Just to recap: we can find what we need, but will we find you?


Traveling without moving by FredArmitage

I have largely suppressed my memories of being UBC’s ‘learning objects project coordinator’ 5 or 6 years back. But I do vividly recall the drive to develop a robust form of “federated search” that would allow users to search across a range of locally-hosted learning resource collections. Were such a thing possible, some of us believed the skies would part and we would enter a blissful state of findability and connectedness. Of course, it wouldn’t be easy. We would need to carefully catalogue our materials using a detailed and often mysterious metadata specification, and our local repositories would similarly be required to meet rigorous standards, perhaps implementing a painstakingly developed communication layer to connect these pieces together. Expensive, arduous, tedious, and confusing. But surely the payoff would be worth it…

Or, we can just let Google do it for us for free with a few minutes work. Here’s a recipe for rich, chocolatey, open educational searchy goodness, adapt it to your own tastes.

* Assemble your ingredients. ZaidLearn saved us a lot of hassle by assembling this handy list of open educational resource (OER) sites.

* You knew that Google already allows you to set up your custom search engine by whatever domains you wanted, right? So Tony Hirst took the ZaidLearn list and used it to quickly create an OER Search Engine. You can put the search box anywhere you want, including right here, just by cutting and pasting a little HTML:

* Then Scott gets it into his disturbingly shaved head to have the list of supported search domains run off of a wiki, so anybody can come in and add resources collections. I added a few bits, including the Creative Commons rich media search, though it might be necessary to paste in some of the specific collections.

It seems to work pretty well. Sure enough, my search on “Willinsky” not only brings up a fantastic presentation John Willinsky gave at UBC last fall that’s hosted on Blip.tv, it also turned me on to this tantalizing set of talks on the economics of open content that I was previously unaware of.

The technology cost is negligible. Someone like Tony can go into the Google Ajax Search API and/or work some Yahoo Pipes magic to do even cooler stuff — but even an idiot like me can do quite a bit with some cutting, pasting and wiki editing.

As far as I know, Zaid Ali Alsagoff, Tony Hirst and Scott Leslie have never met, and there is no coordinating body to facilitate their collaboration. What is required (in addition to Google’s scary hegemonic presence providing a powerful platform) is openness. The resources need to be indexed on the open web, and when people do cool stuff and then blog about it, others can take the work to unexpected places.

A coda. I’m not sure if Google’s Dynamic Feed Control Wizard fits into this picture, but it is kind of nifty. All I have to do is type “zaidlearn, ouseful info, edtechpost” into the “Feeds Expression” field, and Google finds the feeds and quickly generates this customizable display:

Loading…

@import url(“http://www.google.com/uds/solutions/dynamicfeed/gfdynamicfeedcontrol.css”);

function LoadDynamicFeedControl() {
var feeds = [
{title: ‘edtechpost’,
url: ‘http://www.edtechpost.ca/wordpress/feed/’
},
{title: ‘zaidlearn’,
url: ‘http://zaidlearn.blogspot.com/feeds/posts/default’
},
{title: ‘ouseful info’,
url: ‘http://feeds.feedburner.com/ouseful’
}];
var options = {
stacked : true,
horizontal : false,
title : “OER Engine Hero Roll”
}

new GFdynamicFeedControl(feeds, ‘feed-control’, options);
}
// Load the feeds API and set the onload callback.
google.load(‘feeds’, ‘1’);
google.setOnLoadCallback(LoadDynamicFeedControl);

8 thoughts on “Just to recap: we can find what we need, but will we find you?

  1. Amen and pass the Gu!

    (and by the way, do you ever get the sense someone is messing with your heads with the captchas below? The two words I am supposed to type – “Crowley” and “Insane”. Is it a full moon or something?)

