December 2008 Archives

ZDNet's new Community, Incorporated blog (written by recently minted OpenSuse community manager Joe "Zonker" Brockmeier) links to a CNet post about Microsoft applying for a patent on something called metered computing.

The idea is that Microsoft will sell you a really cheap, subsidized PC and then charge you continually for its use. And yes, MS points out in the application that a user could very well pay more in fees than the PC is worth.

From CNet:

Microsoft's patent application does acknowledge that a per-use model of computing would probably increase the cost of ownership over the PC's lifetime. The company argues in its application, however, that "the payments can be deferred and the user can extend the useful life of the computer beyond that of the one-time purchase machine."

The document suggests that "both users and suppliers benefit from this new business model" because "the user is able to migrate the performance level of the computer as needs change over time, while the supplier can develop a revenue stream business that may actually have higher value than the one-time purchase model currently practiced."

"Rather than suffering through less-than-adequate performance for a significant portion of the life of a computer, a user can increase performance level over time, at a slight premium of payments," the application reads. "When the performance level finally reaches its maximum and still better performance is required, then the user may upgrade to a new computer, running at a relatively low performance level, probably with little or no change in the cost of use."

And what if you don't pay up? According to the patent application, there would be a mechanism to turn off your computing faucet.

Remember, this is all speculation; it's just a patent application. But if MS is trying to patent this idea, chances are they're seriously considering implementing it.

Gives new meaning to the term "Microsoft tax," does it not?

I've been thinking about wired vs. wireless networking over the past few days as I plan a new home network in my mind.

And I remembered a device that just might solve a problem you're having.

What if you have a laptop or desktop computer that, for whatever reason, either can't or won't play nice with wireless? Either you can't get a wireless card (PCI or PCMCIA/CardBus) to work with your box, or your OS (even Windows balks at some cards) won't recognize and configure it.

So what do you do? Here are a couple of devices I've never seen offered anywhere else. They're called Wireless Ethernet Adapters, and a company called MacWireless sells them.

The way they work is that you plug an Etnernet cable into the adapter and your PC — and I imagine that this device works with Macs, as well as PCs under Windows, Linux, or any BSD. Anything that uses wired Ethernet can seemingly go wireless with this device. Even an old Sparcstation ...

There are two models available, the MacWireless 11g Ethernet Adapter with 32 milliwatts of power for $99.98 (above right) and the High Power 11g Ethernet Adapter with 400 milliwatts of power for $189.98 (left).

You manage the device with a Web browser (see the PDF instructions for the 11g adapter and the high-power 11g adapter.)

Both of these boxes are expensive. If you can use a $20 Wi-Fi card with your laptop or desktop, that's probably the way to go. But if you have an older computer that just doesn't want to work with Wi-Fi, this is a very legitimate way to bring wireless networking to your Ethernet-equipped computer.

I've known about MacWireless for quite some time. I discovered the company way back in my early This Old Mac days when I was trying to make a Macintosh Powerbook 1400 work in the modern world (under System 7 no less). There are a few helpful Web sites out there on how to modify, expand and generally use older Apple hardware. This one from Penmachine.com led me to the Orinoco WaveLAN PCMCIA card that has been so very helpful to me with just about every laptop I've owned then and since (Linux and all the BSDs LOVE this card). It also pointed me to MacWireless, which is where I discovered these interesting Wi-Fi-to-Ethernet devices.

As far as technological solutions go, this is an idea that you'd think companies like Netgear and D-Link would've picked up on. But thus far, these units from MacWireless are the only things I know of that do what they do.

MacWireless has quite a few other interesting products, including full setups for Wi-Fi routers that can live outdoors and get what's called Power Over Ethernet, meaning they don't need 120-volt electricity to work. They also offer many Wi-Fi adapters for Macintoshes from the G3 and G4 era. This one looks like a good fit for my G4. I wonder if it works with Linux and OpenBSD ...

OpenNTPD, which is OpenBSD's version of the Network Time Protocol isn't just for OpenBSD. There are versions of the software for Linux, FreeBSD, NetBSD, Solaris, Mac OS X, HP-UX, QNX4, IRIX and AIX. (I even recognize a few of those.)

I can't tell you how it works, except that during the OpenBSD install you are asked whether or not you want it. You do.

My clock is now accurate to the second. (I've checked with http://time.gov).

