« March 2004 | Main | May 2004 »
April 7, 2004
Near miss migrating to Linux
Whew! Close one. After talking about it in theory, I almost made a wholesale switch to Linux last night, unplanned.
Most major paradigm changes in my computing style have come as a result of catastrophies. Losing a hard drive, accidentally deleting a partition, someone hacking into my machine, etc. As long as I'm in the process of reconstructing my digital life from scratch, I might as well do it the way I want.
While I was cleaning my office and reminiscing yesterday, I had Partition Magic resizing the NTFS partition to make more room for Gentoo Linux. PM was almost done when it died with an error, and Windows XP was hosed. Blue screen of death every time it booted. 21GB worth of my life's work is on that NTFS partition.
After I picked myself up off the floor, I tried to rescue the NTFS partition using Trinux. Trinux is very cool, and I've used it in the past when my HDD crashed and I had to rescue as much as I could off an NTFS partition. But this time, no go.
As I was formulating how I would blow away the NTFS partition and install Linux instead of XP, I made a last-ditch effort to see if Windows could heal itself. Surprisingly, the Repair feature on the install CD worked well enough to let me do a chkdsk /f, and a mere 45 minutes later XP finished cleaning up the mess and I could boot. But the partition still wasn't resized, and I needed more room for Gentoo.
I immediately tried doing the same thing again, with similar results (but I knew how to fix it this time). 90 minutes later, with a re-re-stored NTFS partition, I decided to try the ntfsresize utility that comes with the open-source NTFSprogs package. Not surprisingly, it actually worked.
Another hour to move files around and expand the ReiserFS partition, and I've got a whole 11GB to grow into with Gentoo. Start some "emerges" and it's off to bed.
Posted by pete at 2:36 AM | Comments (2)
April 6, 2004
Cleaning up old memories
I got a microwave from my wife for my birthday, to put in my office. So today I had to clean the stack of papers off my refrigerator to make room for the new appliance. As long as I was cleaning, I cleaned off my desk, too.
As usually happens, at least when I clean, I find all sorts of interesting relics from the past. Each one reminds me of how good I used to have it, or how much easier things are now, or about a rough time-period that's now long-forgotten. And I drop it in the trash.
After the desk, I decided to tackle a cleaning project long overdue. I inherited several file cabinets full of papers and files when I started three years ago, and never took the time to go through them. They've been moved several times, from an old building to the new one, from one office to another.
The first cabinet was full of papers, minutes, contracts and diagrams going back to 1995, when UEN was looking for a replacement for NSFnet and Internet was still the domain of university researchers. There were early diagrams of SuperNet (used to be the Colorado equivalent of UEN) just after it was acquired by Qwest. There was the RFP for a NSFnet replacement, and the Sprint response that was ultimately accepted, including pricing. The documents included names of some people I have only heard of, and many who I still work with every day. I took my time and kept some of the most interesting papers.
I only made it through one drawer, and I've got about 5 times more papers to go through. I'm excited to see what turns up.
Posted by pete at 11:29 PM
April 5, 2004
Can I make a difference against Microsoft?
(Continued from Saturday's post)
During the last several weeks, my perspective on the Microsoft vs non-Microsoft issue has changed. I'm now thinking seriously about how I could reduce or eliminate my dependency on Microsoft. In particular, how I can work around the mandates of my work-place to use proprietary Microsoft file formats and proprietary Microsoft versions of open standards (I don't get at all why someone writes in cross-platform Java using the proprietary extensions of the Microsoft JVM that's only available on Windows with Internet Explorer).
In early March Eric Raymond released the Halloween X document that showed how Microsoft was secretly funding SCO (contrary to their public statements) to discredit Open Source. In case this evidence wasn't concrete, the person who wrote the email in the document validated it a few weeks later. Mike Anderer talked about how the SCO lawsuit against FOSS (GPL, IBM, Red Hat, AutoZone, etc, etc, etc) was the beginning of what Microsoft hopes will be a long line of lawsuits backed by standards-derived patents to discredit FOSS, or at least limit its spread.
It's one thing when a company succeeds through good, solid business practices, well-marketed products, "good" products (whatever the market thinks that is), fair competition, etc. Their success, at least in theory, contributes to the success of their customers and the economy.
But a company that tries to halt progress, be the spoilsport loser, by scaring people away from potentially superior products, by dampening the success of those products by taking advantage of the legal systems, and by locking customers in and preventing them from making their own choices, that's a company I just don't want to have anything to do with.
It's not like Microsoft will go out of business, but it's becoming more of a matter of principle to me that I don't support a company that does what Microsoft is doing. Maybe in some small way it might make FOSS more successful while it makes Microsoft either change their approach to FOSS or go away, but it'll certainly make me feel better about what I'm doing.
