June 2002 Archives

I've never built a built a full-corporate intranet site, though I've been in a few efforts to build group sites. Even those efforts could have used the information in the article Theo Mandel has written: A solid intranet in eight steps"

He outlines eight steps to getting it right:

• Forget about your Internet site.
• Eliminate frames from your design
• Create intranet guidelines and stand by them
• Put usability before consistency
• Start small and grow iteratively
• Use standard link characteristics
• Evaluate against measurable objectives and criteria
• Make your intranet accessible

From [Column Two] and [Intranet Focus Blog]

Our new interim rector of our church had a wonderful quote this morning. Inscribed on the library of the seminary he attended is the following:

Seek the truth; come whence it may, lead where it will, cost what it may.
 
The Reverend Dr. William Sparrow, 1801 - 1874
Professor of Theology and Christian Evidencer
Virginia Theological Seminary

Steve McConnell is well known for several computer science books. Code Complete is his best known book, but I didn't discover it before I stopped programming for a living. I'm fonder of Rapid Development: Taming Wild Software Schedules, which is a great book of tips about how to keep software projects on track. McConnell has two pages of recommendations on other great software books: a list of top 10 software construction books, and a his top 10 books on software.

I was out at Best Buy today (the infamous one that had had a customer arrested back in April) and noticed an interesting sheet of paper next to the checkout area: it appeared to be a sales projection for today. If I read it correctly, their "sales budget" for today was $119,000.

That's interesting. I'm always curious about what things cost, or how much different businesses make. Let's see what that would imply. Note: aside from the $119k number I saw at the store, all the rest of these numbers are swags (scientific wild-ass guesss, for the unintiated.)

• $119k/day * 365 = $43.4 million/year. That might be high; I'm guessing Saturday is one of their highest sales day. $100k/day would still be around $36 million/year.
• What would the sales force cost? I don't know how many people you need to work a store like that .. let's assume 40 people at $10/hr for 12 hours. That's around $5,000; it may be a little higher because of higher paid people, so assume $6,000.
• I've browsed around trying to find some real number about Best Buy, and they suggest their profit on their goods is about 25%. Assuming $119k, cost of the goods would be around $89k.
• Assume another $2,000 per day for other expenses: electricity, lease on the store, whatever.
• Assuming 119k/day, that still leaves around $22k profit per store/day. (If you assume 100k, you get around 17k/day).

The bottom line: that store could be making between 6.2 and 7.9 million dollars per year for Best Buy. That's about 18% profit margin.

There must be more to the picture, however: Hoover's reports that for the fiscal year ending in Feb '02, Best Buy's net profit margin was 2.9%. Some of that obvious comes from the central operation, but my estimates of the costs per store are probably low in some ways.

(BTW, I did find some page that said that the average sales per store were around $36 million annually. So I'm in the ballpark.)

Sometimes following other people's blogs is like talking to someone who won't shut up: you ask one question, and you're in for a 15 minute answer. Well, it's a little like that, except it's not: it's a lot more interesting. Case in point: I pulled a little piece out of my news aggregator this morning on a k-log pilot experiment, and many hours later, I'm left with a pile on interesting pages scattered around my screen that I'm trying to make sense of. (I can't even remember where I found the reference to the k-log item; it's already gone from my aggregator.)

The k-log item came from David Gammel's High Context k-log. Gammel's log was a font of other interesting stuff, including the Phil Wolff piece on klogging vs the 11 deadly KM sins, which I've already blogged.

It also included this innocent looking little item:

Faceted Classification.

Ease:

Faceted classification of information and The business requirements for classifying content. Here's more. (via IASlash)

These articles look great at a quick glance. I'll pull out some nuggets later after I've had time to review them more closely.

[High Context]

and that lead to a bunch of interesting stuff, including having me pull a book off my shelf that I've had for a while but haven't really read: The Knowledge Managment Toolkit: Practical Techniques for Building a Knowledge Management System In a touch of serendipity, Phil Wolff had a note about the second edition of that book which will be coming out anytime.

