 alias 'etch' on the 8th of April 2007. Users and Distributors have been given a one-year timeframe to upgrade their old installations to the current stable release. Hence, the security support for the old release of 3.1 is going to end in March 2008 as previously announced.

I've heard incredible stories about people running servers with Sarge and having incredible uptimes stretching into full years and beyond. And I'm as loathe to upgrade something that "just works" as much as the next lazy guy, so I understand. Three years seems like a long time ... and if you want more than three years, there's always Red Hat/CentOS/Scientific Linux and Novell's Suse (really just Red Hat Enterprise Linux clones CentOS and Scientific Linux, because what kind of cheap person like myself is going to pay year after year for updates?).

But going three years without needing to do a reinstall is a pretty great thing. And if you start with a Debian release before it goes stable -- like Debian Lenny, which is still in Testing but appears pretty darn reliable to me -- you'll probably get more than three years. At this point, I imagine that most Debian users think of Etch -- the current Stable -- as too old. That's true for desktop users, but if your hardware likes Etch, I really see no reason to move to Lenny unless you want newer versions of all of the packages.

For me, Lenny is working pretty well on the $0 Laptop (Gateway Solo 1450), and Etch is doing great on the $15 Laptop (Compaq Armada 7770dmt). And this desktop/server I just set up? I used Etch just because I know it works. And I know that getting Lenny to perform well on the Gateway means I'll be able to stick with it for what could be four years (but actually might be less because the wait between Etch and Lenny becoming stable is probably going to be much shorter than the wait between Sarge and Etch ... or at least that's what I think is going to happen).

Yeah, I probably won't be running Lenny three years from now ... but you never know. As I said recently, Lenny is looking very, very good.

I haven't hooked up my OpenBSD 4.2 drive and booted it for about a week. The last time I left the box, I was playing around with Apache, and I thought all was well.

Today I hook up the drive and boot OpenBSD.

First of all, instead of a console login, I get an XDM login. That's strange. I don't remember XDM ever showing up before.

Then Internet networking doesn't work. I check all the networking settings. Everything is correct.

I can ping IP addresses on the local network, but nothing is working outside of that. Pinging google.com yields nothing. Since I can get local machines, I know it's not a bad cable.

Back to the OpenBSD FAQ. Instead of doing ifconfig, I check all the files that hold network configuration info. Nothing.

To start networking manually, the FAQ says to do this:

# sh /etc/netstart

An error message comes up. There's an error of some kind in /etc/rc.conf.

Now I know what happened. To start Apache automatically at boot, a line must be edited in /etc/rc.conf. I was trying it, and I must've screwed something up. As root, I edit the file. Sure enough, I had erroneously dropped a linefeed in the middle of the comment line to turn Apache on at boot.

I fixed the line, saved /etc/rc.conf and tried to start networking again from the command line.

It didn't work.

I rebooted.

This time, I got my usual console login. I started X manually. And Internet networking worked.

I also configured an anonymous FTP server. I had to manually change the permissions of the directory and files to root, but everything worked as advertised.

That's the strength of OpenBSD, as well as FreeBSD and NetBSD: the documentation is readable, comprehensive and up to date.

Over the past two days, I did a Debian Etch install in order to compare how all of this server configuration goes in Linux as opposed to OpenBSD.

And this is where the lack of documentation (even the man pages aren't all that up-to-date). At least the apache2 man page for Debian told me about the apache2 command. When httpd and apachectl start did nothing, I was in a bit of a quandary. Luckily I figured out that apache2 start and apache2ctl start would both work. Oh yeah, and the config files aren't where the Debian man page says they are. Instead of being in /usr/local/apache2/conf, they're in /etc/apache2.

I did figure out how to change the default directory for Apache in Debian (editing /etc/apache2/sites-available/default does it).

Part of the problem was that I started with Apache version 1.3 in OpenBSD (which doesn't include Apache 2 for licensing reasons) and had Apache 2.3 in Debian. And sure I don't know quite what I'm doing, but this is all on a local network, not the wide-open Internet, so I'm a bit more free to experiment.

All this underscores the value of good documentation. And when it comes to some distros -- Ubuntu, Red Hat and Suse -- there are doorstop-thick books available. And the good ones are worth their weight in any precious metal you care to name. Luckily the BSDs have great online FAQs to help get you started. And since integration between the kernel, userland and other packages is so tight in the BSDs, and the need for documentation is that much greater, I'm damn glad it's there.

Not that Linux doesn't need something similar, but I don't see any Linux distribution short of Gentoo providing documentation this comprehensive and finely tuned to its users.

Can anybody prove me wrong? I truly, sincerely hope so.

We all know that Linux is a kernel, an operating system, maybe even a socio-political movement (it depends on whom you ask), but in a sense, Linux is about people -- those who create, use and promote it.

