September 2008 Archives

We have a Mac at home, and we have a digital camera, too.

Nothing unusual there, right?

So we use iPhoto, part of the often-free-with-your-Mac iLife suite, to download and manage our digital photos.

OK, Ilene does it. Truth be told, I look to her to help me with most things Mac since the iBook G4 is her main computer, which she uses for everything related to the classes she teaches, from research and preparing PowerPoint presentations to delivering those presentations, creating tests and assignments, logging grades, blogging and way more.

I did rescue the iBook from an ailing hard drive, an operation that took more than a couple hours of sweaty, painstaking work to complete.

Once I set up a backup drive with SuperDuper for this 10.3.9-equipped laptop (which doesn't have and won't be getting OS X 10.5's Time Machine), I figured I was done with Mac crises for the time being.

Then iPhoto "forgot" how to display our many thousand photos.

Upon starting the application, the screen would read "Loading Photos ..." forever. Yet the photos, in their many iPhoto-created folders, were right where they always have been. A check of iPhoto in a smaller account with only a few hundred photos revealed that iPhoto, the application, was working.

It was just the database holding our 3,000+ photos that wasn't working.

I did everything recommended in David Pogue's excellent "iLife '05: The Missing Manual." I threw out Preferences files, ditched database files, relaunched iPhoto a half-dozen times.

Nothing worked.

Finally I resorted to the last remaining tip from "iLife '05: The Missing Manual": The iPhoto Extractor application.

I installed iPhoto Extractor, turned it loose on my "damaged" iPhoto database, extracted the 3,000+ photos, then reimported them into iPhoto.

Sure, any changes we made to individual photos (including rotating them so they were right-side up), any albums we may have created to organize said photos (and yes, we had many) were gone.

But the photos themselves were all there.

I have my idol David Pogue to thank for that (any tech journalist with his own book imprint richly deserves idolatry — and gets it from me, big time).

I also have the people behind iPhoto Extractor to thank for dragging all of those photos out of the many, many folders created by iPhoto in which they're stored and allowing me to see them all, back them up without all that iPhoto baggage and then reimport them into iPhoto, something I did but am not very happy about at all.

I'd be more happy just creating my own system of folders and files which wouldn't be compromised by the application that created them.

For the time being, I'll keep using iPhoto (although I should probably upgrade to a newer version than the Version 4 we have now), but I'll be very, very interested to know what my non-iLife options are for managing and archiving photos in OS X.

I'm giving this a try. Having the screen blank when I close the laptop lid in Ubuntu hasn't worked out so well. That usually hangs the system with the $0 Laptop (Gateway Solo 1450).

I went into the Power Management settings in GNOME and changed them. Now when the lid is closed, the laptop should go into suspend.

So far that seems to be working. Upon opening the lid, I hit the power button to resume the computer.

I'm wondering if my freezing-cursor problem (from which I can only recover with a hard reboot) is somehow related to suspend/resume. That would be logical, since I don't have suspend/resume turned on in Debian Lenny due to that feature not working in that distro.

Irrespective of the cursor-freezing issue, having the machine suspend with lid closing is better than having the screen blank. At least this way X doesn't crash.

I still might be in a position to heap praise upon Ubuntu 8.04 for its performance on the $0 Laptop (Gateway Solo 1450) since I reinstalled it a couple of weeks ago with a separate /home partition and a not-screwed-up UUID scenario.

But I keep getting these freezes in which ctrl-alt-backspace or ctrl-alt-delete won't save me. I have to do a hard reset with the power button.

Now this could be due to the shaky nature of my power connection (the power jack from the laptop's brick doesn't quite meet up with the hacked power plug I installed to make this laptop work after I first acquired it). Having a dead battery doesn't help.

I need to figure out whether my freezes in Ubuntu are due to the OS itself or due to the flaky power situation.

I finally got a replacement power jack at Fry's that I could use on the power brick to make a foolproof connection.

It could be chance, but this freezing problem never happens in Debian Lenny, which has problems of its own (related to X refresh, and chronicled in agonizing detail on this very blog).

I will confirm that suspend/resume continues to work, as does everything else. Except for this cursor-freezing.

Again, I'm not ready to blame Ubuntu and am more inclined to blame the power jack/plug situation. I am keeping an eye on the problem.

Another 150 or so updates rolled into Debian Lenny recently, including new Xorg and Intel video driver packages. For the upteenth time, I'm hoping for the miracle of properly refreshing X. It didn't do so well yesterday just after the updates, but there were some "enhancements" to the Debian login screen, principally the word "Debian" appearing in the upper left portion of the screen.

Again, my hope is that this X problem somehow solves itself and I can continue using Debian on this laptop. Again, no breath being held.

Remember my fit-PC post of a few days ago, in which the announcement of an ultra-small, somewhat-reasonably price, Linux-running PC wasn't accompanied by a working link on where and how to buy it?

Well, reader ludovicus came through with a working link at which to purchase the fit-PC in the U.S. and Canada. (Click here for worldwide shipping).

With the AMD Geode LX800 processor (500MHz), 512 MB of RAM, built-in WiFi, a 60 GB hard drive and preinstalled with Ubuntu 8.04 and Gentoo 2008.1 in a dual-boot configuration, the fit-PC Slim is $295 plus $20 shipping, also known as $315. (I'm not sure why more than a few users would want to dual-boot Ubuntu and Gentoo; I'd rather have one or the other.)

When compared to other small-PC offerings, the fit-PC comes out rather well, given its specs. Having the hard drive, half-megabyte of RAM and WiFi certainly puts it ahead of the competition. The Linutop 2 (pictured at right) has a faster AMD Geode processor (800 MHz), and 1 GB internal flash memory instead of a traditional hard drive (which one is preferable depends on your application). No WiFi, either. Linutop does its own Ubuntu-based distro which fits on the 1 GB flash chip and gives you 400 MB to spare for your files.

The problem with Linutop: It costs 280 euros, which is well north of $500.

The fit-PC is small and fanless, but some might be cautious about running a full Linux (or Windows) desktop in 500 MHz of CPU. Having the 512 MB of RAM (as opposed to 256 or 128 MB) will really help speed things up, but a test of this box is definitely in order. If the overall hardware (including video and sound) is good, performance should be acceptable. The problem with more CPU is the need for more power — and the generation of much additional heat.

In regards to what will and won't run on the fit-PC, start here. As far as what will run and how well, I'd love for Debian to be farther along.

The fit-PC box:

More fit-PC:

• fit-PC slim specifications
  
• The fit-PC Forum
  
• OpenBSD runs on the fit-PC

You might wonder why President Bush is scheduled to address the nation at 9 p.m. EDT (watch here on AP's live feed if you have Internet Explorer).

My guess? It's because "Dancing With the Stars" starts at 8 p.m., when we find out who's getting dumped from the show (aside from Jeff Ross, who I've never heard of and who already got booted, the ABC Web site tells me) in its first week:

What will the president be up against at 9 p.m. Eastern? On ABC, "David Blaine: Dive of Death," in which the quirky magician/illusionist/nutcase presumably dives to what could very well be his death. Quite the metaphor for President Bush.

Luckily (or not), the speech starts at 6 p.m. on the West Coast, so watchers of "Dancing With the Stars" and "David Blaine: Dive of Death" have absolutely nothing to worry about.

President Bush is going up against a plethora of death at 9 p.m. On Fox, the Brad Garrett sitcom " 'Til Death" airs. CBS has "Criminal Minds," and no, I'm not going there, but you are welcome to do so. NBC has "America's Got Talent," where I presume they're not looking for the nation's best CEOs, and the CW has a repeat of the not-old-enough-to-be-in-repeats "Beverly Hills, 91210" remake.

Why the AP's live video plays in Internet Explorer only: Microsoft developed the application for the Associated Press. Can you tell? The whole AP video service used to be IE only, but now the prerecorded videos can be played on any computer with Flash. But the live videos are still IE only. I'm not happy about it, but there it is.

If Microsoft CEO Steve Ballmer were to make a Windows commercial, it would be very much unlike those "crafted" by Bill Gates and Jerry Seinfeld.

In fact, Ballmer has already done this. Watch the above video, if you dare.

And don't think that Seinfeld hasn't done TV ad duty for Apple as well (look for him at the end of this "genius" ad; I think it's a legit ad and not something put together by Apple fanboys to make a point, but I can't be 100 percent sure ... and I really don't know how Seinfeld fits in with Gandhi and Picasso):

Ah, the Linux (or BSD) distro review. They're relatively easy to crank out, they bring the traffic in a major way (especially when the excellent Distrowatch links to you).

But do they mean much? Not really, I think.

Most of the time it's the usual:

• "Here's what happened when I tried/failed/succeeded in installing Distro X on Hardware Y"
• "The installer is good/bad/barbaric"
• "Networking/printing/X was easy/hard/impossible to set up"
• "Package management is like Debian/Red Hat/Slackware and is good/bad/barbaric"
• "Repositories are big/small/good/bad"
• "My favorite apps are present/absent/broken"
• "The default desktop/menus/window manager are good/bad"
• "The community is active/nonexistant/helpful/hostile"

And the list goes on. I feel like writing a shell script that can pose questions and crank out automatic distro reviews.

