Wither CSGNet (was Re: Intro to Javascript)

rsmarken · March 31, 2014, 5:12pm

[From Rick Marken (2014.03.31.1010)]

Adam Matic (2014.3.31.1620 cet)--

Hi Rick,
I see the testing post, but not the post from March 28. I think I have the
same problem with a post from March 29.

Hi Adam (and everyone else on CSGNet).

It turns out that the listserve software at the U of Illinois that
support CSGet was changed on March 28. This was supposed to have no
impact and to a large extent it didn't. But It apparently did change
some settings on individual accounts. For example, I had set up my
connection so that I got sent an acknowledgment from the list server
that it had distributed my posts when I posted them. Apparently this
setting was changed to "no acknowledgment", apparently the default
setting, when the software was changed.

Also, the address to which posts are to be made was changed from

csgnet@listserv.illinois.edu

to

csgnet@lists.illinois.edu

If you post to csgnet@listserv.illinois.edu it still works because it
is automatically forwarded to csgnet@lists.illinois.edu. But to be
completely up to date we should be posting to
csgnet@lists.illinois.edu, as I'm doing with this post.

But I'm wondering if we shouldn't think of moving this whole operation
to a more easily accessible forum, like a social media site. I don't
use social media much but maybe we should think about bringing CSGNet
into the 21st century. What little experience I have had with
discussions on social media has been rather disappointing-- I prefer
writing in email format -- but maybe some of you young people out
there, who are the future of PCT, have some helpful ideas about this.

Best

Rick

···

Adam

On Mon, Mar 31, 2014 at 3:25 AM, Richard Marken <rsmarken@gmail.com> wrote:

[From Rick Marken (2014.03.30.0625)}

Just testing to see if the list is still working. They made some
changes to it, apparently, onMarch 28 and I posted something earlier
and go no acknowledgement.

On Fri, Mar 28, 2014 at 6:24 AM, Adam Matic <adam.matic@gmail.com> wrote:
> [From Adam Matic (2014.03.28 1430 cet)]
>>
>> BA: That's an excellent start. Is the simulation supposed to run in
>> real
>> time? I only see movement when changing a parameter value or dragging
>> the
>> "playerindex" slider.
>
>
> AM:
> No real time simulation yet, only fast time. Changing any of the
> parameters
> on the GUI recalculates the whole simulation, and the 'playerIndex' just
> goes over it.
>
> I considered adding a real-time 'live' mode, that would be fun. To be
> able
> to set reference and disturbance values in real time, and have them
> graphed,
> like in the Live Block Demo
>
>
>>
>> BA: The Javascript syntax looks like an object-oriented version of the
>> C
>> language, (as does Processing). I didn't have any trouble following
>> what
>> the code in basics.js does, but I've had plenty of exposure to
>> object-oriented programming.
>
>
> AM:
> Yeah, looks like OOP. Watch out for pointers to functions, those are
> weird
> at first, but very useful.
>
>>
>> BA: Have you seen Vensim? It allows you to create a diagram with
>> components representing flows (of variables and signals), quantities,
>> etc.
>> And the run it as a simulation. A definite plus is that it includes
>> sophisticated procedures for doing integration, etc. accurately.
>> Apparently
>> there is now a free version available online "for evaluation and
>> educational
>> use," see https://www.shodor.org/tutorials/vensim/pre.php .
>
>
> AM:
> I've downloaded several of the analog circuit simulators of the SPICE
> variety, and there are online tools like the Circuit lab. There is also
> "DESIRE" by Granino A. Korn, author of Electronic Analog Computers book
> that
> Bill mentioned. There is the modern, free and open source SciLab
> recommend
> by control engineers.
>
> I can't say I'm a fan. They are certainly useful to some people, but for
> me,
> there's too many options and elements to choose from, they are big and
> cluttered. The biggest pain is that they lack connection to some kind of
> a
> 'world'. If you want to simulate something with realistic physics, you
> really need to understand the physics of it. That is one reason I like
> robotics - the physics part takes care of it self, I just program the
> control part.
>
> In recent years there was some amazing work on physics engines, mostly
> driven by game developers, and I'm hoping to find a way way of
> connecting
> control systems design with 2D and 3D environments with relatively
> realistic
> physics, and all that in the browser, so people could see it and feel
> what
> happens instantly.
>
> There are great tools to do that naively on a desktop, such as the free
> Microsoft Robotics Simulator or V-Rep studio:
> http://www.coppeliarobotics.com/ Wonderful thing, very fun, but also too
> many options and models and scripts and ways to control the robot. I
> made an
> inverse pendulum in V-Rep, it kinda worked, but never as stable as I
> wanted
> it.
>
>
> Adam

--
Richard S. Marken PhD
www.mindreadings.com

It is difficult to get a man to understand something, when his salary
depends upon his not understanding it. -- Upton Sinclair

--
Richard S. Marken PhD
www.mindreadings.com

It is difficult to get a man to understand something, when his salary
depends upon his not understanding it. -- Upton Sinclair

FredNickols · April 1, 2014, 5:03pm

[From Fred Nickols (2014.04.01.1302 EDT)]

I sure hope this thread isn’t a case of April Fool’s day in action.

