[OT] "Pre-announcement" of Python-based "computing appliance" project.

[Richard Hanson]

Nice to find that my off-topic post was taking seriously even if I
*did* use lots of smileys and such -- as well as positing a possibly
seemingly grandiose set of ideas. :-)

Carlos Ribeiro wrote:

> I really like this kind of stuff. Information management, knowledge
> management, complexity management, you name it -- it's a increasingly
> interesting topic. Appliances also are relevant, as it is seamless
> networking.

Check on all the above -- and many thanks for the kind words and vote
of confidence.

It is my thesis that the world is experiencing a "complexity bomb" if
you will -- a combinatorial explosion of complexity in all things --
and that compsci has been dealing with managing complexity for
decades, and that if the world doesn't turn to the complexity-control
experts (compsci folks), then things may well be grim in the coming
years.

I've long been frustrated, though, that even in the computer industry,
the mainstream has gone the WIMPy way :-) -- instead of continuing to
innovate re HCIs. While I have personal, selfish reasons for working
on my project (such as finding that I am spending more and more time
trying just to *find* stuff on my harddrives, and being increasingly
hampered by a diminishing ability to drive mouses all around the
screen :-) ), I also realize from helping many of my non-computer-type
friends that they are mostly totally bewildered about such
distinctions as harddrives-directories-files, OS-apps, etc. Commonly,
I get frantic calls asking for help to fix their computers, often
with, "I was cleaning off my harddrive and now my computer doesn't
work." Usually, this sort of problem turns out to be that they deleted
the Windows directory, or perhaps that the Program Files directory
seemed a wee bit too cluttered, and so on... :-) The blame points
towards poor design and quite misleading marketeering on the
industry's part, I respectfully submit. The public thought they were
buying a microwave but got a space shuttle... :-/

I realized in my original epiphany perhaps seven or eight years ago,
that something like Pathfinder was possible which would be safe and
easy for Granny, and yet allow me to be much more productive, as well.

> I heed you to check Microsoft's documents on "how Longhorn will be
> great" :-) 

Heh. When they announced such a few years ago, I felt a bit validated
that my ideas were not psychotic. :-) But also, I felt an urgency to
get an open-source version out. As I said in my original post,
however, I had no time at that time, to devote to Pathfinder while I
was involved in another project.

(I also note that MS has recently announced that they are dumping the
database file system from Longhorn as they scramble in the Sysiphean
task of "fixing Windows." :-) )

> Seriously, they're working hard to create a new paradigm
> for the Windows desktop, and some of the things that you talk about
> are discussed there too -- namely, they're hiding folders, drive
> names, etc, from the user view, and creating a new and much simplified
> user interface. 

I am somewhat familiar with MS's theoretical work. However, it is also
my thesis, that open-source software (with a suitable BDFL -- likely
not me, as I'm getting too old; perhaps I could be a pope or
something... ;-) ) won't... uh... screw things up as badly as a
profit-driven megacorp. (MS *does* have the ability to solve the
hardware platform problem -- something OSS may have greater difficulty
dealing with.)

> Google is known to be working on something similar --
> your computer will be just a big Google-indexed repository of
> information. No more trying to file things into folders, Google will
> categorize and locate things back for you in a snap.

Now Google is a horse of a different color. :-) Yet, with their
incredible success, one wonders if profit-motives could possibly
squelch innovation? And now that they're going public, they'll have to
answer to the stockholders -- now *that* cannot be good... ;-)

I know the founders of Google are some big branes, and that Python was
used originally, and still is, in their shop. That's a good sign. :-)

(By the way, ideally, Pathfinder would not use trees of directories --
"flat is better than nested." [Indeed, initial prototypes would have
to kludge a flat structure on top of the directory trees used in
Windows and other file systems, as Windows in particular, really bogs
down with a dir containing ten of thousands of objects, say. And we're
talking about hundreds of thousands of objects, at least.]
Pathfinder's "button-panel" menu-structure *could* be a
"simply-connected" tree, but wouldn't have to be -- the menu-structure
could just as easily be "non-simply-connected" [presuming I have the
nomenclature correct]. Additionally, the menu-structure would be
easily changeable by the user: some may want the same leaf to be
located in several or many branches, or that a leaf be found via
several different routes. I would think that there would be some
standard menu-structures provided, but also that the users may well
want to change the dynamically labeled button-panel menu-structure,
themselves, and such should be made easy to do.)

