Re: How Robots Will Steal Your Job

[Roedy Green]

When we installed central air conditioning in our home about 8 years ago the
electric company put in a little box that allows them to limit
usage a bit when they are running at peak load. I think the
agreement is something like they'll not cut off more than 20 minutes
each hour.

The protocols may already exist. Very low bandwidth is needed, and the
AC wiring is already in place to deliver the signal wherever it is
needed. I think now it would be feasible to have such boxes that
were addressable. Then the electric company could fine tune the
load. You could sign up for varying degrees of Spartan algorithms in
return for reduced rates. Your actual bill could depend both on what
you signed up for, and what hardship you actually endured.

It makes sense different people would be willing to go for differing
degrees of cutback.

Of course in a pinch, the electric company has the option to do a
massive cutback to save a blackout.

Some wag will say "But the Java boxes won't work when the power fails"

One thing concerns me is Thorsten Veblen's conspicuous consumption and
conspicuous waste as methods of attaining social status. Like the
gas-guzzling Humvee, wasting electricity could become even more of an
indicator of social status. Perhaps if the boxes cost more than they
saved ...

 

[Roedy Green]

There is very succsesful visual language based on dragging, dropping and pluging together graphical
elements representing different functions - it is LabView made by National Instruments.

I used one back in the 80s devised by Gould Modicon for process
control. You did everything with a colour touch screen.

It booted from floppy, thumping out a Hiawatha beat, taking forever.

 

[Roedy Green]

(Incidentally a prediction of this theory is that all attempts to create
programs 'visually' by joining little boxes are misguided and doomed to
failure. They are trying to use the wrong bits of the brain.)

The other problem is they are created by analytical types who present
the programs three symbols at a time on screen. The whole point of
visual thinking is to take in the big picture at a single sweep and to
see the problem as a whole. You need a GIANT screen to do this
properly, perhaps LCD "wallpaper" covering every surface of your desk
and partitions.

To make do with small screens, you have to have a topological approach
where you stretch the surface to see more detail rather than panning
over it.

 

[Roedy Green]

I had a large database schema I had to analyze once, so I
printed all the diagrams out on 8x11 paper and taped them
all together to form a picnic table sized diagram.

One of the very first programs I ever wrote was to print out conflict
matrices in high school scheduling. They were too big to fit on a page
,so had to be printed out in pieces then taped together.

PostScript has generic ability to print out large pictures in tiles.

In my own computer language Abundance, I arranged reports so you could
print or reprint any range of pages in a report. It would precalculate
how long the report would be, so that each page was labeled page 3 of
20.

It will be interesting to see what effect cheap giant flat screens
have on programming practice. Spaghetti gets more manageable.

 

[Sudsy]

One thing concerns me is Thorsten Veblen's conspicuous consumption and
conspicuous waste as methods of attaining social status. Like the
gas-guzzling Humvee, wasting electricity could become even more of an
indicator of social status. Perhaps if the boxes cost more than they
saved ...

Roedy,
The boxes exist and the independents claim to want to install the
new meters. But at a cost of CDN$400 per, they want the province to
pay, which means we all pay.
But have you seen the numbers from http://www.theimo.com ? We've
been able to reduce demand from ~24GW to ~19.5GW in one week! A full
20% reduction.

 

[Sudsy]

One thing concerns me is Thorsten Veblen's conspicuous consumption and
conspicuous waste as methods of attaining social status. Like the
gas-guzzling Humvee, wasting electricity could become even more of an
indicator of social status. Perhaps if the boxes cost more than they
saved ...

Roedy,
The boxes exist and the independents claim to want to install the
new meters. But at a cost of CDN$400 per, they want the province to
pay, which means we all pay.
But have you seen the numbers from http://www.theimo.com ? We've
been able to reduce demand from ~24GW to ~19.5GW in one week! A full
20% reduction.

 

[Roedy Green]

The boxes exist and the independents claim to want to install the
new meters. But at a cost of CDN$400 per, they want the province to
pay, which means we all pay.

That price would drop considerably if the box were built into
appliances.

 

[Gerry Quinn]

Isn't that similar to the two-hemisphere theory of the brain?
The artistic side and the analytical side?

Yes, I imagine they all talk about the same thing. However there is a
tendency for people to blather about how the 'artistic' or
'creative' side is undervalued by society, when the truth is more likely
that it is overvalued by a lot of people who don't have much of a bump
on either side. The problem with the 'creative' side is that it's not
so good at evaluating the worth of its own efforts...

Gerry Quinn

 

[Gerry Quinn]

The other problem is they are created by analytical types who present
the programs three symbols at a time on screen. The whole point of
visual thinking is to take in the big picture at a single sweep and to
see the problem as a whole. You need a GIANT screen to do this
properly, perhaps LCD "wallpaper" covering every surface of your desk
and partitions.

