Home » Daynotes Home » Week of 23 May 2005

Photograph of Robert Bruce Thompson
Daynotes Journal

Week of 23 May 2005

Latest Update: Thursday, 26 May 2005 08:40 -0400
Free Speech Online - Blue Ribbon Campaign

Paying for this Place
Visit Barbara's Journal Page

Monday, 23 May 2005
[Daynotes Forums]    [Last Week]   [Mon]  [Tue]  [Wed]  [Thu]  [Fri]  [Sat]  [Sun]   [Next Week]    [HardwareGuys Forums]

08:48 - As much as I like Xandros, it has one drawback. It's a commercial product. For an individual, that's no problem. Xandros licensing allows an individual to install one licensed copy on as many systems as he personally owns. If the price of even one copy is a problem, there's Xandros 3 Open Circulation Edition, which is free for personal use, and does about 98% of what the pay-for versions do. OCE lacks pay-for software like Crossover Office and StarOffice, and has limited support for CD writing and no DVD writing. That last is no real problem, because it's easy enough to install K3b, which is what I use for writing CDs and DVDs anyway.

But Xandros OCE is not licensed for commercial use, even by schools and other non-profits, and that presents a problem. I'd like to have a top-notch desktop Linux distro that I could install on systems I donate to non-profits, but for licensing reasons Xandros OCE isn't it. The non-profits would have to buy one copy of Xandros for each system. So I decided to take a look at some of the other desktop Linux distros that are free for commercial use. I spent some time this weekend looking at several of the most popular free desktop distros. Ubuntu, Mandriva, and MEPIS are the top three on DistroWatch, KNOPPIX is in the top ten, and Kubuntu (the KDE version of Ubuntu) is in the top 25.

All five are excellent distros, but as it turns out none of them are even remotely close to Xandros as a Windows desktop replacement OS. I could use any of them as my primary desktop OS, but the people I want a desktop Linux OS for do not have my skills. For them, any of those five other distros would present many showstoppers, including things as simple as connecting to a Windows network or enabling a printer. MEPIS, for example, didn't find the embedded network adapter in an Intel D865 motherboard, and Ubuntu failed to detect my USB scanner, which every other distro I'd tried did without a problem. Obviously, I could setup networking and shares and scanners for the non-profits, but that'd mean I'd end up the de facto support person, and that's more involvement than I really want to have.

Thinking about it, the solution is obvious enough. Xandros 3 Standard Edition sells for only $50 per system, which even most non-profits can afford.

The other reason I was playing around with distros yesterday is that my UMAX 3450 USB scanner isn't working properly under Xandros 3. It scans well enough, but the output is covered in random colored blotches. I ran into this problem several months ago under Xandros 2.0/2.5, and solved it by installing more recent SANE libraries. Unfortunately, I already have the most recent libraries officially supported by Xandros, and installing more recent software from unsupported sources sometimes borks Xandros.

I know the scanner is okay, because I was able to do a clean scan under MEPIS. Unfortunately, MEPIS has problems of its own, not least that it doesn't recognize the embedded Intel Gigabit Ethernet adapter. I suppose I could scan under MEPIS, burn the files to DVD, and transfer them to my primary system that way, but that seems kludgey. I'd like to get XSANE working properly under Xandros 3.

Brian Bilbrey suggested yesterday that I just install the scanner on ripper, which runs Windows XP. I suppose I could try that, but my experiences with scanners under Windows have been even worse than the problems I've had with Linux. At least under Linux when I get the scanner working it keeps working. Under Windows, I just don't have a lot of luck with scanners. They may work, kind of, for a while, but then they stop working for no apparent reason. And that's been true for me with half a dozen or more different brands and models of scanners, going back to Windows 3.1 days.

I suppose I could install Xandros 2.0/2.5 on another system. I know how to make the scanner work under Xandros 2. Or perhaps I'll just upgrade a test-bed Xandros 3 system from an unsupported repository to the most recent SANE/XSANE versions and see what happens.

Actually, the sad thing is that I'd be happy just to buy a new scanner, if I had any assurance that it'd work as it was supposed to. That'd be cheaper than wasting hours trying to make this one work. Unfortunately, there's no scanner that I can buy with any real certainty that it'll work any better than the UMAX 3450. Support for that scanner under SANE is listed as "complete", and has been for a long time now.