What's harder to write — much harder than the quickie distro review — is a long-term review of a distro after a month or more of heavy use.

For one thing, most of us don't want to spend long periods of time running distros we don't like or aren't familiar with.

And for any given user, most of the 300+ active distros out there won't do anything for our hardware and work patterns that we don't already get from the distros we're currently using.

That's not to say that the many, many dozens of distros out there should just give up and stop trying to do something better and different (even though what they're doing is usually based on an existing distro and often doesn't add much, if any value to what they're already copying).

I'm just saying that after after a year and half of writing this kind of thing, I'm tired of both writing and reading quickie distro reviews that don't really tell the potential user of a given distribution all that much that they can use in making their decision.

I've already done tons of posts on Debian Lenny, and almost every problem has been fixed at some point in the project's long road from Testing to Stable.

So should I do another distro review on the installation, care and feeding of Debian Lenny when it finally does receive its Stable status?

Do I need to reinstall Ubuntu every six months and write about how that goes? OpenBSD?

Never mind that the development of OpenBSD is purposefully more evolutionary than revolutionary, or that a rolling release might be better/worse than one that comes out every six months or at some other regular (or not so much) interval.

I don't quite know how to end this tortuous post except to say that I reserve the right to change my mind. Maybe I'm purposefully shoving my own head in the sand by not embracing your favorite distro (usually Slackware or Mandriva) and sticking to what's been working for me (Ubuntu, Debian, OpenBSD, Puppy ... and that's about it these days).

Maybe it's part of the evolution (or devolution) of me as a writer about technology, but right now I'm convinced that that there's a better way to do all of this that doesn't throw out free, open-source software in favor of what the average guy/gal is using (Windows/Mac) but also does more than preach to the same creaky choir, of which I myself am a warbling member.

Being more truthful, I won't stop reading distro reviews, especially when they're written by writers who know what they're doing. But I plan to be a whole lot more careful about writing them. I've been thinking (and writing) for some time about why it's more than time for me to stabilize my herd of machines and stop the endless process of cranking one distro after another onto their partitions.

The freedom to change distros like underwear, at more than one level, begins to detract from what a computer operating system is supposed to be for, which is getting stuff done. I guess I want things to be more about ends rather than means.

I've been running Debian in one form or another, on one box or another, ever since Etch went stable in April 2007.

Lately most of my work in Debian has been on my Gateway laptop (aka The $0 Laptop), which from a hardware standpoint responds better to Lenny (Testing) than to Etch (Stable), so I'd been using Lenny about 100 percent of the time until I got what I'm calling The Debian Mac, a Power Macintosh G4/466, in a mass giveaway of old Apple hardware.

I tried a Xubuntu live CD, but once I burned a Debian PowerPC image and installed Etch, I knew that this chunk of hardware — especially the ATI video card and LaCie 22-inch monitor — responded to the distro extremely well.

I last ran Etch extensively both on my VIA converted thin client and on the $15 Laptop (Compaq Armada 7770dmt), but since then the VIA has been used to compare Ubuntu and Wolvix, and the Compaq has been running OpenBSD and Puppy.

So it had been a long time since I had worked with Etch.

It really is like a comfortable pair of old shoes.

On the Mac, anyway, everything works perfectly. I didn't have to do anything beyond enter my LaCie electron22blue II monitor's native resolution (1600x1200) when prompted during automatic X configuration while the install was running.

Sure there are a lot of old packages in Etch: OpenOffice 2.0, Gaim instead of Pidgin, Firefox 2 instead of 3.

But again, everything works, and I could really get used to not having to wait for dozens or hundreds of new packages to download and install every week, like I do when running Lenny.

Eventually Lenny will be as old-shoeish as Etch is now, with the rush of revised packages slowing to a familiar trickle, and with the whole system working as well as it ever will (and hopefully that will be well enough.

I already see this process happening in my Ubuntu 8.04 Hardy install on the Gateway, where everything is settling in (and where I recently did a reinstall after what I suspect were UUID conflicts after I monkeyed around with the partitions more than one too many times).

I'll be hoping that for the next 2 1/2 years, the Canonical team can give the same quality of attention to security and bug-patching in 8.04 — the distro's second "long-term support" release — as the Debian Project does for its Stable release and Red Hat does for its RHEL products.

I kind of, sort of planned to remain with Ubuntu 8.04 on the laptop for at least two of the three years for which it will receive support, especially now that I'm continuing to have problems with X in Lenny on that machine that I can't seem to solve. The same problems don't occur in Etch on the laptop, but there are so many things that Lenny does bring to that particular platform that rolling back to Etch isn't an option.

Also, I see using more desktop machines, as opposed to laptops, in my near future, and that breaks things wide open, since the ACPI problems I have with laptops and FOSS operating systems aren't nearly as much of a factor on desktops, where CPU fans generally run all the time and I don't care about suspend/resume.

It's been suggested that I try Slackintosh — an unofficial port of Slackware to the PowerPC — on the G4. That's intriguing. I've always found that KDE runs quicker in Slackware than in any other Linux distribution. KDE probably runs about as fast in Debian, and I could add that desktop (as well as Xfce, Fluxbox, Fvwm, or what have you) to this Etch install.

In my experience, it's easier and better to run a given distribution with its "main" window manager: Debian and Ubuntu with GNOME, Slackware and Mandriva with KDE, etc. Everything seems to "fit" better and work better, too.

I do appreciate that things like the GNOME NetworkManager and Synaptic Package Manager appear across the Ubuntu offshoots (Kubuntu, Xubuntu), unifying the experience somewhat. Going into the KDE version of Debian and using whatever it is that KDE uses to update packages, or into the Xfce version and relying on Aptitude (not a bad alternative; I use and appreciate Aptitude more and more these days) is a bit jarring, and I credit the Ubuntu developers for maintaining a core of helpful applications across their different distros.

And as I've said many times, both GNOME and Debian take a lot of bashing out there in blogland, but I've found GNOME to be extremely fast and quite capable of doing what I want to do, and in tests between Debian and Slackware (running Xfce and Fluxbox), Debian more than holds its own in terms of speed on the desktop.

Not that I wouldn't or won't run Slackware in the future. The fact that Patrick Volkerding and crew seem to maintain Slackware's various versions for years and years says a lot for how long you can safely run it. I don't know if that same longetivity applies to running Slackintosh; I suppose it depends on how many of the packages come from Slackintosh and how many come from Slackware.

But it's hard to argue with "Debian just works." When Lenny goes from Testing status to Stable, Etch will become what Debian calls "Old Stable," and receive security patches for another year. That gives those of us running it — on the desktop, the server, in embedded systems, etc. — ample time to figure out what to do next.

This being a Linux PowerPC box, I wonder how long I'll be happy without the ability to run Flash (which Adobe doesn't code for Linux on this platform; guess I could always dual boot Linux and OS X if I really, really need Flash. Or I could just give up on PowerPC and run i386. I have a promising Pentium 4 machine that ideally will be my next OS test bed, replacing the VIA converted thin client that has spawned dozens of reviews and hundreds of blog posts since I began writing this particular brand of blather at the beginning of 2007.

Taking another tangential trip in this already-too-tangential entry, it's tempting to try each and every new distro, to write a quickie review and reap the substantial boost in traffic that one gets from links in places like Distrowatch.

Truth be told, I've exposed my family to Linux quite a bit. When our iBook was awaiting the 3-hour operation to replace its ailing hard drive, I had a laptop set up dual-booting Ubuntu 8.04 and Debian Lenny (before Lenny's X problem surfaced). My wife prefers Ubuntu, and my daughter prefers Debian, if only because I have her Lenny account set up to quickly access her favorite educational games, which run that much better in Debian (the main difference being that launching TuxPaint through GCompris in Ubuntu results in a lack of the usual sound effects, which return when starting TuxPaint by itself, or starting the program either way in Debian.

And with the Mac's giant, wonderfully-clear monitor, I can see getting a whole lot of work done on this G4. With its 466 MHz CPU, it's not as fast as the 1+ GHz machines I seem to have in abundance, but for a hack writer, it's plenty fast.

And with a drive inside the box set up for and dedicated to backups (something I could do in any i386 desktop box with a case of sufficient capacity), a solid if unsexy distro (Etch), I could get a lot done if I manage not to screw the whole thing up.

(I'm not saying I won't try again to install OpenBSD and actually get it to boot, but it'll be on a different drive so I can preserve this Debian goodness. If I can dual-boot both systems, I will, but it doesn't look good at present.)

All that means is that I won't be dist-upgrading this Mac anytime soon from Etch to Lenny, even if the latter actually goes Stable, as scheduled, some time this month. I might try a cautious test of Lenny on a separate hard drive, or I might hold onto Etch with my cold, dying fingers until the last possible moment.