I'd love to tell you exactly how it works. But I really haven't got a clue.

That's the hallmark of a great technology: It just works, and if you want to know why, it's all there for you to soak up.

I had no expectation that it would work, but I decided to shove my trusty Orinoco WaveLAN Silver PCMCIA 802.11b wireless networking card into the meant-for-Airport-only slot in my Power Macintosh G4/466 running Debian Etch.

I had never heard that this sort of thing would work.

I shoved the card it. It's quite a bit longer than the pricey Airport card, which I've seen go for near $100 on eBay.

The Mac's antenna plug matched. I connected the antenna wire.

I booted Debian. I opened the Desktop -- Administration -- Networking tool.

There it was, eth2, my wireless card.

I configured it for DHCP. It found a network. I now had wireless networking on a Power Macintosh G4 under Debian without having to buy a thing.

One problem: Since the Orinoco WaveLAN Silver card is quite a bit longer than the Airport card this slot was meant for, there's no way I can even close the case of the G4 while using the Wi-Fi card.

That's a bit of a dilemma, no?

Maybe a PCI card will work better? I wonder what might work ... and if I'll have to upgrade to Lenny to increase my chances of this actually working.

But wireless in Linux on a G4. Amazing.

I haven't tried ZenWalk in a very long time, but I'm thinking about it.

When I first started using Linux, ZenWalk was one of the first systems I played around with. I had a nice install at one point, and that particular machine would install the old version of ZenWalk at the time but not whatever the new version happened to be. As a last-ditch effort/experiment, I tried to upgrade the old system, but since ZenWalk pretty much stopped supporting my old system but kept everything in the same repository, the upgrade pretty much bricked the install.

Still, ZenWalk is a super-fast system with excellent hardware detection and less geeky pain than in Slackware, upon which Zenwalk is based.

Then there was/is the controversy about whether or not ZenWalk was complying with the GPL when it didn't make source code available. (I can't find a good item to link to, but the issue was discussed hotly and at extreme length in the LXer forums.)

But with every new ZenWalk release, including the current beta of version 5.4, I'm tempted to give ZenWalk another try.

I have old hardware, love fast systems, love the default Xfce desktop environment (which is right there in Slackware but somewhat of a red-headed stepchild to KDE), love the look of ZenWalk and appreciate its very extensive repository.

When it comes to Slackware-for-the-rest-of-us distros, I've been more than partial to Wolvix, even though I worry about it's relative speed (I get the feeling it's slower than ZenWalk, Vector and plain ol' Slackware, though I have not much to back this feeling up), I worry about whether or not I've updated the kernel properly (it requires some hackery in slapt-get and/or Gslapt that I'm a bit unsure about), and I worry that Wolvix in general isn't as up to date as it could/should be.

But Wolvix — most of it, at leat — continues to be updatable via slapt-get/Gslapt without the whole thing going to hell.

In contrast, if I were to install ZenWalk today, I'd be sure to create a separate partition for /home (which I always do anyway these days) so I could wipe and reinstall between releases. Or I could do the not-unthinkable and dual- or triple-boot a bunch of Slackware-derived distros on a single box and see how I feel about it in a month or so.

I've got a laptop that needs a FOSS OS, and while I've been thinking Debian or Ubuntu because that's my "default" choice, I may give ZenWalk a try just to see how it runs. All of these new features do look like things I'd enjoy having:

Kernel 2.6.27.10 with gspca (supports many USB webcams)

XFCE 4.6 (beta2, already very stable)

Faster boot (tunned init scripts, with realtime I/O scheduler)

PAM authentication has been added to the system

Wicd is becoming the main network configuration tool

Improved suspend/hibernate, with XFCE Power Manager

new Netpkg with orphan dependencies and "offline operation" support

New Zenpanel with integrated Disk Manager, Wifi and Wired Network Manager

Gksu keyring based desktop granting system

New artwork

Many new applications

And it looks (from this 5.2 announcement) that the license-violation issue is at least beginning to be taken care of:

Source repository: many faithful users asked us to provide an online source repository rather than sending source Dvds on-demand. So we have been working on a mechanism to allow the development team to instantly publish source tarballs for any new package we release. This source repository is now 100% ready for ISO packages, and the contributed packages (aka "extra") source repository is being populated actively.