To make do with small screens, you have to have a topological approach
where you stretch the surface to see more detail rather than panning
over it.

You're missing the second shoe of the argument - a big directed graph of
logical connections is NOT using the gestalt portions of the brain.
Topology is not geometry. Unless the two-dimensional Euclidean geometry
of the layout corresponds to some significant feature of the problem
space (as it does when you look at a map, for example) it doesn't mean
anything.

Directed graphs speak to the linear, logical thought processes.
Example: a class hierarchy (in object-oriented programming) is often
presented as a directed graph. But such a hierarchy actually represents
a separation of elements that might be aggregated and connected, pehaps
even visually, if a single final class were defined instead. Creating a
class hierarchy is a process of separating elements, of breaking
irrelevant connections - in other words, a process of analysis.

Usually the sorts of things that flowcharts are used for have no
coherent distance function that can be mapped onto the chart geometry.
When this is the case, you're not seeing anything extra when you see the
geometry of your flowchart. Of course, you do gain from not having to
turn the pages, as somebody has commented, but that's not usually a
killer feature, and you could get the same effect if the data were
presented in another form on a big enough sheet.

Gerry Quinn

 

[Gerry Quinn]

I think you are overlooking the capacity of the visual
system to narrow the focus to just small portions of a
diagram, scanning from one thing to the next, simulating
serial flow between ideas, and possibly emulating the
analytical model. But I'm no cognitive scientist.

Oh, the brain is all wired up so you can feed from one system to another
at any point (have you ever tried plugging one ear, then the other,
while listening to a radio broadcast - it's amazing that there is so
little difference, when the process of understanding speech is known to
be highly assymetric in the brain).

My point is that the high-level processing is taking place in systems
that probably evolved mostly to deal with audio input.

Gerry Quinn

 

[Programmer Dude]

It would precalculate how long the report would be, so that each
page was labeled page 3 of 20.

Wasn't that sorta confusing? :-|

 

[Evgenij Barsukov]

As I posted on another newsgroup recently:

My theory is that there are two main modes of intellect, a visual
'gestalt' mode that deals with two-dimensional images and the
identification of structures in a large mass of data; and an analytical
mode that works on time-varying signals, and deals with logical and
predictive thought. The first can be associated with the sense of
sight, the second with hearing. Clearly programming, chess and music
require particular strength in the analytic mode. Someone strong in the
analytic sphere might - or might not - be strong in the gestalt sphere.

(Incidentally a prediction of this theory is that all attempts to create
programs 'visually' by joining little boxes are misguided and doomed to
failure. They are trying to use the wrong bits of the brain.)

While your argument make sense, I think conclusion is jumping over several
steps. First of all, nobody is _creating_ program by looking at the computer
input (be it in perforated cards, lines of words, or in some graphical form). Program is
created inside your head. The process is not logical but rather intuitive,
and in any case it has nothing to do with the way program will be input in
the computer.

When you sit down and start inputing it, program is _ready_ in your head.
You are just making a mechanical work of translating your thoughts into
computer readable form. And the way how this "form" is realized is a matter
of convinience of dealing with it, not matter of improving thinking process.
Dealing includes "overviewing", "debuging", "searching", "accessing" etc.
Many of this things are not logical but instead directly related to pattern
recognition ability and therefore are better realized on visual level.

Regards,
Evgenij


--

__________________________________________________
*science&fiction*free programs*fine art*phylosophy:
http://sudy_zhenja.tripod.com
----------remove hate_spam to answer--------------

 

[Roedy Green]

Anyway, it was tough going until I made a visual "map" of where
I wanted to go. I then made that map my desktop wallpaper, and
it really made the work easier. Helped me keep the big picture
at my mental "fingertips".

I talk about this is my essay on scids http://mindprod.com/scid.html.

Unfortunately I have not had a chance to play with them other than in
imagination.

My room mate explained that she files things by location. She finds
computers annoying because they don't work that way. Her solution is
to print out material, then lay in out by location to organise it.

I would imagine that a program might place classes in 3D space. so
that related classes were near each other. You could ask, "If I were
to change the definition of method X, what would be affected". Little
winking lights would go off all over showing you the extent of your
proposed change.

 

[Airy R Bean]

Perhaps you are an inexperienced programmer, or the
programs with which you deal are trivial?

An indented IF statement in a high level language, a FOR loop.
or a CASE statement in whatever syntax, indented and using
meaningful variable names is readable and understandable
with no more than a glance. The equivalent in Labview where
you have to follow traces to find the nested blocks is an
exercise in Maze solving and not conducive to commercial
pressures.

 

[Eray Ozkural exa]

The implementation language is irrelevant to the
ideas being expressed within a program; at least
for imperative languages; I cannot speak of
functional languages having no experience
therein.

Implementation language is *highly* relevant to which ideas can be
effectively represented. A poor imperative language like C++ will not
allow you to express ideas beyond a certain complexity in a manageable
way due to its imbecile type system, mockery of genericity, broken
object system, un-orthogonal syntax/semantics, etc. Likewise for Java.

