Re: [uf-new] The Process (was: hAudio case study)

2007-09-12 Thread Mary Hodder

I'd second that.

Even if users are just plopping blobs of data that is recognizable to  
humans when presented in a blog, but unstructured, we can still make  
sense

of those blobs and structure them.

Seeing what people want to do makes more sense that imposing from on  
high.  That is typical of standard's bodies and in practice not often  
followed.


For an example of this, where some standards from on high, are  
compared to some standards that emerged, in video metadata, see this:


http://microformats.org/wiki/media-metadata-examples#Comparison_Table

mary

Mary Hodder
Founder:  Dabble
Blogs:  Dabble.com/blog
Napsterization.org/stories

On Sep 12, 2007, at 8:53 AM, Scott Reynen wrote:


On Sep 12, 2007, at 9:24 AM, Michael Smethurst wrote:

Would it not be better for ufs to standardise markup based on the  
domain

model than waste time wading through flakey html? [Perhaps]


I don't believe looking at current publishing practices is a waste  
of time at all.  Much the opposite, I think it's the most important  
part of the process.  Domain models that don't account for what is  
published on the real web tend to be less useful on the real web.


--
Scott Reynen
MakeDataMakeSense.com


___
microformats-new mailing list
microformats-new@microformats.org
http://microformats.org/mailman/listinfo/microformats-new


___
microformats-new mailing list
microformats-new@microformats.org
http://microformats.org/mailman/listinfo/microformats-new


Re: [uf-new] The Process

2007-09-12 Thread Mary Hodder

I have to second this one as well.

In fact I blogged about this some time ago:

http://napsterization.org/stories/archives/000583.html

nicely of course, but the reality was that for 2 years, I got slammed  
and treated badly by the folks who started it, and then once I made  
them sit down and explain the process, I was able to follow the  
rules.  But it was harsh.  Many people commented privately to me  
after the two blog posts on process that it helped them a lot because  
they had the same experience.


I don't see why hazing helps the process.  You can have high  
standards without being mean.


mary

On Sep 12, 2007, at 8:50 AM, Frances Berriman wrote:


On 12/09/2007, Manu Sporny [EMAIL PROTECTED] wrote:

I think we should call it what it is:

The Accepted Limitations of Microformats


Yeah, that would work.


That harshness has also sent good people elsewhere... you don't  
have to

be harsh to prevent frivolous formats or unwanted behavior. In fact,
this goes against the Microformats be nice principle.



I agree - and I think that these new pieces of documentation will help
everyone be more supportive rather than exclusive in any way.


--
Frances Berriman
http://fberriman.com
___
microformats-new mailing list
microformats-new@microformats.org
http://microformats.org/mailman/listinfo/microformats-new


___
microformats-new mailing list
microformats-new@microformats.org
http://microformats.org/mailman/listinfo/microformats-new


Re: [uf-new] Launching the hVideo exploratory discussion!

2007-09-11 Thread Mary Hodder

Hi Manu,
Thanks for sharing your list.

Though I would say it appears to reflect what is located in APIs or  
feeds or html pages from a small subset of video hosting companies,  
not what we see overall, or what users publish on their own.


Is the video microformat use case one that is about video hosters or  
about what users do, or both?


At Dabble, as we take in metadata from both sources, we see users and  
hosters publishing titles, thumbnail (users don't call it image  
summary and actually, the hosters don't either much), category AND  
tags, uploader, director or creator, license, description, duration,  
sometimes size, and almost always, urls for html page, with some  
people giving multiple download and streaming urls. In fact there is  
almost always a permalink url (how did you come up with 58% on that  
one?) because no one would be able to get to it if that didn't exist.


What we never see is published (what does that mean?) or email  
video url in a feed or api.. sometimes that's on a page and often  
buried in flash or dhtml .. but it wouldn't be very helpful, as it's  
actually the same as the html url usually with extra code for email.


Also most individual users never publish popularity rating or  
number of views etc.


I thought the point of the microformat was for users and publishing  
companies to be able to put out data that would be easily readable in  
both an RSS feed and via spidering an html page, that was commonly  
published online?


I think the basic set of commonly published items includes:

title
html url (sometimes same and sometimes different from the url to  
embed, or link directly to the video)
video url (if existing - could be multiple, see Blip or Revver or  
soon to be Youtube)

category and / or tags
description
license
creator or uploader or both (many youtube videos have both, with  
director as the creator designation)

embed code
size
duration

We see these all elements highly published.

We don't typically see any comments come through feeds or apis, but  
they are certainly on the page at the site, and if say, like Youtube,  
they can be spidered, then yes.. there are comments.


If the goal is to make microformats for publishers, then it might  
make sense to include:

number of views
comments
number of comments
popularity ratings

Also, the sense I get from your write up here:
http://microformats.org/wiki/video-info-examples

is that you are more interested in the use-case of microformats for  
selling video or other content.


But I don't think most video on the web is about that, nor will it  
be.  I expect much of that will die out over the next two years, as  
some of what was there two years ago died out (see google video's  
retraction of for-pay video, or the increase in video that iTunes  
carries for free).  And I don't see a lot of it now.


my 2 cents.. i hope that's helpful.

mary

Mary Hodder

On Sep 10, 2007, at 8:27 PM, Manu Sporny wrote:


The video-info exploratory discussion has been created on the wiki.
There are currently 22 examples that have been analyzed so far:

http://microformats.org/wiki/video-info-examples

Here are the property leaders with the current set of data:

# title: 95.65%
# comments: 86.96%
# image summary: 86.96%
# published: 86.96%
# category: 82.61%
# contributor: 78.26%
# description: 78.26%
# flash player: 78.26%
# popularity rating: 73.91%
# number of views: 69.57%
# email video url: 60.87%
# related content: 56.52%
# number of comments: 56.52%
# permalink url: 52.17%

Example collection and analysis was performed using Microformalyze:

http://wiki.digitalbazaar.com/en/Microformalyze

The tool allowed analysis to be performed at a rate of around 10 sites
per hour. The rate was around 3 sites per hour when working on hAudio.

We could really use some help with more links to video sites. If  
you use
a video site, please add it to the list of video sites awaiting  
analysis:


http://microformats.org/wiki/video-info- 
examples#Video_Sites_Needing_Analysis


-- manu
___
microformats-new mailing list
microformats-new@microformats.org
http://microformats.org/mailman/listinfo/microformats-new


___
microformats-new mailing list
microformats-new@microformats.org
http://microformats.org/mailman/listinfo/microformats-new