Jim's Random Notes - April 2001

Well I settled on a new short-term project.  Now that the Kylix book is done, I can play with the Linux system to learn a little more about the OS.  I'm going to see just how hard it is to get a Linux system up and running from scratch.  Here's what I'm going to try:

From my Windows box, download a bootable Linux diskette image.  There's probably somewhere I can get such a thing along with a program that will create a bootable diskette.  I'll then pop that into my P200 and power up.  That should boot Linux on the machine.  Can I reformat the hard drive from this point, or do I need more stuff first?  I want a minimal system:  just enough so that I can download enough stuff to recompile the kernel.  Once I get the kernel optimized for my system, I'll download and install whatever programs I need to create the system that I want.   This will include a mail program (sendmail, or similar program), an FTP server and client, Perl, Apache web server, etc.  As much as possible, I want to work from the lowest level possible--the source.  I want to see how easy it is to do this.

I expect that experiment to take some time.  I can probably get a working system with a customized kernel running in a weekend.  But finding, downloading, and installing everything else from sources I expect to take quite a while.  Obviously, I'll have to download an executable bash shell, gcc, make, and a few other utilities.  From there, I should be able to rebuild all of my tools.

It'll be an interesting experiment.

Debra and I joined her Master Gardener group today for a trip to Bamberger Ranch, in Blanco County south of Austin.  If you've ever been to a typical Central or West Texas ranch, this one will surprise you.  Where you'd normally see cedar breaks, large stands of mesquite trees, lots of prickly pear cactus, and acres of bare soil or very little vegetation, on this 5,500 acres you see native grasses, running streams, and many different kinds of native hardwood trees.  This is what the Hill Country looked like before ranchers' practices of over grazing and fire suppression destroyed the native grasses.  When David Bamberger bought the Ranch in 1969, the State agricultural agent estimated that it could support one head of cattle for every 41 acres.  Now it will support one head every 18 acres.  In 1969 they cataloged 48 different species of birds on the Ranch.  Today they have over 150 species, including many that are considered endangered.  These bird species were not artificially reintroduced to the area.

There's so much to see and learn on the Ranch that I could easily spend a week there and not get bored.  They have a Chiroptorium (a man-made cave built to house a million bats), and the largest known herd of Scimitar-horned Oryx in existence anywhere today.  This species of Oryx, native to Africa, is extinct in the wild.  The Ranch, in connection with the Species Survival Plan Program, has set aside 640 acres for the Oryx with the intention of reintroducing it into its native habitat.  Sadly, environmental conditions in its native African habitat are so bad that the species' survival there is doubtful. 

I am beginning to realize that we humans can have a very large impact on our environment.  But it doesn't have to be that way.  Like anything else, the quick and easy route is fine in the short term, but makes for long term problems.  By studying and working with natural processes, we can have self-sustaining farms and ranches that are much more productive in the long term. 

I've been reading the book Galileo's Daughter by Dava Sobel; the author of Longitude, which I mentioned on March 9.  The book is billed as "a historical memoir of science, faith, and love."  The book examines the life and work of of Galileo Galilei, weaving into the story many of the letters that his daughter sent to him from her convent throughout her life.  The book is almost a biography of Galileo, although very little is said about his childhood and there are some gaps in areas of his adult life.  More interesting, though, it its discussion of Galileo's discoveries and publications.  The most interesting (and disturbing) to me is the idea of science, especially in Italy, being subject  to the whim of the Church and the Holy Office of the Inquisition.  A person could be tortured and killed for holding beliefs or publishing information (true or not) contrary to the Church's official position.  Galileo's publication of his Dialogue, a thinly-veiled approval of Copernicus'heliocentric world view, almost cost him his life.  A 17th century scientist or philosopher had to tread very lightly indeed.  Galileo's Daughter is a well-researched and informative book written in a very engaging style.  Highly recommended.

