April 2002 Archives

This is a few months old, but it's still a revolutionary idea in transportation. In January GM previewed technology  it's working on towards hydrogen fuel cell cars.  What was most interesting was the expression of that idea in a technology GM calls AUTOnomy. AUTOnomy divides the vehicle into two parts: the body, or upper part of the car that we sit in, and the platform, which sits underneath the body.  The platform is the revolutionary part: it's designed to be produced in mass, last perhaps 20 years, and allow different kinds of vehicle bodies to sit on top of the platform.

GM calls the platform the skateboard, and it looks like one.  Think of a skateboard as long and as wide as a normal car, with four tires at the corners, but no more than 6 inches high.  The skateboard contains the fuel cells, the drive train the brakes - everything that makes the car go.  A regular car has mechanical linkages - the brake pedal pushes down on something that in turn connects to something that squeezes your brakes - but in GM's concept, it's all electronic, or drive-by-wire. The body in turn "docks" with the skateboard, making an electronic connection between the controls you use to drive the car (brakes, steering, etc) and the drive-by-wire system in the skateboard.  You press the brake pedal, which sends a message to the skateboard to apply the brakes a certain amount.

Right now GM has a "non-operating" prototype, but it will be fascinating to see if this takes off.  Cars have changed a great deal over the past 30 years, but how all the parts fit together hasn't changed nearly as much.

Scott Johnson's Marketing Software When You Are a Small Company.  is a splendid little essay on marketing, and it's applicable to more than marketing software.

(Picked up from [Scripting News]).

"If you spend any time at all writing on the web in text boxes and you use IE, then your gonna love ieSpell. It will spell check any text box for you, like the ones i use in MovableType. Very cool and worth the download." [...useless miscellany]

In return for her gracious notes on Gator, I wanted to post this one for masu so that she can download it May 1st, when the account will hopefully be unfrozen. And Eric, I couldn't find an alternate download site, either.  :-

Managing.com. What impact will networked economy have on the day to day challenges of managing? As big business takes up the entrepreneurial challenge of new technologies and new enterprises face up to the fundamentals of business, managers everywhere are doing things differently. Now, all companies will become web companies and learn how to communicate with customers through digital interfaces.Managing.com combines the energy and opportunity of starting up, with the practical challenges of leading people and business in times of turbulent change. A mix of tactical insight, personal effectiveness and inspiration-this is a handbook for wired managers and entrepreneurial companies everywhere. [O'Reilly Safari]

How To Run Successful Projects III: The Silver Bullet. Many project managers-especially in software-go their entire professional lives in ignorance of the factors behind the success or failure of their projects. Permanently in a state of agitation and worry, they just can’t explain why some projects work out and others don’t.It doesn’t have to be like this. There is a method underlying all successful projects, and if you follow this method your project is guaranteed to succeed. We have called this method Structured Project Management. The cornerstone of Structured Project Management is the ’Ten Steps’-the first five steps are to do with planning your project and the other five with implementing the plan and achieving the goal. How to Run Successful Projects III-The Silver Bullet builds on the success of the first and second editions and reminds us all, in the post dot com era, just how important good project management practices are

Here's a pointer to '97 article on making use of complexity in the workplace.  I'm not sure I believe it all, but it's worthwhile thinking material.  For that matter, I think almost any theories you come up with about how people work together can be best regarded as heurustics

From Paul Plsek with hyperlinks to the full text on his web site:

Some emerging principles of complexity

1. View your system through the lens of complexity (rather than the metaphor of a machine or a military organization).
2. Build a good enough vision (rather than trying to plan out every little detail).
3. When life is far from certain, lead from the edge, with clockware and swarmware in tandem (that is, balance data and intuition, planning and acting, safety and risk, giving due honor to each).
4. Tune your place to the "edge of chaos" by fostering the "right" level or degree of: information flow, diversity and difference, connection among agents, power differential, and anxiety (instead of controlling information, forcing agreement, dealing separately with contentious groups, working systematically down all the layers of the heirarchy in sequence, and seeking comfort).
5. Uncover and work paradox and tension (rather than shying away from them as if they were unnatural).
6. Go for multiple actions at the fringes, let direction arise (rather than believing that you must be "sure" before you proceed with anything).
7. Create strategy and new sources of value through generative relations (you can never tell what will happen when agents come together).
8. Listen to the shadow system (that is, realize that informal relationships, gossip, rumor, and hallway conversations contribute significantly to agents' mental models and subsequent actions).
9. Grow complex systems by chunking (that is, allow complex systems to emerge out of the links among simple systems that work well and are capable of operating independently).
10. Nice, forgiving, tough, and clear guys finish first (that is, balance cooperation and competition via the tit-for-tat strategy).
11. Build a space-a community-to convey, concentrate, create and learn together (rather than always being a lone ranger).

- goto
Posted by marpie1@home.com on 6/5/01; 5:11:58 AM from the Complexity (CAS) dept.

You're not doing your job as a writer if you don't tick people off.  So I must be doing something right: Paul Beard took umbrage at my assertion that the iMac my wife sometimes uses is as slow as the P133 in my basement.  He compared my reasoning to "claiming a brand of car is slow because you can't find a good radio station."

I may have been a little careless in extrapolating from my small sample to connecting it to the story about iMac 2's being slow; after all, the iMac 2 uses the latest in Apple Technology.  The Mac in our house is at least two-three years off the pace. 

Still, I stand by the comparison.   I did actually go to three computers in house, stopwatch in hand, and try browsing the same pages from each browser.  I browsed each site once and then quit the browser to make sure all the cacheable pieces were downloaded, restarted the browser, and then timed the next page load for each of 5 sites. I tried IE and NS 6.2 on each platform.  And the numbers were still consistent: The iMac was no faster than my 1996 vintage P133 running Windows 98. 

Does that mean that the Mac is slower than the PC at everything?  Of course not.  The iMac can edit digital video from my DV camera; even new PCs still have trouble with that.  And there are certainly other areas where the Mac is superior. But the original story about the iMac 2 claimed it was slow at surfing the web, which is probably the primary application for most home machines, so that's what I looked at.

Paul also claimed I was a long time Mac basher/Windows worshiper.  That's possible, but I come by it honestly.  My first job was with Xerox working on the Star.  We scoffed at Windows 1.0 and 2.0. We also scoffed at the original 1984 Mac with no hard drive and a tiny screen, but by the time I left Xerox in 1989, the Mac was clearly ahead, and the Xerox Star was on the way out.  The first machine I bought with my money was a Mac SE. 