One of those people is Orv Beach, publicity chairman for SCALE 6X -- the Southern California Linux Expo -- being held Feb. 8-10 in Los Angeles. Since I'm covering the convention for Click, I took the opportunity to interview Orv after hearing from him about getting press credentials for the event, which I wouldn't miss, by the way. And if you do plan on attending, Orv told me that using the promo code CAST when registering for SCALE can get you 40 percent off of admission.

Orv, where do you live, how old are you, and what do you do for a living?
I live in Simi Valley, California, with my wife Beth. I'm 58, and I have four grown kids and four wonderful grandkids. Professionally, I'm the IT director at Simi Valley Hospital.

How did you first discover open-source software, and what part does it play in your work and home life today?
I've been interested in technology all my life. I got my amateur radio license when I was 17, and enjoyed building radio equipment as much as operating.

I got my first computer in about 1979, and when amateur packet radio was authorized by the FCC, it was a natural to use a computer with it. A popular packet radio program at the time was TNOS, written by Brian Lantz. It ran under DOS, and was a communications program & BBS. Brian had an active users group and was happy to add features to TNOS. As it grew in size, the C compiler he was using had more and more difficulties compiling it (It was Borland Turbo C, I think). So he moved TNOS over to Linux to use GCC as the compiler, and a large percentage of his users followed him.

I got Linux from a programmer at work. At that time it was 16 floppies, and that minimal version didn't include X Windows. I ran it on a 40 MHz 386 with 8 Megs of RAM. I've been using Linux steadily ever since and moved my desktop computer over to it full time about six years ago, and my wife's about four years ago.

At work, while Adventist Health isn't a full-blown user of open-source software, they're edging that way. The web programmers at our corporate office seem to have fallen in love with Plone. Some of the programming groups are moving to Project.Net for project management, too. Locally, I use Nagios to monitor over a hundred devices on our hospital network, and we use ZoneMinder to monitor some video cameras.

Now that SCALE is in its sixth year, how big was the convention the first time around, and what kind of growth has it seen? How many exhibitors, speakers and attendees do you expect this year?
SCALE is an offshoot of the "LUGFests" that SCLUG (the Simi-Conejo Linux Users Group - http://sclug.org) held every 6 months where they met at the Nortel building in Simi Valley. They were miniconferences, with people demonstrating open source software and even a few commercial vendors. Even as limited as they were, they drew Linux users from all over Southern California. SCLUG held 4 of them before Nortel closed down that building. (There's an article on LUGFest III here).

The last LUGFest, LUGFest IV, drew 400 people over two days. Based on the response to the LUGFests, we knew we were filling a need for information and education on open-source software.

So after a hiatus of a year or so, SCLUG, UCLALUG and USCLUG jointly started SCALE. The first was held in the Davidson Conference Center at USC. It was one day, with two session tracks. We had 11 speakers spots and a panel, and it was a struggle to fill them. That first Linux Expo drew 400 attendees.

Contrast that with SCALE 6X, which will be held in February, five years later: The main Expo is now on Saturday and Sunday, has 32 speaker slots and two keynotes spread over four session tracks per day. You'd think that number of topics and speakers would be impossible to come up with. Yet we received over 105 submissions to our call for papers! Whittling them down was difficult, and it was painful, as we had to turn down lots of good proposals. We expect to have about 1,500 attendees for SCALE 6X. The Westin hotel will be bursting at the seams.

Ubuntu 6.06 LTS -- the distribution's first "long term support" release -- now has a new installer that incorporates some 600 bug and security fixes and makes installation easier, especially on servers.

It's no secret that Canonical, the company that runs Ubuntu, is making a big play both for the desktop and more-lucrative server markets, and a big part of that play is the LTS release. And even though the next Ubuntu release -- 8.04 (due 4/08 ... also known as April 2008) -- is going to be a Long Term Support release, with fixes, patches and the like for three years on the desktop, five years on the server, there's still quite a bit of time left for the current Ubuntu LTS, which will be supported until June 2009 on the desktop and June 2011 on the server.

The new installer -- you don't really need it if you can successfully use the old installer, already have a 6.06 LTS install (like I do) and have done all the updates -- underscores Canonical's commitment to the LTS concept. While the twice-yearly releases of Ubuntu get most of the light and heat in the uber-geek community, there are many who depend on the relative stability of the LTS release to keep their hardware running. That's especially true on servers, where major upgrades every six months are impractical at best and detrimential at worst -- nobody wants to break a system that's been running well.

And the LTS is vital as a counterweight to Red Hat Enterprise Linux and SUSE Linux Enterprise Server/Desktop, both of which are supported for years on end.

I'd like to say that Debian Stable (currently Etch) and Old Stable (Sarge) are equivalents, but since you can't pin down a date certain for length of their support, there is a bit of an unknown factor there, although once the Stable release goes to Old Stable, you pretty much know that the new Stable release won't give you too many problems.

