Sub Ubi

From λ the Ultimate:

Of the programming languages that were written for teaching ankle biters (e.g. smalltalk, python), only logo seems to have remained focused on that goal.

Neither Smalltalk nor Python were ever developed as teaching languages. I think it is safe to say that LOGO was developed with pedagogic goals in mind. Like many languages, words were wasted on how “teachable” the language would be after it was designed, but neither Python nor Smalltalk were developed from the start for the purpose of teaching programming to novices. Period.

Digging deeper:

• Smalltalk has an interesting history, and the Squeak environment is certainly a well-designed pedagogic environment, but this is successful in part because of the consistency of the Smalltalk programming language.
• Despite Python being influenced by ABC (elch, and ugh), I don’t think it is right to say that Python was developed for teaching programming to everybody.

I know of very few studies predicated on the notion of designing a programming language for beginners.^[1,2] Everything else is just hot air. I, for one, want research to drive the development of a programming language for novices, not ad-hoc observations. As it stands, I’ll be dead and decomposing before computer scientists (in general) acknowledge that design is a process, and linguistic design is necessarily one that must be driven by empirical research and structured interaction with the end-user–especially when you are attempting to build a usable, teachable, and learnable language.

[1] My First Programming Language(TM)
[2] Related papers

It gets dark at 4:30 PM. I mean, the sun is below the horizon at 4:30 PM, not “oh, it might be dark around 6″ dark. And this afternoon, we had a rather decent fog settle in, which combined with the genuine dark makes for some serious dark-dark. Like, “Hey, is that a building over there?” dark.

Four days left until we make for the States.

Found via Slashdot, Role Fragmentation by Robin Sharp.

The article just makes me think, and that’s enough said.

The last three days were a bit diverting. I was inadvertently pointed at a potential source of funding from Microsoft, which was an unrestricted grant fund for curricular development projects on this side of the ocean. Grants were expected to fall between 5K and 30K euros.

I pitched the CSCS group on the possibility of going for this. We came up with an idea. We looked around and talked about it, on IRC and the mailing list.

Basically, the kinds of things we considered are hard. While the idea of getting money to have toys to play with, the CSCS group already has a very cool idea for a project for next term. Going for something like this grant might split the group artificially. Next year, it might kill interest for new students, knowing they have little input or say in what the group does. Who knows?

As it is, the group’s dynamic is OK, but nobody knows anyone else. Hell, people still don’t know me, and I’m arguably leading this thing. In short, we don’t have any tangible notion of trust—in abilities, commitment—anything. And I expect that to take time. Right now, it’s possible that a diversion like a grant attempt might be a diversion. And at this point, large diversions could seriously derail the group from it’s natural development process as a whole group focused on one goal that is inclusive of a number of ideas and individual desires.

The CSCS experiment is interesting, primarily because it’s success depends so much on the actions of a loosely coupled group of individuals. There aren’t any central cliques in the group, there isn’t a driving force with regards to leadership (I don’t force anyone to do anything, nor do I have any power of any sort over the participants). I actually don’t know exactly why students keep coming back—I have my ideas, but there’s something about the group that feels different than my experiences working with students in the past. Perhaps this is an artifact of not having grown up in the British educational system? Who knows.

At the end of the day, I hope the group continues to hang on, and do some cool stuff in computer science, Microsoft money or no. I think it will be fun either way.

Time for bed. Goodnight, moon.

Outsourcing rears it’s ugly head:

From Philip Greenspun’s weblog.

I would have paid money (I don’t have) to have been in that lecture hall. Just imagine the look on students’ faces as they put the pieces together, realizing that their University’s flagship educational content initiative, OpenCourseware, was developed not only out-of-house, but out-of-country.

Nifty.