I didn't touch a PC until I went to CICNet in 1990, and when I did, I had a revelation: the PC that I had reviled was just a computer, no worse in some respects from many of the machines I used when I was younger.  (CP/M, PDP-8, DEC 10).  It was a tool with strengths and weaknesses, and had to be treated as such.  You use the machine and tools appropriate to the job at hand. Every place I worked from then on used Windows as a base, and so long as I had a UNIX box around, I could live with it. Early on I jumped on Windows NT as being a more reliable base than Win95/98/et. al.

When my wife brought her Mac home a few years ago, I was interested to see how it would work; I hadn't used a Mac regularly for many years.  I was dismayed to see that unlike the technology I was used to using (Windows NT), the Mac still froze up regularly, browsed the web slowly, and in general hadn't progressed as far as I'd hoped. 

I'd love to try living with an OS X box for a while.  Given a choice between the UNIX base of OS X and Windows, I'll take UNIX.  In fact, given a choice I like to have lots of machines. At CNN I had four machines in my office: an iMac, an older PC running Linux, a desktop Win2k box, and a Win2k laptop. For that matter, we have six machines at home right now: the iMac, a P133 Linux server, two P3 500s running Win2k, and an Athlon XP running XP home. I like 'em all.

Spring has arrived in Atlanta.  A  week ago I noted the difficulty of finding sun tea jars, and now they've shown up in our local Kroger.  Glass ones! 

We bought three.

Update from April 2003: a year later, one of the has broke, and the other two both leak from the spout. We also found and tried a plastic one. It doesn't leak - I think - and it doesn't seem to taste any different. The plastic ones do melt if you put them in the dishwasher.

Update from August 2007: I've gone over to the dark side; I've been using the plastic ones for a few years.  They no longer melt.  They also don't break, and are much lighter than the glass ones.  Plus, the glass ones are still hard to find, and the plastic ones show up with cute designs every spring.

Can ZOË make my mail intertwingly?. Via Prof Avery, I just found ZOË, a program that aims to make your email more accessible by indexing it and giving you a variety of ways to get at it.  From the ZOË manifesto:

The goal here is to do for email (starting with your personal mailbox) what Google did for the web... The Google principle: It doesn't matter where information is because I can get to it with a keystroke.

So what is Zoe? Think about it as a sort of librarian, tirelessly, continuously, processing, slicing, indexing, organizing, your messages. The end result is this intertwingled web of information. Messages put in context. Your very own knowledge base accessible at your fingertip. No more "attending to" your messages. The messages organization is done automatically for you so as to not have the need to "manage" your email. Because once information is available at a keystroke, it doesn't matter in which folder you happened to file it two years ago. There is no folder. The information is always there. Accessible when you need it. In context.

Thanks to [0xDECAFBAD: Because a day without caffeine is no day at all.]

I've downloaded it, but not tried it yet; making anything work with Java is a pain.  I have no less than five copies of java.exe on my Win2k system.

(Zoe (forget the weird characters) looks like a clever idea, but all the documentation that exists is a bare FAQ.   Perhaps this will point the way towards someone else who will do the same thing in a more useful way, preferably in open source, and something that might work on a Linux box.

I do love the reference to Ted Nelson: as he says, "Everything is deeply intertwingled."

"I am in consumer heaven! Now I get to relive all my 80's childhood memories through old advertisements. Heck, I can even check out the stuff they were trying to sell my Mom back in the 50's! Oh, they included ads for my Dad too!" [Memepool]

Things you didn't even know you wanted to know about Google: "The second rendition of the mega Google Webmasters FAQ by webmasters for webmasters."

For example: I was told to "google" a date - is this like a nooner? [WebmasterWorld]

Last Friday my wife volunteered me to drive our daughter out to a farm to pick up some fertilized eggs for a school project. We almost ran afoul of Cobb County law enforcement.

Currently Subscribed To, the Sequel

"We've done a bit more tweaking to Jon's currently subscribed to, and think that we've got it pretty much where we want it.

Find the tweaked-up script here .

Here's what we've changed:

• Just a little bit of the table formatting.  Obviously, you are free to tinker with that as much as you please.
  
• More important: we noticed that occasionally our subs() script would get hosed by the addition of a new channel.  All of a sudden, our macro would fail, and we'd get a bizarre error message:
  
  "Can't evaluate the expression because the name 'channeltitle' hasn't been defined."

Turns out that the error was a genuine one -- but in the RSS feed, not in the script.  We haven't kept careful track of RSS version numbers, and don't know the schema well enough to get haughty about it, but apparently some people are putting feeds out there that don't include the channeltitle element.  Solution: just wrap the code that addresses the channeltitle element in an if that makes sure channeltitle is defined." [The Boulder Inquisition]

If you've looked on the right at all, you've noticed that I've had this error message on my site for quite some time now. I'll have to give this one a shot to see if it fixes the problem.

Phil does wonderfully amazing things over at a klog apart... things I wish I could pass off here as my own ideas. First off, he's displaying above each post the categories to which it belongs. I was thinking of doing this using Radio's shortcuts, but Phil has another way of doing it. He kindly shared it with me, although I haven't had time to play with it.

Now, however, I notice that he's gone and added Daypop searches filtered by keyword at the top of each category, including synonyms for the category subject. He's also got the pre-configured Google translation links in the right-hand column. Damn he's good!

All things I want to implement, all things I don't have time to implement right now. All good evolutionary steps for usability in blogs.

WebWord has some fine new articles, including one about The Voyeur Web.

"You might not believe me, but I know what other people are thinking. I can't read minds of course but I can observe people. I know where they are going, what they are searching for, what they email other people, and what they link to. I'm not joking. By the end of this article you will be a mind reader too. You won't know why they are doing what they are doing, but at least you'll be able to watch their behavior. You'll know what they are doing....

I am not saying that I spy on people. Instead, I am saying that several web tools are mature enough that anyone with a little effort can understand the collective web."

John then goes on to highlight sites such as Yahoo Most Emailed Content, Yahoo Buzz Index, Google Zeitgeist, and other tools that I've already mentioned in the past. I would add to the 25 Most E-mailed Articles from the New York Times. It's too bad they don't provide an RSS feed for that service, because I would add it to my aggregator lickety-split. Another missed opportunity, especially since the NYT is already partnering with Radio (yeah, I know it's not open to everyone, but it would be a start).

Sure, I could try to scrape it with Stapler or RssDistiller, but I'm starting to feel like I'll be creating feeds for half of the freaking web at this rate. It would help if at least the big-time publishers realized that they would be providing a valuable service if they created the channels themselves.

Wired News asks Why Do New iMacs Surf So Slowly? which reports that the new iMac 2's are not barnburners in the surfing department.  The article blames OS X, but whether it's hardware or software, I've known for years that iMacs are slow.  I did a little test around our house, and the Win98/PI/133/64MB running IE 5.5 in the basement surfs at about the same speed as the 400mhz/128MB G3 iMac OS 9.1 iMac in the dining room.  (And the P133 was faster at a few sites.)