But like many, it only takes a little promise of something better to change my mind. If Gnash would play Flash video with any kind of consistency, I'd jump this box to Lenny in a second. If Gnash gets that good, will somebody let me know?

The answer is yes: Tripling the memory on a Power Macintosh G4/466 running Debian Etch makes a dramatic difference in the responsiveness of the system.

It makes the system way, way more usable, cutting down on swapping tremendously and boosting free memory, also tremendously, when doing normal desktop tasks.

They're getting smaller and smaller.

I want one.

There's the Fit-PC Slim, which measures 4.3 x 3.9 x 1.2 inches and uses 4 to 6 watts of power:

and the previous Fit-PC, measuring 4.7 x 4.6 x 1.6 inches and using 3 to 5 watts of power:

Amazingly, both of these boxes will house a 2.5-inch laptop-sized hard drive.

Prices: $220-$335 for the Fit-PC Slim, $285 for the original Fit-PC.

The only problem: The two Linux Devices articles look good, but there's no way to actually buy these two little boxes.

If I could get my hands on one, a few or a bunch of these, I'd be happy to let some or all of my bulkier hardware go ...

After setting up the mail-client portion of Opera, I haven't had much occasion to use it, since I am quite comfortable and happy with Thunderbird in Windows.

The Opera browser is fast in relation to other browsers, and its mail client is fast, too.

I have it set up with IMAP so I can access my mail from any number of places (unless you use one computer all the time ... forever, POP is barbaric). It's a fairly simple client, but it works great, and if you spend a lot of time in front of the Opera browser, it keeps your e-mail just that much closer to you.

Not that I'm an expert, because I'm far, far from it, but setting up a system to use the Movable Type blogging system ain't easy.

You need a server on which you create a database, configure Web-server software, implement PHP and somehow get all the permissions to work right.

Putting together a Movable Type installation should be easier now than ever, as I read in Matt Asay's Open Road blog, now that Movable Type parent company Six Apart has partnered with Jump Box, a virtual-machine company, to offer a fully virtualized MT bundle.

If I read this correctly, it means that through virtualization, you can have an instant Movable Type setup without having to do all that much configuration on your own.

Since the Power Macintosh G4/466 has enough space for four hard drives, I decided to put a second one in the box.

I originally planned to dual-boot Debian Etch and OpenBSD, but I still can't get the system to even boot OpenBSD after an install, so I abandoned that plan. After two successful OpenBSD installs on i386, I figured I could handle a MacPPC install, but it was not to be.

And since Debian Etch installs — and runs — so well, I'm keeping it.

Before the last unsuccessful OpenBSD install, I set the second drive's jumpers to make it a "slave." Then I used an extra IDE cable as an "extension cable" so I could plug the second IDE input from the too-short motherboard cable into the second drive.

I'm pretty accustomed to chaining IDE cables together for longer runs — I did it for over a year with my VIA thin client.

Anyway, I booted into Debian Etch, ran the GNOME Partition Editor to create an ext3 filesystem on the backup drive, then mounted it and did a few tests with rsync.

Yep, I'm using rsync to do the backups. I first learned about it in Carla Schroder's great "Linux Cookbook" (which all of you should own ... and which O'Reilly should beg Carla and pay her well to revise immediately).

Rsync is in the default Ubuntu install, but in Debian Etch, you have to add it:

# aptitude install rsync

In order to use the backup drive, I created a directory called hdb1 in /media and used it to mount the drive:

# mkdir /media/hdb1

Instead of mounting it from the command line, I put a line in /etc/fstab.

Here's what /etc/fstab looks like on this machine:

# /etc/fstab: static file system information.
#
#
proc /proc proc defaults 0 0
/dev/hda3 / ext3 defaults,errors=remount-ro 0 1
/dev/hda5 /home ext3 defaults 0 2
/dev/hda4 none swap sw 0 0
/dev/hdb1 /media/hdb1 ext3 defaults 1 2
/dev/hdc /media/cdrom0 udf,iso9660 user,noauto 0 0

I made a directory called homebackup on /media/hdb1, which is where my backups of /home will go:

# mkdir /media/hdb1/homebackup

To run the backup from the command line, here's what I did (at a root prompt):

# rsync -av --delete /home /media/hdb1/homebackup

(A note on rsync: The FIRST time I ran rsync with the line above, a home directory was created in /media/hdb1/homebackup. Subsequent uses of the same rsync command just update the directory; they don't create it again.)

The beauty of rsync: I really didn't explain rsync all that well. Basically, it's a great backup utility that, when used multiple times, only backs up the changes between the source and the destination directories.

What the switches mean in rsync -av --delete:

-a stands for archive mode. It is a catch-all switch that turns on a variety of things in rsync, most of which I don't understand. I use all of these switches because Carla Schroder told me to. For more (or less) clarity, check out the man page for rsync.

-v stands for verbose so rsync outputs to the console exactly what it's doing.

--delete tells rsync to delete files in the destination directory that have been previously deleted from the source directory. Without this switch, the backup directory would just get bigger and bigger. You might want that; without --delete, everything you delete would still be in the backup directory. But I don't, so I'm using this switch.

in order not to have to remember to type my entire rsync command with the switches and the directory information, I put it in a very simple shell script in /usr/local/bin.

I called the shell script backup — can't forget that.

Here's what is in the /usr/local/bin/backup file:

#!/bin/bash

# This file backs up the /home partition to the secondary drive with rsync
rsync -av --delete /home /media/hdb1/homebackup

To make the file executable, I ran:

# chmod a+x /usr/local/bin/backup

I'm not a big fan of using su to root, so I added my user account to the sudoers file (after I su'd to root):

# visudo

When you invoke visudo from a root prompt, the system opens the sudoers file with the default editor. Usually that editor is vi, but in Debian, the default editor, as configured, is nano. I've been using vi a lot lately in OpenBSD (with the i386 systems on which I can actually successfully install it). But for me it's easier to use nano, with it's F3 to save, CTRL-x to exit, and just moving the cursor around with arrow keys and not having to worry about going from command mode to insert mode as in vi.

Now that my little vi rant is over, here's what I did in the sudoers file after visudo brought it up in nano:

I gave my user account the same privileges as root. I wouldn't do this for other users; I'd figure out how to more finely grain the permissions, which is what sudo is really good at. But I'm not running a huge multiuser box, so giving my user account those privileges is good enough for now.

Here's what the relevant section of the sudoers file looks like:

# User privilege specification
root ALL=(ALL) ALL
steven ALL=(ALL) ALL

Back to the backups: This script will run without sudo or su privileges, but the --delete switch will not work because the non-root user doesn't have privileges in the lost+found directory in /home. So if you want to back up the entire /home directory, including all the files for all users, you should run it this way:

$ sudo backup

or after su to root:

# backup

Of course, if you want to set this up for your individual users, you can rewrite the rsync line to only back up their own /home files:

rsync -av --delete /home/steven /media/hdb1/homebackup

Then they could run it from a regular prompt, without root privileges, and it will work.

Since I'm the only user of this box, I made the rsync command cover the entire /home folder, which means, for me, using sudo to make it work.

What about a cron job? Sure, you could do this as a cron job, but I'm not in the habit of leaving the box on all the time, so cron (or even anacron) doesn't fit in with my computing habits. I guess anacron, which runs jobs after a certain interval of time has passed (and not only at certain, specified times), would work, but I'd rather just run the shell script periodically and see the jobs scroll down my terminal window.

But like anything in Unix, there's more than one way to skin any given cat.

Ubuntu note: In Debian, the "main" user does not automatically have sudo privileges. But in Ubuntu, that user does have sudo ability. You don't have to use visudo at all for that user; just use sudo from the get-go. I'm not sure about subesequent users in Ubuntu, but I sure hope they don't have sudo privileges by default; I think they don't but I don't have a Ubuntu box handy to check.

The trade-off for Ubuntu users is that they can't su to root. Well, they can ... Ubuntu encourages you to use sudo, but you can get a root shell this way:

$ sudo su

After you type in your password, you will have a root prompt:

#

As I said, I'm not a fan of using su to root, and I'd rather use sudo, but there are some things that sudo can't do, and that's when the Ubuntu su trick is necessary. I'd like to thank whoever it is who passed that one along to me.

Conclusion: If you hae a place to back up your files, using rsync is a great way to make those backups. Rsync also excels at backups over SSH, which Carla Schroder goes into in great detail in her book. She also shows you how to set up an rsync server.

Just buy the book already, will ya?

I'm among those profoundly affected by the Whole Earth Catalog, which oddly enough I encountered in not the '70s but the '80s, when a new edition (probably the last) was published. It was called "The Next Whole Earth Catalog," and I pretty much wore it out over the years after I bought it.

If any book was a precursor to Google, Craigslist or the Web itself, it was "The Whole Earth Catalog."

On BoingBoing, I found this link to an oral history of the WEC.