Sure, many desktop users generally want something more cutting-edge, mainly something like the regular Ubuntu releases, but there are many people -- and many situations -- that warrant hanging on to a Linux installation as long as possible. Over the time I've used Ubuntu and Xubuntu (from 6.06 LTS through 6.10, 7.04 and 7.10), I've seen some parts of the installation improve dramatically, I've seen hardware work better, then worse, and occasionally not at all.

And we all know an individual or organization that hates doing major upgrades, ever. Those coming from a Windows or Macintosh background aren't all used to major OS upgrades. In the case of Windows AND Mac's OS X, major upgrades almost always cost money. $129 for an OS X upgrade might not sound like much, but paying that much every couple of years when your computer runs just fine the way it is? No thanks. That's why I'm still running OS X 10.3 on my Mac. And Windows? I have a disc for Windows 2000, and I'm not about to pay ANYTHING for the privilege of upgrading my sole Windows box (which I boot maybe twice a year) to XP.

And in Linux, just because we can change out distros 10 times a day if we wish, it doesn't mean that we have to -- or should. For people who crave the stability of long-term releases, one thing generally drives upgrade: newer software they need to get their work done, and new hardware that needs new software to run properly.

I did this most recent Ubuntu 6.06 LTS installation for testing purposes, but I've stuck with it because it just works. On this test box, it's flawless. On my Gateway Solo 1450 laptop, it manages the fan as well as 7.10 (i.e. not at all without a cron job; but well with said cron job), but less well than 7.04 (which has the ACPI working with no coding needed). (Note: I'm not currently running Ubuntu at all on the Gateway laptop, which is currently dual-booting the Slackware 11-based Wolvix Hunter 1.1.0 and Debian Lenny, which I upgraded from the stable Etch.)

Using Ubuntu 6.06 LTS on this test box, sure I'm stuck with Firefox 1.5, OpenOffice 2.0, GMOME 2.14.3 and Evolution 2.6.1, but everything works. And there's nothing I do that I can't do with applications of this "vintage." If I this machine had wireless and it didn't work with 6.06, I might feel differently about LTS, but with the hardware I have now, LTS is a good fit.

So if you're looking for stable, supported releases, especially ones that won't cost you anything, it's nice to have Ubuntu LTS as a choice along with CentOS and Scientific Linux (both free versions of Red Hat Enterprise Linux), SUSE, Debian and Slackware.

As far as stable, long-term releases go, I have run CentOS (3.9. 4.2 and 5), Debian (Etch and Lenny) and Slackware (12), as well as Ubuntu LTS, and Ubuntu holds up very well on the desktop in this crowd. It's more flexible, as far as adding software, than CentOS and Slackware -- it doesn't have as many packages as Debian, but it does have plenty -- and the desktop and menus are a bit more tame than Debian's, with a better out-of-the-box experience, especially for inexperienced users.

And the support available from other Ubuntu users is a major component of the distro's success. All the advice may not be of the best quality, but there's just so much of it that you're bound to find the right answer to whatever it is you're asking. Not that the Debian community isn't helpful (I love DebianHELP and the Debian User Forums, but they just don't have the sheer volume of the Ubuntu Forums. Like I said, there's a lot more noise among the Ubuntu people ... but that's the price you pay, I guess.

And since Ubuntu is based on Debian, what you learn in one community is more often than not directly applicable in the other.

Another thing I discovered today: I enjoy reading the Planet Debian blog posts from Debian developers, and I had no idea that there's a Planet Ubuntu as well. Both are more than worth adding to your favorites and checking on from time to time.

Over the past year, I've used both Debian and Ubuntu extensively, and I always say that Debian isn't as "hard" to use as some would make it appear. Nor is Ubuntu a relative cakewalk. Both require, at times, a bit of wading into the muck to make things work. As far as installation goes, Debian's installer -- upon which Ubuntu's "alternate" installer is very closely based, is quite good, and has succeeded for me many more times than Ubuntu's live CD and alternate-CD discs, but Ubuntu works often enough.

What Ubuntu has that Debian lacks is a marketing plan. For some -- especially the average Linux user (read: geek) -- having no marketing plan is, in and of itself, a marketing plan of sorts. Nobody's trying to make Debian "cool," or giving you reasons why you should or shouldn't run it. And while there are a few Debian evangelists out there, and a few for Slackware as well, there's nothing approaching the fervor over Ubuntu.

That might be good, or bad, depending on how you look at it.

A lot of people are running Debian and Slackware -- they're just quieter about it, I guess.

Anyhow, this post has gone on for far too long. All I want to say is that I'm in favor of long-term, "stable" releases with defined periods of support and a smooth upgrade path, and I'm glad that Ubuntu has pretty big foot in this very door.

And I like the fact that 6.06 LTS will be supported for over a year after the next LTS -- 8.04 -- is released a few months from now.