bitsifter
friday, july 19


[rant] I suffer from stress-related waking nightmares.  They run their course in the middle of the night when I wake up and am unable to go back to sleep until I solve some type of technical problem.  Usually, the problem is something completely non-sensical such as "how many times are you going to hit the @ key in the year 2000?" It's tricky, but, then again, it's a nightmare.  The finacee has no clue what is going on as I toss'n'turn in my irrational stupor and, usually, after an hour or so of trying to solve impossible problems, I go back to sleep.

A nightmare from several weeks ago went like this: Wake up and someone has left the computer on. Who left it on? Wait, why do I know it's on?  What components inside of my computer are making that noise? How does someone make a computer silent without turning it off?

I sweated that one for 55 minutes before I went back to sleep, but upon waking the question was still there? Why does my computer have to make noise? How many people would leave their computer on all the time if it was absolutely quiet?  (I'm talking monitor, too, folks, none of this high whine crap)

There are reasons to leave you're computer on. If I was techinical I'd tell you it was wear'n'tear on your computer to get jolted on and off, day in, day out, but I'm not. What I do know is that a good portion of America's working force gets up, grabs the paper, lets the goldren retriever in, and sits to scan the morning news.

Let's save these folks some time. How about if they wake up, tap the spacebar to disengage the screen saver, and sit down to read their customized news report from Pointcast? Sure, the dog is pissed, but we've added value to the morning experience and saved a tree.

The reason someone won't do this now is because they're computer is too loud.  Even if it's in the other room, folks are going to turn their computer off because it's begging for attention with it's murming power supply fan, annoying high pitched monitor whine, and ricky-ticky hard drive clicks.

Thin-computing is promising cheap network access which is going to only aggravate this situation, if they want to get my home-computer dollar, they better make it silent.


[rant] I knew all about domain names before the brother-in-law, but he had the bright idea to register his last name as a domain several months ago.  Hence, he now receives mail at [email protected]. I'm all envious so I do the same.

With thousands of domain names being registered a day, it was obvious the cyberscape was going to get a little crowded. The problem is being exacerbated by the fact that the de-facto naming standard is the COM domain -- with little legal precendant and even less responsible parties with sufficient power to make decisions about trademark infringement, it's a mess.

In comes Norfolk Island, a little land mass off the coast of Australia. Their spin on the whole domain name nightmare is this -- swallow your pride and register in their domain .NF -- take any name you like, they'll even send you a certficate of ownership. It sounds silly, but only until you realize Norfolk's other contribution to western civilization, Mutiny on the Bounty.


[rant] The co-worker was sitting in my cube this morning complaining that her significant other was going to be able to live her because he had a horrible allergy to dust mites, a tiny bug that lives in your carpet and feeds on your dead skin. It was sounding bleak, so I turned to my computer, type "dust mites" into AltaVista and immediately found a page which describes common ways of dealing with the little buggers.

"Where'd you find that?" she asks. Hey, I'm information empowered.

Search engines employ Asimovian sounding technologies called "web robots" which crawl around the web and index all the information they find.  There are lots of different strategies for building this index, some only look at the first 200 words of a page, others will index everything they find, but there number one problem is they are extremely dull.

You're going to think I'm on Bill's payroll, but the gee whiz factor of the net lies in the fact that it is active. The initial NCSA model for the web was to mimic how a magazine is displayed and, to a large extent, we've achieved that and gone further. No longer is a page a few paragraphs of text and a some well placed images... chances are you're liable to stumble upon a framed web page populated with GIF 89a images, Java applets, and plug-ins... and this is where the search engine indexing strategy fails, the web robots know absolutely nothing about the content.

Lets take the example of a framed page. If you search Infoseek for the word "bitsifter", you'll notice the description of the pages is "The BitSifter page makes heavy use of Netscape 3.0+ features -- why not use it?".   If you're familiar with frames, you'll immediately know what occurred when Infoseek's web robot wandered across the Digest -- it didn't know what to do with a framed page.

The intelligent use of frames is the first big step towards web pages becoming applications rather than magazine clones, but web robots are liable to be left in the dust because they've no idea that one page could be immediately related to another.  The task of indexing the web is a daunting one, but as web pages grow in complexity, the tools which monitor the web will need to, as well.