Zenwalk's Community Spirit: As the community thrives, Zenwalk now has a web-accessible Package Database and a conveniently arranged User Repository. Please also have a look at the Zenwalk Companion - a guide to the extra packages available to Zenwalk. Please see the zenwalk.org website for more information, and welcome to support.zenwalk.org for bug reports and friendly discussion between Linux purists :)

Of course while I love systems that are updated forever (as in Wolvix), I'd love even more to see a new version of Wolvix to appear — and if that happened, I'd be a very happy camper, indeed.

It's been a long time since I ran ZenWalk, and I can't say how its application mix would meet my needs, but the thing I like so much about Wolvix is that it has virtually everything I want or need as far as applications go, and its installer and control panel also match my needs better than just about everything else out there, too.

Not that I've run Wolvix in the past six or more months, due to a combination of the issues I raised above, then the time I spent running Debian and Ubuntu, and now having my "main" laptop run OpenBSD, which I've been quite happy with by the way.

But some time back in the Linux world, and not necessarily in Ubuntu or Debian, is starting to sound pretty good.

http://lowendmac.com/mail/0807mb/0710.html

http://www.mklinux.org/

http://www.google.com/search?q=MiBoot+ISO+image&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a

http://nubus-pmac.sourceforge.net/

I'm not one to be enthusiastic about Web-site redesigns. I've gotten used to what happened at ZDNet, and now I even like it. Not so much with O'Reilly, which used to have everything organized into categorized, named blogs but which now just dumps it all in one bucket and relies on "popular topics" tags and a topics tag cloud to parse it for the reader.

Still, O'Reilly continues to publish the best tech books out there — including a new Drupal title I'm dying to get my hands on. I wish they'd update their OS books more often, but I guess there's a lot more of a market for books on how to write code (for the Web or elsewhere) than there is for how to set up and run operating systems. I understand that.

What O'Reilly does have is a fairly steady flow of good, provocative blog posts that really make you think (and often teach you something, too). I took a look today and want to read all of the following:

• How I Ended Up in the Cloud
• Practice, Play and Computers
• The $5 Self-Publishing Workshop
• But What Exactly "Is" Cloud Computing?
• The Beatles of Programming Languages
• Craig Newmark Interview: A Brief History of Craigslist
• Everything You Wanted to Know About Drupal but Were Afraid to Ask
• The New Newspaper Editor: Your Neighbors and Some Python Code
• Why Are Newspapers Dying?
• Is Apple OS X More Secure than Windows?
• Hard Work and Practice in Programming

Clearly that's a whole lot of intriguing entries, and it means that I need to get to the O'Reilly Web site a lot more often. Maybe if I use my Jedi skills, they'll do a revised edition of Carla Schroder's "Linux Cookbook." .. If I knew what was good for me, I, too would be focusing on books about coding and not OS implementation ... so O'Reilly is probably more of an oracle (not to be confused with Oracle) than I'd care to admit at any given moment.

What role does the Internet Explorer Web browser play in your life? In recent days, new vulnerabilities in the flagship Windows browser have come to light.

Alas, the fix is in, but pundits continue to suggest that running IE is just asking for trouble.

I'm not ready to say IE is such a security risk that instead browsing the Web with Firefox, Google's new Chrome, the super-quick Opera or even Apple's cross-platform Safari is enough to save your digital bacon.

Nope, it's all about what you do, where you go and what computing platform you choose to do it with.

The fast is that i386-based Windows PCs continue to be the most vulnerable platforms out there because of both their ubiquity and relative lack of built-in security when compared to Macintosh OS X and the vast number of Unix-like OSes out there (including Linux, the BSDs and Sun's offerings).

If you make a habit of downloading executable files (they're easy to spot in Windows because they end in .exe) without being absolutely sure they're totally legitimate and then double-clicking on them, bad things may very well happen.

Don't get me wrong. Searching for free software for Windows computers is something I do, too. Not often, but I do it. That's how I found some of my very favorite applications on any platform, including the terrific image viewer/editor IrfanView, the fast AbiWord word processor and Notepad++, the best Windows-native text editor ever.

I was pleased as the proverbial punch to receive a link in the latest Distrowatch Weekly, and I didn't even have to stoop to my usual begging. (Thanks Caitlyn!)

What I also saw in the column was the announcement of a new distribution, CentServer, which is based on CentOS. For those who might not know, CentOS is itself a free clone of Red Hat Enterprise Linux.