My favorite quote, especially since we just bought an eMachines box to replace an iMac:

I spent $1,800 on a computer that's slower than the $400 eMachine it replaced," one iMac user wrote in an e-mail.

A few conclusions from my non-scientific tests here at home:

• Mac IE 5 and Netscape 6.2 were about the same speed for more tasks, with a few exceptions.  (Loading my tax return off the Turbotax web site was twice as fast in NS.)
• The P133 was as fast or faster than the Mac on almost all tests.
• My P3/500/256MB running Windows 2000 and IE 6.0 was about twice as fast at rendering pages as the iMac or the P133.

What Does This Word Mean?

"Christopher Schmitt has a cool and useful bookmarklet called Highlight Word Bookmarklet that looks up a definition of a word on a page. It took me less than one minute to grab the bookmarklet and to figure out how to make it work. No more having to type www.dictionary.com to get the definition of a word." [meryl's notes]

Highly recommended!

"John Lennon's 1966 assertion that "The Beatles are bigger than Jesus," may have been true then, but not now. At Google, roughly 170,000 people a month search for "The Beatles," while 850,000 a month search for Jesus. (830,000 search for "Beatles.")

I found these numbers using Google's Adwords program, which uses historical data to estimate how many times a "keyword" is sought in a day, week and month. Joanne Jacobs notes that many people have not used Adwords yet, so here's the crib sheet: a) configure an ad and press "continue" b) insert the name of your favorite pop icon in the "Keyword Matching" field and press "Update Keyword Estimates" button c) note the result and start again....

Ogling Google, or, for short, Oogling, could make a great new bar game. With some imaginative programming, Oogling could put marketing research and polling companies out of business.

The Oogle's absolute data is much more engaging than the zeitgeist "hit list" numbers that make relative comparisons among popular searches. With the raw numbers, we can peel back the consumer's skull and watch the synapses fire.

Yes, Jesus (850,000) beats The Beatles. But even "cheese" gets 750,100. Harry Potter gets 920,000. Clinton gets 1,030,100. Jennifer Lopez gets 1,135,100. Eminem gets 1,235,100. And Britney Spears gets 2,540,200.

Britney is nearly twice as big as God, who gets just 1,295,100 searches. But she's smaller than football, which gets 2,605,200. (All these numbers were researched in early April.)

Beyond celebrity watching, there are brands. Who would have guessed that Adobe gets 2,570,200 searches a month compared to 3,590,300 for Oracle and 6,495,600 for Microsoft. Coke pulls only 216,800 requests....

Proving that many people still don't understand the Internet, the word "Google" itself gets searched for on Google 1,685,200 times a month and www.google.com gets 140,000 searches. That compares to 4,785,400 searches for Yahoo and 350,000 for www.yahoo.com.

If you were BMW (2,305,100) and had just spent $10 million on an adblitz, wouldn't you like to see how much your Oogle jumped relative to Mercedes (1,303,900) or Porsche (927,700) and relative to the Oogle of other premium words like champagne(195,800)?

If you're really ambitious marketeer, you'll want to know how your word measures against the world's hottest. But chances are you aren't within an order of magnitude of sex, which pulls 32,657,300 searches a month." [Pressflex, via Scripting News]

Amercian Scientist: "Terabyte Territory" by Brian Hayes is a splendid article on the history of the magnetic disk drive. 

Hayes notes that if we extrapolate the growth rate of disk drives over the last five years, today's 120 gigabyte drive will be a 120 terabyte drive in 2012.  Hayes asks what you would do with that much space:

One certainty is that you will not fill the void with personal jottings or reading matter. In round numbers, a book is a megabyte. If you read one book a day, every day of your life, for 80 years, your personal library will amount to less than 30 gigabytes, which still leaves you with more than 119 terabytes of empty space. To fill any appreciable fraction of the drive with text, you’ll need to acquire a major research library. The Library of Congress would be a good candidate. It is said to hold 24 million volumes, which would take up a fifth of your disk (or even more if you choose a fancier format than plain text).

Other kinds of information are bulkier than text. A picture, for example, is worth much more than a thousand words; for high-resolution images a round-number allocation might be 10 megabytes each. How many such pictures can a person look at in a lifetime? I can only guess, but 100 images a day certainly ought to be enough for a family album. After 80 years, that collection of snapshots would add up to 30 terabytes.

What about music? MP3 audio files run a megabyte a minute, more or less. At that rate, a lifetime of listening—24 hours a day, 7 days a week for 80 years—would consume 42 terabytes of disk space.

The one kind of content that might possibly overflow a 120-terabyte disk is video. In the format used on DVDs, the data rate is about 2 gigabytes per hour. Thus the 120-terabyte disk will hold some 60,000 hours worth of movies; if you want to watch them all day and all night without a break for popcorn, they will last somewhat less than seven years. (For a full lifetime of video, you’ll have to wait for the petabyte drive.) 

I heard something this morning that sounded odd, but in the years to come will be so familiar that we'll all forget when we first heard it.  It was a story on NPR about trying to reverse the 'sunset' on the new tax rates that will occur in 2011.  The odd part the way the politician and the reporter said 2011 out loud: they said "twenty-eleven."

Up to now, that's not what people say: for 2002 and 2000, we say "two-thousand-and-two" and "two-thousand".  Even for years further out, that's the way I've heard it up until now: "The social security trust fund will run out of money in two-thousand and twenty-three."  But over time the shorter way will win, and chances are people in the future will reverse the way we talk about the decade we're in now: "The World Trade Center attack occurred in twenty-oh-one".

"Pinch me, I'm dreaming. Syndic8 has just released a few directories of RSS feeds. You can now find feeds using Headline Viewer Categories, ODP categories, NewsisFree Categories, and NewsisFree Assignment of ODP categories.

I have been waiting for something like this for quite a while. If you haven't gotten into RSS feeds yet, what are you waiting for. Not only does it save time, but they are so freakin' easy to use. I am sure Jenny is thrilled by this news." [via Library Stuff via Research Buzz]

Not a pinch, although I'm definitely hearing Fred Astaire singing, "Heaven, I'm in heaven..." in my head! Thanks for pointing this out, Steven!

Searchable RSS

From the Shifted Librarian

"You can specify a search query and get search results as RSS.

For instance, say you want to subscribe to this site, but you only want news items that mention Java. You’d subscribe to this URL:

http://www.ranchero.com/xml/rss.xml?q=java" [via a klog apart]

Does this mean what I think it means? Keyword filtering for my hand-picked feeds? Get out! So I can watch for NY Times: Book reviews of titles by my favorite authors? Watch Amazon's list for titles that contain a specific word? Many, many possibilities!

