Internet

Bij mijn willen.be domein hoort natuurlijk ook een algemene 'www' website, en die wordt voornamelijk gebruikt voor familieberichtjes. Ik gebruikte daar eerst een URL forwarder voor, naar een simpel Blogger webblogje, maar hier kreeg ik van mijn familie hier en daar opmerkingen door dat men graag iets interactiever wenste te werken. Daarnaast is het natuurlijk handiger en leuker dat verscheidene personen als auteur aan de slag kunnen. Je familie is tenslotte je eerste echte community ;) en daar hoort dus een multi-author website bij. Ik kon dus moeilijk doof blijven voor deze customer requirements, en vandaar de zoektocht naar een nieuwe blog engine.

Probleem is dat mijn hosting abo slechts één databank toelaat, en dat beperkt de mogelijkheden natuurlijk serieus; anders was het gewoon een tweede Drupal instance op de server zwieren, en klaar is kees. Er zijn enkele blog engines die met SQLite werken, maar da's tegen de commerce van de hosting providers gerekend, die liever multi-db accounts verkopen natuurlijk. Vandaar dat ook mijn hoster een njet antwoordde op m'n vraag om libdbd-sqlite2-perl te installeren. Zeer jammer, want ik was druk bezig om Movable Type uit te testen, wat er redelijk goed uitzag, maar toch problemen opleverde met zijn grootte.

Pivot dan maar. File based, mooie interface, maar wel met een antiek en onhandig template systeem dat je gelijk moet houden voor alle soorten pagina's die je site rijk is. Waarom er niet met blocks gewerkt wordt a la Drupal of PHPNuke is me een raadsel. Maar voor de rest geen kwaad woord van Pivot : familie leden raken er snel mee overweg, theme support is redelijk, en de mogelijkheid bestaat tot private pagina's, die enkel geregistreerde gebruikers kunnen raadplegen.

A new member in the periodoc tables list : the periodic table of the internet, featuring search engines, blogs and many more. For those interested, Internet has already been drawn as a world map, a metro map, a mindmap, a network map & a logo map. even an alternative periodic table exists.

Since yesterday, Belgian internet provider Skynet has removed access to the alt.binaries newsgroup on their news server. Skynet closed about 18 months ago the mp3 binary newsgroups under pressure of the ISPA, and now the complete alt.binaries tree has been removed. The binary news groups contained mostly porn and multimedia content. I guess an era ends for some people...
Rumour has it that the true reason behind this move is that the removal of multimedia content would be advantageous for their digital TV business.

Are you happy ? Then mark it on the map of Happiness !

There's much ado about the Google.be case : two newspapers won a lawsuit against Google for archiving their news stories, which conflicted with their business model, which wants subscription for older archived articles. The court favoured the Belgian newspapers and ordered Google.be to put up a legal note about this on their frontpage, and to remove the articles from their archives.

A lot has been said about this case: that it could have been avoided with robots.txt, that it was all about the money, etc. Francois Planque has a nice roundup. The case is IMO quite stupid, will cost the two Belgian newspapers quite some visitors (the number of people visiting sites through Google is unneglectable) and indicates a problem with court rulings and understanding of the internet. If I download a site page, this page is copied in at least three places : my browser cache, my local Squid proxy cache, and probably on the cache of my ISP's transparent proxy. Must all these versions be deleted after the page has been archived on the original site ?

However, I personally don't think this could so easily be resolved by the use of the robots.txt file : this file is only a 'recommendation' for internet spiders what to index or not. Many bots don't even scan the robots.txt file. I think that lots of website builders aren't aware how to properly display guidelines how to cache site contents : simply by use of the 'HTTP header: Expires' as stated in RFC2616 (read part 14.9 about Cache-Control). The Belgian newspapers easily could have implemented this by using an expiration period of let's say 24 hours, and still have a working business model.

The time that I read Slashdots comments is long gone. There's just too much noise, even with a +5 filter. In the case you really want to read the best ./ comments, head over to Seen on Slash, which accumulates the most relevant comments.

I reworked the main site of this domain, which serves as a memo board for my family members. Previously, the site was generated on a local copy of Postnuke, from which I generated static html pages. There was a time in the past that I enjoyed created websites from bare HTML, but that time is long ago. It revived shortly when I discovered the joy of CSS, but I still hate digging into HTML and Javascript code, and besides, getting IE to render code 'correctly' is such a drag anyway.

Unfortunately, I cleaned up my old computer a bit too thoroughly, when I removed the Postnuke database. So I tried to set up a recent Postnuke version on my new PC, but apparently Postnuke doesn't play well in a modern LAMP enviroment (Apache2, PHP5, ...). Postnuke seems a dead rotten corpse too, if you're searching for some decent themes, so I knew I had to abandon this crapware. I tried Blosxom, but that hasn't decent theming support, and I still didn't want to create a new theme myself, as it meant diving into HTML again.

So I decided to give up serving webpages myself, and to create a weblog on a blog provider. Skynetblogs seems to be sunken in a pool of advertisements, so I created the whole bunch on Blogsome, which uses Wordpress to serve content. Looks good, albeit a bit slow. I had to debug the 'Happy Birthday' calculator, as it seemed that it was ridden with Y2K bugs in Internet Explorer. (or IE would be stricter in processing Javascript code, which wouldn't surprise me).

This is cool : The Inquisitor is a Google frontend, which is an AJAX-style web application that retrieves web results and suggestions as-you-type. So you don't have to wait for the Google request to finish, and you get the first three relevant Google results while you're typing your search term.

In November 2004, Microsoft's second-in-command Steve Ballmer made some headlines by mentioning that Chairman Bill Gates was getting four million spams per day. At the time, Jef Poskanzer was dealing with a little spam problem of his own - he was getting around a million spams per day. He found it a little comforting that his problem wasn't quite as bad as Bill's. However, a couple of weeks later Ballmer corrected himself, saying he mis-remembered the stat and Gates actually gets four million per year. This means he was getting one hundred times as much spam as Bill Gates. He has written a tutorial explaining why he gets so much crapmail and how he dealt with it.

Google offers an interesting timeline of the Usenet archives, going way back into the past, and thereby offering a nice overview of some historical IT events. I once used the Deja archives to dig out my first post on the Usenet, but came no further than this post about my radio program at Radio Scorpio, dating from Jan 1st 1998.
Update : found my first post about Ultima7, dating from November 23rd, 1993 !