I've run CentOS as a desktop system, and lately especially (since 5.2) been extremely pleased.

But CentServer, and those behind it, care not. They want a SERVER OS, with none of the trimmings. No X (because real admins don't use X ... or is it because of the hit on performance?).

Here are CentServer's features:

• Optmized for server
• No X window
• Minimal server components included
• Automate installation support (without any interaction)
• 1 CD (384MB)
• Setup from scratch in 5 minutes

I like. A single CD, automatic install (although real server geeks like the fine-grained control you get in the Slackware installer, where you can choose what will install package by package). Five-minute setup — great for getting a box up quickly.

Here's another distro I'll be watching. I don't exactly know what makes it more secure than plain-vanilla CentOS, and I'd certainly like to find out.

CentServer's big problem: This might not be a problem for you (the global you) but it's a problem for me: CentServer is 64-bit only. It runs on both AMD and Intel 64-bit CPUs, but I don't have a single 64-bit CPU in my herd. ... OK I might have a old AMD laptop in the junk pile, but I'm not about to turn it into a server ...

I was impressed with this easy explanation in MDLog:/sysadmin of how to upgrade a box from Debian Etch (currently Stable) to Lenny (hopefully soon to be Stable).

One of the great things about Debian (which does trickle down to those distros based on it, including Ubuntu) is the ability to continually upgrade without needing to reinstall the whole OS.

I've done this about a half-dozen times, and it has worked every time.

I haven't booted the Power Mac G4/466 running Debian Etch in a while, but I did so today because I'm about to move the box and its massive LaCie electron22blue monitor. So I wanted to power it up, do a software update and get it on the cart.

This is a nice box on which to run Debian. I've complained at length at how poorly Fedora 9 installed and autoconfigured on this box and how startlingly better Debian Etch did with that same task. Sorry to repeat that, but it bears repeating.

Since I've set up this box, I've discovered both an original set of PowerPC G4 Macintosh install CDs, which I suspect are OS 9, with a slew of equally original discs for Classic Mac applications, everything from Adobe Pagemaker and Illustrator to MS Office.

I'm not about to install Mac OS (but at least I'd get Flash support from Adobe, which sees some kind of screwed-up wisdom in supporting the all-but-dead PowerPC OS 9 but not PowerPC Linux, which isn't exactly a front-burner OS but at least is currently supported and would get more use if Flash and a modern version of Java ran on it).

I suppose I'd consider throwing OS 9 or OS X on this box, but with Debian running great, I just don't see it happening.

I had trouble when I tried to install OpenBSD on this PowerPC box, but now I think I have a handle on how to get it to boot:

If I'm not wrong, I can make the disk bootable with:

# fdisk -u wd0

On a not-totally-unrelated, our photo-department systems guru Roger Vargo keeps Macintoshes of many vintages and OSes running as well as I've ever seen them, and he's got a handful of Power Mac G4s running OS 9.

I was surprised recently to see a G4 running OS X 10.4 and doing it very quickly. The last G4 I saw running OS X before this was a total disaster, with any action on the user's part taking many seconds to even begin taking effect.

But this G4 was as fast as you'd want it to be.

It did have dual CPUs — maybe 400 MHz each — and at least 1 GB of RAM. Yep, you can stuff those G4's with up to 1.5 GB, I believe. It screams fairly well in OS X. Could you imagine getting near 1 GHz of CPU and 1 GB of RAM on the PowerPC platform in Debian, OpenBSD (and at that level, maybe even in Fedora)?

And they tend to have DVD-ROM/CD-R drives, plus gigabit Ethernet built it. Apple had gigabit Ethernet in the late '90s? Yep, it seems they did.

And the ATI video card built into my G4 does a great job with this huge, hefty LaCie monitor. The generic onboard video circuitry in my el-cheapo Maxspeed Maxterm thin client delivers a much fainter image on the same CRT monitor (and didn't do well at all when I hooked it up to an LCD monitor an age ago). But this G4 delivers superb graphics in Debian.

In other words, if you have a G4 or G5 at your disposal (and Flash isn't important to you or what you do), you might want to go off the reservation and try GNU/Linux or one of the BSD projects on it. (NetBSD, OpenBSD and FreeBSD all maintain PowerPC ports).