Audit Slams Oracle Pact [The Sacramento Bee, via Slashdot]

I was actually posting this in order to make a glib comment about why SLS is abandoning Oracle, but instead I was completely taken in by The Sacramento Bee's navigation and usability features. While the text size of most of the navigation on the page has too high of a squint factor, check out the IHT-like bar on the left-hand side of the page that lets you enlarge or decrease the text size of the article, toggle the typeface between serif and sans-serif, email the page, print the page, find more stories by the writer, sign up for newsletters or wireless alerts, or make the SacBee your home page. Click on the first icon for "info" and a legend appears above the story.

I'd like to see them add an icon to the toolbar for finding the most popular stories, but overall, very slick and nicely done!

Joe Jenett, the notoriously sleep-deprived web designer, is trying to build up my knowledge of HTML layout. As a current advocate of the 'tables are out' design trend, he directed me to this neat CSS placement tutorial published by A List Apart. On the same site, I found a story on the importance of DOCTYPE syntax, and how it influences modern browsers. For a moment, I thought I had solved my radioScan-with-Mozilla problem. Tss... Back to the drawing board... [s l a m]

I created a 'Books' essay a few days ago, with the intent of putting up a page with current books I'm reading.  Instead, I got sidetracked into trying to come up with a nifty way to generate the booklist of a database.  That's a classic trap technical people fall into, and it's a little embarassing because I'm always telling people to eschew jumping into fancy technical solutions and instead just to solve the problem.

AUGMENTING HUMAN INTELLECT: A Conceptual Framework by Douglas Englebart; a work from 1962.  Englebart invented the mouse and was a pioneer in many other areas of computer science.

I ran across a reference to technography, which is essentially using technical tools to move meetings forward.  This article refers to site called CoWorking, but at the moment I can't actually find anything about technography on that site.

In the context of Manila/Frontier/Radio, one technique used in technography is to create real-time meeting notes by using a laptop and an outlining program. 

I've been using that technique for years to take notes for meetings, usually using MS Word.  My main frustration has been that I can't get Word to print an outline in outline form; it insists on printing a flat form marked up with various sized fonts.  I'll swear that once upon a time I was able to make word do this.

http://technography.userland.com also has useful information.

What's strange is that although I find a number of references to technography on coworking.com via Google, all those pages don't seem to exist; when I go to the coworking site, I don't see any useful references.  Similarly, I've seen a couple of references to technography.com, but I get a 404 when I go to that site.  What's going on?

New blogs by traditional columnists Louise Kehoe (Financial Times) and Peter Lewis (Fortune).

On the site Living Reflections of a Dream, there's several pointers to building multi-column pages without tables: A List Apart and gish.com.

To Hell With Bad Browsers makes the case: simple two column sites don't need tables.  (More complicated multi-column sites still do.)

Community Documentation, Scripts and Tools. It's time. As a famous blogger wrote not too long ago, RTFM Won't Work: Documentation As Narrative.

(If you're doing stuff and you're not in there yet, you probably will be. Pester me and you shall be rewarded).

Russ Lipton: Why RTFM Won't Work.  [Scripting News]  Russ Lipton suggests that documentation for projects such as Radio Userland should be a collaborative effort.

I'm not sure I agree with Russ his point - at least not to the extent that the primary means of documentation is going to be community-based.  He's right on one point - "official" documentation tends to be second rate.  (O'Reilly's "the missing manual" series is a good example.) Official documentation can't admit to any faults in a product, and has to try to make sense over compromises in the product.  And there are always compromises.

One other thing I like: Russ refers to the documentation process as a spiral.  I've always said that my thinking process is a spiral.  My first time around is likely to be (very) wide of the mark, but hopefully encompasses the problem.  If I work with others, talk through the problem, we'll end up going over some of the same ground again, but each time getting closer and closer.

It's coming up that time of year: time to get out the sun tea jar.  For the last few years, our problem has been where to find them.  Like the yellow pollen off the pine trees in Atlanta this time of year, there seem to be a few weeks where they're all over town, but outside of that brief period, you can't find them.

Last September we broke one of glass sun tea jars, and after searching around town in vain, we spent two hours searching before finally buying two of them from Kitchen Etc. After all that, they ended up being plastic jugs.  We bought two, and we've managed to melt one in the dishwasher.  (Yeah, yeah, you're not supposed to put them in the dishwasher.)

Last week we found some very ugly glass jars in our local Kroger, and we bought them.  For future reference, they claim to be Golden Harvest Beverage Tapper Containers, made by Hearthmark, Inc, of Muncie Indiana.

Here's our sun tea recipe:
1 gallon cold water in sun tea jar
9 standard sized bags Lipton tea (the ordinary bags)
3 standard sized bags Lipton orange-and-spice tea bags.
Put bags in jar, cover jar with lid, and put in the sun for 5-6 hours. 

This makes tea that may be a bit strong for some; the normal recommendation is 9 normal-sized tea bags brewed for 3-4 hours.

Fair warning: I like this tea pretty well, but I made some for an office Thanksgiving party, and the VP in charge of my group was heard to say that he hated it.  I was laid off 5 weeks later.  Coincidence?

We also like 'American' ice tea bags from Stash Tea, but they're pretty pricey at $4.25 for two gallon bags.  (I guess that's no worse than Coke.)

Too cool! Now if we only had a librarians.weblogs.com against which we could run this type of query! Imagine it for different genres, too. It sounds like Paul is tracking other interesting Amazon services, as well:

"Amazon is offering an XML Platform for developers so they can integrate Amazon's best-sellers into their websites. If you're an associate, log into the associate site for more info. This is an interesting step toward Amazon becoming a Web Service. They're ahead of the curve, and really understand how the Web works. I can't wait to see how they expand this.

An independent developer has already written a Perl module called Business::Associates that works with the new platform."

as well as:

"DayPop is tracking Amazon Wish Lists. (I have one of those.) The next step is to be able to filter a list like that by my friends—or by groups of domain experts in various subjects. (imagine: this is what the top 50 web designers [as voted by their peers] are wanting to read. or doctors. or indy musicians. or anthropology students. etc.)"

Too much to think about at the moment, but I'd like to spend some time on it someday.

Curling Up with a Good Book Search Engine

"Looking for an interesting read? These book search engines can help you find new titles and authors based on your personal tastes and preferences.

Back in the days when I had free time, I was a voracious reader. To find new books I'd constantly pester librarians and friends for recommendations, and devour publications like the New York Review of Books and Publishers Weekly.

These days, reading time is a scarce commodity, and I'm more selective about the books I choose to read. I'm also generally looking for a book I'm in the mood to read, often in some way related to a book I've previously enjoyed -- something like Neal Stephenson's Snow Crash, or Carolyn Chute's The Beans of Egypt, Maine, for example.