The Internet Engineering Task Force (IETF)  Network Working Group this week released RFC2821 (Simple Mail Transport Protocol) and RFC2822 (Internet Message Format).  These RFCs obsolete the 20-year-old RFC821 and RFC822, respectively.  RFC2821, according to the Abstract, "is a self-contained specification of the basic protocol for the Internet electronic mail transport. It consolidates, updates and clarifies, but doesn't add new or change existing functionality of the following."  It goes on to list the RFCs that it incorporates.  Similarly, RFC2822 "specifies a syntax for text messages that are sent between computer users, within the framework of 'electronic mail' messages. This standard supersedes the one specified in Request For Comments (RFC) 822, 'Standard for the Format of ARPA Internet Text Messages', updating it to reflect current practice and incorporating incremental changes that were specified in other RFCs."

I'm happy to see consolidation of the multiple SMTP RFCs into a single document.  One of the most difficult parts of understanding current standards is simply locating all of the relevant documents and understanding how they fit together.  Consolidation of this sort makes that task much easier.

Some people will believe anything.  Some time ago I ran across www.fixedearth.com, which bills itself as "the non-moving Earth and anti-evolution web page."  The web site is one big advertisement for their book, "The Earth is not Moving."  My knee jerk reaction was to dismiss these guys as crackpots and move on, but I try to keep an open mind.  So I studied the information on the site to see if they have some real evidence.  After reading the entire site, I can safely say that they are crackpots.  All they have is a rather selective interpretation of the Bible, and a lot of anti-science rhetoric that bashes evolution theory, and by extension all science.  There is no objective evidence either supporting a geocentric view, or refuting the heliocentric view.  Of course, when you "prove" that all science and especially mathematics is the work of Satan and therefore False, you have nothing left on which to base objective evidence.

My friend Jeff Duntemann tells me that this is a rather mild form of fundamentalism.  I guess I don't want to see the more extreme versions.

I broke out the mountain bike today to take a ride with a co worker on the Barton Creek Greenbelt.  It's a 15 minute ride from the office to the trail head.  The trail itself follows Barton Creek for about 7 miles from Zilker Park upstream to the bottom of a very large hill, the top of which is in a subdivision that's about a 15 minute ride from the office.  Conveniently located, that office where I work.

The Greenbelt is an enjoyable ride, but not very difficult.  The terrain is mostly flat, and there aren't too many places with rocks or roots to make things technical.  The biggest hazard is the many hikers, runners, and people out strolling with their families, including unleashed dogs.  Most of the trail is forested, and there are many squirrels and other wildlife--especially at night.  It was a perfect day for a ride, except for one thing.  There's water in Barton Creek.  And not just a little bit.  There were four places today where I had to pick up the bike and wade through waist-deep water.  Of course I'll be wishing for water in a couple of months, when the creek dries out and it's over 100 degrees outside.

I said that the terrain is mostly flat.  The hill at the end of the trail is called "The Hill of Life."  I'm not sure why it's called that.  I think "Life Sucker Hill" or "The Hill of Death" would be more appropriate.  The hill is something like 1/2 mile long and over 300 feet high.  It's as wide as a 2-lane road, but very rocky with washouts and erosion barriers that make riding up it something of a technical challenge.  This hill would be difficult enough if it was paved, but the technical work combined with the steep grade makes it nearly impossible to climb without stopping at least once; either because you miss a step or because you're sucking wind.  The descent sure is fun, although you have to be careful in some areas.

I really must get a cheap digital camera to carry around.  I'm not about to take the Digital ELPH through the creek, but I'd love to get some pictures before everything dries up.

Not to worry.  Five minutes with my old Advanced COBOL book, and I decided that I really don't want to get involved with COBOL again.  Not that I have anything against the language, per se, but I'd rather work with a more modern tool.  I've come to like strong typing, objects, and local variables.  I know that the COBOL 85 standard and the new proposed standard add some of that (mostly support for nested programs, including local variables), but the language just isn't suited for the kind of work that interests me these days.  If I was writing business applications again, then maybe.  I've yet to find a language more suited for the banking industry than COBOL, but COBOL doesn't handle GUI applications, parsing, or web transactions very well.  I think I can find something more interesting and challenging than writing a compiler for a 40-year-old computer language that I hope never to use again.