More information:

Not that anything approaching brain-surgery-level thinking was in any way involved here, but I figured out why and how it's easy to get paragraphs to automatically indent when writing in the Geany text editor.

First of all, it's not called automatic tabbing or paragraph inentation. The correct term for what I'm enjoying so much is auto-indentation and it can be turned on and off under the Document menu in Geany. The defaults for auto-indentation can also be set in the Edit menu under Preferences--Editor.

When writing for print, where I don't need — and can't stand — having two returns between paragraphs. After transferring the file from this laptop to my newspaper's print publishing system, those double-returns demand that I delete one of them. That's because in most non-Web publishing, indented first lines make paragraphs distinct from one another, not extra linefeeds.

So having the indents on the first line of every paragraph helps me seen where each paragraph begins.

I know that programmers use indents to help structure their code. But when something so right for coding in C also helps hacks like me, making traditional word processing applications less needed, everybody wins.

My inability to do more-than-simple mathematics at times has really put a cramp in my computing style.

In the case of OpenBSD and the $15 Laptop — the Compaq Armada 7770dmt — it has cost me 32 MB of RAM ever since I upgraded from 64 MB to the maximum 144 MB.

The reason is that in some Compaq's OpenBSD will not address more than 16 MB of RAM without some intervention on the part of the user.

This intervention is described in OpenBSD's installation FAQ. In order to have the system recognize the additional memory in the affected Compaqs, you must create a file called /etc/boot.conf containing one line, which tells the OS about your additional memory.

The example in the OpenBSD FAQ is for a system with 64 MB of RAM but with the aforementioned 16 MB being recognized.

To get the additional 48 MB recognized by the system, the following line should be in your new /etc/boot.conf file:

machine mem +0x3000000@0x1000000

That worked perfectly when I, also had 64 MB of RAM.

But when I added 128 MB to the 16 MB on the motherboard (removing the 48 MB of RAM that the laptop had when I originally got it working), I had to change the value in /etc/boot.conf.

And I didn't do my math correctly. This is what I had:

machine mem +0x6000000@0x1000000

I suspected that the machine wasn't using the entire 144 MB, but I'm not conversant enough in OpenBSD as opposed to Linux to a) figure out exactly how much memory the system was recognizing (a number available in the dmesg output) and how much the system was using and how much was still "free" (which can be seen in the output of the top utility.

I've been looking at OpenBSD's installation FAQ again recently because I'm eager to try the OS on my new/old Power Macintosh G4/466, and I looked at the section on Compaq memory again and realized my error.

I realized that if 48 MB is recognized with +0x3000000, then each +0x1000000 represents 16 MB of additional memory. And following that logic, 48 MB is 16 MB times three, yielding the FAQ example of +0x3000000.

And since I was adding 128 MB, that was 16 MB times ... how many? Eight is the answer. So instead of the +0x6000000, I needed the following line in /etc/boot.conf:

machine mem +0x8000000@0x1000000

I booted OpenBSD 4.2 (I don't have enough space in /usr to upgrade to 4.3 — such is the problem with making that partition only 1 GB in size when it needs to be, as the FAQ says, 5 GB). Then I changed the value in /etc/boot.conf, rebooted and saw the dmesg roll by during the reboot with the entire 144 MB being recognized.

Missing 32 MB all this time represents a very large performance hit, and I'll be anxious to see how gaining access to that memory changes performance under X.

For those keeping score, here's the relevant part of my OpenBSD dmesg AFTER doing the /etc/boot.conf fix:

real mem = 150564864 (143MB)
avail mem = 137732096 (131MB)

I've taken to calling my Power Macintosh G4/466 the Debian Mac.

I continue to be amazed at how well Debian Etch runs on this thing with 466 MHz of PowerPC CPU and a smallish 128 MB of RAM. (I'll take this opportunity to repeat that on this box, Etch runs many a ring around Fedora 9's PowerPC port).

The best thing I could do for the usability of this box is to up the RAM. It'd be nice to have 512 MB of RAM in here. The box will take up to 1.5 GB, but I have yet to find any PC-133 RAM sticks in my possession that will work.

I had a sweet 512 MB module that, in all fairness, I've never been able to test, and it didn't work in the G4. I had a bunch of smaller modules (32 MB to 128 MB), none of which worked either.

There's a lone 256 MB module in my VIA test box that I might try, but I probably will be reduced to going on eBay and looking for SIMMs that somebody pulled from an existing Mac.

With the huge 22-inch LaCie electron22blue II monitor and, with Debian Etch, a very well-matched OS, this is a really nice box to work on.

If I didn't mention it before (and I know I did), the biggest impediment in using Linux or BSD with a PowerPC-based machine is the lack of Flash support. Since Flash is so insidious and must be written for your exact CPU, which it taxes greatly, by the way, there is a Flash plugin for Mac OS on PowerPC, for Linux on i386, but not for Linux (or OpenBSD) on PowerPC.

You can get around the Flash problem in OpenBSD on i386 by using a Linux browser (I use Opera for that purpose) and OpenBSD's excellent Linux compatibility feature, but there's no easy way to play Flash content in Linux on PowerPC.

I installed swf-dec, but that has yet to do one Flash-y thing for me.

Gnash might work in some situations for PowerPC Linux — and it pretty much represents our only hope for this platform — but it is not part of the Etch distribution, and I'm not desperate enough to backport it.

Actually, if I was convinced that Gnash would work, I'd upgrade to Lenny immediately. But I'd miss a) the stability and b) not having to install 100 updates a week in Etch (like I seem to do on my Lenny laptop).

Coming up: I add a backup drive to the Debian Mac — and create a very simple shell script to facilitate the backups we all should be making ... but only after I fail once again at installing OpenBSD.

(This post was originally written on May 22, 2008; since that time, I've added the RAM, and it does indeed make a difference. It's still not easy to live with 144 MB of RAM and 233 MHz of CPU, but it's easier than having less than half of that M. What I can say is that 500 MHz of CPU and 256 MB of RAM is positively picnic-ish. Also, I finally did the OpenBSD 4.2-to-4.3 upgrade on the VIA box. It wasn't easy, but I did get it done.)

If the question is "how low can you go" in terms of computer memory, it's all about applications.

If you stayed in the Linux console and never ran X, just about anybody could be happy with 32 MB of RAM. It might be hard to actually run Linux or a BSD in 16 MB, but I've heard of Linux distributions that will do it, Damn Small Linux, Tom's RtBt (is that the right spelling?) and DeLi Linux among them.

But as much as the hard-core users talk about how they stay at the command line all the time, it's hard to get much done strictly in a console when you're a regular person. Sure you can use Lynx for text-only Web browsing, you can set up Mutt (and Postfix/Sendmail/msmtp/esmtp, Procmail and whatever other helper apps are needed) with highly customized configuration files designed to handle and filter multiple mail accounts, use Vi or Emacs for text editing and all that.

But the bottom line for me is that I need a Web browser. A "real" Web browser, something that works with Movable Type and Google Docs, and that pretty much means Firefox or some Iceweaselish derivative.

I don't tend to use OpenOffice very much (although it runs better in Debian with 64 MB that you'd think), I barely even use AbiWord these days. I'm not saying that I won't need OpenOffice in the future, but at present I'm most comfortable using various X text editors, including Geany in most Linuxes and BSDs, Gedit when I'm in GNOME, and Google Docs half the time just for the easy portability of my copy.

And while Geany doesn't load super quickly from a "traditionally" installed distribution (but is quite quick when loaded into memory as it is in Puppy Linux, once it's loaded it runs very well indeed.

And the Dillo Web browser -- which looks better in its OpenBSD incarnation than it does anywhere else -- performs quite well in 64 MB of RAM. The only problem is that Dillo can't do everything I need to do on the Web. At least the Dillo in Puppy and DSL has https support. That's not turned on in OpenBSD, and the app needs to be recompiled to add it. I can manage to turn on cookies in OpenBSD, which helps me with some sites, but for anything remotely complicated, Firefox is essential.

And while Firefox will run in 64 MB of RAM, it does so very poorly. There just isn't enough memory to keep the program from swapping to the drive incessantly whenever doing just about anything.

In this very 64 MB, I've run just about everything that will load on this Compaq laptop: Puppy, DSL, Debian (the Xfce install, plus a "standard" install with Fluxbox), Slackware (without KDE) and OpenBSD.

Truth be told, Almost all of these OSes run just about the same. Damn Small Linux has a bit of an edge, and if DSL 4.3 ran as well as 4.0, its inclusion of Firefox 2 would put it over the top. As it is, I've lost my desktop wallpaper, and I can't figure out how to display the menu in Fluxbox (even though I prefer to run JWM).

Puppy definitely needs more memory, especially to run the Mozilla-derived Seamonkey Web suite.

Debian Etch was OK. While the Xfce install is odd in many ways, as I say, I was surprised to see OpenOffice run at all -- and not too badly at that. Iceweasel was, again, an exercise in frustration. But Debian remains a distinct possibility for this machine.