Here are several search engines that offer really good suggestions based on personal, quirky factors." [SearchDay, via LISNews.com]

Chris Sherman highlights AllReaders, Book Forager, the Waterboro Public Library's Book Lists and Bibliographies, Book Finder, and isbn.nu. I'll just add yet another reminder to contact your local public library to talk books and get further recommendations. I still think we should start Emergent Books....

Wow Scott Johnson has been writing some great Radio tutorials. Off to dinner now. I'll read them tomorrow.  [Scripting News]

Here are some resources. http://www.fuzzygroup.com/writing/radiouserland_part01.htm http://www.fuzzygroup.com/writing/radiouserland_part02.htm http://www.fuzzygroup.com/writing/radiouserland_faq.htm http://www.fuzzygroup.com/writing/radiouserland_rcs_stepbystep_01.htm http://www.fuzzygroup.com/writing/radiouserland_radioexposed.htm

Some are still draft but there is a lot of information. See Russ's stuff too. I'm more documentation, he's more big picture, conceptual.

The bandwidth issues that I have seen are tied to Instant Outlining. I generated over 5,000 hits PER DAY on an html page on my own site when I stuck in my IO. It was a php routine that generated OPML of PHP questions.

If you have trouble finding stuff in the posts, here is a tool.

http://www.fuzzygroup.com/radiosearch/ (google boxes preset to search only Radio or only UserLand).

Here's a useful page: SGML for NT describes how to get XML/SGML tools running under Windows NT/2k/XP.  It looks like a very through tutorial; I'll try it later.

The New York Times has an article about a site called TextArc.org  From the site:

A TextArc is a visual represention of a text—the entire text (twice!) on a single page. Some funny combination of an index, concordance, and summary, it uses the viewer's eye to help uncover meaning. A more detailed overview is available.

From the Times article:

The texts, which range from Lewis Carroll's "Alice's Adventures in Wonderland" to Balzac's "Z. Marcas," are too tiny to read around the perimeter. Behind the computer glass, though, Mr. Paley's online software is counting each word and noting its location every time it is used. The oval's black center soon fills with legibly larger versions of every word from the source text. Different stories look different. As a result, Mr. Paley's software effectively turns any prose into concrete poetry in which a word's size and location are as important to its meaning as how it is used.

Once TextArc slices and dices a story, the most frequently used words are the brightest. So in the Carroll work, "Alice" glows at the center. And each word's location in this linguistic constellation is determined by its exact locations in the story text. "Cheshire," for instance, is near the bottom, close to the two middle chapters in which the cat materializes. Roll the cursor over a word, and lines pop up that connect it to all the points in the outer circle where the word is used.

I thought the site would be overloaded - and perhaps it will be - but it was up when I tried it.  The effect is fascinating: I tried Edward Abbott's Flatland, the early 20th century fable about people who live in a land of two dimensions.  Words like flatland, circle, and women are more towards the center; the word sphere is more towards one side, indicating that it's used more in one section of the book.

The web site gives no details of how this is implemented, but the actual application runs in Java on your own machine.

When you launch textarc on a text, it starts drawing it as a concentric spiral around the screen, and I'm guessing that during that time the application is actually downloading the full text of whatever you're looking at.  (The texts are taken from Project Gutenberg) .Looking at my memory usage on my Internet Explorer app, it grew to 76meg when I downloaded Jane Austin's Pride and Prejudice. 

TextArc claims that their application runs quicker in Netscape 6.2, and that may be true, but it crashed in NS 6.2, and didn't crash Internet Explorer.  I'll try Mozilla 0.99 next.

Fixing Your Site with the Right DocType

"You've done all the right things, but your site doesn’t look or work as it should in the latest browsers.

You’ve written valid XHTML and CSS. You’ve used the W3C standard Document Object Model (DOM) to manipulate dynamic page elements. Yet, in browsers designed to support these very standards, your site is failing. A faulty DOCTYPE is likely to blame.

This little article will provide you with DOCTYPEs that work, and explain the practical, real–world effect of these seemingly abstract tags." [A List Apart]

The OnFocus weblog points out the Weblog Bookwatch Top 10:

i thought it would be interesting to see which books are being mentioned most frequently on weblogs. Weblog BookWatch keeps track of weblogs that flow through the recently changed list at weblogs.com and searches for links to Amazon.com. Then it looks at the ISBN in the link's URL, and counts the link as a mention of that book. The most fequently mentioned books show up on the Top 10 list, with references to the weblogs that mentioned them. It's only looking for books right now (not CDs or other products), and only looking for links to Amazon.com.

Thanks to weblogs.com for the great service (and for offering their list in XML). And to Amazon.com for the book info.

This URL lets me get at a Filemaker DB and output it in XML:

http://127.0.0.1:8100/simple/fmpro?-db=books.fp5&-format=-dso_xml&-max=5&-find=&

Returns:

 <?xml version="1.0" encoding="UTF-8" ?> 
- <FMPDSORESULT xmlns="http://www.filemaker.com/fmpdsoresult">
  <ERRORCODE>0</ERRORCODE> 
  <DATABASE>books.fp5</DATABASE> 
  <LAYOUT /> 
- <ROW MODID="0" RECORDID="1">
  <title>Physics of Baseball</title> 
  <isbn>0-155-2558-225</isbn> 
  <comment>This is a good book</comment> 
  </ROW>
- <ROW MODID="0" RECORDID="2">
  <title>XTML: The best</title> 
  <isbn>05-88554-48525-255</isbn> 
  <comment>This is also very good</comment> 
  </ROW>
- <ROW MODID="0" RECORDID="3">
  <title>The Stand</title> 
  <isbn>78-4185-11</isbn> 
  <comment>My books</comment> 
  </ROW>
  </FMPDSORESULT>

So I can use Filemaker to maintain my list of books, query it to get an XML file, and then use XSL or something else to format up a nice looking page than I can drop in this weblog.  I could also drop the XML itself into the Radio WWW directory and have it stream up to the server, for whatever that's worth.

And if there was a real DTD that I want to stay current with, I guess I could also use XSL to convert my Filemaker XML to that form.

It all sounds great, but it means I have some work to do; I'm only familar with most of these technologies by name alone.

I've been sitting here starting to try to build a page for books - books I'm reading, books I want to read, mini-reviews of books I've read.  I've spent the last 30 minutes struggling with the HTML to build a decent looking page, when it struck me: I shouldn't be putting my list in HTML, but in XML.  So: is there a DTD (and perhaps tools) for keeping a home library? I don't really need a full-up DTD of the type that might be useful for a real library. 

I have a feeling this kind of project would be a morass.  Ideally, I'd like to use some simple-to-use graphical database, such as Filemaker, and use it to generate XML.  There's a page about XML database products, and a related page about XML and databases, but it's not clear this would be an easy thing to do.

