Back in May, I gave a presentation down at UMW on Just-in-time Learning, where I argued that the skills typically associated with the liberal arts were particularly suited to our modern challenges. In short, what we call 21st century skills are 19th century skills, rethought through network lenses.

I gave the example of trying to work through a claim about Meatless Mondays and CO2 — what do you need to be able to do to parse through that? What skills? What knowledge?

It turns out of course, that you need a little bit of everything: some knowledge for a starting point, search skills, an understanding of what constitutes trustworthy information on the web, a basic understanding of how formulas work, etc. You need, particularly, an understanding of the gotchas associated with this sort of thing.

This morning I heard a neat K-12 example to illustrate this on NPR  – students in a primary school using iPods to identify trees around their school.

If you think about the mish-mash of stuff you need to do this it’s pretty amazing.

The task reminded me of a similar real life task, when Dave Cormier’s daughter had eaten a number of berries, and he tweeted a picture to people to find out if they were poisonous. I was able to find it quickly, but it involved intersecting sets of skills and knowledge. The first realization was for that task Google Images was the correct tool. The second task was linguistic — someone must have described it before — what words would they have used? Red stem, maybe, white berries, shiny? That led to a page with a lot of options, so after clicking through one or two I realized what I really wanted to know was whether the berries were poisonous — and I was happier to get a false positive than a false negative.

(Science geeks will immediately see the connection between Type I and Type II errors in hypothesis testing).

Given that preference, I typed in something like

“red stem” “white berries” “black dot” poisonous.

and got nothing, so I scaled it back to

shiny white berries poisonous

and lo, a page from Minnesota identified them as white baneberry. Checking the wikipedia page I found that they were a) extremely poisonous, and b) native to the Northeast. Close enough, I tweeted Dave back.

That’s a really simple task, partially because we have a tolerance for false positives, but largely because we only need to get to visual identification, and we are set. Once we have the right picture to compare we let millions of years of evolution of the eye kick in and do the rest for us automatically (I don’t think it’s that far a stretch to say that evolution favored people that could accurately compare berries pretty heavily).

It’s easy (and tempting) to think that all tasks are more or less like that, but my next story indicates that — well, not so much.

I was presenting to a class last week on the nature of political blogs, and this question kept coming up — how do you know what you are reading is true, isn’t funded by a company, etc., etc.

There’s a bunch of ways you can answer that (certainly it’s odd for people to say this is a blog problem when the Washington Post is basically owned by a for-profit education company, and yet reports on education reform).

But the positive answer is it’s really freaking hard to know what blogs to trust, and it takes time and a ton of background knowledge. Not just knowledge of the blogs and how to figure out what the network trusts, but also of the current ideological divisions in the U.S., the players involved with funding politics, the inside baseball stories of the progressive democrats vs. the centrists,  the tea partiers’ relation to the GOP and also to certain astroturf organizations, what astroturfing is, the nature of sockpuppets, trolling, etc. You probably have to know a thing or two about the failures of mass media over the past decade, and understand why cable news covers the things it does and why that’s going to differ from blog topics. You certainly have to understand the Drudge phenomenon, and what a talking points memo/fax is.

Or you have to have a friend or expert you trust that can work out these things for you, at least until you work them out, progressively, for yourself.

This has nothing to do with the nature of blogs, it’s really always been true. It was true in the age of newspapers and the age of network news.

But I was struck by how much people wanted a quick checklist that would allow them to operate at an expert level on this. There’s not really a shortcut around the knowledge you need to do these things. There are certainly decent entry points, but the fact that I can do Just-in-time research on a political issue effectively depends on a whole bunch of knowledge I’ve accumulated over the past four years…

I’m not sure what the point here is I suppose — maybe that all the techno-utopian talk we sometimes engage in leaves a lot of people with the idea that this network stuff makes everything easier. I think, on net, it makes it better, deeper, more inclusive, self correcting and productive. It makes a lot of things possible that just weren’t possible. But when you get deep into some of these tasks, it’s hard stuff.  It’s not going to let you skip background knowledge, evaluation, hypothesis testing, close reading, etc.  You still need those things, you just need to do them with the network context in mind.