Forum .LRN Q&A: Teaching with Images

Posted by Michael Feldstein on 03/23/05 09:15 PM

There's been quite a lot of talk in the SUNY community lately about teaching with images, stimulated by Flickr. In particular, there's been a lot of excitement about using the text annotation for teaching purposes. It turns out that there's an open specification for this called "fotonotes." Given OACS' heritage and strengths with image handling, this might be an area where dotLRN could take the lead. An image notation tool would be a big win, especially if it worked smoothly with LORS.

2: Re: Teaching with Images (response to 1)

Posted by Michael Feldstein on 03/23/05 09:16 PM

Sorry, the example link I wanted to give on Flickr was actually http://www.flickr.com/photos/ha112/901654/.

3: Re: Teaching with Images (response to 1)

Posted by Michael Feldstein on 03/23/05 09:40 PM

It looks like the fotonotes wiki has been subject to some vandalism. You might find this reference site more useful:

http://scribbling.net/projects/fotonotesrolloverviewer/

4: Re: Teaching with Images (response to 1)

Posted by Andrew Grumet on 03/23/05 10:51 PM

> Sorry, the example link I wanted to give on Flickr was actually http://www.flickr.com/photos/ha112/901654/.

Fascinating!

6: Re: Teaching with Images (response to 1)

Posted by Bruce Spear on 03/24/05 12:48 AM

I think the ability to develop a commentary on images, and parts of them, is powerful stuff, Images can sometimes be more compelling and accessible than words. I used to teach a lot of art history, and I quickly learned to let the projected slides do the work: sequencing them carefully so that the conversation built on one will be amplified or extended by those that follow, was a fine way to get students talking. So, I'd add to the examples cited here, if I understand them right, more than captions: I think the real return on investment would be when some sort of threading functionality kicks in, like integrating annotation with forum functionality. Or? -B

7: Re: Teaching with Images (response to 1)

Posted by Michael Feldstein on 03/24/05 01:05 AM

Yeah, forum integration would help a great deal. When brainstorming the ultimate tool with a few people at FIT, here are the basics of what we came up with:

Users could draw annotation areas with rollovers, as Flickr allows.
Clicking on the annotation box takes you to a separate web page with a blow-up of the box and a discussion thread.
The blow-up page would also have trackback functionality so that students who have their own blogs could comment on the image, store that comment on their blog (which also functions as a simple portfolio), and still share their comment with the class.
Page layout would be set up so that students could keep the image on screen while scrolling through the commentary text.

5: Re: Teaching with Images (response to 1)

Posted by Carl Robert Blesius on 03/24/05 01:14 AM

Here is a close up from the same piece:
http://www.flickr.com/photos/ha112/901666/

Thanks for the link to FotoNotes specification Michael.

Like your SUNY folk I got excited about this too (although when I imagine it implemented in an my installs of .LRN/OpenACS I see wonderfully annotated medical images of bumps, rashes, crusts, scales, and warts).

It is nice to know that there is a searchable/semantic web friendly way to do this. Something worth exploring.

It looks like this is something that could be provided as part of a RSS feed as well. Here is an example:
http://fotonotes.net/spec/index.php/Current_FNAtom_Schema
Although, I do find the idea of adding regional annotations to the metadata fields in the images themselves more appealing.

Carl

8: Re: Teaching with Images (response to 1)

Posted by Rafael Calvo on 03/24/05 04:20 AM

Thanks for your ideas Michael.
Flickr has a bunch a cool features.

In somewhat similar lines, Ernie has started builing an interewsting application to integrate photos with GPS data:

http://www.weg.ee.usyd.edu.au/people/ernieg/mediaalbum.jpg

cheers

Rafael

9: Re: Teaching with Images (response to 1)

Posted by Nick Carroll on 03/31/05 04:43 AM

I've been using this DHTML JavaScript library for tooltips in dotFOLIO...

http://www.walterzorn.com/tooltip/tooltip_e.htm

It is worth noting that it can be applied easily to image maps, so you can have the same functionality as fotonotes with very little effort. The library is also LGPL, so we can distribute it with openacs.

10: Re: Teaching with Images (response to 1)

Posted by Malte Sussdorff on 03/31/05 08:25 AM

Nick, this looks good. Would it be possible to extend it so you could mark text you are reading and add a tooltip to it?

This is part of my idea for a personal note taking system, where you could read a text and add your own (personal/public) comments to part of the text, just like WORD annotations or the tooltip system you are using.

11: Re: Teaching with Images (response to 1)

Posted by Nick Carroll on 03/31/05 12:49 PM

Malte, I think all you have to do is wrap the selected text with span tags with the onmouseover pointing to the javascript tooltip library, using the comments as input to the escape function.

If the text is static then you can pre-wrap the text with the span tags, and pull the comments out of the database which you'll use to pass into the escape function. Otherwise you'll have to rely editing HTML if you do it through a form.

12: Re: Teaching with Images (response to 1)

Posted by Michael Feldstein on 03/31/05 03:06 PM

I suppose my main concern for using tooltips alone to solve the problem that my faculty advocates are interested in is that it's an overly general solution. My layperson's understanding of photonotes is that it's an attempt to specify a standard for storing rich metadata about images. The fact that its main current use is drawing boxes with rollovers doesn't necessarily mean that it will always be so. It seems to me that supporting the broader (larval) standard would be important.

Having said that, it's not entirely clear to me that the two are incompatible. Again, my layperson's understanding of photonotes is that it's mainly a format for metadata. The tools for displaying that metadata are just reference implementations. Could the tooltips code be adapted to display the information from the photonotes spec?

13: Re: Teaching with Images (response to 1)

Posted by Nick Carroll on 03/31/05 05:31 PM

I can't see why not. I'm fairly positive that the fotonotes implementation uses image maps with a javascript tooltip library. The tooltip isn't an IE tooltip, which are limited to a single line text box. The javascript tooltip is rendered using CSS, and you can even insert HTML and images into the javascript tooltip.

I'm not sure how you create a fotonote, in that how does one specify where a note should be placed on an image? Is this left to the web developer to code? Or is there some sort of functionality that lets the user do this from a UI?

14: Re: Teaching with Images (response to 1)

Posted by Nick Carroll on 03/31/05 05:53 PM

Had a closer look at the fotonotes spec. It is just an XML schema. So implementing fotonotes requires parsing a fotonotes xml document, search for imreg tags that specify a region on the image, convert that to image map tags in HTML. Then pull out the description for a popup region from the fotonotes spec and pass that to the javascript tooltips function. That should cover the presentation implementation of fotonotes. As to what you want to do with the meta-data associated with the fotonote, well that can be pulled out and inserted into LORS.