I've been following the Echo ProjectThis essay contains a number of links to a Wiki. The dynamic nature of a Wiki means you may not see what I saw. Caveat lector. Oh, and the name of the project may change, too.(And thanks to Dorothea Salo for pointing me to the proper latin phrase.) as best I can. Conceptually, it seems like a good idea: the current state of affairs with respect to RSS is pretty messy and this effort would give us a fresh start. I've already voiced support for the idea.
But I'm really startled to see that what started with a pretty simple conceptual model consisting of author, date, link, and some content has blossomed into something deeply complex with titles, subtitles, links, multiple dates, multiple authors, contributors, and content variations with explicit MIME types.
Some of these discussions have actually gotten tangled up in the differences between CDATA and other forms of escaping markup in XML. It's tick season, folks, if you're that deep in the weeds make sure you check yourselves carefully. :-) (That's a smiley folks. You know, humor. No flames, please; no disrespect intended.)
Perhaps I completely misunderstood the goal of this exercise, but I was expecting the Echo core to consist of about five elements:
<entry xmlns="some/appropriate/uri"> <publisher>Norman Walsh</publisher> <link>some/permalink/uri</link> <date>2003-06-26</date> <description>Well-formed XML.</description> </entry>
End of story. (More or less, anyway.)
Now, make no mistake, I want other stuff in there, but I thought that extension modules were for that purpose. Some people have a deep aversion to namespaces and that perhaps drives them to put “everything that might reasonably be needed” in the core. Uhm. No, thank you.
I would expect some small amount of standardization around each of the extension modules, but I'm happy to let a thousand flowers bloom there.
Let's standardize the core today. We can come back and standardize the extensions later, after we know which ones are really useful.