Groklaw writes today about a similar dilemma that Sun is going through.
Posted by pete at 2:30 PM
April 3, 2004
Does one vote against Microsoft count?
For the past year or so, I've been thinking about whether it's worth the effort for me to ditch Microsoft, or at least reduce my Microsoft dependency except where it's mandated. I tried Mac, but it seemed to introduce as many barriers as it solved problems, at least for me (I haven't blogged about that experience yet).
The two big questions for me, besides the amount of effort it would take to use non-Microsoft alternatives: is it worth the fight, is the amount of effort it'll take to not use Microsoft really worth the inconvenience; and, does it really make a difference whether I use Microsoft products or not? I frame these in the context of assuming that the longer I wait, the easier it'll get to switch away from Microsoft products.
The area of greatest friction for trying to avoid Microsoft is at work. Though I work for a university, where I would hope there is a great deal of personal freedom, there isn't on critical applications. The official office file format (I've heard) is proprietary Microsoft Office format, and the official calendaring system is GroupWise (though a weak version of GroupWise is available via Web and a slightly less-weak beta version of GroupWise is available for Linux and OS X). There are a number of critical specialty applications that work only with Microsoft's Java or Internet Explorer.
So even though for almost everything I do, I could use Linux or OS X with non-Microsoft applications, for these few mandated requirements I have to use Windows and other Microsoft software or find work-arounds (and the burden's on me to make sure they work seamlessly with other Microsoft users). I guess that's the way Microsoft wants it, that I pay for their software whether I really need it or not, just because other people "need" me to have it.
These are the issues I've thought about and discussed and tried to sort through the last year. I'm sure they are very similar to issues many others are dealing with. And I have to say that when it was just these issues, I decided it wasn't worth the fight. But things seem to be very different now. More on that on Monday.
Posted by pete at 10:37 AM
April 2, 2004
Amazon Top 100 predicts 2004 Presidential Race?
If the books people purchase are any indication of what the 2004 Presidential Race is going to be like, looks like President Bush is in for a fight.
The Amazon Top 100 Bestsellers list includes six anti-Bush books, and only one pro-Bush book. In the top 25 alone, there are 4 anti-Bush, one at the top of the list, and only 1 pro-Bush (and I doubt that Karen Hughes carries the same weight that Richard Clarke, Paul O'Neil, John Dean do).
Average rating of pro-Bush books: 3.5 stars
Average rating of anti-Bush books: 4.17 stars
I have to hand it to the editors/publishers of the anti-Bush books. Not only are there three times as many to choose from as pro-Bush, but the titles are far more intriguing than Ten Minutes from Normal.
Anti-Bush
1. Against All Enemies: Inside America's War on Terror
13. Lies and the Lying Liars Who Tell Them: A Fair and Balanced Look at the Right
21. The Price of Loyalty: George W. Bush, the White House, and the Education of Paul O'Neill
23. Worse Than Watergate: The Secret Presidency of George W. Bush
35. American Dynasty: Aristocracy, Fortune, and the Politics of Deceit in the House of Bush
37. House of Bush, House of Saud: The Secret Relationship Between the World's Two Most Powerful Dynasties
Pro-Bush
12. Ten Minutes from Normal
Posted by pete at 11:44 AM
April 1, 2004
Learning to teach by thinking like a learner
Jim writes about learning Perl programming from me as a tutor.
One of the most frustrating experiences for me is teaching, especially something I know very well. It's not that I don't enjoy it, or that I can't be patient as people try to learn what I'm teaching. In fact, I enjoy all of that.
What's frustrating is thinking like a learner. Most things I know well, it's been so long since I learned them, I have forgotten how I learned them. I've forgotten all of the details of how I first got interested, how I learned the fundamentals, how I internalized the important information, how I improved as I gained experience.
Jim likes to focus initially on how he should think about something, instead of starting with how he should do it. I found this invaluable in understanding what I should be teaching, and it made it easier to go back to my formative experiences with Perl. Focusing on the student's cognitive processes, instead of their demonstrative, lets me see what they "get" and don't "get" and focus my attention on the "don't get" part. Helping Jim learn how to think about Perl helped me formulate a Perl cirriculum (that's probably doing it more justice than it deserves) that I think would work for other first-time Perl programmers.
I'll have an opportunity to test this new approach out in the coming weeks. I've had a lot of requests from others at UEN to teach Perl (I'm calling it PEON: Perl for Engineering and Operating Networks) so they can be more effective at what they do.
Posted by pete at 10:06 PM | Comments (1)