It's main OS for awhile has been OpenBSD, with a partition set aside for the Linux files generated by the Puppy and DSL live CDs.

OpenBSD runs pretty well, but as I said, Firefox remains an issue.

The question: Will things improve with the boost of RAM from 64 MB to the Compaq Armada 7770dmt's maximum 144 MB? From my past experience, I know that Puppy can run in 128 MB if you have swap space, and DSL is certainly comfortable with 128 MB.

To answer the question, I could reduce the memory in my Via test box from 256 MB to 128 MB and see how OpenBSD (now version 4.3) runs in that configuration. But I'd have to pull the cover from my converted thin client and find a 128 MB SIMM. I've probably got one ... somewhere.

Better to just wait for my Compaq memory to come in the mail (luckily it's cheap).

I've know for awhile that 256 MB is a significant sweet spot for Linux, but I'd love for 144 MB to be just sweet enough to give this laptop a new lease on open-source life.

And while I managed to upgrade my VIA box from OpenBSD 4.2 to 4.3, it takes a lot more work than a simple apt-get, and I'm reluctant to do it

(This post was originally written on April 24, 2008. In the days following, I was able to tweak xorg.conf and run Wolvix on the $15 Laptop (Compaq Armada 7770dmt, on which it runs quite well.)

Finding my long-lost Wolvix post got me itching to run it again. I haven't had it on the $0 Laptop (Gateway Solo 1450) in some time. It did a good job there, but I wasn't able to turn off the annoying tap-to-click feature on the laptop's Alps touchpad, and I've been pretty happy with how Debian Lenny is doing that and more, so my use of Wolvix has dropped quite a bit.

If I could manage to get X configured properly on the $15 Laptop (Compaq Armada 7770dmt) now running OpenBSD 4.2, I'd be running the smaller Wolvix Cub (smaller than Wolvix Hunter, which I run otherwise) on the aging PC right now. The combination of the Xfce and Fluxbox window managers, plus the excellent choice of apps (it has pretty much everything I use day to day) makes the Slackware 11-based Wolvix Cub and Hunter two of the best choices out there — for me anyway.

Adding to Wolvix's flexibility, it can run as a live CD, or be installed in a traditional or "frugal" manner. I've chosen traditional installs, and the install process in Wolvix is excellent. It's easy to create as many partitions as you need, you get a choice of GRUB or LILO bootloaders, and once you do have it installed, slapt-get and Gslapt are ready to bring all the apps up to date from both the Wolvix and Slackware repositories. And the Slackware team continues to update version 11, which was first released in October 2006. Even Slackware 8 (circa 2002) seems to get an update every once in awhile.

All of this makes me very comfortable running Slackware and ... well ... Wolvix, because the two "big" Slackware-derived distros, Zenwalk and Vector, use their own repositories, and they don't support a given release for very long before moving on to the next one. This burned me pretty good one time; I was running Zenwalk 4.4.1, and it was working great, but an unfortunate invocation of the distro's update manager completely broke the thing after 4.6 came out. My machine wouldn't boot the 4.6 CD, so it was the end of Zenwalk for me.

I like Vector, but I like Wolvix even more. And I really like having Slackware's security team watching out and sending patches that flow right into my Wolvix installs with a few easy clicks.

I also like a Slackware-based setup that doesn't depend on KDE. If you have a fast-enough box, KDE is great. My Gateway laptop handles KDE very well, even though I prefer to use GNOME on it. But on my older, slower, less-memory-rich boxes, KDE is a bit too sluggish, and I like a well-appointed Xfce or Fluxbox setup much better. (Note on 9/15/08: My most recent Slackware 12.1 install on the Gateway, though not without problems related to X configuration, again showed KDE to be quite spry and Slack's default lineup of applications to be more than adequate.)

Ever try installing Slackware without KDE? It's easy to do. In fact, the excellent Slackware installer offers complete control, on a package by package basis, over what gets put on your box during the intallation. If you elect to leave out all of KDE, you can run Fluxbox or Xfce, but you get very few apps. And the install still takes up over 2 GB of space. No Abiword, no OpenOffice. I guess that would be a good time to add one of the many GNOME-for-Slackware packages out there, like Dropline GNOME (which wouldn't work on this non-686 CPU, I think, but one of the others will).

But again, Slackware's KDE-centricity doesn't leave the Xfce or Fluxbox user with a whole lot of applications. Sure you can build the system you want with Slackbuilds, Linux Packages or Robby Workman's packages, but I don't think it's a coincidence that there are three major Slackware-based distros (Wolvix, Vector and Zenwalk) that use Xfce as the primary desktop environment.

And being able to use a Wolvix CD to get literally dozens of applications I know and love makes things that much easier. Add to that my seeming inability to get GRUB to boot Slackware 12 (I'm sure using Slack's own GRUB package and script would solve my problem), and Wolvix has solved quite a few problems for me.

(Note on 09/15/08: While Wolvix does an excellent job setting up the LILO or GRUB bootloaders, I've since adopted a policy for dual- and triple-booting in which one distro, using GRUB, controls the Master Boot Record and chainloads into all the other bootable partitions on the drive, with every one of those OSes using their own bootloaders — whatever they may be — on their own root partitions. It makes swapping OSes easier and makes automatic updates of the various /boot/grub/menu.lst files work every time — not always the case with everything stuffed into the /boot/grub/menu.lst controlling the MBR.)

(This post was originally written on April 24, 2008; since then, I've bumped the system up to 144 MB. This entry should set the scene for how much better things are working with the additional memory).

When you're not running X, 64 MB of RAM is plenty. In OpenBSD, or just about any version of Linux for that matter, you just don't need a lot of memory to use the console. Of course, you can't do a whole lot either.

I know, I KNOW, that real geeks use the command line as much as possible. E-mail with Mutt or Pine (and fetchmail, procmail, sendmail, procmail ... did I miss anything (maybe msmtp, which I prefer, or esmtp, exim, postfix ...), text entry with vi (or nano, joe, emacs), text-only Web browsing with Lynx or Elinks.

OK, I do all this stuff, though I did give up on Mutt; it just didn't work for me as well as I needed, and while I put in plenty of time on the configuration, I needed to be way more of an expert than I'll probably ever be). But I really prefer to run X. I get the apps I want, real Web browsing, and a whole lot more overall productivity.

But X takes memory, and while OpenBSD with the Fvwm window manager can run in 64 MB, things take forever and a day due to all the swapping. Unfortunately, my 1999-era laptop -- a Compaq Armada 7770dmt -- maxes out at 144 MB. That's 16 MB on the motherboard, plus two 64 MB EDO SODIMMs.

The memory is on the way (I hope). Right now I have two smaller SODIMMs, a 16 MB and a 32 MB, in the laptop. And yes, I had to do the Compaq memory fix from the OpenBSD FAQ to make the OS recognize the "extra" memory. But it does work.

Anyway, I'm hopeful that OpenBSD will perform dramatically better in X with 144 MB. Since this is a pre-ACPI laptop, I don't have the problems that plague me with my Gateway Solo 1450, on which only Linux, it seems, will turn the fan on and off in response to CPU temperature. In OpenBSD, it's all on. In FreeBSD, it works for a day, and then that's the end of it. Can't figure out that one.

But I'd love to have a laptop devoted to OpenBSD (I'm using vi now ... but I miss Geany, Firefox and the rest of the junk I've got loaded on here). And if OpenBSD can work well on the desktop with only 144 MB, that will be a significant achievement for all of us with hardware in the 10-year-old range.

I'd love to roll OpenBSD onto my 10-year old PC that now runs Windows 2000. It would Do OK with Linux, for sure, but getting OpenBSD on there would be really great. And I have a full 256 MB of RAM on that box. I'm already running that much memory on my test box in the office, and I have no complaints there when it comes to running X apps in OpenBSD.

I started X to finish this post. First I ran Firefox, even though this laptop has only wireless 802.11b networking (and no wired Ethernet, although I've been meaning to get a PCMCIA Ethernet card). Yep, still takes a dog's age to start Firefox, and it's not all that responsive when it's running.

Again, I would love for that NOT to be the case after the memory upgrade.

I started Geany to continue writing. Geany runs pretty well with this 233 MHz processor and 64 MB of RAM.

So does the Dillo browser. And everytime I write about Dillo in OpenBSD, I like to mention that, for some reason, the Dillo menus and buttons look way better in the OpenBSD version of the app (I'm using the package, not the port) than they do in any other operating system in which I've tried it. And I've tried many.

Long story short, I decided to wipe Ubuntu 8.04, which had been stalling during the boot process for an intolerable length of time, from the $0 Laptop's (Gateway Solo 1450) hard drive.

The problem had something to do with the ATA controller, and I think I know what was "wrong."

Back to that later.

I decided to throw Slackware 12.1 on first. I did this. X wouldn't work, and none of my little tweaks to xorg.conf would work.

