I would voice a few worries... 1) Dublin Core spent YEARS looking at metadata about 'things' and they narrowed things down to a handful of properties. It has been brought-up before to possibly model DC terms.
2) when you look for 'real-world' examples you will find "things" everywhere. There are no shortage of "things"/"items" on the internet, and while it would be great to model them, it is almost back to the "boiling the oceans" senario. I'd suggest a more narrow topic than "things" or "items" - otherwise i'd bet you'd spent alot more than "tens of hours" on the project. -brian On 11/16/06, David Janes <[EMAIL PROTECTED]> wrote:
Any takers? I'm willing to start putting together an -examples and -brainstorming page, but I don't want to invest tens of hours in the project if people think it's semi-stupid or "we consider that but then we figured out it was a bad idea". hItem is probably a better name indeed. Regards, etc... David On 11/14/06, James Darling <[EMAIL PROTECTED]> wrote: > I have been having such thoughts, although I had hItem in my mind. > I pretty much had the same justifications in my head as David has > laid out. I was also thinking I wasn't sure much more base was > covered by this then hListing and hReview, as selling and reviewing > are, from what I can think of right now, are the only two things > you'd do with an item description on the internet. > On the other hand, it does seem nice in my head to mark something as > an item first, and then add reviews or prices or hcards onto that > item, linking them together. But maybe it's just a weird desire > rather than an actually helpful thing. > So this is basically saying I've had the same kind of thoughts, but I > personally hadn't come to a conclusion about it. > What are other's thoughts? (I have googled, but I would not be > surprised to be told this has all been discussed before, in which > case, any links to a justification against?) > > James > > On 14 Nov 2006, at 13:27, David Janes wrote: > > > I'd like to broach the topic of a "thing" microformat (or design > > pattern). I don't believe this is a "boiling the ocean" type of topic > > and I'll outline what I've been thinking about and how it would be > > used > > > > (1) This would be a simple container "hthing" with all the elements > > from vCard that would make sense -- fn, url and photo in particular; > > geo and adr also for things that are in a particular place. I.e. this > > proposal is fully based on hCard > > > > (2) This would be the basis for future "thing" microformats. For > > example, this morning a "wine" microformat was proposed. If there was > > a hThing microformat, there would be a solid basis for just extending > > that with information specific to the wine world; likewise, one could > > see this being of great use for making car microformats or housing > > microformats that would be used with hListing [1]. > > > > (3) This would fit nicely into existing microformats and may even be > > an improvement. For example, in hReview [2] there's an explicit > > container for the people and places being reviewed, but not for > > things. By having such a container, we ensure that attributes of the > > thing being reviewed associate with the thing rather than the review > > (that is, a car being reviewed has 4-wheel drive, not the review has 4 > > wheel drive; although this is obvious to us it may not be obvious to a > > machine reader) > > > > (4) This may just be a design pattern -- that is, a template for > > create new microformats in the future. We do see this pattern in use > > in several places, including the microformats mentioned above. > > However, I would argue that there's a real benefit for today's > > parser's knowing there's a thing there, if if future progress means > > the parser doesn't know what all the attributes are (just the most > > important ones) > > > > (5) If it is a microformat, how we identify the type of thing is an > > open question -- we could do it like phone number types in hCard or it > > could be a new class element (or both). The type option is nice > > because it means we can cover a lot of ground without creating new > > microformats. > > > > (6) There's obviously lots of things that could be added to a hThing > > microformat that are reusable across many types of things (just > > sketching here): > > > > <span class="hthing"> > > <span class="fn"> > > <span class="company">Rosemount Estates</a> > > <span class="product">Diamond Label</a> > > <span class="generic">Cabernet Sauvignon</a> > > (<span class="type">wine</span>) > > </span> > > </span> > > > > <span class="hthing"> > > <span class="fn"> > > <span class="company">Mazda</a> > > <span class="product">Tribute</a> > > <span class="generic">SUV</a> > > (<span class="type">automobile</span>) > > </span> > > </span> > > > > Regards, etc... > > David > > > > [1] http://microformats.org/wiki/hlisting > > [2] http://microformats.org/wiki/hreview > > _______________________________________________ > > microformats-discuss mailing list > > microformats-discuss@microformats.org > > http://microformats.org/mailman/listinfo/microformats-discuss > > _______________________________________________ > microformats-discuss mailing list > microformats-discuss@microformats.org > http://microformats.org/mailman/listinfo/microformats-discuss > -- David Janes Founder, BlogMatrix http://www.blogmatrix.com http://www.onamine.com http://blogmatrix.blogmatrix.com _______________________________________________ microformats-discuss mailing list microformats-discuss@microformats.org http://microformats.org/mailman/listinfo/microformats-discuss
-- brian suda http://suda.co.uk _______________________________________________ microformats-discuss mailing list microformats-discuss@microformats.org http://microformats.org/mailman/listinfo/microformats-discuss
