[uf-discuss] Need a figure markup

Greg Elin greg at fotonotes.net
Sun Oct 30 10:21:40 PST 2005

Well said Chris. And Ryan's link to FXIF extension is quite interesting.

Here's a good example for re-thinking the <img> tag. Let's say you  
want to include the title/caption for the image (from a database or  
from EXIF field) below an image and have the caption wrap nicely to  
the width of the image. To do so, it is necessary to wrap the entire  
image inside <div> or <table> and depending on the situation,  
coordinate the dimensions of the block based on details of the image  
size. End result: lots of non-standardized CSS and javascript  
approaches to deal with the issue.

(And while I am making noise, wouldn't it be nice if my browser that  
did such a nice job of selecting a block of text fundamentally let me  
click-drag around a region of a photo and report back the coordinates?)

To your second point, Chris: browsers DO render image data so it does  
not seem much of a leap and would make plenty of sense for the  
browser's power, including CSS, to reach into that data.  Adding some  
CSS or HTML control over what's inside that image file makes what's  
inside more accessible and transparent, and the history of the Web is  
that more accessibility and transparency is a good thing.

Greg Elin

On Oct 30, 2005, at 12:46 PM, Chris Messina wrote:

> The image properties that Opera provides didn't seem entirely useful.
> Not like the Firefox extension anyway.
> To Greg's point, there does seem to be some value in being able to
> render or extract image metadata from images that contain that
> information. FXIF looks awesome and has a lot of potential -- I wonder
> what would be possible, for example, if Flock shipped with that built
> in -- and what having that data might allow web developers or
> extension designers to be able to accomplish.
> Though I am all for visible metadata, if the photos contain this
> information already, it might be possible, therefore, to start to
> review this data with CSS:
> img[exif="true"]:after {
>   content: "Taken with a " attr(Manufacturer);
> }
> Or something like that. I can imagine some amazing Greasemonkey uses
> for having data available like that, as well as fulfilling Eris'
> one-time request when shopping for a camera: "show me photos taken
> with these five prospective purchases".
> So I guess I'll look to Tantek on this, since I know he's got browser
> experience... is it the browser's place to pull in EXIF data and make
> it available? I presume there are numerous security concerns around
> doing so, but it would seem that the potential benefits (and personal
> media revolution going on) suggest a great need for browsers to become
> more multi-media friendly and aware.
> Chris
> P.S. We've strayed from the original figure discussion, but this is
> still certainly a worthwhile thread!
> On 10/30/05, Andreas Haugstrup <solitude at solitude.dk> wrote:
>> On Sun, 30 Oct 2005 16:30:50 +0100, Scott Reynen  
>> <scott at randomchaos.com>
>> wrote:
>>>> It is a great loss that browsers have the power to crack open  
>>>> the JPEG
>>>> file and uncompress binary data to display data but entirely  
>>>> ignores
>>>> embedded metadata text like EXIF.
>>> Not entirely true:
>>> http://ted.mielczarek.org/code/mozilla/fxif/
>> In addition Opera displays EXIF data out of the box (right click,  
>> Image
>> Properties). I think this was added in version 8.
>> - Andreas
>> --
>> <URL: http://www.solitude.dk/ >
>> Commentary on media, communication, culture and technology.
>> _______________________________________________
>> microformats-discuss mailing list
>> microformats-discuss at microformats.org
>> http://microformats.org/mailman/listinfo/microformats-discuss

More information about the microformats-discuss mailing list