Solution: Run Wolvix Hunter as a live CD and copy that xorg.conf to the Slackware 12.1 install.

That worked.

I had the full Slackware desktop at my disposal.

If you're going to use KDE, I really see no faster place (barring the cut-down PCLinuxOS spins) than in Slackware.

I'm still not crazy about KDE, but I did find a KDE text editor (can't remember which one) that loaded in a second.

I had planned to stick with Slackware for a couple of weeks at least, but I started getting the same kind of screen artifacts that have been plaguing me in Debian Lenny.

So I turned back and reinstalled Ubuntu 8.04, this time with a separate /home partition to make reinstalls less painful.

It works great.

I think my previous problem had something to do with how Ubuntu recognizes drives, and when I deleted and changed partitions after the install, the UUID numbers were somehow screwed up. Not too technical, my explanation, but that's what I suspect at the moment.

To get sound to come out of the front speaker — which isn't the default, by they way — on the Power Mac G4 (mine is a 466 MHz) while using Debian Etch, open a terminal.

run alsamixer:

$ alsamixer

Arrow over to PC Speak and type m to unmute the speaker. Use the left and right arrows to navigate and the up and down arrows to adjust the levels of Master, Bass, Treble and PCM as you wish. It helps to have some audio playing while you do this so you can hear the effects of your changes.

When you're done, hit esc to exit alsamixer.

Then, still at a prompt in the terminal, you must save your settings:

$ alsactl store

My settings still weren't saved when I rebooted, so I used su to become root and put the following line in /etc/rc.local:

alsactl restore

Once I modified /etc/rc.local, I could reboot and have the sound coming out of my front speaker without having to go into alsamixer every time.

It's a bit hacky, but whatever works, right?

I turned on touchpad tapping today in Debian Lenny on the $0 Laptop (Gateway Solo 1450), and I had forgotten the one thing that makes the tap-to-click function work so well in Lenny:

Double-tapping.

By that I mean it takes two taps to count as a left-click. That way, every time I use the touchpad to move around the screen, I'm not mistakenly left-clicking on things and screwing up everything I'm doing

I also turned on touchpad scrolling, by which sliding a finger down the far right of the touchpad mimics scrolling up and down with the arrow keys.

Note: Double-tapping only seems to work if the settings are just right. I will report later on just what those settings are.

Robin Harris of ZDNet's Storage Bits blog always has excellent advice on backups and the best ways to do them. Here he is with How to archive on disk drives, which explains why putting something on a hard drive and then powering it off for extended periods of time isn't such a great idea:

The reason: the bits on the hard drive will gradually lose their magnetism, sometimes in as little as 12 months. Disks automatically rewrite marginal data blocks on the fly.

That doesn't happen when the disk is powered off.

ITWire is one of those mildly cheeky tech news sites that I enjoy reading very much, despite its annoying habit of breaking longer items into as many as six different screens.

Lately its writers have been waging a good-humored battle on why you might want to use Windows Vista or Linux, giving numbered reasons for one position or the other.

Actually, numbered lists are somewhat pandemic in the blog world, and I usually take little note of them. But these from iTWire are worth taking a look at:

• 40 reasons to lose Linux and vote Vista! by Alex Zaharov-Reutt
• 20 reasons to shed the Microsoft yoke and use Linux by Sam Varghese
• 10 reasons to stick with Vista and forget Linux by Alex Zaharov-Reutt
• 5 reasons to upgrade from Windows Vista to Linux by David M Williams

When we left off, Fedora was taking quite a long time to install over the network on this Power Mac G4/466.

I returned to the office to find the Fedora 9 install finished. I rebooted.

No X.

After Debian Etch installed with no problems whatsoever, I don't know why I expected Fedora 9 to do the same thing, but I did.

I didn't think to save the xorg.conf from Debian to help me configure X in Fedora. I thought, if they can't autoconfigure X on a Power Macintosh G4, which likely represents a huge chunk of all PowerPC machines that might try to run Fedora, what can they autoconfigure?

I found some of the information I needed online, and I was able to get X to work with 1024 x768 resolution.

I didn't take the extra time needed to flesh out the meager xorg.conf to get my monitor's native 1600x1200 resolution.

Why?

Fedora 9 was really choking on the G4/466. Yes, the 466 stands for 466 MHz. And yes, I only have 128 MB of RAM.

But Debian ran so comparatively well, I couldn't see staying with Fedora longer than the half-hour it took me to finally get X to work.

In Fedora, everything was slow. Getting menus to open, apps to start.

Before I reinstalled Debian Etch — which I did do, by the way — I tried to install OpenBSD 4.3.

The OpenBSD instructions for MacPPC were less than stellar. It's pretty much just a rambling supplement to the i386 instructions in the FAQ.

I went through the install, but when it came time to boot from the hard drive, that didn't happen.

Otherwise, it was a typical OpenBSD install, whatever that means. I must've neglected to make the proper partition bootable.

I'll have to study both the regular FAQ and the MacPPC instructions and see what I did wrong.

Part of this is a lack of understanding on my part regarding how the Mac boots, i.e. with yaboot and not the i386 bootloaders GRUB and LILO.

I could install OpenBSD on the same drive as Debian, or I could swap in another drive. I don't think I'm confident enough to dual-boot OpenBSD as yet.

Anyhow, I wanted to actually use the box, so I reinstalled Debian Etch, and once again everything works perfectly and acceptably fast. Not super-speedy fast, but remember, this is a 466 MHz/128 MB box. If I could somehow find a couple of memory sticks that would work in this thing, I'm sure I could improve the performance immensely.

And I'm not giving up on the OpenBSD install, either. I don't know how well OpenBSD will run on this box, but I really need to find out. I could always throw NetBSD or FreeBSD on it; both have PowerPC ports.

Huge stopper for PowerPC: I never thought about it before, but I'm sure thinking about it now: There is no Flash player for Linux (or any BSDs, for that matter) on PowerPC.

I tried swfdec, but I have yet to find a single thing it will work with.

Gnash is not in the Etch repositories (although it's probably in Backports), but it is in Lenny. I've never previously gotten it to work, but maybe it's getting better. That's a potential solution for Linux and OpenBSD.

In my shop we work with Flash quite a bit. I'm going to leave this loveliness right now and go to a Mac to create something in Flash. The fact that Flash is resource-heavy, doesn't run on all platforms and is basically a pain in the ass doesn't appear on the radar of most people.

Getting some kind of open-source player, whether it be Gnash or something else, is essential if Flash is going to continue dominating the online video space. Having a way of creating Flash in open source is another thing that we sorely need.

If Sun figured out that we all need the tools to create and run Java, maybe Adobe will eventually feel the same way about Flash. I'm not holding my breath.

Back to the box: So I'm running Debian Etch again, and it's quite a testament to everybody behind Debian that it runs so much better on this hardware than Fedora.

This box is a pretty good candidate for Lenny. I don't know if I'll do an upgrade yet because I'm getting tired of having to constantly download and install so many new packages every few days on my Lenny laptop. If Etch ran better on that Gateway, I'd be using it, but it doesn't.

I have a good feeling about Lenny and this G4, but I'm going to wait until it goes stable, and then maybe wait some more, before doing an upgrade.

Another day, another new kernel and Xorg package for Debian Lenny. I keep hoping that one of these updates on the road to Lenny going Stable is going to magically solve my video problem (those screen artifacts I get with the Gateway Solo 1450).

Again, if things magically change, I'll be squacking about it big time.

I also added the Fvwm window manager. As it installs, Fvwm isn't anywhere near as well-configured as it is in OpenBSD, where Fvwm is the default window manager.

But it is Fvwm, and I've gotten kind of used to it.

Have you noticed that in most distros, things work way better in the default window manager then they ever do in the many other window managers you can add to the system?

I also finally burned a CD of Debian Etch and a Half, which somehow melds Etch's packages with a newer kernel.

I don't know if Etch and a Half will perform better or worse than standard Etch or Lenny in its current state, but I intend to find out.

I'm using Google Chrome again today in XP, and I continue to be impressed by its speed if nothing else.

So far, I'm having a lot of luck with pages being rendered correctly, and did I mention that it's fast?

I like being able to shuffle the tabs around. I never did that before, and in Chrome they just look like they should be shuffled. Then I tried the same thing in Firefox. You can shuffle the tabs there, too. Kinda funny, huh? Something about Chrome made me think about doing something that I could've been doing in Firefox all along.

I won't really be able to evaluate Google Chrome until they release the Linux and Mac ports. I wonder how long it will be.

When I first installed Google Chrome, I imported my Firefox bookmarks. All 900 and something of them.

As far as managing those bookmarks, all I can seemingly do is drag them around. Right-clicking on a given bookmark allows for some options, but it's just not enough to manage hundreds of bookmarks and folders.

I guess it's all part of making the app intuitive, like the Mac, for which nobody ever wants to read any kind of manual or docs at any time.