As it is, I can see this G4 being my main home box in our office, should we ever get all the accumulated junk removed enough to return the space to genuine office use. I kind of, sort of need Flash, but it's not a total deal-breaker.

I can only hope that upgrading the G4 from Etch to Lenny keeps all of the Debian goodness I've been enjoying so much. And there's always that next install of OpenBSD.

Endnote: Since we're not allowed to keep boxes (computer or otherwise) on the floor at the Daily News' new digs, I've had a desk packed with boxes (computer and otherwise) ever since we moved here. I hate to take the G4 down, but right now the G4 case has served me better as a Post-It bulletin board than as a working computer, and I hope to somehow rectify that with this change of CPU scenery.

And this: I'd love to try Slackintosh, the Slackware port to PowerPC, on this box.

So I'm working on a not-so-complicated (but not plain text) document that began its life some time ago in Microsoft Word, which means it's a .doc file that got uploaded to Google Docs.

It sort of, kind of looked OK in Google Docs, except that in a few places I couldn't get the fonts and the margins right.

And outputting the Google Docs document back into .doc or .odt (OpenDocument) was a real mess, with a mix of Web styles, Word styles, strange margins, etc.

After getting nowhere fast in Google Docs, I finally tried to remove all formatting and start over. But I couldn't even get the line spacing right.

I'm sure a little CSS hackery could have made things right, but I'm not in any mode to do that.

So I exported the document in .odt format and worked on it in OpenOffice Writer. Now I can save it as an HTML, MS Word or RTF document, or better yet export it as a PDF.

I love having Google Docs enable me to work on things anywhere, at any time, but I've found that the cloud-based app works best when documents originate in Google Docs and stay there. Converting them to .odt, RTF and .doc format causes the formatting to break down.

I've blogged in the past about how poorly Google Docs offline with Gears worked for me.

So at this point, what would work better for my situation would be cloud-based files accessed by apps on my local client.

And I'd like to see the ability to access networked files in the cloud be available from every application, meaning the feature would be integrated in the operating system or desktop environment and not be part of a single application.

Just a thought. I'll feel better about Google Docs later this week when I get back to what I mostly use it for. I'm dropping code and documentation into it all the time and sharing those files with my co-workers. That's one thing that Google Docs does better than anything else I've seen.

And I will try to create a heavily formatted document in Docs. I just wish it could be the SAME document shared between Docs and OpenOffice. Maybe a Docs-formatted document would play more nicely in OO than a Word-formatted document turned into a Docs document, then an OO document.

Hey teeming masses, don't say Steve Ballmer, Bill Gates (and the Professor and Mary-Ann) never did nothing for you.

In a totally roundabout way, I learned that Microsoft is giving away — I say giving away — Microsoft Word. OK ... Microsoft Word for DOS. Remember that? I do. I actually used to run Word for DOS a bit back in the day. It was slower than ... everything else — including the admittedly harder-to-use but wildly popular WordPerfect.

At the time I couldn't imagine Word overtaking and crushing WordPerfect. But it happened.

Back to our twisted tale. I found this Lifehacker entry where one of the very best tech-book writers today, Keir Thomas (get one of his Ubuntu books already!), has an excerpt of his new "Ubuntu Kung Fu" book.

The Unix-like OS world is awash in console-based text editors. There are literally hundreds, from vi and nano to joe and emacs. But is there an actual word processor for the Unix/Linux console? Nope.

Thomas suggests running the freely downloadable Microsoft Word for DOS and using the DOSBox MS-DOS emulator to run Word as a command-line word processor.

(For specific instructions, just go to the link and scroll down until you see "Get a High-Quality (and Free) Command-Line Word Processor with Microsoft Word.").

If you just want to get your free Word for DOS, click this link and the almighty Microsoft download deity will cause it to appear on your PC. It's a self-extracting archive, so in Windows double-click it, or at a DOS prompt in the environment of your choice (A command line in windows, or the DOSBox environment in Unix/Linux) and it will turn into the executable for Microsoft Word 5.5 for DOS.

Oh the humanity!

I always wondered why there wasn't a bona fide word processor for the Unix/Linux command line. I'm still wondering, but I'm laughing (inside ...) about this hack.

Did I forget to mention that Keir Thomas is a genius? He's right up there with David Pogue, Chris Negus, Carla Schroder and Mark Sobell in my pantheon of "world's greatest tech writers."