Somewhere along the way I found a pointer to a Wiki on kblogs and that was a rabbit hole all by itself.

Six hours and a lot of blog posts later .. where has the day gone?

The mention of old PIM programs in my last post sent me to google to find any page that mentioned Lotus Agenda, InfoSelect, and Ecco. I came up with a page about Software for research using Windows which looks interested. I'll have to look later.

I picked a reference to something called faceted classification from High Context. The back credits on where this comes from are getting a little deep for more (more on that later), so I'll just quote the item:

Faceted Classification.

Ease:

Faceted classification of information and The business requirements for classifying content. Here's more. (via IASlash)

These articles look great at a quick glance. I'll pull out some nuggets later after I've had time to review them more closely.

[High Context]

I've had a look, and I think there's good stuff there. It's a little hard to tell on first glance; things like this can be real mind stretchers. (Or after a while they may turn out like a popsicle that you've held in your hand: just sticky and not very much fun after all.)

The notion of classification fascinates me. Long ago I toyed with the idea of classifying my then growing collection of computer science articles by the the ACM's computing classification scheme. (This must have been after 1982, because they came out with a revision of the original 1964 scheme them, and I'm sure that's what set me off.) In the end, I don't think I did anything with this; when you get right down to it, I'm just not organized and dedicated enough (anal enough?) to be that organized.

But something infected me way back then in this general area; it lead to a whole series of side effects/symptoms which included a long time fascination with PIMs (Personal Information Managers). Remember Lotus Agenda? InfoSelect? Ecco? Don't even get me started ....

Phil Wolff has posted a wonderful little piece called Klogging vs. the 11 Deadly KM Sins. It's Phil's suggestion about how klogging overcomes previously experienced problems with KM. My favorite meme:

By reading your colleague's klogs, you crawl inside a little of their day. Almost as fun as Being John Malkovich.

[Phil Wolff, via High Context]

The New York Times has a nice article called "Swab, Then Scour: How to Sell a House. There a couple of useful looking pointers in the article. A couple of excerpts with pointers:

Clearly, a seller may need to embrace a whole new universe of heavy-duty cleaning supplies for this job. I had in mind the sort of products that I last saw used by the school custodian — and that's exactly what I found online at www.atmosphereproducts.com. The site, which sells more than 3,000 products that fall into hard-core-cleaning categories like "Floor Maintenance Equipment" and "Matting and Utility Cleaning Tools," specializes in selling items in bulk at wholesale prices and will ship any order up to 19 pounds for $7.95, said Howard Hurwitz, owner of Atmosphere Products.

....

One annoying chore on my list the Internet could simplify: replacing some unattractive old switch plates. Although switch plates are a commonly stocked item, even large offline stores have frustratingly idiosyncratic selections that don't take into consideration the fact that nearly every room in my house calls for multiple sizes of plates.

If I go in looking for, say, six white single cover plates, four combo wall plates and a triple rocker, I might end up finding only half of what I need in stock. And I never like the jarring effect of mixing brands in a single room. One solution proved to be cornerhardware.com, which stocks a wide selection of sizes at competitive prices in chrome, ceramic, plastic and wood.

It's interesting how your choice of tools is influenced by the environment you're working in. I worked at a university-based ISP for years (CICNet) before I went to CNN: I used Pine for all my email, some Emacs, some MS Word, etc. When I went to CNN, I started using a new set of tools to manage my email and work life: Microsoft Outlook had my calendar, address book, and all my email. Now that I'm working back at a university, my choice of tools is moving back towards what I used to use at CICNet: Pine and Emacs.

Part of this is based on a lack of resources. I'm currently using a PII/266 laptop with 128mb of memory. Pine uses around 4mb of real memory on average on my box; emacs around 6mb. Fire up Outlook, and immediately 20mb of real memory is gone. (To be fair, the biggest pig is Mozilla: I regularly have 40mb Mozilla process running around.)

