Wednesday, May 16, 2007

Squeak: Tinker-Toy Software

Despite my evident cynicism, I'm still completely gob-smacked by posts like "Virtual Tinker Toys" by Doug Johnson, which starts in response to a comment by Sylvia Martinez:

I'm wondering if there is another criteria - something like "extensibility". The old idea of "no floor, no ceiling." I hate learning something new and then finding that there is no way to extend it, add options, go under the hood, etc. Not that everyone wants that, but in an educational setting, having tools that you only have to learn once yet can accommodate different learners and different styles seems like important criteria. Maybe there's a better word for it?

Of course, there are at least two good words we can use to describe this concept, or at least a big chunk of it: freedom and openness. I'm not going to belabor that point; follow the links if you're confused.

After an exchange of comments, Sylvia states that some activities which have what she's looking for are HyperCard (entirely defunct), HyperStudio (it lives!), Flash (which is both expensive and not designed for K-12 students at all), Microworlds (a commercial Logo which I'm sure is lovely) and good old coding HTML by hand (yes, it is good to know the rudiments).

Doug continues, incongruously:

First, it was a good reminder that teachers' resistance to technology is less about technology itself but more about unfamiliar ways of teaching. Learning HyperStudio is not an issue; adopting a new teaching philosophy in which students learn by creating instead of absorbing is an issue.

Perhaps he didn't notice that her list is really pretty weak, and the best candidate in there, MicroWorlds, isn't exactly universally available to teachers. We really should have better tools for this, and we do. And they are free (some libre, some just gratis). The problem is that they aren't well known.

I could point to various projects by leading researchers in our field like Mitchel Resnick at MIT or Randy Pausch at CMU, but the best example for what Sylvia is talking about is Squeak, which was started at Apple in 1995 by Alan Kay as an extension of many of the ideas first seen in HyperCard. The project was booted from Apple to Disney and then into the void, but because it is more or less open source software, the project has survived and grown, despite the fact that the US ed-tech community has chosen to ignore it. Squeak is highly platform and OS independent. It is widely used in classrooms in Spain and is integrated into the OLPC environment. Its foundations in the SmallTalk programming language are sufficiently robust that it has been extended to be used for a variety of new and cutting edge projects like Scratch and Croquet, the not quite ready for prime time open alternative to Second Life.

I'm simply baffled by the argument that the quality of the tools schools currently have is sufficient, particularly when there are superior, mature, free alternatives that aren't even being considered.

3 comments:

Sylvia said...

Hi Tom,
It's a good discussion to have, and I agree, my examples are weak. I thought those products would be fairly well known and simply would help explain the idea of extensible, tinker-toy software. I was not attempting to create a comprehensive list.

I'd love to work on that with you - want to start a tinker-toy-wiki?

I know Squeak is supposed to be ready for prime time, and I'm sure that its exposure to a large audience via the OLPC will do it good and make it stonger. At this point, though, I defer to educators like Bill Kerr (http://billkerr2.blogspot.com/search/label/squeak )
who advocate making games with kids. He likes GameMaker as a tool and writes about his struggles with incomprehensible Squeak documentation and weak user support.

Finally, I completely disagree that "no floor, no ceiling" is synonymous with freedom or openness. That may be your viewpoint, but it's not what I am saying.

While freedom and openness may be good things, they are very different than having a tool that offers both feature accessibility and growth potential. You acknowledge that MicroWorlds is the "best candidate" on my list but dismiss it because it's not free. The cost of software is not equivalent to its usefulness. Sure, free is nice, but what if I actually want to pay for something? There are many reasons - like superior design or customer support. I don't see why that option should be dismissed.

Tom Hoffman said...

Hi Sylvia,
I wasn't so much alarmed by your list, because it is quite difficult to think of "tinker-toy" software that is actually being used in schools right now. I was more taken aback by Doug robotically regurgitating the line about how the technology isn't a limiting factor. Clearly, the technology IS a limiting factor.

I do agree with Bill that actually using Squeak is exasperating, and if we were having a face to face conversation it would have continued with "Squeak is designed to do exactly what you want, and it is fantastic technology, but... you should probably try Scratch." But the thing is, if you look at Bill's critique of Squeak, much of his concern is with the size of the community (the lousy website, etc., are just manifestations of that). If we simply pretend it doesn't exist, that won't be fixed.

What is the role of a leader in this situation? You're looking at software of the highest quality (truly!) from an engineering point of view, which is if anything hampered by an overabundance of pedagogical ambition. It is as relevant and ahead of its time today as the day it was started, but has mostly languished due to a lack of attention. Isn't it the role of a leader to point to it and say "This is worth saving, what can we do to make this work?"

Regarding the role of software freedom, I am probably taking a longer and larger view than you. I am not only thinking as a customer considering an individual product to fit my short term needs.

Tom Hoffman said...

Jim,
What I actually wrote was really only the tip of the iceberg of what I was thinking, which wasn't just about Squeak, but I had to keep things to a reasonable length.