I'm feeling deja vu about this idea.  A couple of years back I wanted to be able keep a list of projects in a simple database - Filemaker again, because it's very simple to work with - and then periodically publish that database to a webview.  At the time, I was stymied over something stupid: the inability of Filemaker to take a URL and render it on an HTML page as a clickable item.   I spent quite a bit of time messing with Filemaker and even MS Access before giving up on it.  I have a bad feeling that the state of XML tools might be such that the best tool would be something like Emacs.  I love Emacs, but it's not the best tool for managing lots of data.

• system.verbs.builtins.radio.html.commentLink:

  Replaced &p by &p in the definition of commentPageUrl.

• system.verbs.buitins.radio.weblog.render:

  Changed &c by &c if adrblog^.prefs.flCommentLinksEnabled in bodytext for the commentCount script tag.

• In weblogData.root:

  In the archiveLinkImgTag string, added an alt="Permalink" attribute to the img tag.

  In itemPermaLinkImgTag, modified the calling macro like this:
  <%radio.macros.imageref ("images/woodsItemLink.gif",\ "",\ "",\ "",\ "",\ "Permalink")%>.

Here's a spendid little story from The Shifted Librarian about Kate and her 16-year old daughter Clare.  Clare and Kate gave up their land-line for a shared-minutes cell plan.  You can guess what happened next ... they ran over their minutes, and Kate restricted Clare from using her cell phone. 

If you have or know a teenager (or perhaps even were one), being disconnected from friends is the worst punishment of all.

If you open the History bar under IE 6.0 (control-H will get you there), there's a Search button at the top.  If you click it, you can search through the pages in your history.  More to the point, the search goes against the content of the pages, not just the title or URL.  I assume you must also have the page in your local cache. 

Still, this is a pretty powerful tool.  I haven't tried it on earlier versions of IE. I note that Netscape 6.2 on the Mac has a history search, but it only covers page titles and URL components.

I found this in, of all places, a book called The Usborne Guide to Homework on the Internet. Usborne books are children's books that are sold mostly through home shows. (Some titles are also available through commercial booksellers, though I didn't happen to see this title yet.)

The item from Slashdot about old computers reminds me of one of the first systems I was ever around: a PDP-8/E running OS/8 off a DECtape.  The OS and all the applications were stored on the DECtape, which was this little (4") tape that had a file system stored on, much like a floppy disk. 

What I remember about it was hitting control-c to interupt whatever program was running, and then waiting for the tape to rewind back to the beginning to reload the command prompt.

We're all spoiled now.

The Computer History Simulation Project [Slashdot: News for nerds, stuff that matters]  "The Computer History Simulation Project is a loose Internet-based collective of people interested in restoring historically significant computer hardware and software systems by simulation."

This is an interesting idea.  Of course, even if you've got a simulator, you need to get the programs themselves that ran on these machines, and that means dealing with all kinds of now-ancient media.

When I was going to college back at UC Irvine '78-'82, we used a system called the Terak that used USCD Pascal as the OS.  It's barely possible that I still have programs for the Terak down in the basement someplace - but I haven't seen an 8" floppy disk drive in years, so I don't know how I'd ever get them back again.

About the only thing I've been able to keep moving forward is email.  When I left Xerox in 1989, I took copies of some of my mail files.  I have no idea what form I took them in then, but ever since then, they've ended up as files in various /home directories on various UNIX and Linux systems, and I've been able to move them from one system to another over the Net.   By now I must have those same files on multiple ZIP disks and now CDs. It's like storing things on a boomarang - throw it out into the net for long enough (but not too long) and then repackage it when it comes back around again.

My original idea for developing this site was to keep well below radar coverage, that is until I was ready to launch. Dave made me modify my plans in a hurry yesterday morning. Well, it's nice to have company :-) For readers who may be wondering what I'm up to, here is a short explanation. No doubt it will change over time, so I've placed a link in the home section of s l a m 's sidebar. [s l a m]

Library cats in the United States.  [Scripting News]

Anyone who knows me would know that that I had to post this item.

Following up on Jon Udell's discussion of reducing the size of RSS summaries, and later Dave Winer's implementation of a callback to make this possible, and then my note on CNN.com's practice of using the first 'graph as a summary, I believe I've made it all work.

As Dave suggested, I defined my own callback in weblogData.callbacks.rssFilterDescription:

on firstPara(description, adrpost) {
 return (html.getOneTagValue(description,"P"))} 

The description is stored in HTML, so getOneTagValue conveniently does exactly what I want.

Of course, if the first tag in my description isn't a P, this won't do what I want it to do.  I'll work on that another time. 

Side note on figuring this out: I spent what seemed like hours trying to make this work with regex.split (which basically worked if you split at </P>) and regex.extract (which never seemed to work.)  Extract was particularly annoying: I had a pattern that looked something like "<P>.*</P>".  Although the documentation of regex.extract suggested that the * would grab the minimum number of characters to make a match work, the * in fact seemed to be greedy: it appeard to match the entire description, which of course starts with a <P> (for the first 'graph) and ends with a </P> (which is actually part of the last 'graph, not the first.)

Shortening RSS descriptions to lead sentences. I'm really enjoying my ability to scan a lot of sources in my Radio news aggregator. [Jon's Radio]

Jon Udell writes about how he hacked the code that writes his RSS file to include only the first sentence instead of the entire item.  He contrasts this to another approach he's seen that cuts off the description after an arbritrary number of characters.  (I've seen that one, and I don't like it.)

I'd rather have it another way.  CNN.com has a product they referred to as "Quick News."  It's a syndication product, suitable for sending out to WAP phones, airplane seatbacks, or any other place that might want short CNN headlines.  Up until early last year, the Quick News product was produced by a separate staff who re-wrote a certain number of CNN.com stories into this shorter format.  This was deemed as not the most efficient use of staff, so instead they sought to modify the CNN.com internally-built CMS tools to take the first paragraph and turn that into a quick news story.   (They also had to train staff to make sure the first 'graph was something that could be syndicated.) 

I think that approach would work very well here.  If you want a single sentence as your description, great; if you need a couple of sentences, that would work, too.

How To Back-Up Important Radio Files. Not unlike your favorite uncle, Radio is partly but not entirely rational. Radio will automatically back-up some of your important files (if you ask nicely) but not all.

Taking care of business is important if you want to experiment with all that nifty scripting stuff, but it is equally vital if you ever need to make a complete reinstall of Radio.

(This topic fills a pothole left within my earlier Install Radio series. Real Soon, I will finish rewriting that and make a .pdf file for all newbies. A topic on re-installing Radio wouldn't hurt either ...). 

David Brown and I have corresponded a bit. I forgot to mention in my Learning Scripting topic that I have indeed not only downloaded his Python IDE Tool for Radio but ... it works like a charm with Python 2.2! 

