Sam Ruby recently launched a wiki for the Echo Project (which needs to be renamed) which has the following goals

  • We want a weblog authoring tool to be able to post log entries to all sort of weblog engines. Prerequesites:
    • common API to the weblog engine
    • common markup for the entries
    • common meta information such as author, time, place, etc.
  • We want to read weblogs using a variety of means, including various transformations by software. Prerequesites:
    • standard output formats such as RSS with a specified list of optional well-defined modules (and plain HTML, of course)
    • common API if the queries are allowed (eg. all log entries in a specific time period)

Although people like Sjoerd Visscher and Ben Trott have mentioned why the "Echo: The Syndication Format" is different from RSS, no one has actually bothered to state why the why "Echo: The Weblog API" which is supposedly the primary reason for Echo existing would be better than the status quo (the MetaWeblog API and others)

Below is a more in-depth exposition of my earlier post describing the limitations of current weblog posting technologies and the problems Project Echo is supposed to fix. Specifically, I will tackle the MetaWeblog APIs since certain parties who are personally invested in it have engaged in a FUD campaign and have given the impression that there is little if anything fundamentally wrong with it.




June 28, 2003
@ 12:58 AM

It seems Simon Phipps has stared a list of known bloggers that work at Sun Microsystems which led me to Norm Walsh's blog. Yet another one of the XML gods has started blogging. Sweet. By the way, Simon's list makes a decent companion to Joshua's list of bloggers who work in the B0rg Cube. I wonder when someone will start similar lists for IBM, Apple and Sun folks.

So far Sam Ruby is the only IBM blogger I've seen and the most visible Apple blogger I've seen, David Hyatt, has hard to turn off comments in his blog due to being flooded with bug reports. Interestingly enough I haven't seen similar things happen on the blogs of the various B0rg folk including mine. Not that I'm asking people to post bug reports in my diary. ;)




I stumbled on the The Big Picture of the XML Family of Specifications which lists a large number of technologies that are related to XML in one way shape or form. It seems some people take a look at the diagram and it gives them the impression that XML is too complex after all, just look at all those specs. This is amusing given that one could probably run out of printer paper if a diagram of all the specs that were related to or dependent on ASCII were placed in a diagram.

An interesting fall out of this has been that some fellow B0rg have posted their opinions on what they consider the core of XML. Unsurprisingly, I disagree with some aspects of both their posts. My opinions on the what constitute the core technologies for XML development are below.




June 20, 2003
@ 12:58 AM

TITLE: On the GPL and the Devaluation of Software




June 19, 2003
@ 12:58 AM

Based on some comments by Beth Goza at a meeting yesterday I noticed that a number of the blog kids are engaging in a What is a Weblog? debate which was kicked up a notch with a decision by the grand arbiter of funk to step into the fray.

I remember the first time I heard the term "weblog" about four or five years ago and it was used specifically to describe Slashdot. Come to think of it this whole debate sounds familiar...

Here we go, people were having it four years ago. Here's a Jon Katz piece on Slashdot about weblogs being the "new gated community of the WWW" and it links to a four and a half year old article called Anatomy of a Weblog which seems identical to the one or two pointlesss philosophical meanderings I've seen in the most recent incarnation of the debate.

A bunch of links and quick opinions below.




A few days ago I posted an entry about spending 15 minutes writing an XSLT stylesheet to save 1 minute of effort. So it turns out the 1 minute of effort gave me the wrong results. The problem was how to add 4 months and 20 days to a column containing dates in an Excel spreadsheet. It turns out the change one value, click and drag approach doesn't work because it adds one day to each subsequent date instead of adding the changed amount to each date. This means I get a column of dates in chronological order instead of having X days added to each date in the column.