GNU Screen

I've also started using GNU Screen, a tool I haven't used in five or six years. Screen lets you keep multiple "virtual" sessions going over a single telnet/SSH connection. These days you'd tend to just fire up another xterm/SSH window, but Screen has one important benefit over that approach: you can "detach" from a group a virtual screen sessions, then re-attach later without losing any state. I typically have 3-4 virtual sessions running on screen on my home Linux box: a Pine session for my non-work email; one session running Emacs, and another misc shell session or two. I don't have X installed on my laptop, so when I get to work, I start up a single SSH session back to my Linux box, fire up Screen with all these sessions. When it's time to go home, I can detach from these screen sessions all as a group, shutdown my laptop and go home. At home I can fire up my laptop again, login to my Linux box again, re-attach my screen session, and find my email, emacs, and shell sessions right where I left them.

Even better, if I lose my connection to my Linux box, Screen will keep my sessions intact, and I can re-attach to them when I get my connection back. Compare that to your normal xterm/SSH Linux box scenario: you lose your connection for any reason, and that job you were running for the last N minutes is killed.

The only downside of Screen is that it traps Ctrl-A by default. I can change it to something else, but if you're an Emacs user, anything you pick is going to be used somewhere. (Screen can of course send a Ctrl-a to your application, but training your fingers to remember to use that isn't easy.

Pine

I've never completely got away from Pine; even when I used Outlook at CNN, I would still use Pine from home: on a dial-up connection, it was much faster to start Pine and point it at the Exchange server than it was to start Outlook. (On a dial-up connection, Outlook could take many minutes to fire up.)

But Pine has it's pleasures. The interface is moded, but easy to use; you can do everything in Pine without taking your hand of the keyboard.

The main downside to Pine is the lack of rich display for HTML email. Pine can now grok HTML, and presents a readable view of HTML email, but it's kind of like being stuck using a Gopher client in the age of HTML and the web.

Emacs

I'm also back to my old friend, Emacs. Emacs was where I lived my life when I worked at the CERT: I used MH-E to read my email, GNUs for reading News, and I believe the CERT might have still been using Scribe for text processing, which could be edited very nicely under Emacs.

Using Emacs again was actually pushed by my almost complete conversion to Mozilla. Radio Userland uses the very nice Microsoft in-line HTML editor, but that tool only works if you're accessing Radio from Internet Explorer. From Mozilla, you need to do straight HTML. Movable Type, which I'm now using for a work weblog, also wants to deal in straight HTML. And so Emacs is back again as my HTML editor. I'd probably prefer a WYSIWYG editor; at home I use Dreamweaver. I tried a couple of freeware Windows HTML editors, but all of them ended up being inferior to Emacs. I run Emacs locally under Windows 2000; compared to the other Windows based tools, Emacs is nice and lightweight, and since I'm only using simple HTML markup, Emacs benefits as a tool for slinging around text come to the fore again. (Ah, the irony of Emacs now being considered "lightweight!")

What sealed the deal for Emacs was get a spelling checker working again. It took some messing around, and I ultimately have to run Emacs from under Cygwin, but I've got ispell working again, so all's right with the world.

I've started to use Movable Type for a work-focused weblog.  It's an interesting experience: I started this web log when I was between jobs, so I didn't have to think about the division between work and non-work life.  But I must have been infected by all the walk about K-Blogs and knowledge managment, so I'm going to give it a try.  Starting another blog made sense, because the all the details of the projects I'm working on at Tech don't really fit well with the more personal, observational orientation of my Radio blog.

I'm not using Radio for a couple of reasons:

• I'm still a temp (no benefits) employee at Georgia Tech, which means I don't have many resouces.  Movable Type is free, so I don't have to worry about getting someone to reimburse me for a copy of Radio. 
• Second, since this is a work blog, I don't really want to expose it to the world at large.  Radio seems a little more appropriate for something that wants to fit into a larger community. I'm not going to link it here for that reason.
• Finally, I just wanted a chance to try Movable Type.  It's a nice system.

I'll have more reaction to Movable Type when I get some experience under my belt, but so far I have a couple of first impressions.  Radio seems to be easier to get started with, but Movable Type seems to be more straightforward to do simple customizations on.  Radio is probably ultimately more powerful in that area, but MT has a reasonably-well documented template language that gives you a fair amount of power.