Tuesday, 24 May 2005
[Daynotes Forums]    [Last Week]   [Mon]  [Tue]  [Wed]  [Thu]  [Fri]  [Sat]  [Sun]   [Next Week]    [HardwareGuys Forums]

10:57 - Someone commented to me the other day that she wasn't concerned about malware on her Windows system at home because the computer "wasn't doing anything strange." It struck me then that the average user's understanding of malware is dangerously obsolete. Years ago, malware was written mostly by teenage punks, whose only real goal was to vandalize systems by corrupting files, formatting hard drives, or displaying stupid banners.

Those days are long gone. Much malware nowadays is the result of an unholy alliance between spammers and organized crime. Their goal is to make malware pay, and the last thing in the world they want to do is make it obvious to you that your system is infected. They want to keep using your system, unbeknownst to you, as a spam relay or bot. They want to log your keystrokes, record your passwords, and steal your identity. What they don't want is for you to figure out what's going on.

Some recent malware even defends your system against being infected by other malware. After all, once someone owns your system, he doesn't want someone else's malware trying to take it away from him. It's ironic to think about the invisible Darwinian battle going on right now on millions of PCs, with only the fittest malware surviving. It's more than ironic; it's frightening.

And what's truly frightening is that nearly all computer owners are more or less clueless about what's happening, and even those who have a dim understanding are pretty much powerless to do anything effective to stop it. Give Microsoft their due. They've belatedly realized the problem, and thrown huge resources into the battle against malware. Unfortunately, what they're doing is all for naught because they refuse to acknowledge the real problem, which is that Outlook, Windows, and Internet Explorer, what I call OWIE, are fundamentally broken from a security standpoint. Short of completely rewriting their operating system and applications from the ground up, there's ultimately not much Microsoft can do to fix the problem.

Microsoft touts SP2 and autoupdate, not because they're truly effective, but because they're all Microsoft has to offer. A Windows XP system with SP2 applied and autoupdate enabled is more secure only in the sense that it is protected against older exploits. But many post-SP2 exploits exist, and locking the front door doesn't help much if you leave the back door standing wide open. The fundamental design flaws of Windows remain, and all the patches in the world won't change that fact.

Here's the simple truth. Every time I look at a Windows system owned by a regular person, regardless of how up-to-date the patches are, that system has been infected by at least one malware program, and usually several. Every time I look at a Linux system owned by a regular person, even those that have never been patched, that system is free of malware. You can argue the whys and wherefores all you want, but that simple fact remains: Windows=Infected and Linux=Clean. A regular person who runs Windows is playing with fire, period.

My readers are a technically-savvy lot, of course. Most of them are capable of securing Windows adequately on their own systems, although I wouldn't be surprised if even some of my readers have a malware app or two lurking on their hard drives. But even technically-savvy people can't secure Windows sufficiently well to allow it to be used safely by non-technical users without regular supervision and intervention.

My solution to setting up a safe system for family and friends is to install Xandros Linux. The OCE version is free for personal use, and is more than sufficient for about 99% of what most people want to do most of the time. For those who absolutely need Windows, for example for running games, my solution is to setup the system to dual-boot Xandros and Windows and to warn the user to run Windows only when absolutely, positively necessary. Ideally, I'll disable the network adapter under Windows, which is about the only way to make Windows truly secure. And even then I have my doubts.


Wednesday, 25 May 2005
[Daynotes Forums]    [Last Week]   [Mon]  [Tue]  [Wed]  [Thu]  [Fri]  [Sat]  [Sun]   [Next Week]    [HardwareGuys Forums]

11:52 - I've been doing some more experimenting with DVD rips, or, more accurately, with burning DVD rips to DVD+R discs. In the past, I'd simply burned the DVD+R discs at 8X in the Plextor PX-708A burner, using Verbatim 8X DVD+R discs. That never worked properly. On a four-episode DVD, the first episode would generally play back properly, the second had some video and audio artifacts, the third was extremely jerky in video and audio, and the fourth wouldn't play at all.