I think the worst imperative language is C++, followed by C and Java
(and C# recently)

For the example of a purely imperative language that really is
designed: Algol 68, Eiffel, Oberon, etc... I'm not really interested
in purely imperative languages any more, but I think the success of C
is simple to explain. Because it's a very awkward and incoherent
language and most programmers are not people who can really think well
owing to their beind dumb as a screwdriver, they like all sorts of
quirks and stupidities like macro processors :> And maybe C has just
the right kind of semantics that a stupid person would like: a
language that is only ad hoc. A mechanism that does not lead to
mathematical elegance but consists of simple pieces that can be
understood independently although they do not work together neatly.
So, I guess that's the rosary path of uneducated intuition.

Agreed, Java is a "cleansed" design derived from C++, but it is still
not a realistic PL in my opinion. Although C++ is a worse design, it
is more capable. I don't know, speaking of these languages make me
want to throw up. Really.

 

[Airy R Bean]

Give me an example of ideas which cannot be
effectively represented in C (A subset of C++ to
all intents and purposes) and I will return it to you
as an example of ideas which cannot be effectively
represented as a computer program.....

(Remember that C is a minimal high-level representation
of machine language.)

Once that you have the ability to represent information of
whatever source in some encoding of 32-bit integers and
arrays thereto, and once you have the high level structured
constructs of selection, sequence and iteration, and once
you have the nicety of long and meaningful variable names, I posit
that there is no essential difference between the capabilities of C,
Pascal, Basic, Algol, Fortran, PL/M, Coral, RTL/2, or
indeed any other HLL that you care to specify. The design
of a program is not in the language in which you choose to
code it. That coding is a tedious formality only; an interim
and ephemeral representation on the way to executable
code much as are the internal tokens generated by the
compiler.

 

[David Longley]

. I'm not really interested
in purely imperative languages any more, but I think the success of C
is simple to explain. Because it's a very awkward and incoherent
language and most programmers are not people who can really think well
owing to their beind dumb as a screwdriver, they like all sorts of
quirks and stupidities like macro processors :> And maybe C has just
the right kind of semantics that a stupid person would like: a
language that is only ad hoc. A mechanism that does not lead to
mathematical elegance but consists of simple pieces that can be
understood independently although they do not work together neatly.
So, I guess that's the rosary path of uneducated intuition.

Does this apply to folks' penchant for cognitivism perhaps? I suspect
that if one spends a long time with in the austere and extensional
environment of Computer Science and related disciplines, one must find
Cognitive Science and "all sorts of quirks and stupidities" promulgated
by its practitioners quite a relief. Alternatively, it *could* just be
an eccentric, mild psychopathology, something like pica (where women
start eating bits of coal etc when they're pregnant)..

Agreed, Java is a "cleansed" design derived from C++, but it is still
not a realistic PL in my opinion. Although C++ is a worse design, it
is more capable. I don't know, speaking of these languages make me
want to throw up. Really.

Hmmm... sounds like the holiday wasn't long enough or was otherwise
therapeutically interrupted. You didn't take any Quine or Skinner books
with you did you? Reading them (or that crackpot Longley's posts) would
have turned a much needed break into something of a busman's holiday!
You'd be much better off with something by Fodor - he even cracks the
odd joke or two, interspersed with little stories about his granny.

Now that's the sort of stuff that makes *me* want to throw up. Really.


(PS. Have you read the one about the handsome cognitivist and the wicked
behaviorist?)

 

[dan michaels]

(PS. Have you read the one about the handsome cognitivist and the wicked

behaviorist?)

I suspect this would be Steven Pinker and J.B Watson. Pinker's just as
pretty as genetic endowment can make one, and a 41 YO Watson
impregnated his 19 YO lab assistant. The union got Watson kicked out
of academia [and into a career in advertising - ha!], and also
resulted in a kid named Albert, who Watson named after his favorite
lab rat.

[.. or did you really mean Chomsky and Skinner, in Fodor's Precis on
Modularity? My examples are much better than Fodor's, obviously]

 

[CBFalconer]

ignored due to persistent top-posting.

 

[David Longley]

(PS. Have you read the one about the handsome cognitivist and the wicked
behaviorist?)

I suspect this would be Steven Pinker and J.B Watson. Pinker's just as
pretty as genetic endowment can make one, and a 41 YO Watson
impregnated his 19 YO lab assistant. The union got Watson kicked out
of academia [and into a career in advertising - ha!], and also
resulted in a kid named Albert, who Watson named after his favorite
lab rat.

[.. or did you really mean Chomsky and Skinner, in Fodor's Precis on
Modularity? My examples are much better than Fodor's, obviously]

You'd just better keep reading your Fodor references.

And whilst you're at it - read up on the genetic fallacy.