I heard an interesting stat today at a meeting here at Georgia Tech (where I now work): there are about 75,000 living alumni of Georgia Tech. That's fewer than I thought there would be. There were 15,576 students at Tech in 2001, so 75k is only about five times more than the number of folks currently enrolled.

A few relevant stats from the Tech 2001 "mini fact book:"

• Number of graduates in 2001: 3,370 (2,035 Bachelor's, 1,080 Master's, 255 Ph.D.'s
• Number of students in 2001: 15,576 (11,043 undergrad, 2,359 master's, 2,204 Ph.D.)

So in 2001, about 20% of the students enrolled graduated that year.

Now that I think about it, 75k alumni is about 22 years worth of graduates at the current rate. Perhaps that 75k number is the number of alumni on Tech's mailing list of alumni. That would make more sense.

I wonder which school has the most living alumni?

I don't know how I missed this one, but Xerox PARC is no longer attached to Xerox. They're now just PARC - the Palo Alto Research Center; they're now an independent research house. John Seely Brown was the Director of PARC, but his name isn't in the list of management people, so I'm not sure what's happened to him. (His title was "Chief Scientist of Xerox," so I suppose he could have stayed with the parent company.)

This is a bit of sad news. Back in '79 or '80 when I was in the computer science program at UC Irvine, I was fascinated by Xerox was doing up at PARC. I found out that PARC had a tech report series, so I wrote off to PARC to request some of them. I still remember the day when I got a rather large box with two dozen of the famous blue-and-white covered reports. (I was amazed that they would send some random college sophomore so much paper!) I was so interested in what PARC was doing that I joined Xerox in El Segundo after I graduated from college. I spent 7 years working on the Xerox Star, a product that drew heavily on research from PARC.

The other half of the Xerox Star team was up in Palo Alto at buildings some miles from PARC, but I still found an excuse to go over and visit PARC once or twice. I do remember they had a wonderful cafeteria ...

I still have most of those old blue and white reports in my basement.

Slashdot has a story about where to get parts for building a PC from scratch. Two sites come up a lot: Newegg is recommended over and over again as a good place to work with for cheap (but high quality) parts. Others mention PriceWatch as a place to compare prices, but warn that the stuff that comes in with the lowest prices may not be worth having.

Here's something useful in these days when everyone is talking about copyright and fair use: a page from the University System of Georgia with the University's interpretation of copyright and fair use. The page also contains a pointer to the Digital Millennium Copyright Act. Interesting opinion from the University's lawers and librarians: "Based upon our initial review of the Digital Millennium Copyright Act ("DMCA"), we do not believe that the act affects the guide's fair use analysis." The page was created back in 11/97, presumably right after the DCMA was passed.

The University's analysis contains a particularly useful section with a variety of examples illustrating what is and isn't fair use.

agmweb.ca allows you to search 7 online book stores for books and prices. All information is available in XML-RPC and RSS feeds for you to syndicate on your site."

Three Knowledge Management Primers for the Millennium : Primers include:
- What is Knowledge Management, by Hubert Saint-Onge
- The Sveiby Toolkit, by Karl Erik Sveiby
- Putting the Engine of Innovation to Work, by Jef Staes
* Go to the Primers
* While you are there, checkout the Knowledge Innovation Timeline graphic

An excellent set of resources courtesy of SynapShots, an excellent resources in its own right

But what do you do if you don't run your own mailserver? Well, if you're running OS X, you can install SpamAssassin locally and have it prune your mail on your own computer. Ben "Movable Type" Trott has written an excellent tutorial on running SpamAssassin under OS X. Link Discuss (Thanks, Merlin!) [Boing Boing Blog]

Interactive shells uber alles!. Taking a stab at learning XSLT. TestXSLT from Marc Liyanage is making me happy and helping, for many of the same reasons I enjoy learning Python, and have been somewhat enjoying learning Common LISP again.... [0xDECAFBAD]

Appears to be Mac/Cocoa only.