---

There also is a developer in Italy with an open-source project that
features many of Pathfinder's ideas. While his work is good, and his
work further validated mine -- indeed, some of his writing reads
eerily like my own copious notes on Pathfinder -- his project is not
the complete "paradigm shift" that Pathfinder is envisioned to be,
particularly re accessibility issues.

> A project of this size is really big and ambitious. A suggestion is to
> create a new mailing list and to start discussing it there. If it's in
> Python, I'm sure you can thrown baits here at c.l.py from time to time
> to hook new people :-)

Thanks for that suggestion!

(Perhaps I should also hook up with the Italian developer mentioned
above...? I've lost the link to his website, but maybe I'll find time
to relocate it via Google.)

But back to your excellent suggestion: My concern is that I may not be
around along enough to even get the project launched in the envisioned
direction. (Thus, my "dumping" of my off-topic post in here,
yesterday.)

But, you have *greatly* encouraged me; perhaps I *can* get a prototype
working -- such would speak volumes louder (to mangle metaphors :-) )
than just a posting of some ideas by a newcomer to c.l.py
(active-posting-wise; I've been lurking here and elsewhere for years).

If I am successful in getting some reliable hardware rebuilt (I use
laptops, preferably with touchscreens, due to physical handicaps), I
may well be able to get a prototype of Pathfinder implemented. Python
makes the development of even big projects so much easier than the
myriad of languages I've used prior, that there is a slight chance
that I could get something working, perhaps...

(I have hesitated to mention my physical limitations even though I
hinted at my age, and obtusely to health problems, in my email
address, as such can often be taken for whiny complaining. :-) But,
ultimately, my physical limitations have been a blessing in disguise
for many, many reasons. Indeed, it is the arthritis in my arms and
hands which has been some of my main impetus for the "button-panel"
replacement of the WIMP interface which I elaborated on more in my
original Pathfinder posting, yesterday. And, in many other ways, I
have found much silver within the clouds. So I am decidedly *not*
complaining, nor even remotely bitter, nor whining, etc. about my
limitations. I want to be absolutely clear about *this* point.
However, accessibility issues *are* related to HCI issues, so perhaps
it *is* appropriate that I mention my own situation at this juncture.)

---

The more I think about it, the more I like your mailing-list idea.
I'll see if I can figure out how to start one which is mirrored-by and
postable-to the newsserver Gmane -- I like their resistance to
disclosing private email addresses which spammers could harvest.

---

I'm more of a generalist; it is slightly possible that I am not
deluded in my thinking that my background in programming *and* art may
well be the requisite background that folks bent on improving the HCI
will need. My being intimately familiar with the accessibility issues
of physically handicapped folks may also be of benefit in improving
the HCI situation. (And, of course, laziness is a *very* good
motivation for innovation... ;-) )

Well, I've rambled way too long, again. :-)

I will post any notes pertaining to this project's progress, i.e.,
becoming more concrete, even if it's just an announcement of a
mailing-list as you suggest, here in c.l.py.

Again, thanks very much for your comments. I shall continue to forge
ahead as time and energy allows. At least now, some distillations of
my notes from years of work (mostly back-burner stuff till very
recently) is in the public record.

And, once again, thanks very much if you or anyone has read through
this far. 

(I'm afraid if I keep posting, I *could* turn into a "postingbot"
rivaling some of c.l.py's best-loved and most-prolific writers. :-) ) 


the-grandeur-of-delusions'ly y'rs,
Richard Hanson

-- 
sick<PERI0D>old<P0INT>fart<PIE-DEC0-SYMB0L>newsguy<MARK>com