You can drag bookmarks and bookmark folders from the list of bookmarks to the bookmark bar. And you can shuffle them around. Again, you can probably do this sort of thing in Firefox, too, but I never thought to do so.

I didn't have any complaints about the way Debian Etch performed on my new/old Power Macintosh G4/466. The install went smoothly, the system performed better than I had reason to expect with only 128 MB of RAM, and I can unreservedly recommend Etch to anybody with a box of this pedigree (PowerPC) and vintage (circa 2001).

But since this was my first PowerPC install, I can't leave things where they are without taking a few more distros for a spin.

Right now I'm installing Fedora 9. I've been wanting to try it for awhile, and the fact that it is made for both PowerPC and i386 means it's something I could run on the Mac and my laptop, if I decide to go that way.

Update: I was working the nightshift, and I started the installation about 5 p.m. It was still downloading packages when I left at midnight.

I've had great things to say about the graphical Anaconda installer when I was installing CentOS, but the text-mode version on this Fedora-for-PPC network-install disc could be much better.

I guess if I hadn't done maybe 20 Debian and about as many Ubuntu installs over the past two years, I might look less favorably on the text-based installer for both. But I do look on the Debian installer favorably, as I do the installer for CentOS. So it's not about being overly familiar with Debian and not so much with Fedora.

This text-based installer for Fedora 9 on PowerPC is, for lack of a better word, barbaric.

I'm using the network install ISO, and that meant the packages would come over the Internet. I made the mistake the first time of saying that the packages were on the CD, after which the installer told me that no, there were no packages on the disc. It took me a couple of times to figure out that I had to tell the installer that the packages were indeed coming via network.

Couldn't Anaconda somehow figure out that I'm using the network-install disc?

Once I selected network install, I was prompted for information on my network connection; nothing out of the ordinary there.

But then I had to select a mirror. Unlike the Debian installer, in Fedora you don't get a list from which to choose a mirror. You either have the information written down, or you go to another computer and start digging into the documentation to find a mirror.

I did the latter and finally found a proper mirror.

I'm still not all that experienced in rolling my own partitions in PowerPC, so I let the installer set them up for me. It looks like Fedora's default is to go with logical volume management. This might be a good time for me to get schooled a bit on how to work with LVM.

Once I went forward in the install process and chose the desktop selection of packages, the process began. That was a couple of hours ago, and it looks like it's going to be a long wait for all 920 packages to download and install.

I don't know how well Fedora will perform on the G4 with 128 MB of RAM (I've been meaning to hunt down some more memory ...), and that's where this tale will resume.

I always seem to have trouble with Ubuntu. On the $0 Laptop — the Gateway Solo 1450 — there comes a time in every Ubuntu install when the thing either won't boot or runs so slowly that I have to wipe the thing off the drive and start over.

It could be something particular to this laptop, the hard drive in it, or my constant dual- and triple-booting of Linux and BSD operating systems in a constantly shifting array.

When I use recovery mode to boot Ubuntu 8.04 and see the messages scrolling across the screen, I can see the point where it stalls. Something about ATA 2.01 is pausing for 5 seconds to look for devices. This pause used to be only 5 minutes, but today it appeared to stretch forever.

I had (and have) work to do, so I ctrl-alt-deleted out of there and booted Debian Lenny. I'll take the annoying screen artifacts problem in Lenny any day over not being able to boot at all in Ubuntu.

The Ubuntu problem began after an aborted installation of FreeBSD about a month ago. And even though I wiped that partition right away and have reformatted it a few times, Ubuntu still stalls during the boot sequence.

Now that I sort of, kind of know how to use rsync to backup my /home files, I need to delete the Ubuntu partition and start again. I have a funny feeling that I'll still have a problem. It could be the hard drive. I have an old 30 GB Toshiba drive in here that I bought on eBay, and it's probably not the ideal drive for daily use, it being old and all, but it's what I've got, and I've never had a problem before. ... Except for these Ubuntu problems (7.04 and 7.10 didn't fare too well in this respect; I thought that 8.04 would be OK, but that hasn't turned out to be the case).

Anyway, gotta get back to work, so I'll be auditioning distros soon enough to see what's going to work for me. I'm almost at the point of throwing CentOS on the box. I'm worried that I'll be missing packages and codecs that I need, and I'm nowhere near good enough with RPM repositories and packages to figure it all out. That's what I count on the people from Debian and Ubuntu for ...

I've really enjoyed using Ubuntu this go 'round. Everything has worked better than ever ... except for this not being able to boot. That's quite an "except," don't you think?

Update: The Ubuntu partition does boot; it just takes a long time.

I've been meaning to try LogMeInFree, a program/service that allows you to access a Windows (and now Mac) box from a remote location through a Web browser.

It's like GoToMyPC but with the promise of being free. Forever, presumably.

I tried it out, and it works.

So far, I've controlled one XP machine via Firefox on a Ubuntu-equipped laptop.

Seamless is the word I'm looking for.

What prompted me to finally try LogMeIn Free? Two recent ZDNet blog posts:

The highlight of the Daily News moving from its huge, windowless box to a smaller office has been all the old equipment that has been flying out the doors.

Aside from the Power Macintosh G4/466 (not a 450, as I initially thought), I plucked a couple of trashed-looking old laptops from the junk heap.

Neither had batteries or power supplies. Luckily, the power brick for my Gateway worked in one, a Compaq Armada 1125.

The damn thing actually booted ... to a Windows 95 desktop.

It doesn't look like a great candidate for Linux or BSD, unless you're taking pure command line or the barest X desktop possible.

The specs:

Pentium 100 MHz processor
24 MB RAM (the machine's maximum)
640x480, 16-bit color display
3.5-inch floppy drive
1.2 GB hard drive
PCMCIA telephone modem card
Windows 95

What's missing? Enough memory to do much of anything with, a CD drive, easy networking (although I might have an Ethernet card that works).

So what should I do with this thing? Clean it up a bit and see what the intelligent masses on eBay give me for it? Hey, the damn thing boots, which is more than I could say for a lot of gear I come across.

Note: The photo above isn't this exact Compaq Armada 1125, just a representative image plucked from the Web.

Update: Since all I've got is a floppy drive, I pulled my Linux-on-floppy discs and loaded up the two-floppy Basic Linux.

The Compaq booted, and after the second floppy loaded, I was even able to use X.

Among the applications I used during my test were vi, another text editor called wp (with pico keybindings) and the Links text-only browser, all in an xterm window.

I don't yet have networking up, but I'm working on it.

More Basic Linux:

Besides yesterday's overly long post on the new Google Chrome browser, I also cranked out a shorter, better-written piece for the Daily News. Here's an excerpt:

On the surface, Google Chrome seems simple, almost Spartan in appearance. There aren't tons of menus. That might frustrate "advanced" users but will be welcomed by average Web surfers.

I don't think Chrome will stay simple for long. Expect Google to continue tweaking it in the months ahead, adding features and functionality at a rapid pace. Can Firefox and Internet Explorer compete with Chrome? It won't be easy, but it is possible. It's hard to keep up with the world's king of search. And, like it or not, Google's bold move with Chrome might represent the beginning of a new era on the desktop and the Internet.

Or it could be just another Web browser.

I know that quite a few of those who use Debian on the desktop favor the Testing distribution, which is currently Lenny, over Stable (now Etch).

The primary reason is that once the Stable version of Debian is set, so are the packages that come with it. Even though they get security fixes and bug patches, you'll be stuck with, let's say OpenOffice 2.0 when others are "enjoying" version 2.4.

I know there are always backports, and if there are a few packages that you absolutely, positively must have in the latest versions, that's a great way to stay with Debian Stable.

But for most, Testing is the way to get something that runs pretty well for the most part along with much newer packages.

On this Gateway Solo 1450 laptop, however, the reason I'm running Debian Lenny (aka Testing) is that I picked up a great deal of hardware compatibility by moving from Etch to Lenny.

Quite a few things have broken along the way. At various times, there have been problems with the Epiphany and Iceweasel browsers (in both instances the problems were really with NetworkManager, but they affected use of the Web browsers just the same), plus issues with sound and video.

I'm still having the problem with video. The screen just doesn't seem to want to refresh all the way during the computing session, and as a result there can be a lot of "artifacts" on the screen at various times.

Running xrefresh in a terminal window usually helps, as does running the Update Manager. But should I really have to worry about this kind of thing, especially when the problem doesn't occur on this hardware with every other Linux distribution I've used on it?

The problem is that I really like Debian.

From the way the community that puts it together is structured, to the commitment to non-i386 ports such as PowerPC and Sparc, to the way the desktop distribution is structured and the speed with which it runs, there is a whole lot that keeps me using Debian. The many thousands of packages are another great thing.

And I actually like GNOME.

Getting Debian to achieve what, for this laptop anyway, is the hardware holy grail, meaning suspend/resume, would tie it with Ubuntu for most effective OS on this platform.

That and this annoying video refresh problem.

