Three Threads Of Squeak

Justin Walsh jwalsh at bigpond.net.au
Sat Nov 10 21:47:54 UTC 2001


Yes Daniel you are so right.
It is getting a bit boring: going over the same old thing.
On balance, there is as much in what you have said that is correct than what
is not.
To save boring the List, and you embarrassing yourself any further, I'm
happy to write and receive directly.
Yours respectfully
Justin  jwalsh at bigpond.net.au

----- Original Message -----
From: "Daniel Joyce" <daniel.a.joyce at worldnet.att.net>
To: <squeak-dev at lists.squeakfoundation.org>
Sent: Sunday, November 11, 2001 5:17 AM
Subject: Re: Three Threads Of Squeak


> On Tuesday 06 November 2001 05:33 am, you wrote:
> > Re: Three Threads Of SqueakDaniel H. H. Ingalls wrote:
> >
> > "The purpose of the Smalltalk project is to provide computer support for
> > the creative spirit in everyone."..
> >
> > Justin:
> >
> > As was elaborated in "Three threads .." there is more than one level
> > involved in the actual use of a system prot-typing language like
Smalltalk.
> >
> > Smalltalk is seen by Ingalls to be a "tool" for the creative spirit.
> >
> > However there are in fact three levels of the creative spirit:
> >
> > Conceptual No reliable tools available
>
> So write them Justin, and stop yammering. Put yer code where your mouth
is.
> Nothing in the language prevents the writing of such tools. In fact, ST,
is
> the best platform I've seen for it...
>
> >
> > Logistical No reliable tools available
>
> Same here. Though, if by Logistal you mean code management, several such
> tools exist.
>
> >
> > Construction This is where Smalltalk, the tool, was originaly targeted.
and
> > yes! no other tools can beat it. Its idea is to encourage children to
play
> > with computers and thereby learn.
> >
> > Play as a (dubious) method for the development of children is not
> > satisfactory in the adult world of business. Here, Smalltalk is hard
> > pressed surviving. Cincom is marketing VisualWorks as a serious business
> > modeling tool. Unless it deals with original design short fall it will
> > surely fail.
>
> Every pyschologist out there will strongly disagree with you that "Play is
> dubious" for child development.  ;)
>
> Currently, there are no "serious" business modeling tools out there. If
you
> say Rational Rose, I'll LAUGH!
>
> If you mean converting pretty UML pictures to code, you can always do that
> by hand.  It's been done before.
>
> If you want to make it easier, write the tools. Smalltalk's underlying
> framework is so general and flexible, that writing such tools would be
pretty
> easy. Just lots of gruntwork though.
>
> >
> > "Our work flows from a vision that includes a creative individual and
the
> > best computing hardware available. "..
> >
> > Justin:
> >
> > The Vision Statement(Alans actually), the foundation stone of any
System,
> > clearly says that Smalltalk was meant to be a tool for the creative
> > individual. The architypal creative individual was Piagets (the child
> > behaviorist) child.
> >
> > Piaget:
> >
> > http://encarta.msn.com/find/Concise.asp?z=1&pg=2&ti=761557692&cid=10#p10
> >
> > "... Behaviorists encouraged experimental studies and were responsible
for
> > moving child psychology into the mainstream of psychology. Although they
> > contributed much to the study of children, their concepts eventually
were
> > viewed as being overly narrow...."
> >
> > Justin: Please note "overly narrow"
> >
> > By rights, if Smalltalk is be more than a tool for the manipulation of
> > children by computer programmers then it must be redesigned to
incorporate
> > the Conceptual and Logistal needs of adults that have used it ie Wall
St,
> > the CIA, Bankers Trust, Macquarie Bank etc, from the begining.
> >
> > All the hacking and patching in the world will not achieve that.
>
> Uhm, the language is flexible enough to accomplish all of these. Heck,
> Squeak can pretty much extend itself automatically in some cases.
>
> Again, I say, if you think the tools are lacking, write them. Heck, this
can
> be done for ANY Language, including musty old C. But I suspect it's easier
> for ST.
>
> Oh, and the CIA uses SmallTalk, The Analyst, a large spreadsheet/data
> management system, with a list of tricks and capabilities that makes Excel
> look bad, and was written in the late 70s...
>
> >
> >
> > "We have chosen to concentrate on two principle areas of research: a
> > language of description (programming language) that serves as an
interface
> > between the models in the human mind and those in computing
hardware,"...
> >
> > Justin:
> >
> > It is clearly not succeeding because of the reasons previously given.
> >
> > "and a language of interaction (user interface) that matches the human
> > communication system to that of the computer. Our work has followed a
two-
> > to four-year cycle that can be seen to parallel the scientific method:
> >
> > Build an application program within the current system (make an
> > observation)
> >
> > Based on that experience, redesign the language (formulate a theory)
>
> Well, it's been a lot already. ST is the product of 40 yrs of computer
> language development, and Squeak currently makes use of several
extensions.
> Plus the fact you have direct access to the compiler, you can add and test
> easily.
>
> Again, write code....
> >
> > Build a new system based on the new design (make a prediction that can
be
> > tested) "
>
> Keep rewriting the language, and never get anywhere. Eventually, a
> programming language becomes "Good Enough". You have to stop rewriting it,
> and start writing apps.
>
> Small Talk is Good Enough. Whether or not it supports Logistics or
Concepts
> is a  HIGH LEVEL problem. They are programs. You could write them in C, or
> Cobol. You could write them in assembly (which is what all programs are
> eventually).
>
> So again, if Squeak doesn't have the tools, I say WRITE them, and stop
> boring us with your weird writings.
>
> Write a UML editor, a Database code management system, etc etc.
>
> The lacking of these items in Squeak has nothing to do with the language,
> but the people willing to write them. They are TOOLS, not language
> constructs. In all your writings, you seem to have this mixed up.
>
> Just because Squeak lacks these tools, does not mean it is a failure.
Heck,
> look at C, it has NOTHING unless you use umpteen gazillion libs, and it
runs
> on nearly anything.
>
> You seem eager enough, write them!
>
> And BTW, if you really think the basic Squeak dialect needs to be
extended,
> then look in Compiler, and read up on Primitives.
>
> >
> > Justin:
> >
> > The above empirical scientific method is questionable.
> >
> > If you study the method closely is parallels the actual empirical method
of
> > the behaviorist (above).."their concepts eventually were viewed as being
> > overly narrow...."
> >
> > The rest follows pretty much in the same vein.
> >
> > The Smalltalk-80 system marks our fifth time through this cycle. In this
> > article, I present some of the general principles we have observed in
the
> > course of our work. While the presentation frequently touches on
Smalltalk
> > "motherhood", the principles themselves are more general and should
prove
> > useful in evaluating other systems and in guiding future work.
> >
> > Just to get warmed up, I'll start with a principle that is more social
than
> > technical and that is largely responsible for the particular bias of the
> > Smalltalk project: ....
> >
>
> Ugh, more hot air? Please no.
>
> Maybe stop writing weird essays, and start writing tools????
>
> -Daniel
>
>





More information about the Squeak-dev mailing list