|
|
|
23 May 2005
FOAF, not Neutrino
Phil Wilson explains why one should use FOAF and not Neutrino. I do not think his comments on OPML are very good. Because it is already used for a BlogRoll, you can not use it for a PersonalFeedRoll. It is indeed a problem that OPML can be used for anything. But then prohibit it’s usage is yet another thing. Maybe they should not have used OPML in the first place. But the fact that is that it can be used. Danny Ayers talks about OPML autodiscovery. That seems to fix OPML even further. Maybe it is time to talk about BlogRollOPML.
Phil Wilson likes FOAF for a PersonalFeedRoll. I agree with him there. I see FOAF more as a personal profile. But his comments on using autodiscovery instead of the PersonalFeedRoll I do not understand. I like autodiscovery to designate the associated feeds, but it is not a method to present the all the feeds a users has (I have 10 feeds Atom and RSS). The argument that the URL might change is bullshit. Anything might change. Time to have a PermaURI, which generates a FOAF-file based on the latest information.
In a comment to that post Philip Ringnalda defends Neutrino, mainly with the idea that one should be able to subscribe to the PersonalBlogRoll. But then it should go to my FOAF. Is equally good and a more preferred method. Anything that I want to publish should be in my FOAF and anybody that wants to know something about me should look there.
Categories/tags:
format
microcontent; foaf; neutrino; blogroll; opml
PermaLink •
(2)
Comments •
TrackBacks •
Trackback URL
Comments
You’re right: my comments on OPML are not very good, I left them as such deliberately because I have a personal dislike for it and didn’t want to spend more time talking about it (e.g. it is impossible to validate because you can have arbitrary attributes, has no normative schema, etc.).
I do think that autodiscovery is a good thing. In use on this site, you have it set up in exactly the way that it most useful for users.
I don’t really understand what you mean when you say “The argument that the URL might change is bullshit. Anything might change.”. What I say is that if you use autodiscovery then you never have to update your Neutrino file yourself and any feed URL change becomes transparent. This assumes that the Neutrino file and FOAF file don’t move. If they do, then if you have control of your server you can just use an Apache redirect to point at the new version, but if you don’t, you can use rdf:seeAlso in your old FOAF file to point at the new version; you can’t do this for a Neutrino file (I’m not sure I’ve answered your point though).
OK, sorry. I was a bit harsh there. I like the method of autodiscovery for various files. I guess we will see it more often. Maybe it should be a separate file at the root of a website (autodiscovery.htm). Or is that the point you are making?