SpamAssassin sounds like great way to kill spam. Simson Garfinkel writes glowingly about it, and notes that Dave Farber is a convert. SpamAssassin only runs on UNIX for right now, but that works for me.

Patrick Delaney has written a splendid essay on using blog technology to help children learn to research and write.  The title of the project is an instant magnet: Find It, Read It, Write It.  It's a wonderful idea, but I confess after I read it I'm still trying to fit it into my world model.  It grabs at me in a number of ways:

• The project is an outgrowth of the Bay Area Writer's Project.  The BAWP's co-director is - or was, I don't know - Rebekah Kaplan.  Ms. Kaplan was one of my english teachers at Foothill High in Pleasanton, CA.  I do remember Ms. Kaplan, but I can't honestly remember a lot about her class.
• The approach sounds reminiscent of the things Montessori schools try to do in getting kids to learn to do research.  Montessori focuses in part on asking questions: before you go off to do research, the child needs to generate a list of questions they'd like to answer about a topic.  This hopefully encourages the child to go out with an inquiring mind, and not just a shovel ready to scoop up whatever they find.  (It works: last year a group of children from our Montessori school went up to the Tennessee Aquarium in Chattanooga.  The Aquarium staff was most impressed by the kids: they had rarely seen a group with so many questions.)
• The project title reminds me of a couple of books by William Zinsser: Writing to Learn and On Writing Well, both of which espouse the idea that writing is one way to go about learning something.

And it all pulls at me for the same reason I started this weblog: the sense that writing is something that can help pull it all together.

It's a great idea, but one I haven't taken to heart.  I've been at my new job at Georgia Tech here for almost three weeks, and I haven't written much down yet.  I'm unsure of where and how to go about it.  At my last job I started an internal Wiki to help document what I was finding out.  Here I could use this blog, but I'm sensitive to the line between work and home, which is another way of saying I'm still new in this job, and don't want to screw it up.  I don't think I know much dirty laundry about Tech yet, but I still want to be careful.

Paolo Valdemarin's weblog (excerpted below) suggests that the sources for news aggregators would include "reports generated by your accounting software, the status of your servers ..updates from your co-workers workflow management software." At first blush, this sounds great. But how do you prioritize those kinds of items? At what point does your news aggregator start to resemble the fire hose that email has turned into?

At least an aggregator only pulls in the sources that you select, but we've only removed some of the random junk from the information flow: we still have to make sense of it.  Radio's aggregator at least groups content by source, but it still harkens back to the early days of email: when you get enough stuff coming at you, it all starts to blur.

Scripting News: A news aggregator is "software that periodically reads a set of news sources, in one of several XML-based formats, finds the new bits, and displays them in reverse-chronological order on a single page." 

[Paolo comments] It's important to consider that "set of news sources" could also mean reports generated by your accounting software, status of your servers, posts in a discussion group, orders from your e-commerce site, updates from your co workers workflow management software ... got it?

[Paolo Valdemarin], found via [The Shifted Librarian].

Up until this year, EPSN had a neat feature that displayed the various baseball division races as a graph. The graph typically showed how many games over .500 each team was. It was a very clear visual way to see how a team's fortunes rise and fall over the season: you could really see it when I team made a sustained run. ESPN dropped that feature this season, but now Sportsline has picked it up. Check out the Sportsline MLB standings, and click 'Race Chart' in the header of one of the divisions.

Caveats: it's a Java app, and further more it seems to lock up Mozilla when I try to close the window. Works ok in IE, though.  [Later note:] What seems to kill Mozilla is trying to deal with the popup window.  If you use this direct link to the Race Charts, Mozilla seems to be able to cope when you close the window.

9/10/03: You can find race charts at http://www.baseballgraphs.com/. They don't seem to be kept up to date every day, but it's better than nothing. Thanks to Nick Adami for finding this and pointing it out to me.

11/28/04: And the baseball season is over, but ESPN had the race charts this year. Hopefully they'll come back again next year ...

