Dave’s made it to the web with his OPML Validator Beta. On the surface it seems a little more strict than Randy’s validator.

Dave’s linked to the caveats of the beta-ness of the tool, as well, which is pretty nice – especially the presence of the valid and invalid test files. As I linked previously, he’s noted some prior art in this area that he’s used in this tool in the Guidelines for Validation.

In my case I had cribbed someone else’s use of the outline type “title” for outlines that contain outlines and had to refer to Randy’s work on common types. As Dave noted, he’s used that for his work on the opml.org validator. Without referencing Randy’s notes I would have been hard pressed to make a determination as to what to do to fix the “invalid” types – that last in quotes because by Dave’s own admission in the comments to the Caveats post, the types are still up for discussion. It would, I think, be nice if Dave pointed to out-of-the-norm practices such as this as warnings (as opposed to what amount to errors), and to leave the error condition for things that actually violate the spec, which is silent on acceptable type-values (the spec says: “type is a string, it says how the other attributes of the <outline> are interpreted.”). Either that or Dave could point to the source he’s using for valid types in the actual error message. But these are small things, and I swiftly modified my OPML generator to output OPML that’s correct per Dave’s validator, which has an undeniable stamp of authority.

Thanks, Dave!

update: small copy-edits

This entry was posted on Friday, October 28th, 2005 at 9:52 pm and is filed under OPML, Technology, development, xml .
You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.