Where is it said that idle hands are the Devil's tools?  Having finished the Kylix book, I find my hands somewhat idle.  I've been dabbling with some minor projects off and on during the book, and will likely finish one or two of them and nuke the rest.  (On a side note, if you decide to cancel a project, back it up onto a CD or something and then delete it from your hard drive.  Otherwise, you'll be tempted to fiddle with it at inappropriate times.  Not that I'm speaking from experience or anything.)  At any rate, I ran across the Tiny COBOL project and actually started thinking about joining that effort.  I'm a member of that apparently rare breed of programmer who actually knows COBOL and C.  Interesting idea, but I'm not sure I'm ready to immerse myself into COBOL again.

Perhaps I'll take a break from programming for a while.  At least at home (I do have a programming job).  I've been wondering what it takes to write publishable fiction.  Maybe I'll take a stab at that.  And there's always stuff to do in the yard and around the house--mulching under the trees, repainting a room, creating a new garden bed, etc.  I could even get more involved in amateur radio.  I've had my license for 10 years, but haven't actually used a radio for 5.  There are plenty of things to do.  I just need to figure out what I want to do next.

My friend Jeff Duntemann once told me that being bored with life is like starving to death in a supermarket.  It's possible, but you have to work at it. 

Another thing that drives me nuts is those electric hand dryers in some (all too many) restrooms.  When I was a kid, these things scared me.  Push the button and a very loud monster starts blowing in your face.  As an adult, I've seen more than one kid come screaming out of the bathroom at McDonald's, terrified of the hand dryer.

"To serve you better," says the label on one model, "we have installed pollution-free electric hand dryers."  Better for who?  I don't know one person who thinks these devices constitute better service.  They only serve the management by eliminating the need for paper towels.  I especially like the instructions on one model:

1.  Push button.
2.  Rub hands under warm air.
3.  Stops automatically.

And some wit added:

4.  Dry hands on pants.

Which is what I and most other people end up doing.  Why don't theatre owners just remove the silly hand dryers and install a sign:  "To make things more convenient for us, we request that you dry your hands on your pants."  

"Air fresheners."  Yuk!  As if mold, rotting garbage, and sewers weren't bad enough.  People install these silly devices that supposedly freshen the air, but in reality do a very poor job of masking the real odor.  A pine scent in the bathroom, for example, makes the room smell like somebody pooped a pine tree.  The original odor was bad, true, but adding a cloying sweet smell on top of it is worse.  Whatever is in the air stinkifier is certainly more persistent than most other odors, and succeeds very well in keeping bad odors around.  Do you want your moldy garbage to smell like moldy cheap perfume?  Put one of those stick-on air fresheners near the garbage pail.

No amount of air freshener will get rid of a persistent odor like that.  The problem is lack of ventilation.  Well, that and laziness:  people would rather spend money masking an odor than taking the trash out every night.  Ventilation is a problem because people want their houses to be energy efficient.  They seal up every crack so there is very little air exchange with the outside.  It's energy efficient, for sure, but this practice leads to lingering odors and (according to some) elevated indoor radon levels.  I think I'd rather burn a few extra kilowatts. 

I picked this one up on slashdot.  A woman in California sued a company called Kozmo.com in small claims court for sending commercial unsolicited email.  You can read her story, or view the entire slashdot thread.  Nobody got rich:  she was awarded $50 plus $27.50 in court costs.  With Kozmo gone the way of the dodo, it's likely she'll never see any of it.  She and the anti-spam crowd are calling it a moral victory.

Isn't it funny that the most vocal spam busters are also the loudest voices screaming for anonymity protection?  It's OK to clog message boards and email boxes with anonymous unsolicited crap, but any commercial exercise of that privilege should be against the law?  It sounds to me like these people want the right to say whatever their hearts desire with impunity, but don't want the inconvenience of having to filter through other peoples' garbage.  The California anti-spam laws apply specifically to unsolicited commercial emails, which I take to mean that it's OK to send non-commercial spam.  So the anti-spammers' real enemy is business, not spam?  I know that isn't the case, but a superficial read of the situation could lead one to that conclusion.

Hormel, the maker of SPAM Luncheon Meat, has gotten into the act by publishing a note that doesn't really endorse the use of the term "spam" to describe unsolicited commercial email, but does say:  "We do not object to use of this slang term to describe UCE, although we do object to the use of our product image in association with that term."