  2. While I agree that it is best to have servers/machines/google/HAL do the work searching and harvesting content given certain parameters that we input, I’d like to throw my 2 cents in for those lone ghosts who help ensure that the content is structured or formatted in such a way that it can be indexed, trolled, google-botted, and returned in those search queries. I’m speaking, of course, of those hard-working, controlling, organizationally-minded tweeds otherwise known as librarians, information architects, or content managers. Details and metadata, standards and structure, though painful and expensive they are to implement and manage, often produce content that is richer, findable, and easier to reuse than the unstructured and undescribed kind. Using the Google custom search box above, I looked for resources on Canadian History and came up with a wide variety of content, much of which was not related to my query. This is one of the key problems with both the wide variety of content available online and the mysterious ways in which google ingests and indexes it. For some unrelated hits, see here and here.

    An example in which well-described structured content has been automatically harvested and reused is a project I worked on at the BBC called Topics, a directory of sorts that pulls in various content from across the site related to places, people, or subjects. It represents the best of both worlds: content management and dynamic content generation. No human interference necessary after programming the initial search queries. But the content that is returned is rich, and more importantly, relevant.

    Cheers,

    pj

  3. What’s the digital equivalent of pulling your tongue, phhwwt? well phhwwt and double phhwwt!

    “painful and expensive they are to implement and manage…” that’s the whole frickin’ point! IT’S NOT GOING TO GET ANY BETTER IT’S GOING TO GET WORSE. What we need is to stop pulling content out of it’s CONTEXT OF USE and let that serve as the way we find it, exactly the way google works. It’s the 80/20 thing again, and I’m quite happy with 80% (especially for free).

    No doubt indeed, the existence of some structured or properly curated metadata does help the search engines, and in limited cases (read: your library catalog) may provide the expert searcher better results. For now. but don’t go betting your pension on it. And you’re right, we can use other things like tags, social search and the like to add to the pure search engine approach.

    (oh, and you’re still invited to our place this summer, that way I can pull my tongue in person πŸ˜‰

  4. There’s a bit more to the story, in fact…

    http://www.downes.ca/post/45032
    Downes writing about Zaid’s post in OLDaily:
    “Zaid has updated his list of free online learning resource sites. This is a great resource – and something to spur me to create something (using RSS or something) that will aggregate and make available for search the collected resources from those sites.”

    The CSE hack was actually a response to Stephen’s ‘note to self’…

  5. What a great reflective post πŸ™‚

    Yes, I have never met or have collaborated with Tony Hirst and Scott Leslie (I hope to one day!). But we seem to have done something interesting together, which hopefully someone out there find useful.

    Also, Stephen Downes (or Joseph Hart) I suppose is the ORACLE that have linked us together on this mission to find great OER and OCW.

    Although, I really admire Tony Hirst’s great idea to use Google Custom search for this post, I did actually create an OER search using Google’s master piece way back in January (2008). URL: http://zaidlearn.blogspot.com/2008/01/google-custom-search-for-openfree.html

    Google Custom search can be a useful tool (if we we feed it with a lot of relevant links) to search OER and OCW (or whatever!), especially if we use the filters, too.

    At least it is cheap, I mean free and fast to develop πŸ™‚

    Combine Google Custom Search with human generated repository and course collections, and we are really going to make it easier for the rest of the world to find great OER and OCW.

    Have a great learning week and thanks for great reflective post πŸ™‚

    Warm Regards,

    Zaid

  6. What a great reflective post πŸ™‚

    Yes, I have never met or have collaborated with Tony Hirst and Scott Leslie (I hope to one day!). But we seem to have done something interesting together, which hopefully someone out there find useful.

    Also, Stephen Downes (or Joseph Hart) I suppose is the ORACLE that have linked us together on this mission to find great OER and OCW.

    Although, I really admire Tony Hirst’s great idea to use Google Custom search for this post, I did actually create an OER search using Google’s master piece way back in January (2008). URL: http://zaidlearn.blogspot.com/2008/01/google-custom-search-for-openfree.html

    Google Custom search can be a useful tool (if we we feed it with a lot of relevant links) to search OER and OCW (or whatever!), especially if we use the filters, too.

    At least it is cheap, I mean free and fast to develop πŸ™‚

    Combine Google Custom Search with human generated repository and course collections, and we are really going to make it easier for the rest of the world to find great OER and OCW.

    Have a great learning week and thanks for great reflective post πŸ™‚

    Warm Regards,

    Zaid

Leave a Reply

Your email address will not be published. Required fields are marked *