While David's IDE can't (and doesn't try to) compete with Python's more-or-less standardly available editor IDLE, David says that he is on the track of some code that might help get him closer to the real goal: (reasonably) direct manipulation of Radio's object database in Python. Yes!

Meanwhile, if you intend to fool around with Python as I have urged, why not do it within Radio using David's Python tool?

(Not so parenthetically, one of Radio's many innovative characteristics is the ease with which developers can add their own tools directly to the Radio platform with a radical minimum of fuss and zero inteference with Radio's own internals).

So if Google hasn't indexed me directly, how come the Google toolbar claims the PageRank of my weblog is 4/10?  It can't show me a cached copy of my page, or any backward links ....

A little later ... I believe what I'm seeing is differences between different parts of Google's network. A little while ago I did a search on Google, and got my web page .. I did it again a while later, and didn't. My guess: the databases with the web page matches are updated on a rotating schedule, but the pagerank information is centralized.

I think Americans like to savor situations: "One down, bottom of the 9th, one run game, first and third, left-handed batter, right hand reliever, infield at double play depth. Here's the pitch ..."

Scoring in hockey - it seems to come out of nowhere! "Lepetia passes to Huckenchuck, who skates past the blue line .. Huckenchuck, of course, was traded from Winnepeg for a case of Labatts after sitting out the last season with -- Oh my God, he scores!

-- Vice President" John Hoynes on The West Wing.

If you want proof, see this transcript of Vin Scully's call of Sandy Koufax's September 9, 1965 perfect game.  Read it out loud for the full effect.

It's interesting to see how pages and sites get sucked up into Google.  If you search 'Paul Holbrook radio userland' on google right now, you only come up with one relevant entry:

On the other hand, if you do the same query on Yahoo right now, which uses Google for web page matches, you get these results:

1. Radio UserLand : Discussion Group
   ... Post vs publish? Paul Holbrook, 3, 18, 4/9/2002 ... Mozilla/Netscape on Mac, Paul Eliasberg,
   1, 15, 4/5 ... 2002 UserLand Software, Inc. Radio UserLand and Radio are ...
   http://radio.userland.com/discuss/

2. Paul Holbrook's Radio Weblog
   ... 55 PM. Copyright 2002 Paul Holbrook. Click here to
   visit the Radio UserLand website. April 2002. Sun, ...
   http://radio.weblogs.com/0106188/

3. Weblogs.Com: Recently Changed Weblogs
   ... 2:30 AM. 32. Paul Holbrook's Radio Weblog, 2:30 AM. ... 03 AM. 220. Phil Ackley's Radio
   Thingumabob, 12:03 AM. ... AM. Copyright 1999-2002 UserLand Software, Inc. ...
   http://www.weblogs.com/

4. Referer rankings for Wired News
   ... Radio Weblog, 1. 18. Paul Holbrook's Radio Weblog, 1. 19 ... 26. Brian Tol's Radio Weblog,
   1. 27. xio, 1. ... Copyright 2001-2002 UserLand Software, Inc. Last update ...
   http://subhonker6.userland.com/rcsPublic/referers?site=Wired%20News&group=rss
   More Results From: subhonker6.userland.com

5. Vacuum Weblog by Edward Vielmetti
   ... Paul Holbrook has a new weblog running using Radio Userland. I've known Paul since
   our days at CICnet together, where we wrangled Gopher servers. We share ...
   http://www-personal.umich.edu/~emv/project/vacuum/

6. InfoWorld's Next-Generation Web Services Conference
   ... http://radio.weblogs.com/0101359 ... Cape Clear Software. *Steve Holbrook, Web Services
   Tech ... David Winer (Userland) – Keynote Address, 09 ... Paul Holland, Venture Partner ...
   http://nextgen.infoworld.com/presentations.asp

Clearly, Yahoo is using a different crawl cycle and a different data set for its search results. Yahoo has found my web site, but Google hasn't.

Dave has just published a great story describing the notification model used by Radio with rss and opml subscriptions. He also outlines the way Radio will shortly interact with Instant Messaging servers to enable notification even behind firewalls and NAT. [s l a m]

I've added standard URL compliance and expanded/collapsed state rendering to renderCss, my previous attempt at using Frontier macros to render opml outlines. renderCss is born again as opmlRender, here is a full tutorial on how to use it. URL compliance can be fun: whose outlines are those ?

From Tim O'Reilly in the O'Reilly Network:

"The future is here. It's just not widely distributed yet." I recently came across that quote from science-fiction writer William Gibson, and I've been repeating it ever since.

In Inventing the Future, O'Reilly offers his take on what's already here and and coming up fast.  Among his takes: wireless (as in community 802.11b networks), next generation search engines, web logs, instant messanging (between programs, not just people), and the one he spend most of the article on: spidering.

From [The Shifted Librarian, via Ian's Messy Desk]:

Lego Church.  Most lego structures are objects you view from the outside, but what's amazing about the church is that it's an interior, and it feels like a church - all light and space.  Wonderful.  (And it's dedicated to a cat.)

And if you like Monty Python and Legos, then you must see this aninimation of the Song of the Knights of Camelot.  And if you'd like the lyrics to that song, and a number of other Python sketches as well, check out Georg Rehm's Monty Python Archive.

Total OS X Preview. Macworld magazine and O'Reilly team up to produce a special Mac OS X issue. The package includes a beefy edition of Macworld mag, a separate tips and tricks guide, and two full CDs.  The link also describes several books ORA is working on about OS X.[O'Reilly Network Articles]

I used an early version of OS X while I had a iMac in my office, but I didn't really have a beefy enough Mac to make it fly.  As I understand it, OS X wants 512 to 768Mb or more to really work.  Now, I've heard that this all works very nicely on the new iMac 2 ...

Unlike every other UNIX system out there, OS X doesn't have a local X server running the display.  (Having a local X server is very useful if you're trying to manage other local UNIX/Linux boxes.)  I asked several friends who are running OS X whether or not they'd tried to get a local X server running.  They both noted that XFree86 can be made to work.  One of them found it somewhat less than satifactory, and the other hadn't really use it much.  Sounds like it's not quite ready for prime time.

I found this years ago in an O'Reilly Oracle book, and I've always liked it.

One simple model of classifying requirements is known by the acryonym MoSCoW:

Must have
Should have
Could have
Won't have

The last one is particularly important: a clear design must be just as clear about what it won't cover as what it will.

Denise Howell is another lawyer publishing a web log.  There's a stereotype that lawyers can't write, but that isn't true for Denise.

Wiki sites are getting bit by the RSS bug.  A number of Wiki implementations now support RSS files of changes.  My personal favorite Wiki implementation is UseModWiki.  RSS code isn't built into the current release, but there are various patches available.  