I agree that spam is a nuisance, and I'd sure like not having to worry about it.  In my case, email filters make it much less of a nuisance, and the few messages that get past my filters are easy enough to delete.  It's certainly no more of a problem for me than the junk that the postman delivers every day.  But it's apparently a huge problem on the Internet:  clogging mail servers and wasting bandwidth.  Legislation isn't the way to combat it, though.  Getting governments involved will undoubtedly lead to more problems than solutions.

How to combat spam?  spam.abuse.net is the definitive anti-spam site.  Read and take action.  But please leave government out of it.

Written in 1994, The Unix Haters Handbook looks at the darker side of Unix:  security problems, sendmail problems, file system problems, The X Windows Disaster, and many other parts of Unix that give people no end of problems.  The authors are all Unix users who have come to loathe the operating system (a term they believe doesn't apply to Unix).  The book contains much commentary and is interspersed with mail messages that were sent to the UNIX-HATERS mailing list.

My understanding is that many of the problems discussed in the book have been fixed since the book was published, but others still exist in all flavors of Unix (including Linux and the BSDs).  After reading the book, I wonder if I really want anything to do with Linux or any other Unix knock-off.  But then I ask myself if Windows is any better.  I honestly don't know the answer to either question.

But the book is an entertaining read, and mostly informative.  Sure, the authors go on rants now and then but they do have valid points.  People working on Linux, BSDs, and any other flavor of Unix would be well served to ensure that the problems outlined in this book are solved.  I borrowed the copy that I read, but will probably end up buying the book.

Reader Robert Esguerra sent me a note today about searching the faqs.org site (see April 14) using Google's advanced search.  Works like a charm.  He sent me a URL that will return all of the hits, not stopping at 100 like the faqs.org site search does.  I was able to refine the Google search to find just those RFCs that have "SMTP" in the title.  I've been using Google for well over a year and never played with the advanced search page.  It's good stuff, and I'll make good use of it in the future. My drive to work takes me about 10 miles down Loop 1 (called the MoPac Expressway because it follows the Missouri Pacific railroad) every morning.  As you would expect, there often are police cars parked under the overpasses, ostensibly looking for speeders.  But there's a catch:  the cars are empty!  I can't remember the last time I actually saw a police officer sitting in one of those cars.  But they do have a calming effect on traffic.  Every day we'll be clipping along at 75 MPH until the guy in front can see the empty police car sitting under the bridge.  Sure, he probably knows as well as I do that there's nobody in the car, but he can't be sure.  Presto, instant speed control.  It's an interesting lesson in how peoples' expectations control their behavior.

Debra and I watched Fantasia 2000 on DVD this weekend.  Although it didn't have the impact on me that the original Fantasia had back in 1973, I still consider it an excellent picture--probably the best I've seen in years.  I only wish that I could see it on IMAX, as it was originally intended.

What happens when you give a yo-yo to a flock of flamingos?  Just that one piece was worth the price of the movie rental.

If you're looking for Internet RFCs or standards documents, the best place I know of to find them is http://www.faqs.org/rfcs/.  They have almost every RFC dating back to 1969 (some were paper documents that apparently have been lost), and good information about the RFC process.  A separate Standards section indexes documents (mostly RFCs) that are the currently accepted Standards.  The site also includes FYI (For Your Information) and BCP (Best Current Practices) indexes that are invaluable if you're trying to write something that conforms to current Standards and Practices.

There are some problems with the RFC process, though.  First, they're still publishing using stone age tools.  I realize that ASCII text is "universal," but it wouldn't hurt to publish the documents in PDF or some other format that's easier to print and to read.  Not many of us are still stuck with 80-column dot-matrix printers.  It'd be nice to have RFC documents that use real fonts and formatting.  Heck, a printer-friendly HTML format would be fantastic.

Another problem is that you can't update an RFC.  Once it's published, it's set in stone.  The only way to "revise" it is to issue a new RFC that obsoletes the old one.  This is good and bad.  Good because there remains a history of changes so that we can go back and see where we've been.  Bad because it's sometimes hard to find the most current RFC for any given topic.  I would think that it would be more effective to assign a single RFC number and then append revision numbers or dates.

