Several new and interesting XFN-related posts and a new XFN tool prompts this new entry on XFN right after my previous post, Friends, XFN, and Hyperlinks.

[Integrated Web Design: Building the New Breed of Designer & Developer, by Molly Holzschlag. Published September 2002 by New Riders.]

XFN, FOAF, and More Info

09:50 am, pdt 8 April, 2004 Comments, Trackbacks (2) ·';}?>

Categories: Development, Information Architecture, Internet, Weblogs


Comments, Trackbacks: 2 so far. Add yours!

  1. Yeah, I looked at the Eric Meyer article your mentioned but he didn’t really have an argument ether way. He noted

    FOAF makes it possible for other people to provide information about an author, whether or not said information is accurate.”

    but this is also valid against XFN. Friendster has had this problem of people adding people as friends who aren’t, but since this is a controlled environment it can require consent from both parties before a 'relationship' is established. This is impossible for XFN and FOAF. Reciprocating relationships is a solution but this becomes clumsy as the relationship information is spread across two locations doubling the chance that it will become inaccurate and increasing work load on both ends (there are tools to help you manage this but why would someone want to add another step to the development process? and what about pages generated by CGI? These tool are of no help).

    This seems to be a 'hammer before a nail idea' and until there are any real advantaged to XFN I don’t see way people would make the effort. It just further complicates the HTML standard. If you were to follow the CSS or RSS (RDF) examples you would separate the 'network semantic' from the presentation. You’d have your external links in a separate file which describes these relationships and use a link in the header to point to it, but isn’t that way FOAF is?

    The advantage I see in FOAF is that people can provide a URI that says who they are and is centralised so instead of membership forms people could provide a URI that has all the information that they want to share and will be easier for the person to keep their contact information up to date for everyone who uses it.

    The XFN idea seems reasonable, although the 'rel' attribute seems to be to help describe a documents relationship to others not its authors. So it’s a corruption of the HTML standards 'intent'. This is not bad in itself as I think standard should support good ideas not fight them, its just what is the long term vision behind XFN and is this the best way to achieve it?

    11:34 pm, pdt 9 April, 2004Comment by CpILL

    comment #1 permalink ·'; else echo '·'; ?>

  2. tantek’s post quotes Molly who quotes Matthew

    “Mullenweg looks to the simplicity of XFN... you don’t have to learn... or mess with strange files.”

    Matt, praising XFN’s simplicity, in the next breath offers “his new tool?” Why would we need a tool if it was simple? The answer is in the “you don’t have to learn” part quoted before this offer which seems to appeal to the frontend-types aka 'designers' who move to the web and don’t want to learn anything new (particularly if they have come from print, I find). XFN could be seen as social networking for dummys. But dummys are important too, that’s what democracy is all about, isn’t it?.

    tantek also goes on to quote The Underground Economist

    “I don’t yet have a decent mechanism for keaping my FoaF file up to date...”

    which says a lot. I think the complexity of CSS is a big problem and makes its uptake a problem amoung designers. If people looked at CSS, or RSS for that matter, with the same view where would we be now (probably at the same point as no one was consulted when they were thought of, but hey, good idea bad execution)?

    The point is that learning is good and shouldn’t be avoided (all you creative types). If there were more learning and less... no-learning the world would be a better place. For sure!

    12:10 am, pdt10 April, 2004anonymous comment

    comment #2 permalink ·'; else echo '·'; ?>

This discussion has been closed. Thanks to all who participated.

*/ ?>