Wikis and RSS should fit together naturally - Wikis generate list of recent changes (e.g,. Meatball's recent changes), but I've been subscribed to Meatball's changes as an RSS channel, and for some reason I've found it less than useful.  I think part of the reason is that Recent Changes gives you some context; typically, changes are listed by date with author, and by seeing the last few days of changes at once, you can get a sense of where the Wiki is going right now.  Having the changes trickle in once per hour is much less useful.

Slashdot has a good thread on the upcoming Debian 3.0 release. I use Debian on my home linux server.  This thread brings out what's good and bad about Debian: Debian doesn't always have the latest packages, and sometimes the installation procedures can be convoluted, but the people working on it are committed to bringing out a high quality release.

My sense is that Debian is the Linux distribution that is closest to the pure GNU/Richard Stallman philosophy - "We won't do it if it's not right (and we know what's right.)"  Quality and purity over user-friendliness.

Lately I've been using VMware on my W2k box to play with different Linux installations and with FreeBSD; I'll have to write that.

Managing Einsteins:Leading High-Tech Workers in the Digital Age

"In many workplaces, especially high-tech ones, managers and those they manage are operating on parallel tracks, with different sets of motivations, expectations and rewards. How to keep tech workers happy, given that they likely don't want the same things as their bosses, and certainly would choose different ways to achieve them? The long-suffering Jim Richards submitted this review of Managing Einsteins, a book which attempts to inject some sanity into the situation by clueing managers in on what it is their programmers and other tech workers might actually want in a workplace. Read on for his review." [Slashdot]

From the review itself:

"The mains themes throughout the book are:

• Managers should be honest with their workers about the company's successes and failures
• The point of management is to guide and suggest not to be autocratic (the metaphor of herding cats was used to illustrate this)
• Let the Einsteins have freedom in work environment (location - there is a whole chapter on telecommuting, hours and style)
• Einsteins are project-focused, not job-focused
• They value training and education highly
• They require a stimulating and fun work place."

Reading the review reminded me of Bruce Tulgan's Managing Generation X: How to Bring Out the Best in Young Talent. Many of the principles outlined between the two of them provide a roadmap for how to keep me happy and working at my best. I imagine a lot of the folks in libraries who have become the "techie" by default would appreciate these themes, too.

I love lists of books.  Here are a couple of them:

Ed Vielmetti's list

Ed has one of them best book lists around, and some of the best shelves as well. Ed also had the idea of taking photos of bookshelves, which is the kind of thing that makes me want to drive up to Ann Arbor, knock on his door, and ask if I can browse.
 

The Shifted Librarian

I confess I haven't looked closely at this list, but it there's some overlap with books that I've read. I also spot a couple on this list that I didn't like.  The Social Life of Information, for example.  I wanted to the book; it was by John Seeley Brown of PARC, but I found it too ponderous.  (I sold it as a used book on Amazon a while back.)

And another one, also from the local Atlanta Clear Channel outlet (WMAX, 105.3)   Yes, the same station where the jocks can't do math.

They're running a promotion around tax time.  The deal: enter their contest by faxing them a copy of your completed tax return, and you may win $80. 

I'm guessing they could make more than $80 for each return for that social security number and all the other information on your return.

Clear Channel communications, the owner of piles of radio stations across the country, was running a 'Win a million dollars' promotion.  One of the jocks at the local Atlanta station demonstrated  a serious lack of basic arithmetic skills when she came up with this:

So, you like the idea of riding on the Concorde?  Over to London in first class?  That costs, what, $40,000 a trip?  Well, if you win a million dollars from us, you can ride the Concorde every other day for the next 50 years ...

Umm ... do the math.  If you assume a 40% tax bite, she may be off by a factor of as much as 600.

Our useless digital archives, from WriteTheWeb. The BBC created a 'Domesday' project in 1986 to document the state of the empire.  Now, 15 years later, this article claims that the discs that were created for the project are unreadable.

It's almost certainly true that the data is completely gone; we're not talking about 60's era magtapes that fall apart when you touch them.  But it's probably true that it would take some investment of time and money to get what's off them.  It's probably not worth it; even though the original project cost 2.5m pounds.

I'm convinced that Digital Archeology will be a big field in the years to come, given all the multimedia we've created over the last 50 years that requires machinery to decode it.

The WELL's Gopher server is still up!

I helped the WELL staff install their Gopher server, but beyond that, the only part I played was in helping write The WELL Gopher Manifesto.  I wrote something about this on MeatBall wiki some time back.

A few people who's stuff I like a lot:

• Ed Vielmetti (Vacuum web log)
  I worked with Ed when I was back at CICNet; he was into Gopher, the Web, and if my recollection is right, conference software long before it was fashionable. I get Ed's periodic Vacuum emails that keep me up with what he's doing.  I relate to Ed more these days, because Ed was out of a job for quite some time after he was laid off from Cisco. But he's starting this next week at Arbor Networks, so perhaps there's hope for me as well.
• Jon Udell (Jon's Radio)
  Jon has been on the collaborative technology trail for ages, and he writes lots of good stuff.
• Lou Rosenfeld (Bloug)
  I probably met Lou when I worked at CICNet.  He's the co-author of  Information Architecture for the World Wide Web, which next to Jakob Nielson's book design web simplicity is one of the best books on web design.  Plus, as you would expect, his site is just plain beautiful.
• Jakob Nielson (useit.com)
  You won't find much personal about Jakob on his site, but he has more practical information about building usable web sites than almost anyone else I've read.
• Bob Frankston and friends (SATN)
  A very nicely done weblog updated by Bob Frankston (co-inventor of VisiCalc) and friends.  Very well written.

Lou Rosenfeld and Peter Morville will be publishing a new edition of Information Architecture for the World Wide Web in July 2002.  Peter has a slow-loading PDF brochure describing the book on his site.

I was motivated to try Radio UserLand by Jon Udell's article on "Instant Outlining" on the O'Reilly network site. 

I've been interested in Jon's vision of collaborative technologies for some time.  I bought two copies of Jon's book Practical Internet Groupware; one of them went missing at work, and I liked the ideas so much that I bought it again.  But as with many of these things, Jon's approach required too much buy in to specific technologies.  Jon has acknowledged this himself in his article.

I've tried plenty of collaborative software packages. I belonged to The Well for a time; I think even paid for it.   I helped them get their Gopher server running.  I worked for one of the original regional ISPs called CICNet, and I had them buy the Caucus text-based conferencing system, which was based in turn on the Confer conferencing system built at the University of Michigan.  (Or was it PicoSpan?  PicoSpan was used by The Well, but I'll have to look up it's origins.)  Over the last year I've been somewhat into Wiki software, and starting using at CNN while I worked there. 

So I'm a bit of a skeptic when it comes to this general area.  We'll see how this works.