Thanks Bret
Yes "experimentation" should be put on an equal footing with "simulation" (I would even tend to give the nod to the former).
A blast from the past (where there is some evidence for the following story being other than just a great story) is from a piece I wrote for teachers:
Galileo didn't have home video cameras and computers and Squeak to come up with the model. He did his discovery "with no money" by being very diligent about observation and noticing until he found a way to pin down what was happening crisply enough to map it with mathematics.
How did he do it? There doesn't seem to be an absolutely definitive answer to this, but there are many stories about it which have been pieced together from Galileo's notes and writings. Galileo's father was a professional musician and Galileo had an excellent reputation as a musical amateur on a number of instruments including the flute and the lute.
He had been doing many experiments with inclined planes using uniformly sized balls made from different materials and having different weights. He discovered that same sized balls of different weights appeared to go the inclined plane at the same rate of increase of speed regardless of angle.
One day he may have for fun idly rolled a ball or two down the neck of his lute. You can see that the frets of lutes and guitars are not evenly spaced.
At some point he noticed that the clicks of the ball on the frets were almost regular and realized that the wider spacing of the frets was compensating for the increase in speed of the ball. Now a wonderful thing about lutes is that, unlike guitars, their frets are made of the same gut that is used from the strings and are simply tied on. So Galileo could move them. He started to move them until he could hear an absolutely regular sequence of clicks (at some point he probably started to tie the fret material across his inclined plane).
When he got perfectly regular clicks, he measured the distances and found that the increase of speed (the acceleration) was constant!
One of the important conclusions here is that there are many interesting real science probes that can be done with materials at hand if the teacher understands the real science. This is one way to do this investigation "with no money", and rolling a toy truck down the inclined plane carrying a baggie of ink with a pinhole, is another.
Don't let the lack of a computer or equipment slow you down. Science and math are all around us. The world we live in is a vast lab full of equipment, if it can be noticed. There are free public libraries even in the most disadvantaged parts of the US that contain books about how to do all of this: the knowledge doesn't cost money, but it does cost time and interest and focus.
Cheers
Alan
From: Bret Victor <****************>
To: Alan Kay <****************>
Cc: Communications Design Group <****************>
Sent: Thursday, July 14, 2016 4:45 PM
Subject: Re: [cdg] particle systems and physics of particle systems from Khan Academy
re dropped objects for gravity models, pucks for momentum models, etc
One challenge is that current computing systems place a wall between the computational space and physical reality. Regardless of the quality of the programming environment, it's always much easier to program a simulated falling object "inside the computer" than it is to record and analyze a real one "outside the computer". The design of computers as boxes with an "inside" and "outside" (and a little mail slot called "I/O") is taken for granted. To the extent that people are encouraged to model "within" the computer, they become estranged from the world that they're supposedly modeling.
My group is building a non-boxed computing system that is grounded in and participant in the physical world, observing and responding, with close commingling of physical and computational processes. If you want the position of a falling object over time, you might just pick up some object and drop it. The time signal resulting from this physical movement is just as "immediately available" as the result of a simulation, and can be viewed and used identically. Observed signals can feed into computations and simulations; simulated signals can feed into real-world actuation. All this is live, continuous, visible, etc.
One goal is to put "experimentation" on equal footing with "simulation", to encourage "getting the actual models from nature" as you put it. But not in the sense of "setting up an experiment" (which itself creates a contrived artificial space that feels separate from nature) but rather by working in an augmented space which allows for observation and analysis of nature in vivo.
Hopefully this should be a good environment for doing the sorts of "real science" that you have in mind.
On Jul 14, 2016, at 6:42 AM, 'Alan Kay' via ALL <****************> wrote:Hi JohnI'm just getting back to "thinking about things". So I'm just ruminating in public at this point.I should mention that I originally came up with the scheme for the Galilean Gravity project with junior high schoolers in mind -- it has quite a few pieces to it. BJ (the teacher we worked with) was interested in trying it in 5th grade, and my advice was that we should try to stretch the project out over a much longer time (this is a general great heuristic for helping younger kids do amazing things -- they need a lot more time).(We saw this a lot earlier at the Open School where the principal would do full length Broadway musicals with the 5th graders and get fantastic results with about 4 months of rehearsals.)Similarly, we spread the parts of the gravity project over about 4 months and hoped that the iconic nature of some of the "things to remember" that we'd baked into the earlier parts would carry over into the payoffs with the dropped objects (this worked much better than I'd hoped for, and I was quite lucky to be sitting there in the classroom when one of the kids suddenly "saw and remembered" what was needed (that was a very exciting moment!))I just loved how well this actually worked!And I later wanted to do the same thing with a camera for 1D and 2D momentum models (Alex helped with this, but air tables are not so easy to get into classrooms, and there were no "air pucks" then -- now this would be pretty easy to do.) This also gets us to "triangle math" and a lot of important ideas that are handled so badly -- even almost viciously -- in regular schooling.One of the big questions in any good curriculum/practice is "when should it be easy, and when should it be hard?" The need for the latter is often missed, both in education and in UI and language design, etc. Many of the things that are most important for humans to learn are not well pre-disposed by human genetics -- and thus will be experienced as "hard". Learning something that is both "good" and "hard" means that the learner has radically changed the processes of their "brain/mind" to now be fluent in something that was barely (or not at all) there.Thus, a goal for "real education" is how to help learners deal with "importantly hard" (no matter how hard). On the other hand, "gratuitously hard" is a real time-waster (and most schooling is loaded with this to the exclusion of "what is actually needed".So, the way I interpret "the curriculum matters more than the tool" is that the tool should be *changed to help* with what is "importantly hard" and to diminish "what is gratuitously hard". (I don't subscribe to the idea that it is a programmer's job to make bad ideas in a language work anyway (just because of universality). That's really terrible for children, but has been disastrous for everyone trying to use rigid poor language features.CheersAlan
From: John Maloney <****************>
To: Alan Kay <****************>
Cc: Cathleen Galas <****************>; Communications Design Group <****************>
Sent: Thursday, July 14, 2016 6:20 AM
Subject: Re: [cdg] particle systems and physics of particle systems from Khan Academy
Hi, Alan (and all).