(Here is a screenshot of my Windows XP desktop using PuTTY and Xming to tunnel X over SSH from The Self-Reliant Thin Client running Debian Etch. Click the picture above for a full 1280x1024 image. Clockwise from left, I'm running GNOME's Update Manager, a console with PuTTY and the Rox-filer file manager).

I hadn't used my go-to SSH and X apps in Windows XP to access a Unix-like box in a long while. I actually had to find PuTTY before I could create a shortcut and run it. I already had an Xming shortcut, so I started that and left it in the background until I was ready to begin my X session.

If you use both Windows and Linux/Unix boxes and are not familiar with PuTTY and Xming, you're really missing out. In case it's not totally clear above, PuTTY enables you to run an SSH console session from your networked Unix-like box, and Xming allows you to run X apps over that same connection.

It's all good, clean geeky fun, especially over a local network. One of these days I'm going to experiment with dynamic DNS and figure out how to SSH from a box that isn't in the same building. The stakes there are a bit higher, as the box is out there on the Internet for all to see. But with the proper tools and procedures in place, everything should be secure.

Still, running Linux apps both in the console and in X on a Windows box really never does get old (to me at least).

Related:

While I've experimented with FreeBSD (and offshoots DesktopBSD and PC-BSD), NetBSD and OpenBSD (the latter of which I run the most; including right now), I never really paid much attention to DragonFlyBSD.

A quick perusal of the DragonFlyBSD Web site offers a lot of information on things like its new HAMMER filesystem as well as the operating system's goal of bringing "native clustering support" into the kernel.

It's all a bit over my desktop-using head:

In the 2007-2008 time-frame a new filesystem called HAMMER was developed for DragonFly. HAMMER sees its first light of day in the July 2008 2.0 release. This filesystem has been designed to solve numerous issues and to add many new capabilities to DragonFly, such as fine-grained snapshots, instant crash recovery, and near real-time mirroring. The filesytem is also intended to serve as a basis for the clustering work that makes up the second phase of the project.

The second phase of the project is now upon us. The DragonFly project's ultimate goal is to provide native clustering support in the kernel. This involves the creation of a sophisticated cache management framework for filesystem namespaces, file spaces, and VM spaces, which allows heavily interactive programs to run across multiple machines with cache coherency fully guaranteed in all respects. This also involves being able to chop up resources, including the cpu by way of a controlled VM context, for safe assignment to unsecured third-party clusters over the internet (though the security of such clusters itself might be in doubt, the first and most important thing is for systems donating resources to not be made vulnerable through their donation).

I recently dug out the iPod I've barely used for the past few years and decided to start investigating the many free-software-focused podcasts out there. The one I'm focusing on now is Will Backman's excellent BSD Talk, which provides a great way to learn about all of the BSD-derived operating systems.

Last night I listened to an interview with DragonFlyBSD founder Matthew Dillon.

While all that talk about HAMMER remained firmly over my head, it's nice to know that those who understand the finer points of filesystems are working on ways to continually innovate and bring new features to the operating systems we use.

I decided to take another look at the DragonFlyBSD Web site, and while they don't give it up easily, even a thick-headed guy like me could install and deploy the OS on the server or desktop.

DragonFly — itself derived from FreeBSD — shares the NetBSD package system and as a result has a very extensive repository with thousands of desktop-friendly apps. Like most developers, those who work on DragonFly keep an eye on the other BSDs, and they borrow from FreeBSD and NetBSD when appropriate.

(Side note: There is surprisingly little rivalry among users — and even less than that among developers — of the various BSD systems; code is often freely shared, and this is one of the strengths of free, open-source software that leads to better functionality for everybody.)

At this point in time, DragonFlyBSD is an i386 OS. It's ISOs produce CDs that boot into a live environment. I'm a little fuzzy on the various installation methods, but it looks like the system can be installed via either a console, over a serial connection or even with a Web-based interface.

While right now I'm pretty deep into running OpenBSD, I can see giving DragonFlyBSD a tryout soon.

The beauty of free, open-source operating systems in particular, and the BSDs in general, is that the various projects have different philosophies and emphases. Aside from its emphasis on security and cryptography, what attracts me to OpenBSD is its commitment to multiple platforms (something NetBSD is even more committed to) as well as the extremely high quality of its many packages (more than 4,000 for i386) and ports.

