Internet Alchemy

Grandfather of weblogging Jorn Barger has written another weblogging related essay: An Internet way of self-knowledge:

The use of journalling for self-discovery is a longstanding and well-studied phenomenon, but traditionally its privacy has been seen as its greatest strength. When you write for the public, you naturally censor and glamorise the truth, which largely defeats the purpose of honest self-discovery.

The discipline of logging every web-article you find interesting and spelling out your reaction takes the journal in a very different direction, because unlike the private life- events of a diary, every weblog-reader can share the exact same experience of reading the web-article, and so know exactly what you’re reacting to.

The act of self-censoring also becomes harder to rationalise if you’ve just read an interesting article, and you find yourself thinking “Dare I admit that I enjoyed this?”

I missed this one somehow, a new Internet draft of the JXTA v1.0 Protocols Specification.

The JXTA protocols are a set of six protocols that have been
specifically designed for ad hoc, pervasive, and multi-hop peer-to-
peer (P2P) network computing. Using the JXTA protocols, peers can
cooperate to form self-organized and self-configured peer groups
independent of their positions in the network (edges, firewalls,
network address translators, public vs. private address spaces), and
without the need of a centralized management infrastructure.

A nice unprofessional way to upgrade your website from British Telecom, one of the UK’s largest companies: We are currently upgrading this site. Full service will be restored shortly. Below is a list of useful numbers that may be able to assist you while the site is down…. That’s the on the home page of bt.com! There are currently no links at all from that page, just a list of customer service numbers.

I have been looking at how mod_link can be used to replicate or
replace what I have been developing with OCS. The key function of OCS
is to provide user-agents with enough information to make a choice
about which channels to retrieve and display for the user and how
often. Currently OCS can describe alternates by schema, content type,
encoding and publishing schedule.

Continue Reading »

Now here’s an interesting challenge: take a Basic program from the seventies, riddled from top to toe with gotos and single letter variables and apply modern refactoring techniques to produce a version comprehensible to the casual observer. That was
Alan Hensel’s goal. Looking at his result I think he did a pretty good job.

I thought it was time for a makeover of Internet Alchemy. A few truly minor tweaks to the underlying XHTML and a new stylesheet. Hey presto! Cleaner, easier to read - it could almost be 1992 again!

Just for the fun of it, here’s a look back at Internet Alchemy through the ages (courtesy of the Wayback Machine):

• October 1999 - six months in, still the original design. There used to be a banner along the left hand edge which doesn’t seem to have been archived.
• February 2000 - nearly the anniversary and a change of posting style.
• May 2000 - yet another redesign and the first announcement of work starting on OCS 0.5!
• September 2001 - disaster strikes. I lose control of the openjava.org domain due to a mishap with email. A porn site takes control!
• June 2001 - another redesign, the last until today’s. This is about the time I ported Internet Alchemy to run on the extremely cool (IMHO) XML Engine I developed at Calaba. That’s technology that’s now buried deep within another company, unlikely ever to be resurrected.

It’s interesting to note that there are no archive entries for the whole of 2002. I wonder if they’ve stopped crawling or if they’ve just not placed the 2002 archive online yet?

This list is gleaned from my server logs. If your RSS reader supports ETags or HEAD requests let me know and I’ll add it here:

• AmphetaDesk
• NetNewsWire - latest beta
• Cupcake
• Aggie 1.0 Release Candidate 3
• Radio Userland - testing code, see Dave’s post
• blagg with DJ Adams’ modified wget.
• Wildgrape NewsDesk [added 3 Feb 2003]
• nntp//rss [added 3 Feb 2003]

There seem to be a lot of home-grown aggregators too using Curl on Darwin, WinHttp, Perl libwww, etc.

[updated this list 3 Feb 2003]

Here’s bunch of other XML difference tools:

• IBM AlphaWorks XML Diff and Merge Tool - Java based, uses tree matching based on ID attributes or content
• Dommitt Unordered Diff - ignores order of nodes in document (?!), Java based, flattens DOM trees to arrays then does step by step comparison to generate insert, delete, update and nochange operations.
• diffxml - GPL, Java based, algorithm detailed in author’s dissertation [postscript]
• XML::SemanticDiff - Perl module, artistic license, event driven parser, seems to handle namespace changes
• DeltaXML - commercial application, Java based, tree matching, root elements must be the same

Some other pointers that might be relevant: XUpdate [xmldb] (appears dead), X-Diff Algorithm [citeseer], Algorithm::Diff perl module.

It seems perverse to me that the diff example in Microsoft’s XML Difference language is larger in bytes than the original and final xml added together. I’m looking for an XML diff that is more efficient to transmit than simply sending the new document.

Chris Sells has made his two WTL papers (WTL Makes UI Programming a Joy) available for download from his site after they were removed from DevelopMentor’s. Both are in zip format: Part 1: The Basics and Part 2: The Bells and Whistles. These are excellent introductions to Microsoft’s WTL.

While you’re getting those it’s worth browsing the rest of Chris’ writings.