The other problem with the site is that it's not indexed very well.  Here again, full text search just doesn't do it.  If you search for SMTP, you'll get over 100 hits (the system stops when it exceeds 100) on pages that mention SMTP, but you'll miss some of the important documents.  And although the individual document descriptions indicate whether they obsolete or are obsoleted by other documents, they don't always mention other relevant RFCs.  The description for RFC821 (Simple Mail Transfer Protocol), for example, does not mention RFC1869 (SMTP Service Extensions).

All that aside, the faqs.org site is still very useful.  If you know of a better RFC index, I'm very interested.

So Microsoft can poke fun at itself.  Clippy (the much-hated Office Paperclip) is history, as  Office XP team makes abundantly clear in its Microsoft Clippy web page. Take a look at the movies and listen to the song "It Looks Like You're Writing a Letter."  It was worth a laugh.

Now, whether this actually helps sell Office XP is anybody's guess.  Clippy never really bothered me because I learned how to turn him off about 3 days after I got Office 2000, so his demise makes zero difference to me.  And Office 2000 does more than I want any office package to do, so I'm in no hurry to upgrade to Office XP.

The New York Post ran an article yesterday describing how many radio stations have stopped webcasting because the American Federation of TV and Radio Artists is demanding additional commercial talent fees.  Stations that have stopped web streaming transmissions include all 1,170 Clear Channel stations.  Who is to blame is anybody's guess.  The station owners are doubtless accusing the AFTRA for wanting more money for the same amount of work, and the AFTRA is accusing the station owners of not sharing profits.  This is starting to sound like the conflict between authors and publishers over Internet publishing rights.  My friend Jeff Duntemann wrote a very good piece about that in his web diary last month.

This isn't the only problem faced by Internet radio broadcasts.  Most radio advertisement is local in nature, and advertisers (who ultimately pay for commercial radio broadcasts) aren't willing to pay higher rates for advertisements that are broadcast to a wider audience (as required by most advertising contracts) if that audience isn't interested in their products.  Discount Furniture in Atlanta probably doesn't care that people in Seattle can hear their advertisements.

I went to the bookstore yesterday looking for a book that describes the Internet mail protocols (SMTP, etc.).  I know that I can get the RFCs online, but they make for pretty dry reading and I can't take the computer to bed with me.  Sure, I could print them, but I find paperback books much more convenient and easier to read than single-sided sheets in a 3-ring binder.  I was looking for something (perhaps two books) that contains not only the RFCs, but also English descriptions of how the different protocols work together, possible interpretations of and ambiguities in the RFCs, and other useful information.  The SMTP RFC, for example, was written in 1982 and I was hoping to find a book that has practical advice from somebody who has actually implemented an SMTP server.

So why do I want this information?  I'm seriously considering writing a simple SMTP server for Windows and Linux.  And I'm doing that simply because the available SMTP servers are either broken (Microsoft's Windows 2000 SMTP service is brain-damaged, and sendmail is, well, sendmail) or overkill for what I want (sendmail again).  I want a simple and reliable SMTP transfer agent that can send and receive mail.  I don't need fancy routing rules, the ability to send mail over nonstandard networks, or to support every conceivable mail client.

At any rate, I didn't find what I was looking for, which doesn't mean I won't keep trying.  I did find one book that discusses many different Internet protocols, mail among them.  It had very good brief descriptions of the different mail protocols, but little low-level technical information.  I did, however, run across the book sendmail (the bat book) which, at 1,046 tightly-packed pages, pretty much validates my comments about the unnecessary complexity of the sendmail program for most applications, and the need for a simple and reliable replacement.

Plastic.com posted an article the other day about President Bush's appointment of Scott Evertz to head the White House's AIDS policy.  Even though Evertz is likely well qualified for the job, his appointment has angered both right-wing conservatives (because he is homosexual) and left-wing liberals (because he's a Republican).  I'll leave those politics to somebody else.  What really got me was one poster's statement:

Let's cut to the chase and figure out if AIDS is curable or preventable (other than by modifying behavior), and get on with it. At this rate, I'll have prostate cancer before they cure AIDS.