The Open School Galilean Gravity project was terrific! Part of what made it so effective was that the activities and discovery process were lead by a gifted teacher and carried out as a group activity in a social context over several weeks. Many of those same aspects apply to Doreen's City Building/Design curriculum. Having a gifted teacher/mentor/guide seems like one of the best ways to foster scientific and mathematical thinking -- when it's possible.
Unfortunately, gifted teachers are rare and schools that allow the flexibility to run an activity like the Galilean Gravity project are even rarer.
The question that many of us have been thinking about is how to bring great learning experiences like that to a wider audience. Kahn Academy gets the "wider audience" part, but at the expense of the learning experience. However, we recently met a couple of new Kahn Academy employees who were hired to disrupt Kahn's current practices and create better learning experiences. They are both former Apple designers with a passion for education, and I was unexpectedly impressed by them. They visited Parts and Crafts and spent an afternoon with Mitchel and Natalie. I think they would be excellent collaborators, and could provide a conduit for getting some of our ideas out to the world. I'm terrible with names, but Nagle has their emails, I believe.
Alan, I think you are suggesting that a possible HARC project would be to create examples of learning experiences that foster real scientific and mathematical thinking. For example, for Galilean Gravity we could provide some video frames to analyze so people don't have to drop things off the roof themselves. (The new Apple iPhones have slow-motion video capture, would give students more frames to work from, thus increasing accuracy and perhaps revealing second-order effects like air resistance). Our curriculum could ask students to notice things about the frames, suggest lining them up next to each and measuring the spaces between the falling object in subsequent frames, etc.
If we create a curriculum, I know of two places in the Boston area we could probably test it.
If anyone would like to use GP as the platform for some curriculum, I'd be happy to help. You could also use Lively, something from Alex, Flowsheets, Etoys, one of our other systems or none of them -- the curriculum matters more than the tool.
Is there a subset of us interested in creating some curriculum? If so, maybe we should get together offline to brainstorm. Let me know if you're interested.
-- John
> On Jul 14, 2016, at 8:26 AM, 'Alan Kay' via ALL <****************> wrote:
>
> They aren't really showing particle simulations as simulations. They are showing black boxes and parameters and claims.
>
> The first set of things they are doing is well within range of "detailed understanding from nature" by 5th graders (see the Galilean Gravity project at the Open School). This gets the actual models from nature as an honest form of "real science" for children.
>
> For example, the gravity model is just two lines of "increase by" code in a completely understandable and derivable differential form. Similar "cameras on hockey pucks on an air table" (or "air pucks") will give differential models for 2 D collisions of circular things at angles. (Alex once found an air table for this -- but now there are "air pucks" that don't need an "air table".)
>
> Lots of computing power gets enough collision interactions (it would be interesting to see how well this could be done from scratch in Etoys before having to go to Kedama). Certainly I'd expect to be able to do 50 or so "little blobs" these days..
>
> I think a good way to do this -- especially these days -- would be to get the children to program some level of kinematics simulation with the ideas taken from nature in the same manner as Galilean Gravity was. Once they've done it, a black box optimized simulation system won't be such a bad idea.
>
> One of the basic bugs of the Khan Academy stuff I've seen is that it focuses only on received knowledge rather than on how and why. This makes their approach both pre-scientific and pre-mathematical. It fits really well into "our first 200,000 years" but not at all well into the last 2500, and especially not the last 400.
>
> It's essentially "back to basics with transistors".
>
> Cheers
>
> Alan
>
>
> From: Cathleen Galas <****************>
> To: Alan Kay <****************>
> Cc: Communications Design Group <****************>
> Sent: Thursday, July 14, 2016 5:04 AM
> Subject: Re: [cdg] particle systems and physics of particle systems from Khan Academy
>
> Good to get your reaction Alan!
>
> I haven’t had time to go through them. I do think it’s great that they’re trying to show particle systems as simulations. Unfortunately, if the approach isn’t good and converts people (and kids) to “beliefs”, then maybe it’s better if they hadn’t done anything.
>
> I think it’s good for us to know what is out there and being presented, and then we also can figure out how to do it better!
>
> Kat
>
>
>> On Jul 13, 2016, at 9:58 PM, Alan Kay <****************> wrote:
>>
>> I watched the first few -- and the one on elasticity -- and did not like them at all (but I really don't like the general approach of Khan Academy on most things). Basically a pro-belief non-science approach when they could have gotten what's going on directly from nature. Yecch!
>>
>> Cheers
>>
>> Alan
>>
>>
>> From: Cathleen Galas <****************>
>> To: "****************" <****************>
>> Sent: Wednesday, July 13, 2016 1:18 PM
>> Subject: [cdg] particle systems and physics of particle systems from Khan Academy
>>
>> The new Pixar film “Finding Dory” is the basis for one set of lessons the Khan Academy has developed introducing particle systems
>> https://www.khanacademy.org/partner-content/pixar/effects
>>
>> Pixar in a Box (lessons on special effects)
>> https://www.khanacademy.org/partner-content/pixar/start
>> https://www.khanacademy.org/partner-content/pixar
>> --
>> ************************
>> To post to this group, send email to ****************." target="_blank" href="mailto:****************.">****************.--
************************
To post to this group, send email to ****************.--
************************
************************