After looking at warehouse.com, I was under the impression that memory for a G3 400mhz iMac (the kind we have in the house) was running around $80 for 256mb, and $200 for 512mb. I just found a site called ramseeker; they provide price comparisons from a variety of vendors. According to them, 256mb can be had for between $35 and $50, and 512mb for around $90. That's more like it.

ramseeker doesn't sell memory; they claim to be a free service comparing prices for Mac memory. Very useful.

I still don't want to pay for it. The Mac belongs to my wife's work, so I'd like them to pay for it. I don't think that's real likely right now.

I've been playing around with the ActiveWords application recently, and just wrote up my impressions.  I came across something I initially thought was a bug in ActiveWords, but I've since concluded is a big performance problem in both Mozilla 1.0 and Netscape 6.2: textboxes are very slow.

This came up when I was testing the ActiveWords text substiution feature.  ActiveWords lets you define any word to stand in for any amount of text.  For example, I've set the word 'homeaddress' to stand in for my basic contact info.  That works out to be about 5 short lines of text, about 120 characters.  On the PII/266 laptop I've been using, if I type 'homeaddress' into a text box in Mozilla or NS6.2, it takes about 8 seconds to backspace over the word 'homeaddress' and type in the 5 lines of text.  (The experience was very reminsiscent of using a teletype; ActiveWords was doing about 15 characters per second.)

My first reaction was that something was screwy between Mozilla and ActiveWords, but then I turned off ActiveWords and tried typing very rapidly into a text box, I discovered that Mozilla just can't keep up. 

The problem is probably masked on most modern systems, but a PII/266 isn't that bad at most things.  (At least that's true after I got 128mb into it.  I first got this laptop with 64mb.  Win2k is very unhappy running in only 64mb.)

Follow-up:Nope, it's still slow on somewhat newer systems. My PIII/500 also had a notable lag typing into a Mozilla textbox.

My Radio news aggregator recently spit out a mention of the ActiveWords tool.  ActiveWords is a tool that watches as you type; when you type something it recognizes, it does something: launches a program, fires up a web page, substitutes some text. I was intrigued enough to follow up and install it; I've been playing with it since Friday.  The best tips and descriptions about ActiveWords come from the Ernie the Attorney blog.  A few first impressions:

• If you come from a 'type and remember' background, ActiveWords feels like coming home.  ActiveWords can be recognized anyplace, not just when you're typing into a text field. 
• One of Ernie's first tips was to set ActiveWords to trigger when you hit a double-space.  This is a great idea; it turbo-charges how the program feels.
• Using ActiveWords to launch a program or open a folder is particularly useful.  So is launching to a particular URL.  (I've set 'baseball' to take me to the current baseball scores page at CNNSI.com.)
• ActiveWords normally runs with a "MonitorBar" at the top of the screen.  The 'Ernie' article suggests moving this to the bottom.  I turned it off; I found it distracting, in part because the MonitorBar echos every word you type, including passwords. I don't like that.
• ActiveWords is a memory pig.  I've got an older laptop with 128mb of memory.  Right now ActiveWords has five processes running, using a total of 28mb of memory (!) 128mb is the practical minimum you want on a Win2k system, so 28mb is a lot.
• ActiveWords shows up a performance issue in Mozilla and Netscape 6.2  (More about that in another post.)

Despite some of the downsides, so far I think it's a useful tool.  I've got another 57 days on my trial, so I should have a better idea in a month or so whether ActiveWords is worth the $30/year and the system resources.

Now that Mozilla 1.0 has been released, one of the things that's standing in the way of using it fulltime is the the fact that Radio's nice WYSIWYG HTML editor only works in Internet Explorer.

This is annoying not only the PC, but on the Mac as well, because even IE for the Mac doesn't support this technology.

Anyone have a nice solution that would work for the Mac and PC in Mozilla without paying more money? I know there's at least one commercial Java solution, but this feels like something that UserLand should step up to.

Here's a blog feature comparison table.

