On Oct 15, 2009, at 5:50 PM, alex wrote:
>
>> They certainly aren't - textual symbols are processed
>> by different parts of the brain.
>
> I'm not sure if the "different parts" thing is true, from what I can
> tell from a naive glance at the literature, brain areas are
> multi-talented things. I agree though that reading is different from
> looking at a picture though so we're on the same page.
>
> However, you're wrong to imply that we don't read a max patch. It's
> full of symbols! In max a box is meaningless unless you type into it.
>
> Look at all the words and numbers in this simple patch:
> http://taknight.files.wordpress.com/2009/09/firstnotes1.jpg
>
>>> It's a very different experience
>>> connecting lines and occasionally typing in object names than typing
>>> everything in words and symbols.
>>
>> Sure but it's a different experience writing code in eclipse than it
>> is in emacs, and different again in vi. They all require visual
>> faculties, and some (not PD, and only in a limited way Max) require
>> spatial arrangement.
>>
Yeah, my argument is pretty weak here, but I can only speak from
experience (my own, and others') programming Max, that is just
*feels* different. Wishy-washy, yes, but sometimes you have to trust
your feelings. I'd love to see some sort of study about this, though.
>> That's not the case for most people, especially non-
>> designers. There
>> have been studies that show that their brains recognize words as
>> meaning
>> something, but not as graphics or symbols in their own right (I
>> hope I am
>> explaining this well... wish I could dig up that paper).
>
> No need, I agree. Similarly hearing sounds is perceived differently
> to hearing words (cf sine wave speech and noise speech). But I think
> it gets more complicated than that when we're coding. When we zoom
> out and think about the structure of what we're looking at, we think
> in more spatial terms, and structure of the code on the screen helps
> with that. This is true both with max and with java. With max the
> spatial arrangement is not part of the language. With java, more of
> it is part of the language, with organisation into classes, objects
> etc.
In grad school, a colleague wanted to do a project that drew data
graphs based on the spatial layout of Java programs, so you could
visualize their overall structure. The instructor said "that's not
interesting." I'm still disappointed...
>> For most people,
>> words are words and pictures are pictures, and while they get the
>> hang of
>> the "text-as-symbols" eventually, they need to adapt to them over
>> time.
>
> Sure, but I think we diverge where you say a max patch is a picture.
> The interface might let you make a picture but that is nothing to do
> with the language.
>
Like Michal said:
> This has been brought up in this thread, I think, but the fact that
> the visual aspect is more appealing to some people (especially those
> who are non-programmers) is that one can see the paths the data flows.
> Or at least they think they can because, in reality, in most patches
> (beyond "hello world" stage) one can only a little bit at a time.
> Yet, it is easier to see those connections then to make a mental image
> of them.
Yes, the illusion that the program is more than just a collection of
words is important, I think. And I do mean *illusion* because, as
Dave pointed out, all these bits of the programs could have been
represented in so many different symbolic ways, but choosing boxes
and lines and words like "moses" (and don't forget the really great
Max objects like "anal" and "plur") is so, so arbitrary. Plus, the
patching languages are very bad at real abstraction, that is,
representing large groups of virtual objects (such a bank of
synthesizers).
But the illusion is important, from an experiential standpoint (dare
I mention "marketing?"). It draws people in. In the same way, I like
that in Fluxus, the text sits on top of the 3D world, because that's
more enticing than watching emacs. Sorry, it just is, unless you
have an emacs fetish.
> But yes great point about accessibility. What makes max accessible?
>
> 1/ Live coding
> 2/ You can put words anywhere, it doesn't matter.
> 3/ Great help
> 4/ Easy to share patches
> 5/ You can't make syntax errors (syntax checking happens at mouse-
> click time)
> 6/ You can edit someone else's patch without fully understanding it
> (due to 5/)
> 7/ Nice a/v library functions
> 8/ Some myth about max being more 'visual' means people drop their
> guard
Well put :)
> What makes java accessible?
>
> 1/ ... err
>
Processing!
1/ Complete IDE with nice color scheme
2/ Auto-formatting of text
3/ Great help
4/ Great collection of examples and community
5/ Simplified Java lets you hack others' programs without really
understanding
6/ Nice a/v library functions
7/ Some myth about Processing being "visual" means people drop their
guard ;)
>> and the "accident" of using text to describe
>> complex instructions (computer programs) to machines due to
>> typewriters
>> being so ubiquitous. I'll leave you with that thought...
>
> Oh, that reminds me of something I was reading yesterday by Ted
> Nelson:
> http://geeks-bearing-gifts.com/gbgContents.html
>
> Yes it's strange that seems so solid and fixed about computing is so
> arbitrary and, well ... broken. The qwerty keyboard is a fine example
> of that.
>
Is it strange? Sounds like the state of the entire world these
days... see "The Financial Industry," esp USA
> I don't think the notion of text, as discrete symbols, is an accident
> though. If you are going to organise your ideas about the world into
> discrete categories, to understand it, make predictions etc, then
> you're going to want to give those categories names (symbols), and
> want to combine and manipulate those names with computations.
Something I like to point out is that intelligent people almost
ALWAYS write well. You can tell the best students in class ebcause
they document their work and can express it textually.
Still, its hard to see the other side, because there are no more
Egyptians around who write in hieroglyphics. I would love to do some
livecoding in hieroglyphics, or at least strange images (performing
logic operations on barnyard animals, anyone? That sounds a bit
dirty, actually...)
> alex
>
> --
> http://yaxu.org/
Evan Raskob
ML Studio
4-8 Arcola Street
London E8 2DJ
United Kingdom
http://mlstudio.co.uk
http://pixelist.info
Received on Fri Oct 16 2009 - 11:53:45 BST
