When I render my outline using Joshua's stuff, Les Orchard's transcluded outline is directly included, not linked. Make sense? Yes and no. It can be argued either way. We geeks are going to have infinite amounts of fun with this stuff. [Jon's Radio]

The outline renderer can do both. It should provide the link and dynamically load the transcluded outline when the user expands the link to the opml file.

RFC: MetaWeblog API. "It's time to broaden the XML-RPC pipe that tools use to connect to Radio, and in doing so offer an evolution to the art of scripting weblog tools." [Scripting News]

To add a sprinkle of the semantic web to the api: Instead of the postid, use permalinks. The cool thing is that permalinks are globally unique. So maybe you could call getPost to a community server, which knows where to get the posts. And an edit tool using the api can provide a link to the online version of the post. Blogid could also be changed to the home url of the blog, but I'm not yet sure if that will actually help, or if it is possible for every blogging software.

To clarify my previous post, I made a little demo that transforms OPML to XML. You can see how you can add opml:text and xlink:href attributes to any XML file. Note that OPML is just fine. It's just me enjoying the exploration of theoretical alternatives.

Patrick asks what the difference is between XML and OPML. OPML has the advantage that it has defined semantics for a few attributes, while being able to represent all XML structures. The text attribute is especially useful, because it gives a reasonable textual representation of each node. And that gives nice options for user-interfaces.

XML has the advantage of having all kind of tools and schema languages. And with XLink XML now also has linkink, just like OPML. Which made me think: OPML shouldn't have been an XML format, but a set of attributes, using an OPML namespace. First you have the elements in the head. They can all be attributes on the root node and are all optional. Then each element can have an opml:text attribute. Simply adding the OPML namespace definition and opml:text attributes can make any XML document editable by an outliner. Could that be of any use?

Mozilla 0.9.9 supports MathML. And it looks great, certainly with ClearType enabled in Windows XP. Note that any serious business with XHTML is not possible with the latest version of Internet Explorer. Even when IE6 doesn't choke on the DTD, it has no clue that it is expected to do something with the XHTML data.

This is a test for the autogenerate link option.

I updated the weblog sources page and the OPML Loader. These pages now have their own specific sidebar OPML files, which you can load with the OPML loader. I mailed Dave Winer once how OPML could have been much cooler with a more general approach. But some of the nicest demos on w3future are OPML related, which shows clearly that OPML is more than good enough. "Dare to keep it simple" is something Dave keeps reminding me of.

Sadly the OPML Loader suffers from cutting edge problems. Both IE and Mozilla screw up in completely different ways.

At Q42 I created a color blindness simulator using the filters of Internet Explorer. There are in total 6 layers of filters and it's horribly slow, but it looks very cool on photographs.

On my archive pages I added links to the next and previous archive pages. I also added top, prev and next <link> elements, so you can navigate the archives with the Site Navigation Bar if you use Mozilla. And as these macros aren't site specific, I thought I could just as well make them available for download.

Intentional Programming FAQ. (via comp.compilers)

Intentional Programming (IP) is an extendible programming environment based on Active Source which was being researched by Microsoft. Program source is represented as a data structure rather than plain text. One of the most exciting concepts is that you can implement a system using domain-specific abstractions.

Some of the links look interesting. IP is related to aspect oriented programming and subject oriented programming. Last time these were discussed I mentioned my doubts about their adoption by the programming community. I am still interested to hear what others think about these ideas in general, and any practical experience you might have. [Lambda the Ultimate]

One of the links, "Intentional Programming" A Talk With Charles Simonyi, is very interesting:

It's not the languages that are of value, but only the abstractions that the languages carry. [...] It's just that we don't think of abstractions without languages, because languages used to be the only carriers for abstractions. But if you could create an ecology in which an abstraction could survive independent of everything else, then you would see a much more rapid evolution for abstractions, and you would witness the evolution of much more capable abstractions.

I think this will happen within the next 5 years. It could be the "killer app" of RDF.

Ross Gardler just received this from an Xopus developer: "The next version will be open sourced once it has reached the release level. The next Xopus will be IE5.0 and Netscape 6 compliant." [Sam Ruby's Radio Weblog]

It's cool to read that Sam Ruby is interested in Xopus. Of course Xopus is less cool without IE5.5's contentEditable. But it's nice that Xopus will be cross-platform. And we also already noticed an IE-only product is not really taken serious in the open source world.

Mindblowing!. RPC to clients without a fixed IP address. The aggregator posting its stories to a chat room so anyone can read them 'live'. A remote command line interface with Radio Userland in your chatclient. A good programmable client (and server?) like Radio is exactly what Jabber needed.

In the long run the scripting culture has to win. Tightly coupled and brittle interfaces aren't particularly enduring. We also need a protocol which enables people to initially develop simple interfaces and then gradually evolve them to be more adaptive without breaking deployed clients and servers. [Sam Ruby's Radio Weblog]

I could not agree more with this. The software world is becoming more dynamic, not less. We need more runtime tools. Here's some interesting reading along those lines. [Patrick Logan's Radio Weblog]

We need more freedom, that's true. But type freedom the way Javascript has is not the solution. Type filtering is something I'm looking for. In a far from perfect way this is possible in XSLT. For example a template that requires an x element with an a subelement, would have a match="x[a]" attribute. If then later x elements need more subelements, the template won't break. And if the template needs a subelement, say b, you change the match to match="x[a][b]". And then x elements without b will simply not match anymore. There's no "b is not defined" error.

Ps I am aware of Sjoerd Visscher's OPML Loader web app but he hasn't responded to my e-mails. [weblog.masukomi.org]

Oops. Well, better late then never, so the OPML Loader is now available for download.

And of course, most of them were from Radio sources that use 0.92 RSS and don't send an item.link at all, at all. [jbond's blog at voidstar.com]

It's probably on Userland's ToDo list. But I fixed it for my installation. It's the most convenient solution for bloggers who would want to blog my blogging. However, if I'm blogging about another webpage (which is usually the case), the link should actually contain the link to that item. But in blogging practice, the comments are more important than the original item. (And the original link is usually in the quoted part.) Or maybe that's a bit too general. But there certainly is a big difference between blogging news or blogging blogs.