Idea: Stanley Jordan to Host A Video Series on Array Languages
Stanley Jordan
documentary or masterclass series on Video/CD-ROM/DVD/Quicktime, etc.
I am a musician, primarily jazz, with a relatively high degree of
notoriety in the general public. I am a huge APL fan/advocate, and
have written lots of APL code. Most people are surprised when they see
how advanced my software is considering that I'm a musician. I think
that makes an important point about APL and productivity. There is no
way I could have done all that in a traditional programming language
such as Java or C.
I might just go ahead and do this on my own anyway, but if the APL
community rallied around such a project it could be so much better. It
would benefit from high-quality video production which, although
cheaper than making an MTV-style music video, is not exactly free. I
could use assistance in writing and organizing the material, preparing
budgets, hiring production crews, etc. Vendors could supply
interpreters and tutorial assistance, so that a number of dialects and
products could be represented. Perhaps I could interview some
luminaries in the APL community and go on-site to places where array
languages are being used. It would also help if we can find people who
are well-respected in computer science in general who have a fondness
for APL.
This could even be a whole series, or a number of different versions
for different audiences. One for a more general audience, including
laypersons with a general interest in science and technology, or MIS
managers and other corporate administrators who have authority to
bring APL into their companies. Another version could be for students
and would actually teach them the language. Or how about a highly
technical presentation for programmers who are proficient or
knowledgeable in computer science but are currently using other
languages?
Yet another version could focus on my own projects. In the 80s, I was
featured in a segment of a Novus series on technology that was aired
on PBS. I demonstrated some computer-aided instructional software for
music that I had written in FORTH on a Commodore 64. I still get
comments from people who saw that. My current projects are way more
advanced, so I would think the impact today could be even greater,
particularly if the presentation is more in-depth and is
well-researched, well-written and well-produced.
Obviously, this is an idea in its formative stages. Any help
developing it would be greatly appreciated.
Cheers,
-Stanley Jordan
http://www.stanleyjordan.com
http://www.stanleyjordan.com/Links/Computing.html#APL
http://www.stanleyjordan.com/Technology/technology.html
email responses can be directed to a...@stanleyjordan.com
François GAILLARD
> I am a musician, primarily jazz, with a relatively high degree of
> notoriety in the general public.
I remember the 1st time I heard you (15~20 years ago?), I wondered how many
fingers you should have got.
> I am a huge APL fan/advocate, and
> have written lots of APL code.
I would not have bet a cent on that!
[snip...]
Overall, I think it is a very good idea and could leverage a dynamics for
APLs.
The tough part seems to get lots of individuals on the same boat. (APL
vendors for instance)
> http://www.stanleyjordan.com/Links/Computing.html#APL
The link to Eric Lescasse's APL Links site is broken. He has moved from cold
fusion to ASP.
It's now http://www.lescasse.com/Links.asp
--
François GAILLARD (former guitarist)
Paris, FRANCE
fg[@]fgaillard.com www.fgaillard.com
Look at my poor man's memory leak tracker:
http://www.fgaillard.com/English/prod.htm
"Stanley Jordan" <s...@stanleyjordan.com> wrote in message
news:f6f9a220.02033...@posting.google.com...
Kevin R. Weaver
This is a great idea and I'm discussing it with a few people. My other hat
(other than APL) is quite different as yours is. I have produced some 500
+/- fitness TV shows on ESPN and ESPN2 and some 75+ fitness video
programs...so, I'm quite familiar with production, crews, studio, editing,
etc.
More later after a few of us run this up the flagpole.
Kevin
Olivier Lefevre
write a MIDI APL2 auxiliary processor. I forgot his name and
although I must have archived the exchange, I am separated from
my backup CDs at the moment. Does somebody know if anything came
out of that project?
Regards,
-- O.L.
David Liebtag
Tomas Gustafsson
---
I've written an AP which drives the Roland MPU-401 MIDI interface
directly from the DOS version of APL2/PC (the version written by
Madrid/Winchester Sci. Center). I'm interested in locating a business
partner(s) who'd be interested in migrating it to other APL systems
(Dyalog, APL2000, APL2/PC, J, et al.) and marketing it. I think the
current version does work with IBM's TryAPL2. This AP packs quite a bit
of functionality including MTC/SMPTE support which is included in the
enhanced OEM versions of the MPU-401. All of the SoundBlaster and
compatible sound cards include support for the MPU-401 in UART mode.
Please direct serious inquiries to efr...@ms.com and
efr...@sprintmail.com.