Yes, I've tweaked just about everything I can think of in xorg.conf. No, none of that has had any affect whatsoever.

I'd love to magically change horizontal sync and vertical refresh and have the problem go away. And if anybody has a clue as to how I should do this, I'm more than ready to receive your wisdom.

And since Ubuntu still runs exceedingly well on this laptop, better than Debian even though the former is based on the latter, I should probably just migrate to it and stop whining about Debian so much.

Since Debian so far runs exceedingly well on my newest acquisition, a circa-1999 Power Macintosh G4/450, I can let Debian go on the Gateway and still have a place to run it.

With two hard-drive bays and less worries about ACPI, the G4 is a great test bed for anything and everything that runs on a PowerPC chip, and that frees up the Gateway to be more of a working computer with a lot less dual- and triple booting.

While on that subject, I'm moving toward chainloading everything I can out of the main GRUB in the master boot record. That way I don't have to rely on an update to one distro in a dual- or triple-boot system to rewrite /boot/grub/menu.lst in every installation on the drive.

I'm currently chainloading to an experimental Debian install from the GRUB controlled by Ubuntu on this laptop, and it's working out quite well. Chainloading -- it's what's for dinner.

As is customary when a brand-new product makes its debut with great fanfare, developing an opinion about what it means now and for the future is at turns extremely easy and all but impossible.

And yes, I do have an article about my first impressions of the Google Chrome browser coming out in Wednesday's Daily News. I also have a long entry here outlining my initial opinion of the app and what it means to Google, the industry and the world in general.

On one level, it's just a Web browser. We've got tons of them.

But on another level, it's a Web browser. The most-used computer application on the planet, and Google's now making one.

Chrome has the potential to give Google unprecendented control over how its advertising and Web-based applications are delivered and used.

Is that a good thing?

Applications and the data that go with them are moving to the cloud — to servers out there somewhere, managed by giants like Amazon, IBM and Google — and the way we interact with those applications is, by and large, the Web browser.

It doesn't have to be that way. An e-mail client isn't a Web browser. Nor is an IM client. But we use Web browsers a lot. If there was ever a Swiss Army knife of applications, it went from being an office suite to a Web browser long ago.

Under it all, Google Chrome is meant to be a pipeline for data, a framework for applications, and most importantly, a way to reimagine the browser with all-new, non-legacy code.

So why does it look so much like ... a Web browser?

Can't answer that one. But if the under-the-hood innovation promised by Google — features to quash malware and phishing, an emphasis on multiple threads, closer integration with other Web-fed applications like Google Docs and Gears ... if all this proves compelling in some way, and if pure speed comes along with it, Google can win this war.

And if rumors are true, part of that war might involve the purchase of Mozilla.

Others, including Matt Asay of Cnet (and Alfresco), think that Mozilla's superior leverage in the community — of users and open-source developers vs. Google's paltry showing in this area means that Mozilla won't fold so easily.

And Mozilla has plenty of ideas of its own about where users interaction with the Internet and the desktop are going.

Notice how I didn't mention Microsoft? I don't see MS out-innovating anyone. It follows, and does so brilliantly and profitably, but it does not lead when it comes to technological innovation.

But as great as Mozilla's community might be, a concerted effort to develop the Chrome browser and any other technology that comes out of it really cannot be matched.

And if in some way Chrome manages to bring in more green for Google, I can see many, many resources, indeed, being heaped upon this latest venture.

The Internet — and the rest of the press, newspapers included — has been buzzing loudly with the initially leaked and now confirmed news of Google's new Web browser.

After the leak, Kara Swisher of the Wall Street Journal reported on the new Google Chrome application — currently available for Windows only but with planned versions for the Mac and Linux to follow.

Chrome doesn't represent the first time Google has taken on the technology world's big dog, which in case you didn't guess is Microsoft.

Here's the Wall Street Journal's main story on Chrome, which includes a bit of helpful analysis about Google and Chrome's place in the computing universe.

This all got out of the proverbial bag when a HUGE comic book leaked on the Web (yes, a comic book; that's how the kids do things these days, I guess). If you want to know a lot about Google Chrome, I suggest you click the link and read this long — 38 pages in all — comic book. Or is it a graphic novel?

Before I get into all of that, let me recount my morning:

I got in, and while Google Chrome was supposed to be available for download, there was no indication of that anywhere on Google. Search revealed that there was a Google Chrome page, but clicking on the link led back to the main Google page.

Then came the announcement of a noon PDT news conference at the Googleplex in Northern California, at which time Google Chrome would again be available for download.

I went to the Google Chrome site about 12:30 p.m. PDT and found an actual link to download the browser.

Have you ever tried to download a "marquee" application on its first day of availability? The most recent was Firefox 3, and Mozilla's effort to set a bogus Guinness World Record, combined with all-around clamor for the new browser, made it impossible to download at first.

The same thing happens with the popular Ubuntu GNU/Linux operating system. Whenever there's a new version of it, good luck on getting it any time during the first week; the servers get hammered.

But with Google Chrome, I clicked the link and had a small installer app in literally 5 seconds. I clicked on that installer icon and had the full Google Chrome browser in about a minute. I do have a fast connection here at the Daily News, but this was still quicker than I'd ever expect.

So if you want to download Google Chrome, I don't suspect you'll have any trouble at all.

And if the title of this post has already left your mind, let me repeat: I'm using Google Chrome right now.

When you run Google Chrome, opening up a new tab tells you that you are running Google Chrome Beta, meaning this isn't considered a "final" release. But Google is notorious for tagging its products as "beta" for extended periods of time. Google Docs is still in beta, and it's been out for over a year. Google Groups finally lost its beta tag, but did carry the beta designation for an awfully long time.

But even though Google Chrome is technically a beta app, it looks polished, albeit very basic, in its first public release. Clicking on the little "wrench" icon near the upper right portion of the Google Chrome window, then clicking on About Google Chrome, yields the following version information:

Google Chrome 0.2.149.27

As is customary during a pre-release, Chrome hasn't yet hit the 1.0 in its version number. That will happen when it leaves beta.

But does any of that matter? So far I've had quite a bit of luck with Google Chrome when it comes to rendering Web pages. It works flawlessly with Movable Type. According to the Google comic book (yeah, I'm using a comic book to write this ... lovely, eh?), one of the things that the Google engineers are pushing in Chrome is a faster Javascript rendering engine. And things do, indeed, seem to be appearing faster.

Where does Google Chrome come from?

It is based on the open-source WebKit layout engine, which is what Apple based its Safari browser on. What else uses WebKit? Apple's iPhone and Google's open-source Android mobile-phone operating system do.

And the Epiphany Web browser, which is based on Mozilla's Gecko layout engine, has been planning to move to WebKit for some time. The super-fast Konqueror browser which is an integral part of the KDE desktop environment that's very popular with Linux users, has also adopted WebKit. Not so ironically, WebKit itself is based on the KHTML rendering engine that used to power Konqueror.

One thing that comes out of all this: WebKit and Gecko, Firefox and Konqueror, and now Google Chrome are all open-source projects. That means anybody can see how they're made and add to them if they wish. They can even fork them and create their own project, just so long as proper credit is given and their code is made as freely available as that upon which it's based.

Security: Google claims that the Chrome browser is very secure and by the nature of the way it's put together doesn't allow malicious entities on the Internet the same access to your PC as "traditional" applications.

One way Google does this is in the way Chrome is installed.

"Normal" applications, including the Firefox and Internet Explorer Web browsers, are installed in the C:\Program Files directory, where all of your applications tend to live. Chrome, however, is installed here: C:\Documents and Settings\user\Local Settings\Application Data\Google\Chrome\Application\

If I'm understanding this correctly (and it's entirely possible that I do not), being installed under "Documents and Settings" instead of "Program Files" is one way of keeping Chrome, or anything interfacing with it, in its own area, if you will. Chrome is intent on keeping outside entities away from your PC's most critical files.

Cool Google Chrome features for geeks:

Debian Lenny — on this laptop anyway — just got the 2.6.26 kernel, along with X packages that, once again, have a chance of making Debian Lenny's ghosting problem go away when I run X.

Basically, I've concluded that for some reason the screen isn't properly refreshing during the course of the X session. (The hardware in question is a Gateway Solo 1450 with the Intel i810 video driver).

Lately I've been able to run xrefresh in a terminal and clear things up, but since this problem only exhibits itself in Debian Lenny and not in Debian Etch, Ubuntu, or anything else I've ever used on this laptop, I'd like to believe that said problem will somehow be miraculously solved, like the other half-dozen or more potential deal-breaking issues I've had with Debian Lenny over the course of its use on this PC.

What I'm saying is that after a whole lot of fiddling with xorg.conf, I can't seem to fix the problem myself, and I hope the fine Debian Developers out there will do it for me.

Once again, everything's looking good in my preliminary testing, but that optimism seems to fade rather quickly as things turn to their usual crap as the X session wears on.

As always, I'll report back when I know more.

Later: My X refresh problems continue as before.