As a follow-up to my post on small numbers of people doing tagging, as of today, for our NMC Regional Conference held last week in San Antonio, we have tagged in flickr 393 photos from 11 individuals, more than we had at our summer conference in Cleveland where the attendence was 3 times as large as San Antonio. See all tagged photos.
And we were able to use the slide show as the screen display prior to our closing keynote session, which was well enjoyed by the folks coming into the auditorium.
I am not ready to draw any statistical conclusions, but will toss some anecdotal ones. Okay, maybe nmc2006reg was not the greatest tag, and some folks got it switched around, but the only promotion for this was the mention in our program. Secondly, via this, I picked up 3 new flickr contacts, folks I did not know before. Lastly, photos are the things people will tag the most.
On a suspicious note, given than there were 11 taggers here, 13 at the summer conference, 19 at the EDUCAUSE annual conference… I wonder if it is just the same people 😉
On a technical note, when you create a flickr badge for a public tag, as I used on our tag this conference page, the form does not provide the option to choose at random from the tag set, just the most recent. So just in a tinkering mode, I compared the output HTML from what you can do by making a badge for tagged photos from your own photos (where “random” is an option), I found it does work on the public tags, simply by changing the display=xxxx
in the JavaScript portion of the of the code from:
<script type=”text/javascript”
src=”http://www.flickr.com/badge_code_v2.gne?count=5
&display=latest&size=s&layout=h
&source=all_tag&tag=nmc2006reg”>
</script>
to
<script type=”text/javascript”
src=”http://www.flickr.com/badge_code_v2.gne?count=5
&display=random&size=s&layout=h
&source=all_tag&tag=nmc2006reg”>
</script>
On the blog post front, I was frustrated by Technorati’s either refusal, or multi day lag in picking up blog posts with proper tags, even when you pinged them manually. A day before the conference, I had switched to using the Google blog search and its RSS feed, which picked up at that time more than Technorati would, but it too did not seem to pick up new postings quickly enough, and for some reason, technorati started acting like it should. Compare the results from the Technorati report for the nmc2006reg tag (12 results) to Google’s blog search on nmc2006reg (6 results). Again, hardly conclusive, but to me, blog post tagging is much more fraught with holes that photos and bookmark searches.
Lastly, someone at the conference had asked in one of the Web 2.0 sessions if there were services that could look across multiple sites for tags. I could not think of it then, but did come across one earlier this week in my research for our Horizon Report project– TagFetch which looks for tags across 8 different services– Google Blogs, technorati, IceRocket, Topix, Digg, Feedsterm del.icio.us and flickr, (and features a logo much like my own!):
See the results of tagfetch for the nmc2006reg tag. It’s not perfect, it takes some time, but it may get to a better set of results than relying on one source.
You might be interested in the taggregation we’ve worked up over at EDUCAUSE too.
Check out …
http://connect.educause.edu/term_view/drupal
See the elsewhere on the net section.
I’d definitely be interested in feedback.
Cheers,
Matt
Interesting! It’s pulling in a lot of content– is it via RSS, or do you have to create the little nodes for your community to point to EDUCAUSE content??
I need to poke a bit more about it– Since I am heading sown a drupal path, I’d like to hear more how you set this up….
We have the nptech meta feed – which is a rss feed that splices together the tag feeds from aobut 15 different places looking for the nptech tag
http://feeds.feedburner.com/NPTechMetaFeed – would that work for your purposes? I write weekly summaries of what has been posted for netsquared.
Great post!
What about collaborative search? Have you looked at that?
Most of it is pulling in via RSS or ATOM. For Flickr, I believe we’re using serialized PHP.
You can get a feel for modules we’re using via …
http://connect.educause.edu/colophon
I’m also very interested in collaborative search, but alas, only so much time.
Shoot me an email if you’d like to put in a quick call about what we’re doing. If we can ever get caught up, we want to contribute it back to the drupal community.