10/16/2005; 1:26:32 AM
I've converted my rss 2.0 feed from <content:encoded> to the more bandwidth and xpath friendly <xhtml:body>. It looks like gotdotnet and blogx users will soon follow. Hopefully the owners of the wellformedweb and w3future weblogs will take notice.
The updated feed is valid, and it uses namespaces in exactly the way that rss 2.0 and xhtml intend. I've tested it with radio and syndirella. [Sam Ruby]
Very interesting, this certainly is an improvement, and very much in the spirit of RSS 2.0. I tried to implement this with Radio's RSS callback feature, but only string values (and attributes) are allowed. One small note: Maybe xhtml:div is better, because xhtml:body only allows block-like elements.
My only remaining concern is that Sean's piece could be seen to be adding to the misconception that RDF is especially difficult. In his response to Sean's piece, Uche goes for the throat (in the nicest possible way). Classic quote, easily worth bold and italics :
"...anyone who can't get RDF can't get relational databases or any other sort of formal information modeling, and they can't get code..."
I'm wondering how many people actually get what relational modeling is about. A lot of people use SQL tables, but a lot of them do that by following some rules they learnt at school. But there's a huge difference between XML or relational databases and RDF. If the XML model or the relational model works for you, you're done. Even if the design goes against all rules. But the whole point of RDF is that the model has meaning. If you screw up, and the meaning conveyed by your RDF data is different from what you want to say, the data is effectively useless.
I really think that creating an RDF model is difficult. You not only have to wonder “does this model work for me”, but also “does this model convey what I want to say”. RDF is the only standard format that conveys meaning. That's what is great about it, and the only thing that's great about it. And your RDF is useless if it does not convey your meaning. Maybe it's not even so hard to use RDF to express what you want to say, but it is hard to say what you mean. People constantly make mistakes and don't say what to mean. (“You didn't read your e-mail?” “No, I didn't.”)
Sean McGrath: “I think it's time for the Semantic Web proponents to stop trying to teach us all to think at their level of abstraction.”
Great article. Of course, the proponents are quick to reply. Danny Ayers seems to forget that using RDF is not enough to interoperate. You'll need a common vocabulary. If all parties want to use this vocabulary internaly, they'll argue for years to get their little details in, which most other parties don't care about. If on the other hand all parties create their own internal format, they'll be able to get everything up and running. The experience grows, and it'll be a lot easier to agree on a common vocabulary of the data that needs to be shared among the parties.
Uche Ogbuji thinks that Sean is arguing to quit RDF. I don't think he does. Some tasks (like gripping monkeys) don't have any better tools than monkey wrenches. If you acknowledge this, then it's obvious that development should focus on tools that allow everyone to use what they understand, instead of explaining RDF over and over again.
Searching for URLs in Feedster and subscribing to it works great! I'm subscribed to the results for “http://w3future.com”, and there are more results than in the referrer statistics. Some blogs don't generate referrers because they are low traffic, and/or they repost my whole post so there's no reason to go to this site.
On Thursday, March 20, the U.S. State Department released a list of 42 nations it describes as the "Coalition for the Immediate Disarmament of Iraq.". In the Netherlands we got a little behind the scenes look today of how weak this coalition actually is. I watched the briefing of general Franks today. And to my huge surprise he was flanked by a Dutch officer! The Netherlands support the U.S. politicaly, but are not participating in the war. So I wondered what the hell this guy was doing there. As it turns out, the Dutch ministry of defense was asked if he was allowed to “attend a meeting”. He is a communications officer helping out with patriot installations in Turkey. That the U.S. needs him to be at this briefing shows how desparate they are to show that they have a coalition.
The road to XHTML 2.0. I'm positioning myself to be “Mr. XHTML 2.0”. Because, you know, what the world really needs is more irony. (83 words) [dive into mark]
Doh! And I thought I could be Mr. XHTML 2.0!
Thanks, Joeri! I enter the story halfway the Basel 2003 page, where it looks like I'm thinking very hard while drinking coffee. But it's just that my eyes are still in winter mode.
I just spent an hour looking for this document: Common HTTP Implementation Problems. I post it here to be able to find it back the next time. With Google I rarely have to search longer than a minute, so I was terribly frustrated I couldn't find it. I even started wondering if I just only dreamt it. It turned out my memory of the document was simply too vague. Maybe if I link to it with my own keywords, the next time it might be ranked a little higher in Google: Content Management URI design.
I'm impressed. I installed Mandrake 9.1SP2 on my laptop, and that went very smooth. The installer even resized the Windows XP partition. KDE looks nice. But I can't get used to the fonts. And Mozilla doesn't render pages as well as in Windows. OpenOffice is installed, which is handy as I don't have MsOffice. And finally I can read PostScript files without hassle. I'm off, toying with a few drivers.
Scott Johnson (maker of Feedster) mailed me to say that the search results already are available in RSS. It's the XML icon at the bottom of the result page. That's just great!
Towards structuring blogging, Sebastien Paquet: I believe a critical element to get a sustainable system is for people to get reasonably quick feedback in return for the extra effort expended in creating metadata. [Sam Ruby]
Actually I'm more and more loosing faith in anything structured in blogging. It's too hard to do right, and too much work. Also it is hard to agree on categorisations and topics. (How often did you reorder your bookmarks? I've given up.)
But I've not given up on the semantic web. It's just that the software needs to be smart, not the data. But in this case that's not even necessary. To know if two post are about the same subject is to see if they link to the same location. For example: to know which blog posts talk about “Towards structuring blogging”, you can simply search for it's URL with Feedster. If Feedster could provide an RSS service for search results, you get functionality comparable to what is suggested in “Towards structuring blogging”. And that is without any change in blogging tools. And the blog posts improve too, when they use more links.
One problem remains, a problem in general for the semantic web: Often there's not one unique URL for one topic. Try to find the URL for Radiohead's album “Kid A”. If everyone does that, we'll probably end up with about 10 different URLs. And at least half of them will break within one year. If the semantic web is ever going to work, we'll need to actively advocate the creation of permalinks for all popular topics.
And for those who don't believe in the power of linking: it's exactly what your brain does.
Today at Q42 Laurens looked into Microsoft Office 2003 beta 2. After reading John Udell's article about InfoPath we thought Microsoft migth have caught up with Xopus. Thankfully this does not seem to be the case:
- Infopath 2003 uses its own templating language. It can only export to XSLT.
- In Word 2003 you can edit a raw XML file and save it. You can also apply an XSLT transformation and edit the styled result, but you can no longer save the XML data separately after that.
Office 2003 is actually a good developer tool to use with Xopus. Infopath is a nice editor for data schemas and data views. The schemas and views can then be used with Xopus to edit XML online in the browser.
I'm currently enjoying a week vacation. I'm staying in Basel, Switserland, at Joeri's place. Marco is here too. It's perfect weather and we're going to take a little tour through Basel. Joeri and Marco both have a camera with them. I hope they can get some pictures online soon.