Huh?  I'd say that the most effective means of promoting public health is by modifying behavior.  Getting doctors to wash their hands before operating on patients did more to reduce post-surgical mortality than anything else.  Washing your own hands regularly, especially after using the bathroom and before eating, is a very effective way to reduce your chances of contracting many communicable diseases.  Brushing and flossing your teeth regularly is the most effective way to prevent tooth loss and other oral health problems.  All behavior modification.

We know where most people get AIDS:  sexual intercourse.  The disease is almost 100% preventable, and we could virtually eliminate it in a single generation if people would take very simple, effective, inexpensive, and readily available preventive measures.  Is using a condom really so terrible?

Certainly, a cure would be nice.  But we'd be way ahead of the disease if we spent those billions of research dollars educating the public and advocating responsible behavior.  Instead, we have spent billions of dollars, two decades of research, and millions of lives on research that has yet to bring us a cure.

I guess I'm behind the times.  Back in January,  David Horowitz published an article entitled Ten Reasons Why Reparations for Blacks is a Bad Idea for Blacks - and Racist Too.  Last month, he tried to get the article published as an advertisement in university newspapers around the country.  Most wouldn't run the ad, which I think is cowardly, but beside the point.  Of the three that did run the ad, two later printed apologies--another cowardly act.  At one university, students who disagreed with the contents of the ad went around collecting and destroying every copy of the newspaper that they could find.  I guess those students don't believe in open debate.

Horowitz has made a huge mistake and damaged his credibility by withholding payment ($1,007.50) for the ad that ran in Princeton University's student newspaper.  The reason?  Because on the same day the paper ran an editorial that attempts to discredit the contents of the advertisement.  The paper ran the ad as agreed--Horowitz should pay up.

Although I don't agree with all of Horowitz's statements, I do agree that reparations for slavery is a Bad Idea for everybody, especially if paid for with tax dollars.  There's absolutely no logical reason for taxpayers, most of whose ancestors arrived after the Civil War, to pay these claims.  Yes, Blacks were enslaved.  Yes, it was an unspeakable thing to do.  But there's no way that we can punish the guilty parties (the small minority of slave-holding citizens), and there's no way to compensate (or even identify) the victims, their children, grand children, or even great grand children.  Racial reparations are hardly fair, as not all Whites supported slavery and not all Blacks were slaves.

I think this is just a money-grabbing scheme cooked up by lawyers who will ultimately benefit far more than the clients they claim to represent.  Don't believe me?  What small group ended up with the largest single piece of the 15.3 billion dollar Texas tobacco settlement?  How about 25% to the lawyers? 

Today was the annual Ride for the Roses to benefit the Lance Armstrong Foundation.  Over 8,000 riders participated in rides of 10, 25, and 62 miles.  The Ride is the culmination of a full weekend of events that includes a bicycle and health expo, the Downtown Criterium bicycle race on Saturday night, and some other events.  All proceeds from the events benefit the LAF.

62 miles is about 10 miles more than I was really ready to ride today, but they didn't have a 50 mile course.  About 50 miles into the ride I got a good laugh at Farmer Bob and his home made protest sign.  Believe me, when you're bucking a 15 MPH headwind at the end of a long ride, something like this can keep you amused for miles.   "Take your spandex/spoke parade to China or France and stay."  When I rode by, Farmer Bob had a few other people with him.  The irony of it got me laughing:  here we are, 8,000 people helping to raise money for cancer research, and there sits Farmer Bob (who could use some exercise) with his buddies, smoking cigarettes and cussing us because they can't go ripping down the back country roads at 80 MPH on a Sunday morning.  What a hoot.  How would you like to see him in biking shorts and jersey?

In all fairness, I should mention that this guy's small group and maybe one or two people who passed us in trucks are the only people I saw who were unhappy with our ride.  There were many people camped out in their front yards cheering us on and supporting the ride.  The presence of the few protesters just made me better appreciate the supporters.