Erik S. Friis
---
/ Tomas
"Olivier Lefevre" <Olivier...@genedata.com> skrev i meddelandet
news:3cab...@news.core.genedata.com...
Stanley Jordan
Assembler for maximum speed, the AP sends and receives MIDI data via
the MIDI Manager.
The newer Macs have USB ports and as far as I know there is no USB
driver for the MIDI Manager, so I have updated my MIDI AP to utilize
QuickTime Musical Intruments. This system not only sends and receives
MIDI but also plays sound through the computer's built-in speakers and
audio outputs using stored samples of instrument sounds.
I have also written other multimedia APs, including interfaces between
APL and Apple's Sound Manager, Speech Manager and many other parts of
QuickTime.
One nice feature of the Quicktime interface is the ability to open an
existing movie, import a sequence of frames as a rank-3 numerical
array, process the array in APL and export the result back into the
movie.
I also helped Erik on the APL2/PC project described below.
-Stanley Jordan
http://www.stanleyjordan.com
http://www.stanleyjordan.com/Links/Computing.html#APL
http://www.stanleyjordan.com/Technology/technology.html
"Tomas Gustafsson" <we...@sci.fi> wrote in message news:<a8h02k$mvb$1...@tron.sci.fi>...
Stanley Jordan
responses via email. I've been pleased with both the quantity and
quality of the replies. I will be getting back to the respondents some
time next week.
I didn't mean to imply that I was only interested in email responses,
because they can of course be posted here as well.
But if you do reply by email, please address it to:
a...@stanleyjordan.com
Thanks to all,
-Stanley Jordan
http://www.stanleyjordan.com
http://www.stanleyjordan.com/Links/Computing.html#APL
http://www.stanleyjordan.com/Technology/technology.html
s...@stanleyjordan.com (Stanley Jordan) wrote in message news:<f6f9a220.02033...@posting.google.com>...
Olivier Lefevre
> One nice feature of the Quicktime interface is the ability to open an
> existing movie, import a sequence of frames as a rank-3 numerical
> array, process the array in APL and export the result back into the
> movie.
How are such manipulations performed exactly? In Assembler, I presume.
But do you use the AltiVec subset of the PowerPc ISA or only "vanilla"
instructions? Not an APL question but I am curious nonetheless since
AltiVec and similar add-ons in other chips were designed with precisely
such applications in mind but, unless I am misinformed, are woefully
underused in the real world (no doubt because most C compilers still
can't recognize parallelism and generate such code automatically: you
has to write it and then insert in the right location by hand). From
a marketing point of view (since this is the topic of this thread), an
APL AP that uses the full power of AltiVec would be something to boast of!
Regards,
-- O.L.
Stanley Jordan
QuickTime frame data can be Imported and Exported using the QuickTime
API.
My QuickTime auxiliary processors interface APL to the QuickTime API,
and were written in 68000 Assembler. They work fine up through System
9.
I haven't yet updated the interfaces to PowerPC Assembler, but this
will be necessary for OS X. Do you have a 'Net reference for more
information about the AltiVec subset of the PowerPC ISA? I'd like to
see if it can be of use.
For the most part the APs don't require much sophisticated Assembler
code. Since the QuickTime API is so rich and comprehensive the only
low level code I usually need to write is glue code that calls OS
routines and passes data back and forth between APL and the OS. I do
everything else in APL.
My QuickTime interface is currently working in APL68000 Level II for
Power Mac, but I'm looking ahead to APL/X, the newly-released
successor. APL/X adds {quad}NA, which I'm hoping will eliminate most
of the Assembler code completely. Another nice feature of APL/X is
that it runs under Mac OS, Windows, Linux and AIX .
All the best,
-Stanley Jordan
http://www.stanleyjordan.com
Olivier Lefevre
> interfaces to PowerPC Assembler, but this will be necessary for OS X.
> Do you have a 'Net reference for more information about the AltiVec
> subset of the PowerPC ISA?
The official site is www.motorola.com/SPS/PowerPC/AltiVec but there is
another one at www.altivec.org. As usual Google returns many other
interesting links. AltiVec, btw, is a Motorola thing: IBM has always
shunned it in its own PPc chips.
> For the most part the APs don't require much sophisticated Assembler
> code. Since the QuickTime API is so rich and comprehensive the only
> low level code I usually need to write is glue code that calls OS
> routines and passes data back and forth between APL and the OS. I do
> everything else in APL.
Ah, in that case you may not need have much need for AltiVec indeed:
I believe these vector instruction sets are more oriented towards
computations.
Regards,
-- O.L.