Thursday, 22 August 2002

What a day So, today we have some knob head - Peter Chernin, president of News Corp. - warning that the internet is full of crap, and we have AOL offering exclusive content, like we will all sit round watching music video's over 56k modems. Could some one fax these guys a clue? As long as I have been watching this change - since the BBS days - it has always been about people talking to other people in virtual communities. Why don't these people understand this? Because they can't figure out how to make a buck doing facilitating people talking to each other. Stupid wankers. 4:35:19 PM

A cooks tour of RDF Morbus on Semantic Web. My erstwhile co-blogger, Morbus Iff, is just to damn polite to link to this, but his Semantic Web 123 is... [Content Syndication with XML and RSS] Which links through to Morbus Iff's The Semantic Web: 1-2-3. Good schtuff. 4:23:35 PM

Book Addiction This is so cool. I found this site in my referers list. You can see there which books were recently mentioned on weblogs. I think that this kind of automatic gathering interesting news is a cool concept. Need to do that more. [Krzysztof Kowalczyk's Weblog] [dws.] That rocks much arse. Blog mined book references with links back to the original comments. This is going to get interesting. 3:37:53 PM

Fun to use search engine Kartoo - a strange search engine. KartOO is a new meta search engine with a graphical interface. It's based on a technology developed for 3 years by Laurent Baleydier and his team. KartOO is programmed in Flash although you have an optional traditional HTML interface. What makes KartOO interesting is that it elaborates the so called semantic links between results. Those links are represented by sinuous lines that link the balls. Amidst said lines you find a word that is the one the algorithm considers that links both results semantically. By hovering on top of it you can highlight the related balls. When hovering over the ball you can see all its related semantic links highlighted. http://www.kartoo.com/ [Jon Alsbury's Radio Weblog] » Good find by Jon, this looks interesting. I can't quite work out what it's doing as the results appear on a number of "maps" which don't appear to be connected.  Is it arbitrary as to what appears on each individual map?  Do the size of the result "balls" indicate relevance?  And what do the colours signify. It looks pretty amazing, and I get the feeling there is something powerful going on.  But I need help with this one... [Curiouser and curiouser!] I tried out this search engine using a topic close to my heart - jboss 3 documentation. Following it around I found a bunch of doco over at http://cosi.clarkson.edu/C4C/projects/ub/jboss/. It was definatly fun to surf using this tool. They have some interesting technology over there. This thin client shows the real limitations of HTML forms for user interface design. 3:24:36 PM

Spammers are learning quickly As forshadowed in Paul Graham's writings on spam filtering, spammers wised up. Quicker than I would have expected. Here is a spam I just received: From tlxtxdwb@hotmail.com Thu Aug 22 14:12 EST 2002 Received: from homer (homer.xxx.xxx.xxx [xxx.xxx.xxx.xxx]) by marcie.xxx.xxx.xxx (8.9.1a/8.9.3) with ESMTP id OAA05638 for ; Thu, 22 Aug 2002 14:12:59 +1000 (EST) Received: from mailfilter1.xxx.xxx.xxx (mailfilter1 [xxx.xxx.xxx.xxx]) by mail.xxx.xxx.xxx (iPlanet Messaging Server 5.2 (built Feb 21 2002)) with SMTP id <0H180006G913VC@mail.xxx.xxx.xxx> for bjmorgan@xxx.xxx.xxx (ORCPT brett@xxx.xxx.xxx); Thu, 22 Aug 2002 14:12:40 +1000 (EST) Received: from 12-217-13-186.client.mchsi.com(12.217.13.186) by mailfilter1.xxx.xxx.xxx via csmap id 23473; Thu, 22 Aug 2002 14:14:13 +1000 (EST) Date: Thu, 22 Aug 2002 14:12:40 +1000 (EST) Date-warning: Date header was inserted by mail.xxx.xxx.xxx From: Informer <info@yahoo.com>; Subject: Please check it now .. To: brett@xxx.xxx.xxx Message-id: <0H180006H913VC@mail.xxx.xxx.xxx>; MIME-version: 1.0 Content-Type: multipart/alternative; boundary=-12zz.12 Content-Length: 529 Content-Type: text/plain; charset=us-ascii http://www.rapemux.com/?revshare=02657dca 2:35:05 PM

Radio issues [...] Thank you so much for these helpful instructions. Radio reallly should have this as a standard macro. However I still get this error... [Macro error: Can't evaluate the expression because the name "channeltitle" hasn't been defined.] Any ideas? Did you take the subs.txt macro from here [James Strachan's Radio Weblog] That is the exact issue I am having. Someone mailed me to say that the issue is resolvable by finding the subscription without the channel title defined, and zapping it. I wish I could remember who sent that. :) My problem is that I currently have 150+ subscriptions, and being the coder that I am, I would prefer to be able to write a script to find out which one is broken. After 6 hours of searching in vain through the doco, I gave up. I am almost tempted to roll my own aggregator using Mark Pilgrim's ultra liberal RSS parser ... 11:45:33 AM

Signs of impending Doom Telstra Considers 45,000-Seat Linux Deployment [Slashdot] Hoooboy. Mickeysoft is not going to be happy about this. Or, more likely, this is Telstra doing some license negotiation with Redmond. Because if Telstra, like y'know, changed to Linux, the whole world might come to an end. Or something. ;) 11:27:19 AM

Bootstrapping a distributed community Blog syndication clarfification clarification. Brett noted that using a server for the categories and blog lists would constitute a single point of failure, and a point of authority that may not be so nice. I agree, but what I was *really* thinking was to use JXTA to form a server network so that this is avoided. I only mentioned SOAP because it'd be a decent first test. The next generation would be JXTA based, most likely, to avoid the issues Brett mentions. Touché. ;-) [Random thoughts] Y'know, if you actually used something like jabber, then current Radio weblogs could also play along as well, solving the large bootstrap problem ... :) 11:23:39 AM

© Copyright 2002 Brett Morgan.

[Macro error: Can't call the script because the name "nextDayLink" hasn't been defined.]