Somebody sent me this link today to a page that describes some problems the Debian team is having with the release schedule for their new Linux distribution.  It seems that they're having trouble creating reliable boot floppies. Boot floppies?  My first response was "is this a joke?"  Apparently it isn't.  As the follow-up messages indicate, this is a real issue, and the question is:  "Why?"

A couple of posters spout the silly B.S. that the problem is the "rampant idealism" of the Debian team:  "Debian developers insisting that it be done 'right.'"  "Debian is not driven by marketing-created illusions like 'release deadlines'."  Whereas the lack of a release deadline is fact, I doubt very seriously that it's the cause of the problem.  Much the reverse, I suspect:  there's no release deadline because the developers can't say with any confidence when they'll be able to make the thing work.  And the reason for that is, as one poster put it:

I think a large part of it is that the installer is extremely fragile -- either because it runs in a very unusual environment and performs a lot of low-level tricks to do so, because its code is simply not robust, or both. It seems to break whenever anything else in the base system changes -- and pretty much everything there changes in each release, including this one.

Another poster points out the obvious:  the installer (which makes up a large portion of what goes onto the boot floppies) isn't used and tested on a regular basis like other packages are.  The installer only gets tested when release time comes around.  By then, of course, so many things have changed that the underlying assumptions upon which the installer is built have changed.  The implication is that testing, not development, is the problem.  In his final paragraph, he suggests making the installer simpler in order to "insulate it from a poor testing regime."  As if testing is the problem.

When are programmers going to learn that testing is their job?  A programmer's task isn't finished until he has tested his code in all supported environments.  Volunteer or not, if you accept the task of writing code, then you also incur the responsibility of testing it.  Anybody who thinks he's too good for testing is simply a spoiled brat, and not much of a programmer.  Any marginally bright person can bang out almost-working code all day. 

I'm busy finishing up the book here, sleep deprivation and all.  No diary entries until the book is finished.  Hopefully by Sunday, 'cause I have a bike ride.

We're finishing up the Kylix book (Kylix Power Solutions, with Don Taylor, Jim Mischel, and Tim Gentry, published by the Coriolis Group).  We had originally planned a chapter on interfacing with the KDE and GNOME desktops to provide things like desktop icons, docking apps and such, but that turned out to be a major job worthy of several chapters, if not a book in itself.  Our backup plan was a chapter on providing application help, and it falls to me to write this one.

Stepping up onto soapbox

Like it does with everything else,  the Linux world has about a zillion different ways to display online help.  There's KDE help, HTML help displayed in a web browser, man pages, HyperHelp, and Heaven knows what else.  Perhaps No Help is the most popular.  Also par for the course, this flexibility ends up costing on both ends:  developers have to decide on a help system and ship the viewer with their applications, and users have to install yet another help viewer on their computers.  I fail to see how this is better in any way than Windows which for years had a single standard help system that was guaranteed to be on every machine.  Under Windows, developers are free to supply their own help systems if they like, but most choose to use the built-in WinHelp.  Microsoft complicated matters a few years ago when they released HtmlHelp and "deprecated" WinHelp, but WinHelp is still there and used by many applications.  Users especially like having a single help system interface.  Getting a standard help system for Linux GUI applications is almost as likely as settling on a single desktop standard.  I suspect I'll find a winning lottery ticket in the trash before something like that happens.

Stepping down

Borland did an excellent job supplying Kylix with a flexible help system interface.  The HelpIntfs unit defines a handful of interfaces for developers to derive Viewers for individual help systems.  The online documentation of these interfaces is a bit sketchy, but what the help files don't explain fully, the sample applications (a HyperHelp viewer and a man page viewer) make abundantly clear.  The help viewer interface works behind the scenes of an application, serving up help topics in response to requests from the Help Manager, which is invoked by users' actions.  The application's interface to the Help Manager remains constant regardless of the underlying Help Viewer.  It's a very elegant solution to a thorny problem.  By supplying these interfaces, Borland has made it possible for developers to create applications that can easily switch their help file formats, making it easier to keep up with emerging (ha!) Linux help systems.  

