On Dec 4, 2006, at 9:16 PM, Mike Meiser wrote:

On 12/4/06, Mike Hudack <[EMAIL PROTECTED]> wrote:
Mrss-thumbnail would work for me. Do we want to tie this so explicitly to
MediaRSS though? Maybe just media-thumbnail?

Makes sense to me.

I was thinking about the whole schema

tying to media RSS doesn't seem strong


<span class="mrss">
<img src="http://acmevlog.com/thumbnail.gif"; id="thumbnail">


class, not id (id's need to be unique per page)

<a href="urltovideo" class="type-flash">flash version</a>
</span>

There is an existing HTML way to express the type of a linked file, using MIME types

<a href="urltovideo.swf" type="application/x-shockwave-flash">flash version</a>


see

http://microformats.org/wiki/alternates- brainstorming#Strawman_6_.28lists_.2B_explicit_alternator_.2B_using_exis ting_HTML_idiom.29



The point is we need to make it as flat and simple as possible.

<img src="http://acmevlog.com/thumbnail.gif"; id="media-thumbnail">
<a href="urltovideo.swf" class="media-filetype-flash">flash version</a>
<a href="urltovideo.mov" class="media-filetype-quicktime">flash version</a>

Simple enough?

It's not clear that the files are alternatives - see the discussion at
http://microformats.org/wiki/alternates-brainstorming

<ol class="alternates">
 <img src="http://acmevlog.com/thumbnail.gif"; class="thumbnail">
<li><a href="urltovideo.swf" rel="enclosure alternate" type="application/x-shockwave-flash">Flash version</a></li> <li><a href="urltovideo.mov" rel="enclosure alternate" type="video/quicktime">QuickTime version</a></li>
</ol>


Media is the namespace. Shouldn't be any conflicts there.

Namespaces are an antipattern. Lets find a term that is expressive and converge it.

http://microformats.org/wiki/namespaces-considered-harmful



Should be fairly extenisble.

And as Andreas pointed out you can specify alternative class info.

I.E. <a href="url" class="media-filetype-quicktime cool-links
media-enclosure">cool styles link</a>

Where cool-links might be a CSS style, and media-enclosure and
media-filetype might be attributes of our media spec.

All we're talking about right now is "media-thumbnail" but just
keeping an eye out looking forward.

I think 'thumbnail' alone is OK (can you think of it being confused with markup for false nails or something?)

Looking at http://microformats.org/wiki/existing-classes there isn't an existing classname that fits (you could argue for class="photo summary" perhaps, but I think that is less accurate).


Simple enough?  Or still to convoluted?

I know services like blip can and will pick up on this stuff real
quick once we publish a spec, but it's got to be simple enough for
everday vloggers too.

The end goal is better marked up code in video and audio blog posts
everywhere dealing with media.  Would also be great for images, pdf
and other media too in the future.

See http://microformats.org/wiki/media-info for previous discussions. It seems like a revival of interest here could help us move on to converging schemas and proposing names.


I think the problem this points to is that the magic happens in the
blog post, not in the RSS. By embedding the metadata in the RSS with
mediaRSS we've cut most vloggers and services like blip who interact
by cross posting to blogs off from properly identifying the elements
of their blog post.

Feedburner's been picking videos out and enclosing them using the
rel="enclosure" standard.  And technoratti and others have been
picking out tags using the rel="tag" standard, but it's time to
develop a more robust schema to put the more power back into the blog
posting input box.

And finally, I like that this schema can be flat, like tags
themselves. Allowing multiple atributes to be attributed in the
"class" space. I think people may well get it.

Let's just define ourselves a simple schema or language then, shall we?

This is maybe something could be something that could really catch on
among bloggers and podcasters in the coming years to give them more
control.

It also leaves plenty of room for additional complimentary schema as
well wheras standards like rel="tag" and rel="nofollow" cannot be used
in conjunction with one another.

Yes they can, multiple rel's are legal HTML; rel="tag nofollow" is a little nonsensical (the link is a tag for the page, but don't index it?) but see above for more coherent examples.


Peace,

-Mike
mefeedia.com
mmeiser.com/blog


----- Original Message -----
From: videoblogging@yahoogroups.com <videoblogging@yahoogroups.com>
To: [EMAIL PROTECTED] <[EMAIL PROTECTED]>
Cc: Kevin Marks <[EMAIL PROTECTED]>; Eric Lunt <[EMAIL PROTECTED]>;
[EMAIL PROTECTED] <[EMAIL PROTECTED]>;
videoblogging@yahoogroups.com <videoblogging@yahoogroups.com>
Sent: Mon Dec 04 20:06:33 2006
Subject: Re: [video_vertigo] Re: [videoblogging] Media RSS what?