I Think of It More Like My Brain's "Memory Stick". My Blog, My Outboard Brain by Cory Doctorow

"As a committed infovore, I need to eat roughly six times my weight in information every day or my brain starts to starve and atrophy. I gather information from many sources: print, radio, television, conversation, the Web, RSS feeds, email, chance, and serendipity. I used to bookmark this stuff, but I just ended up with a million bookmarks that I never revisited and could never find anything in.

Theoretically, you can annotate your bookmarks, entering free-form reminders to yourself so that you can remember why you bookmarked this page or that one. I don't know about you, but I never actually got around to doing this -- it's one of those get-to-it-later eat-your-vegetables best-practice housekeeping tasks like defragging your hard drive or squeegeeing your windshield that you know you should do but never get around to.

Until I started blogging. Blogging gave my knowledge-grazing direction and reward. Writing a blog entry about a useful and/or interesting subject forces me to extract the salient features of the link into a two- or three-sentence elevator pitch to my readers, whose decision to follow a link is predicated on my ability to convey its interestingness to them. This exercise fixes the subjects in my head the same way that taking notes at a lecture does, putting them in reliable and easily-accessible mental registers....

Being deprived of my blog right now would be akin to suffering extensive brain-damage. Huge swaths of acquired knowledge would simply vanish. Just as my TiVo frees me from having to watch boring television by watching it for me, my blog frees me up from having to remember the minutae of my life, storing it for me in handy and contextual form."

I find that Cory's reasons for blogging echo my own, and that I am reaping the same benefits. It's the opening up and sharing that makes it possible, whereas before I would send a link to a few select people. It also gives me the opportunity to explore my thoughts and ideas more fully, having to flesh them out into actual concepts I have to articulate. It's helped me connect the dots in my own mind, and luckily it's all centrally stored on my blog.

Hopefully I'll be able to impart this vision and explain the advantages to my colleagues when I start implementing Radio for internal KM.

[The Shifted Librarian]

In my wrap-up Tuesday night, we talked about the notion of having a personal KM strategy. If knowledge is your craft, you have a responsibility to maintain and develop your tools and your craft.

When we talk about learning organizations and about knowledge management practices, it can be easy to lose sight of this personal dimension. We think about the problem in terms of what 'they' ought to be doing. This problem is aggravated by the fact that senior level executives don't have a lot of knowledge management problems of their own. They have assistants and staffs whose fundamental role is to be the executive's KM system. Most of us are not so fortunate.

Tom Davenport wrote an interesting piece on the notion of personal information environments in CIO magazine quite some time back. It's still a good introduction to the notion, although I would take it up a level. Managing the details of your information life is a starting point,  but we need to do more if we take a knowledge perspective.

Blogging is one piece of the puzzle as Cory's comments capture nicely. Not only do you have that link to something that has caught your attention and interest, but you have an opportunity to boil down the 'so what' that warranted that attention.

The other thing that blogging can do for you is create raw material for your learning and reflection.  This works on at least two levels. When you create an entry, you have to do some thinking and reflection. That alone puts you way ahead of most of the pack. And, as you do it over time, your skill at thinking, reflecting, and writing will all improve, which will make you a more effective knowledge worker.

It's the next level, however, that creates a long-term amplifier for your knowledge work productivity. You now have a chronological trace of what you thought at the time. You have something you can examine to understand how your thinking and insights have evolved over time.

Now, there is a question of how much of this you choose to share publicly. Most of what I've said so far works whether you publish your weblog or not. Although there is an advantage of visualizing an audience to help you distill your thinking. Warren McFarlan at the Harvard Business School was one of the professors who dragged me through my doctoral program.  He used to joke that one of the worst aspects of being an academic, especially in a fast-moving field like information technology, was that there was a public record of every dumb idea you'd ever had. On the other hand, if you have the guts to put the ideas out there, you also get the opportunity to test and refine them.

Fundamentally it's the difference between doing real science vs. crank science. The only way to tell the difference in the end is whether you're prepared to open yourself up to criticism. Putting yourself on record is the first step in that process.