Having 300+ GNU/Linux distributions — all using variants and versions of the Linux kernel — is a great thing.

The BSDs offer a compelling alternative. I have hardware that OpenBSD can auto-configure better than most Linux distributions. A monopoly in FOSS operating systems would be as dangerous as the monopoly Microsoft has in proprietary desktop systems (with a Microsoft-Apple duopoly being almost as dangerous and counterproductive).

Innovation, usability and security can't happen in a monopolistic vacuum. So while Linux, along with the GNU tools and the many pieces of software that go into all those hundreds of distributions, is an important counterweight to Windows and Mac OS, having a half-dozen or so BSD-based projects offering alternative FOSS environments is a great thing for everybody who depends on computers. And that's pretty much all of us.

Related:

Even though I do a lot of work in Firefox, where Chris Pedrick's excellent Web Developer add-on helps me code, whenever I'm doing "casual browsing," working in Movable Type, Google Docs, Gmail or any of the various Web-based programs I rely on that allow it, I use the Google Chrome browser.

Why? Speed.

Even though I think a 3 GHz Pentium 4 with 512 MB of RAM is adequate for Windows XP, there's no denying that Chrome is faster to load and run than Firefox (and Firefox leaves Internet Explorer 7 way back in the dust). Chrome is right up there with the Opera browser when it comes to speed, but already Chrome does better in terms of rendering pages.

And basically Google Chrome is a nice, lean, uncomplicated browser.

I made it my default browser because every time I click on a link in an e-mail (usually in Thunderbird, by the way), the machine would open that link in Firefox. And on this box, while I am using Firefox for development, I'm happier doing the rest of my browsing in Chrome.

I haven't yet had the opportunity to run Firefox 3.1, which is supposed to be much faster than 3.0.x.

So what if Chrome had a tool like Web Developer? And what if Chrome ran (and ran well) in non-Windows environments (Linux, BSDs, Mac OS)? Just more world domination for Google (and a faster box for me).

Web Developer or Firebug?: I should probably try to familiarize myself with the Firebug extension for Firefox. Having more than one tool to help with Web development (and I need all the help I can get), isn't a bad thing. I guess I use Web Developer because it was the first of the two that I was able to get working the way I needed it.

Related:

Google Chrome: What does it offer developers?
Chromium Developer Documentation

My ongoing quest for a Sun Sparcstation 20 continues. Why do I want a 13-year-old box that was beyond cool in the mid-'90s but is still ... 13 or so years old?

If you've read this blog over the past 1,000 entries, you probably don't have to ask that question.

(Answer: I enjoy working on old iron. ... but I should probably be thinning the herd, not the other way around.)

If I do manage to acquire a Sparcstation, I could run Solaris (which is what it was made for), possibly Debian but definitely OpenBSD, NetBSD and, I just learned, a version of Slackware called Splack Linux will also run on Sun hardware.

Do you have a Sun box now? Did you have one years ago? What do you remember, good and bad?

Don't let anybody fool you. If you want to do anything in Movable Type, you'll be neck-deep in arcane MT tags and Javascript before you know it. These "instructions" on how to deploy threaded comments are an example of this.

If this works, I'll eat my hat. I'll have to find a hat, which I will then eat.

As I write in this week's print column, I'm getting ready to give the Ubuntu- and CentOS-powered $0 Laptop to our 5-year-old daughter.

I mentioned that I do have a replacement that was working out pretty well. Of course that wellness went considerably south in the past few days (as chronicled in Dark Side of the Laptop), but I remained determined to prep the laptop, which is currently running Ubuntu/Xubuntu 8.04 LTS as its No. 1 distro, for our daughter, who used it tonight to run TuxPaint.

Whether or not my new/old Toshiba (or newer/just-as-old/identical Toshiba) works out, I'm ready to move on. I've got boxes I've set up in the past couple of months (The Self-Reliant Thin Client, The Debian Mac, which I bet I could finally set up with OpenBSD and actually get it to boot) that could be used more, and boxes I haven't yet had time to work on (an old Dell with something in the 1 GHz-ish range and for some reason stuffed with 256 MB of ECC server memory).

I'm also thisclose to getting my hands on a Sun Sparcstation 20, a box that was the envy of every self-respecting geek ... in 1995. That could be a fun project, don't you think?