Yeah,

I like as andreas points out using the class attibute to declare a
thumbnail. Because class is an atribute unlike name that is standard across all manner of elements (images, links ans such) and it is very extensible
allowing the specification of multiple atributes and nearly infinite
namespaces.

I.E. <img src="http://yourvlog.com/thumbnail.gif"; class="mrss-thumbnail">


This is very similar to the technorati tags microformat for specifying tags
on blog posts.

I.E. <a href="http://mefeedia.com/tags/tagword"; rel="tag">tagword</a>


The only real difference is we're using class and to avoid conflicts with
CSS we're using a namespace, "mrss" in addition to "thumbnails"


That's a pretty good start, but simpler suggestions are very welcome.

It's important that not only automated tools and services like blip.tv, moveabletype, blogger.com or wordpress be able to add this markup to a blog post, but also that it become as simple a language as possible so everyday ordinary people can easily learn and use the "mrss" markup language the way
they've learned to adopt technoratti tags.

BTW, should this be called the "mrss" microformat, or does
microformats.orghave their own name space?

We are replicating some of the functionalities of Yahoo's mrss (aka. media RSS) so perhaps it's best to make some attempt to correlate what we're doing
with that.

-Mike
mefeedia.com
mmeiser.com/blog

On 12/4/06, Lisa Rein <[EMAIL PROTECTED]> wrote:
>
> Hi gang,
>
> We'll implement anything that's XHTML compliant.
>
> With that in mind, I vote for the "title" attribute of the "img"
> element, since "alt" is generally used for titles.
>
> There's a "name" attribute too, which is already in use for scripting.
> But perhaps it's *already* being used to say things like "thumbnail"
> for scripting anyway, in which case it would work fine.
>
> http://www.zvon.org/xxl/xhtmlReference/Standard/struct/
> objects.html#edef-IMG
>
> http://www.zvon.org/xxl/xhtmlReference/Standard/struct/
> global.html#adef-title
>
> <img src="http://image.com/image.jpg"; title="thumbnail" />
>
> thanks!
>
> lisa
>
> p.s. how do i get back on the list? :-)
>
> On Dec 3, 2006, at 6:25 PM, Mike Meiser wrote:
>
> > First off, I think whatever email client you're using is marking
> > everything as *****SPAM***** in the subject line. Very detrimental to
> >  conversation. :)
> >
> >  Response below.
> >
> >  On 12/3/06, Mike Hudack <[EMAIL PROTECTED]> wrote:
> > > We currently use rel="enclosure" in cross-posts, but we don't and
> > can't
> > > use rel="thumbnail" because (afaik) images in XHTML don't have a rel
> >  > attribute. I'd be very hesitant to use something like
> > class="thumbnail"
> > > because of potentially conflicting CSS on remote sites (we prefix
> > class
> >  > names in cross-posts with a namespace like "blip_" to avoid
> > conflicts).
> >  > I'm not sure what the proper approach to attaching semantically
> > > important information to images is, but that's all the more reason
> > why
> > > this conversation is more important for the microformats-discuss
> > list
> >  > than the videoblogging list.
> >
> >  Where's andreas of solitude.dk when you need him? :)
> >
> >  Overall, yeah, I agree.
> >
> > But I don't think I'm on this microstandards discussion list for some
> >  reason. Must have been some oversite on my part. You got a url or
> >  were you talking video vertigo?
> >
> >  > The only way that a standard for additional metadata will be
> > adopted is
> > > if FeedBurner is on board. As I said in my previous e-mail, I've > > > discussed this with them and they're gung-ho for the idea but don't
> > have
> > > room in their roadmap right now. At blip we're ready to go with it,
> >  > though. We'll add whatever microformats we have to so that the
> >  > blip->cross-post->FeedBurner->MediaRSS workflow works best.
> >
> >  Well, I think if just mefeedia and blip adopt it'll be mutually
> > beneficial to us both immediately making it worthwhile. Peter is out > > of town at the moment, but this is at the very top of our priority > > list. Those damn thumbnails are sort of important to people and we
> >  can't generate them all.
> >
> > Remember, if it's in the blog post, even if Feedburner doesn't pick it > > up immediately and translate it to mediaRSS it'll get through to the > > RSS anyway and anyone including mefeedia, democracy, dabble or anyone
> >  who chooses to identify and use it can.
> >
> > Right now we're talking about going back in to identify all the videos > > coming from blip and their thumbnails and then cross referencing that
> >  with videos we're picking up from feedburner feeds and using the
> > thumbnails there. It's be much easier for everyone if those thumbnails
> >  (and other meta info from blip) were contained right in those
> > feeburner RSS feeds, either with mediaRSS as they get the chance, or
> >  with microstandards until then.
> >
> >  This is not just about thumbnails of course, that's just at the