Below is an XSLT stylesheet that adds 4 months and 20 days to every column containing a date in an Excel spreadsheet (saved as XML in either Excel from Office XP or Office 2003 beta 2)




Recently I've been bored with working on RSS Bandit and it has begun to seem more like work than fun. I wondered why until I revisited the reasons I had for building RSS Bandit in order of priority

  1. Investigate how hard it was to embed Internet Explorer in a C# forms app.

  2. Write an app that makes it easier to read and post comments to Joshua and Sam's blogs.

  3. See for myself how difficult it is to write a .NET framework application that consumes and produces XML. Note places for improvement in our APIs.
Then I saw the problem, all my itches had been scratched. This was until I saw a screenshot of RSS Bandit in Torsten's blog and was inspired.

Below is an attempt to organize my thoughts around the potential for RSS Information Aggregators and the dual themes of XML Everywhere & Unified Query Data Access.




Yesterday my level of XSLT geekery ratcheted up a notch. I was updating a schedule within a spreadsheet and had to update all the dates in a column by 4 months and 20 days. So after asking a few people who didn't have a solution I decided to save the spreadsheet as XMLSS then wrote an XSLT stylesheet that used my EXSLT extension library (for date functions) to update the value in that particular column. Of course, right after I did this one of my coworkers pointed out that all I had to do was change the value in the first cell in the column, left-click and drag to update every other column by however much I updated the first cell. His way took a few seconds while mine took about 15 minutes and was much more complex to boot. Geek.

Speaking of geeky distractions, you need to try out this very impressive online guessing game.

Ruminations below on the AI Winter, why derivation by restriction should be avoided in XSD, and Slashdot on mass walkouts.




I got back into work with a vengeance this week. I had six meetings on Monday and I just realized I that besides this milestone's spec deadlines I'm supposed to complete articles for both O'Reilly's and XML Journal plus there's this article that I've owed MSDN for months that I just finished this morning. Sheesh!!!

After Mark Pilgrim's hack I decided to get with the program and now strip potentially malicious HTML from RSS items in RSS Bandit. That and a bug fix constitute v. of RSS Bandit (Oh, and I now include the OPML file for Blogs @ GotDotNet with the install). Get it here

Thoughts on George Bush's resume, the Matrix Reloaded (shitty plot, fantastic action scenes), and a blog parody that's making the rounds.




June 11, 2003
@ 12:58 AM

The newest version of RSS Bandit is available. The primary improvements in this version are performance related specifically the app is more responsive, better support for handling feeds with errors (thanks to Justin Rudd's feedback), Outlook 2003-style smart folders, ability to view all items in a particular category, plus lots of drag N drop support for doing things like subscribing to feeds and moving nodes from one category to the other.

Old users will be glad to know the installer no longer requires you to uninstall previous versions of RSS Bandit but is now smart enough to figure out that it should overwrite previous installations. I apologize for not reading the docs in the first place and figuring out how to do this in the past.

You can grab the latest bits here. More details about changes in this release provided below.




It's amazing how you can use a piece of software every day and still not be aware of fundamental yet extremely cool features it has. I just noticed that Torsten added code to RSS Bandit a while ago that highlights in red any blog post that references your homepage link. Here's a screenshot of it in action. More comments below on related RSS Bandit features.




June 8, 2003
@ 12:58 AM

So I'm back. I've posted in general about travelling and even put up some pics. This post contains all the pent up XML geekery that I haven't been able to post over the past few weeks. Thoughts on passing XML around in the .NET Framework applications, links to the slides and code for my TechEd 2003 talk, opinions on Don Box's TechEd 2003 keynote, a note to reporters who are considering asking me about "Microsoft's official position on RSS", RSS Bandit updates and beating the W3C at its own game.




June 7, 2003
@ 12:58 AM

Most of my travelling is done. In the past 3 weeks I've been in Los Angeles, Vancouver, Lagos, London, Dallas, Abuja, as well as brief lay overs in Chicago and Houston. I have a bunch of stuff to blog about from the Nigeria presidential inauguration and the international first class travel to the poor reception of my TechEd 2003 talk (entirely my fault) and RSS Bandit goings on.

I'll probably post a few entries tomorrow to cover everything. For now I have a poll which I'd appreciate if folks answered.

Poll: Most Effective Weight Loss Mechanism You've Tried?