Semantic UMW Rotating Header Image

March, 2009:

What might a linking-game look like?

An interesting conversation in the comments to Cole Camplese‘s post about the importance of having access to data in a university system. Brett Bixler brought up the idea of bringing some game aspects to using the data, and Cole pushed toward the ‘how many clicks’ game:

students are given a random starting topic and a random ending topic and they have to see how many clicks through articles it takes to get from A to B … an example might be, “get from the topic iPod to Kennedy Space Center.” I wonder if we could design something like that uses tag aggregation as a pass through

I’d love to see something like that, might even scrape up the time to work on something like that, but need help figuring out what, exactly, that might look like. Tags might be a great way to start something like this off. Links might be a way, too. How does the game work, though? I don’t really know where to begin with this.

One additional thought. Theoretically, with both tags and links we could come at it from two different directions. One is to have something that focuses on the existing tags/categories and links to make the game go. The other is for players to create their own links and tags/categories on posts they encounter. Since the data in SemanticUMW is in a completely separate database from the UMWBlogs database, this could fairly easily be done, and might even provide an additional set of interesting and useful data.

Anyone with ideas/models for sharpening up this fuzzy idea?

Posts By Day Of Week

After I posted this morning about the number of posts by hour on UMWBlogs, Jim reposted it, and from there D’Arcy Norman at UCalgary did similar digging in his WPMU tables. Since he’s working directly from the database, he could even dig out stats for comments, something that I’m unable to do.

One thing that really struck me was how similar the posts-by-hour data was between and UMWBlogs.

Again, posts by hour at UMWBlogs:

And at UCalgary:

That got me wondering if the data that D’Arcy grabbed about posts by day would be similar. Close, but a little different:



Looks like at UMW we can’t get ourselves back into the blogging mode after the weekend until Tuesday, while the Canadians come out swinging on Monday!

Hopefully, as more and more campus-wide blogging platforms come up, we’ll be able to compare this kind of data across a lot of institutions. I don’t know what might pop out of it, but it should be interesting. (BTW: Cole Camplese wrote just today about the question of why to run a service in-house instead of outsourcing to an external service. What we have here is just another example of his argument–because when its in-house we can get at the really interesting underlying data.)

If nothing else, I think this can start to give some real insights into how students are living their lives, at least in a very broad sense. Along those lines, here’s the parallel data that I fantasize about having:

  • Library patronage
  • Library book checkouts
  • Wikipedia readership
  • Beer consumption
  • TV watched
  • Homework time
  • Pleasure reading

Anything more that you all would like to do the comparisons on? Not that there’s any chance that I can actually dig all this up. I’m just sayin’ the comparisons would be interesting.

“Tis now the very blogging time of night…”: When We’re Posting to UMWBlogs

Tis now the very blogging time of night
When posts do churn and the bava breathes out
Edupunk to this world . . .

(Apologies to Shakespeare)

Over the weekend there was an odd glitch with the scrapers. They’re still down until we figure out exactly what’s going on, so unfortunately I might miss some of the blog data from over the weekend.

But it brought up the interesting question of, if I need to test something, when is it lease likely that people will be posting to UMWBlogs. Of course the intuitive guess is “late at night”. But I’ve got data, so let’s use it.

Here’s a table and simple bar chart of how many posts were written when. NB — many posts get aggregated and republished, which means that they show up in this data twice. Alas, I don’t have a good mechanism for tidying up the data there.

Looking at 16877 posts, for you folks who like that kind of stuff.

12am – 1am 631
1am – 2am 383
2am – 3am 164
3am – 4am 77
4am – 5am 62
5am – 6am 59
6am – 7am 118
7am – 8am 268
8am – 9am 548
9am – 10am 812
10am – 11am 1066
11am – 12pm 1020
12pm – 1pm 1026
1pm – 2pm 980
2pm – 3pm 1267
3pm – 4pm 1257
4pm – 5pm 1112
5pm – 6pm 766
6pm – 7pm 677
7pm – 8pm 856
8pm – 9pm 1014
9pm – 10pm 934
10pm – 11pm 910
11pm – 12am 870
Blogging Times chart

Blogging Times chart

I was actually a bit surprised by how late into night people are blogging — I expected it to taper off a little earlier. I’m also curious about the dip right at 5 and 6 pm. Is that just dinner time? If so, it might be interesting that our blogging practices are so closely tied to that social norm.

And it looks like if I’m going to be testing things, I’ll need lots of coffee.

Whose Link Is It, Anyway?

Continuing my fascination with links, I’ve got a new Exhibit in the spirit of the Link Friends Exhibit. That one is an overall list of common links between posts. So it gives you a list of URLs and the posts that link to them, ordered by the number of posts that link to it.

This one gets blog-specific. For any given blog URI, it digs up all the things it links to, if some other post anywhere in UMWBlogs also links to it. Like so:

You can start here, and just type or copy-paste in the URI of a blog in UMWBlogs, and it’ll dig up the common links. Or, start with the exhibit for this blog. (For blogs with many links, it takes a little while to process the data., for example, pretty much stalls) There are facets for tags on all the resources, though it’s unlikely right now that the common link will have been tagged. In the happiest circumstance, the common link will be to another post within UMWBlogs, in which case it just might have been tagged or categorized.

The pop-up for the common link also gives you an option to go to a page of all the posts and blogs that link to it.

The idea is that if you find a blog that you like, chances are that it will link to other things that you like. And if indeed it links to other things that you like, other, unfamiliar, blogs will also link to it. And so you can traverse through (1) blog you like (possibly your own!) -> (2) what it links to -> (3) new blogs that you might like, based on common linking.

SemanticUMW Celebrates 15000 Posts With Style!

Today, the scrapers scraped in their 15000th post from UMWBlogs. Way to go to everyone at UMW using it!

To celebrate, I’m also ready to finally add some better design and style to the exhibits. Thanks to the hard work and many tears that I caused her, our student aide Serena Epstein has added a prettiness to the exhibits that I would never have been able to accomplish. Thank you Serena!