> > forefront. We're talking about ways to pass around geographic data on > > individual posts, and having ways to continually extend and add new > > semantic data. Data that blip and others are collecting, data that
> >  gets lost in the cross posting.
> >
> > The trick is to allow people to specify as much metadata, or semantic > > data in the post as possible as simple as possible. This is something
> >  everyone can benefit from.
> >
> >  The problem is as it turns out with mediaRSS is most people don't
> >  write their own mediaRSS feeds. They need simple microformats to
> >  specify semantically key meta info right in the blog post.
> >
> >  rel=tag is a great example of this.
> >
> >  We need a way to identify thumbnails next.
> >
> >  Then to start looking at other metainfo that's missing.
> >
> > > We're more than happy to do support this in code before FeedBurner
> >  > does... so if FB agrees to support a microstandards standard at
> > some point
> >  > in the future we'll add support to it in our code immediately.
> >
> >  Rock on. That's what I'm talking about.
> >
> >  Just got to figure out if we can't specify rel=thumbnail on image
> >  source what else we can do.
> >
> > We could use the "alt" space. It's usually used for text describing
> >  the image, but alt=thumbnail might work.
> >
> >  Keep in mind while blip is going to be handling all this stuff
> > automatically we need to keep it very simple to keep it accessible to
> >  everyday people whom very well may be hand coding it.
> >
> >  So... if we can keep it rel=thumbnail that would be best.
> >
> >  How else are people specifying microstandards? Within div tags?
> >
> >  <div rel=thumbnail><img src="http://...";></div>
> >
> >  What about <span>?
> >
> >  I must admit I'm so out of touch. :P
> >
> > What strikes me as a more technical question is how to semantically > > specify alternate video formats like Flash in the source. As far as I
> >  know there's no way to extend rel=enclosure.
> >
> > The only think I can think of is to specify urls in duplicate. For > > example specifying one url for the generic rel="enclosure", specifying
> >  it again with rel="quicktime-enclosure", and another time with
> >  rel="flash-enclosure".
> >
> >  That seems very complex though. There's got to be a simpler way.
> >
> > > If I had my druthers, Kevin Marks from Technorati, Lisa Rein from
> > Dabble
> > > and Eric Lunt from FeedBurner would be part of this conversation
> > and we
> > > could settle on a solution very quickly. So I've gone ahead and CC'd > > > all of them (and also CC'd Video Vertigo, which most of them are
> > on).
> >
> > Sure, this should be pretty cut and dry. It's very simple and yet very
> >  useful stuff.
> >
> >  -Mike
> >  mefeedia.com
> >  mmeiser.com/blog
> >
> >  > Yours,
> >  >
> >  > Mike
> >  > blip.tv
> >  >
> >  > [EMAIL PROTECTED] wrote:
> >  > >
> >  > > Good question jay,
> >  > >
> > > > Does anyone know if Moveabletype, blogger, or Wordpress support > > > > mediaRSS? Has anyone created a plugin perhaps for MT or Wordpress?
> >  > >
> > > > I've realized one of the primary issues with the mediaRSS spec is
> > that
> > > > when you're righting a post there's no quick markup to specify
> >  > > metadata IN the post. Specifically for example I was going to
> > propose
> > > > a series of Microstandards like "rel=thumbnail" so people could
> >  > > semanticly specify metadata in their post.
> >  > >
> > > > For example when you specify the image to represent your video in
> > the
> >  > > page you'd specify it like this.
> >  > >
> >  > > <img src="http://acme.com/yourthumbnail.gif
> >  > > <http://acme.com/yourthumbnail.gif>" rel="thumbnail">
> >  > >
> > > > As some may have noticed one of the biggest issues we have with > > > > mefeedia is generating tens of thousands of email a day. One of
> > the
> >  > > things I've realized is that MOST people are specifying a
> >  > > representative thumbnail in their post, they're just not
> > specified in
> >  > > the mediaRSS.
> >  > >
> > > > I think we're going to adopt this rel=thumbnail standard pretty
> > quick
> >  > > on mefeedia.
> >  > >
> >  > > But if we can also just get a few providers on board like
> > feedburner,
> > > > blip.tv, dabble and a few others we could creat much prettier and
> > more
> >  > > usefull feeds really quick.
> >  > >
> >  > > A couple examples.
> >  > >
> > > > 1) if feedburner jumps on board identifying thumbnails based on
> > the
> > > > rel=thumbnail standard they can also specify those thumbnails in > > > > mediaRSS and aggregatory tools like democracy, itunes, mefeedia, > > > > fireant or whomever recognizes mediaRSS will immediately start
> >  > > displaying the thumbnails you specify in your blog instead of
> > randomly
> >  > > pulling their own.
> >  > >
> > > > 2) This is particularly an interesting issue with people hosting > > > > their videos on blip. Blip does allow you to specify a thumbnail
> > for
> > > > each video, and they DO put it in your blip feed. However, 99% of