Managing My Knowledge

As we conclude the final week of the final quarter of the final year, I've certainly learned quite a bit this year about knowledge management.  I've probably thought more about it than I've ever wanted to.  However, I'm still not sure I have a very good method for managing my own knowledge.  I've learned an immense amount in the past 12 months - far more than I expected to.  However, it will be clearly difficult to draw on my own knowledge base and to actually apply what I've learned.  Models, frameworks, concepts, etc. are all interesting to study but only useful if you can actually apply them. 

I for one am terrible at remembering who did what study, specific terminology, descriptions of models, etc.  I can apply it all in real-time as I am learning it, but it quickly leaves me.  Thus, I'm very conscientious about keeping course summaries that I can refer back to.  My hope is that I will remember that I once learned something about a given topic and that I'll know where to go to get it.  But even that may be a challenge.  If anyone has any good advice for holding on to what we've learned here, I'm listening.  It is amazing the toll that time takes on a memory.  In the end, the most enduring thing I think I will leave with is a refined ability to think, challenge, and critique - and a new found humility that there is still so much that I don't know.  School for me has been analagous to training for a mental marathon.  Training for a marathon, you continually push yourself far beyond your limits and find you can do even more.  Then when the race comes you fully exhaust yourself.  However, you only retain the physical condition if you keep on running.

I think my is in the best condition of my life - now I have the challenge of keeping it from atrophy; it can be a difficult challenge when you work in a task-oriented world, but I will certainly give it a try.

[Greg Harmeyer's KM Weblog]

Business schools now routinely require that their students have a PC. Perhaps they should also require that students start a weblog (if they haven't already done so) and provide some guidance and support about how to do both information management and knowledge management at a personal level.

My oldest son is now in middle school.>Alan Kay - formerly of Xerox PARC, recently of Disney Imagineering, now founder of Viewpoints Research Institute. Start with The Computer Revolution Hasn't Happened Yet and then check out Squeak and SqueakLand.

One theme across all of these thinkers is the learning is a personal phenomenon. There's lots of help you can get, but you have to do the work. None of them are particularly impressed with schools or classrooms as the best place to do that learning. That's why the real value of a place like HBS or Kellogg is the community of smart and motivated people they assemble.

SCORM and Modules

The Chronicle of Higher Education has an article about standards for courseware and CMS allowing "mix and match" development:

For the first time, evolving technical standards for software are making it possible for colleges to customize distance-learning programs by easily mixing online-learning software from multiple companies.

This is of course exactly the approach to a CMS system that makes the most sense to me. One of the standards the article refers to is SCORM, "the Sharable Content Object Reference Mode." Here's a good overview of SCORM. The difficulty is that talking about a standard is one thing; actually adhering to it is another, far more important step.

Stolen straight from Paul Beard's blog:

apache logging refined

Once I worked out how to ignore all the nimda and other MSFT exploit-related requests, I realized I also needed to ignore my own hits. I'm not that interested it seeing how often I post or read comments (could be depressing, actually).

So I set all my host IP addresses to 'dontlog.' And the nimda stuff stays the same. I suppose ignoring a range of addresses is better: I just haven't done it yet.

SetEnvIf Remote_Addr "192\.168\.2\.1" dontlog
SetEnvIf Remote_Addr "192\.168\.2\.2" dontlog
SetEnvIf Remote_Addr "192\.168\.2\.3" dontlog
SetEnvIf Remote_Addr "192\.168\.2\.4" dontlog
SetEnvIf Remote_Addr "192\.168\.2\.5" dontlog
SetEnvIf Remote_Addr "192\.168\.2\.7" dontlog
SetEnvIf Remote_Addr "192\.168\.2\.10" dontlog
SetEnvIf Remote_Addr "127\.0\.0\.1" dontlog
RedirectMatch (.*)/(scripts|root.exe|cmd.exe|default.ida).* /goaway.html
SetEnvIf Request_URI "/(root.exe|cmd.exe|default.ida|goaway.html)" dontlog
CustomLog /usr/local/weblogs/httpd-access.log combined env=!dontlog

Posted by paul at June 01, 2002 09:49 PM