Last month (March 18), I mentioned that I was looking at NEdit as my editor of choice on my Linux system.  I've been working with it for a couple of weeks now, and am mostly happy with its features and functionality.  I don't particularly like some of the key bindings, and I'll probably change them if I decide to keep the editor.  At the moment, that's a pretty big "if."  The editor seems to get confused sometimes and stops responding to Alt key combinations or command keys.  I haven't tracked down the cause yet, but sometimes I'll be switching between applications and all of a sudden I can't operate the editor except with the mouse.  Pressing Alt+F for the File menu results in a "beep" from the computer.  Other key combinations produce similarly bizarre results: control characters in the file, for example.  If I shut down all editor windows and restart the application, the problem goes away.

I haven't taken the time to track down the cause yet, and haven't been to the NEdit web site to report or investigate the problem.  I'm wondering if it's an interaction between Kylix and NEdit.  It wouldn't surprise me.

On the Windows front, I'm happy enough with TextPad that I sent in my $27 registration fee.  It's a wonderful text editor and well worth the price.

The problem of school violence has come to the fore in recent years, with the Columbine shootings in 1999 and the recent incident in Santee, California bringing the issue to everybody's attention.  And everybody has a pet theory as to why the violence is getting worse:  broken homes, global warming, lack of parental involvement, violent computer games, the "bully factor," widespread availability of guns, fluoridated water, or alien abductions.  Who knows the real reason?  Everybody has a pet theory and some reasoning or evidence (real or manufactured) to back it up.

Except the problem isn't getting worse!  According to this article, the Secret Service National Threat Assessment Center began tracking school shooting statistics in 1974.  Since then, there have been something like 100 multiple-victim school shootings in the United States.  Despite the prevailing perception, though, school violence (including gun violence) is decreasing.  A Justice Department report (which I have been unable to find) released last fall shows that school violence has dropped 30 percent since 1992.  Arrests for juvenile homicide have fallen by 56 percent since 1993.  In 1999, the year of the Columbine shootings, violent deaths at school dropped 40 percent from the previous year.

The Justice Department's 1999 Annual Report on School Safety shows that school violence has been on a steady decline for at least a decade.  Homicides at school remain extremely rare events, but the number of multiple victim homicide events has increased: 2 in 1992-93, 0 in 1993-94, 1 in 1994-95, 4 each in 1995-96 and in 1996-97, and 5 in 1997-98.  Those 5 incidents claimed 14 lives.  Page 4 of the report (page 12 in the PDF file) reports that "Students in school today are less likely to be victimized than in previous years."

The perception is very different than the reality.  Yes, violence in school is a problem.  One schoolyard homicide is too many, and we should take steps to prevent any type of crime (violent or non) being perpetrated by or on our children.  But the problem is not nearly as acute as the media would have you believe.  We only think it is because our reporting systems are so much better today than they were 30 years ago. 

The Zilker Botanical Gardens in Austin held its annual Spring Garden Fest this weekend.  This is the primary fund raising event for the Garden Center, where we raise money to pay the Center employees, maintain the building that garden clubs use for their meetings, and fund education programs.  The Garden proper is owned by the City's Parks and Recreation Department (PARD), and the people who maintain the Gardens are PARD employees.  Debra is vendor co-chairman this year:  she and her counterpart got vendors to pay for booth space, and handled the hundreds of details involved in getting the vendors into and out of the park, and keeping them happy while they're there.

The Garden Fest is run almost exclusively by volunteers.  There are only a handful of paid PARD employees who maintain the gardens--everybody else who worked over the weekend is a volunteer.  There were over 400 of us selling tickets, selling drinks (we had vendors for the food), manning the information booth, providing transportation in golf carts, providing information, and generally helping out.  It's heartening to see so many people give their time to help out the Gardens.

Too many people think that charitable giving consists of giving money to a worthy cause.  Although it's true that we can use money, many organizations need time.  If you want to help out, volunteer your time to a cause that interests you.  Not only do you get the satisfaction of "making a difference," but you also might learn something.  I've learned a lot from working with people at the Gardens.  If you're interested in building, volunteer with your local Habitat for Humanity or similar group.  You'll learn more than you'd ever need about building a house.  There are hundreds of organizations that would greatly appreciate whatever time you can give them.  And it's a heck of a lot more interesting (and healthier, too) than sitting in front of the TV all weekend.