> > all
> >  > > people using blip don't use blip's feed, they cross post from
> > blip to
> > > > their video blog where such meta information as the thumbnail is
> > not
> >  > > semantically specified. By semantically specified I mean the
> > image is
> > > > just genrically specified in the page and the aggregator can't
> > assume
> >  > > to know what it is.
> >  > >
> >  > > The point is as a result all the blip feeds look great on
> > mefeedia,
> >  > > but all people's primary feeds have no thumbnails specified.
> >  > >
> >  > > If blip specified the thumbnails using the "rel=thumbnail"
> > standard
> > > > when crossposting to people's vlogs that information would make it > > > > into the RSS feed where it could be picked up by any aggregator.
> >  > >
> > > > The bottom line is this... more semantic data = a prettier and
> > more
> >  > > useful vlogosphere for everyone.
> >  > >
> > > > So... who's with me? Josh K? Mike H., Justin, and the Blippers?
> > Lisa
> >  > > Rien, Mary Hodder and the dabblers? Do we have someone
> > representing
> >  > > feedburner here?
> >  > >
> > > > On top of this I'd like to take a look at what other information
> > is
> >  > > getting specified in blog posts, such as that that blip is
> > collecting,
> >  > > that's not making it into the mediaRSS and why not.
> >  > >
> > > > Again, 99% of vloggers specify a thumbnail in their blog post, but
> >  > > these thumbnails aren't making it into the mediaRSS because
> > there's no
> > > > way to semantically specify it in the blog post. Rel=thumbnail is
> > the
> >  > > simplest way i can think of to accomplish this.
> >  > >
> > > > Jay, you said you recently took a look at mediaRSS. What sort of
> >  > > metadata are you talking about, what metadata do you want to
> > specify?
> >  > > Can you give a few examples of what your clients might find
> >  > > particularly interesting?
> >  > >
> >  > > I'm afraid that other than developing this and other
> > microstandards
> >  > > that can be specified right in the blog post, rich meta
> > information
> > > > will continue to get missed untill the major blogging platforms
> > like
> > > > Blogger, Wordpress, and Moveable type support mediaRSS by default.
> >  > >
> > > > Oh, and you should also note that Yahoo video search will pick up
> >  > > these thumnails too if we can get feedburner to support this
> > proposed
> >  > > RelThumbnail standard.
> >  > >
> >  > > Are there any other major search engines or services I'm
> > forgetting
> >  > > about that aggregate mediaRSS for search and such?
> >  > >
> >  > > -Mike
> >  > > mefeedia.com
> >  > > mmeiser.com/blog
> >  > >
> >  > > On 12/3/06, Jay dedman <[EMAIL PROTECTED]
> >  > > <mailto:jay.dedman%40gmail.com>> wrote:
> >  > > > > Here's the MRSS spec:
> > > > > > <http://search.yahoo.com/mrss <http://search.yahoo.com/mrss>> > > > > > > It was developed by Yahoo! with a lot of collaboration from a > > > > > > community of contributors, including many folks on this list. > > > > > > FeedBurner supports MRSS in a pretty limited way -- really
> > just as an
> > > > > > addition to the enclosure element. Blip.tv includes a lot of
> > MRSS
> >  > > > > metadata in their feeds, including support for media
> > thumbnails and
> >  > > > > alternate versions of each video (FLV, Quicktime, etc.).
> > > > > > Is there something in particular you want to do with MRSS?
> >  > > >
> >  > > >
> >  > > > yep....i saw the spec, but am having a hard time fitting my
> > brain
> >  > > around it.
> >  > > > I am working with a group of Community TV stations that are
> > starting
> > > > > to upload and trade TV programs for playback around the country.
> >  > > >
> > > > > They want to attach a lot of metadata into their posts....so
> > they are
> > > > > asking if Media RSS could help them. Questions I have is....how
> > do
> > > > > they create feeds that attach all this info into their feed?
> >  > > > Do they need to make their feeds by hand?
> >  > > >
> > > > > right now, they are just uploading to their own servers...and
> > using
> >  > > > Drupla to create their feeds.
> >  > > >
> >  > > >
> >  > > > Jay
> >
> http://www.dabble.com
> http://www.onlisareinsradar.com
>
>
>
>


[Non-text portions of this message have been removed]




Yahoo! Groups Links





[Non-text portions of this message have been removed]




Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/videoblogging/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/videoblogging/join
    (Yahoo! ID required)

<*> To change settings via email:
    mailto:[EMAIL PROTECTED]
    mailto:[EMAIL PROTECTED]

<*> To unsubscribe from this group, send an email to:
    [EMAIL PROTECTED]

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/



_______________________________________________
microformats-discuss mailing list
microformats-discuss@microformats.org
http://microformats.org/mailman/listinfo/microformats-discuss

Reply via email to