Someone suggested burning the discs at a slower rate. I tried that once on my previous main system, setting the burn rate to 2X or 4X, but K3b 0.11.15 on that system simply ignored my speed selection and burned at 8X anyway, with predictable results. I have K3b 0.11.23 installed on my new main system, which also has a PX-708A DVD burner, so last night I decided to try again. I set K3b for 2X and started it burning to a Verbatim 8X disc. K3b started up, and told me it was burning at 2.5X, which I thought was an interesting number. The disc completed burning without errors, and I popped it into the DVD player. Barbara and I watched the first two episodes last night, and there were no video or audio problems in either of them. We'll try watching episodes 3 and 4, but I expect they'll be fine as well. So apparently burning at a lower rate does indeed solve the problem.

Here's what I don't understand: when I use the DVD drive to backup data, either to DVD+RW or DVD+R, I write at the maximum rate supported by the drive. (The Verbatim 8X discs also work reliably at 12X and 16X, just as an old spindle I have of 24X Taiyo Yuden CD-R discs works fine at 40X and 48X.) When I backup data, I tell K3b to verify after writing, and the verify operation always returns "all files seem binary equal". So, if data can be written flawlessly at high speed, why does video need to be written at 2X or 4X to be reliably readable?

Barbara has a four-day holiday weekend off from work, so I'm going to take the time off as well. I'll probably do a bit of work, but most of the time we'll spend lying around relaxing. So there's not likely to be much new posted here until next Tuesday.


Thursday, 26 May 2005
[Daynotes Forums]    [Last Week]   [Mon]  [Tue]  [Wed]  [Thu]  [Fri]  [Sat]  [Sun]   [Next Week]    [HardwareGuys Forums]

08:40 - I'm trying to figure out if I'm an A-list blogger or a B-list blogger, not that it really matters much. I came across an article on slashdot yesterday that had a link to lists of the top A-, B-, and C-list bloggers. I recognized several of their A-listers, a few of their B-listers, and a couple of their C-listers, but that was about it.

That got me thinking, so I did a search to learn what qualified a blogger for those categories. I came up with this article, which categorizes blogs by daily page read counts and total reader attention per day per blog. I was surprised to see that only about 2,000 blogs generate 1,000 or more page reads per day. According to that article, 1,000 page reads/day defines the low end of the B-listers, and 15,000/day the bottom of the A-listers. On that basis, I'm solidly in the middle of the B-lister range, which I suppose means I'm in the top 1,000 or so in page reads/day.

But according to the other metric, I must be an A-lister. The article categorizes B-listers as receiving 30 to 100 hours of total reader attention per day, and A-listers 100 to 400 hours. My sites receive roughly 2,000 visits/day on average, so if one assumes that the average visitor spends no more than three minutes/day reading my sites, that still takes me to the bottom of the A-list at 100 hours/day total. Figuring by page count makes the issue even more obvious. In terms of average page reads per day, the average visitor could spend only a minute or so on each page before I exceeded the 100 hour/day total.

Not that any of this is really meaningful. Whenever I see a list of top blogs, I always look for Jerry Pournelle's journal page. It's never listed, despite the fact that Jerry generates between three and four times as much traffic as I do, both in terms of visits and page reads. Jerry often posts very long entries and numerous reader mail messages. With his daily visitor count and page count, he surely exceeds the 400 hours/day which that article lists as being the top of the A-list. Jerry's also solidly A-list in terms of page counts. And yet his journal is never listed. Hmmm.


Friday, 27 May 2005
[Daynotes Forums]    [Last Week]   [Mon]  [Tue]  [Wed]  [Thu]  [Fri]  [Sat]  [Sun]   [Next Week]    [HardwareGuys Forums]

00:00 -


Saturday, 28 May 2005
[Daynotes Forums]    [Last Week]   [Mon]  [Tue]  [Wed]  [Thu]  [Fri]  [Sat]  [Sun]   [Next Week]    [HardwareGuys Forums]

00:00 -


Sunday, 29 May 2005
[Daynotes Forums]    [Last Week]   [Mon]  [Tue]  [Wed]  [Thu]  [Fri]  [Sat]  [Sun]   [Next Week]    [HardwareGuys Forums]

00:00 -


Copyright © 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005 by Robert Bruce Thompson. All Rights Reserved.