In any event, Yahoo groups could be a host. I don’t think you can import the old messages but you could upload the archives as files accessible by the members.

Fred Nickols

···

From: csgnet-request@lists.illinois.edu [mailto:csgnet-request@lists.illinois.edu] On Behalf Of Rupert Young
Sent: Tuesday, April 01, 2014 8:58 PM
To: csgnet@lists.illinois.edu; Richard Marken
Cc: Richard Marken
Subject: Re: Wither CSGNet (was Re: Intro to Javascript)

This is very timely as I was considering posting a message suggesting moving to a modern forum format.

I find the current format quite awkward to read with all the unformatted quoting. A forum format is much easier to read especially when it comes to quoting. And we wouldn’t need to worry about that pesky timestamp! Have a look at a forum I maintain on www. Atheismuk.com for an example.

There are lots of fori around and we could add one to an existing website. The Web host may actually provide one as an addon.

I wonder if we could find one that allows users to still post by email and will import all the messages from csgnet?

Rupert

On 31 March 2014 11:12:14 GMT-06:00, Richard Marken rsmarken@gmail.com wrote:

[From Rick Marken (2014.03.31.1010)]

Adam Matic (2014.3.31.1620 cet)–

Hi Rick,
I see the testing post, but not the post from March 28. I think I have the
same problem with a post from March 29.


Hi Adam (and everyone else on CSGNet).

It turns out that the listserve software at the U of Illinois that
support CSGet was changed on March 28. This was supposed to have no
impact and to a large extent it didn't. But It apparently did change
some settings on individual accounts. For example, I had set up my
connection so that I got sent an acknowledgment from the list server
that it had distributed my posts when I posted them. Apparently this
setting was changed to "no acknowledgment", apparently the default
setting, when the software was changed.

Also, the address to which posts are to be made was changed from

csgnet@listserv.illinois.edu

to

csgnet@lists.illinois.edu

If you post to csgnet@listserv.illinois.edu it still works because it
is automatically forwarded to csgnet@lists.illinois.edu. But to be
completely up to date we should be posting to
csgnet@lists.illinois.edu, as I'm doing with this post.

But I'm wondering if we shouldn't think of moving this whole operation
to a more easily accessible forum, like a social media site. I don't
use social media much but maybe we should think about bringing CSGNet
into the 21st century. What little experience I have had with
discussions on social media has been rather disappointing-- I prefer
writing in email format -- but maybe some of you young people out
there, who are the future of PCT, have some helpful ideas about this.

Best

Rick

Adam

On Mon, Mar 31, 2014 at 3:25 AM, Richard Marken rsmarken@gmail.com wrote:


 [From Rick Marken (2014.03.30.0625)}

 Just testing to see if the list is still working. They made some
 changes to it, apparently, onMarch 28 and I posted something earlier
 and go no acknowledgement.

 On Fri, Mar 28, 2014 at 6:24 AM, Adam Matic <adam.matic@gmail.com> wrote:

[From Adam Matic (2014.03.28 1430 cet)]


 BA: That's an excellent start.  Is the simulation supposed to run in
 real
 time?  I only see movement when changing a parameter value or dragging
 the
 "playerindex" slider.

AM:
No real time simulation yet, only fast time. Changing any of the
parameters
on the GUI recalculates the whole simulation, and the ‘playerIndex’ just
goes over it.

I considered adding a real-time ‘live’ mode, that would be fun. To be
able
to set reference and disturbance values in real time, and have them
graphed,
like in the Live Block Demo


 BA: The Javascript syntax looks like an object-oriented version of the
 C
 language, (as does Processing).  I didn't have any trouble following
 what
 the code in basics.js does, but I've had plenty of exposure to
 object-oriented programming.

AM:
Yeah, looks like OOP. Watch out for pointers to functions, those are
weird
at first, but very useful.


 BA: Have you seen Vensim?  It allows you to create a diagram with
 components representing flows (of variables and signals), quantities,
 etc.
 And the run it as a simulation.  A definite plus is that it includes
 sophisticated procedures for doing integration, etc. accurately.
 Apparently
 there is now a free version available online "for evaluation and
 educational
 use," see [https://www.shodor.org/tutorials/vensim/pre.php](https://www.shodor.org/tutorials/vensim/pre.php) .

AM:
I’ve downloaded several of the analog circuit simulators of the SPICE
variety, and there are online tools like the Circuit lab. There is also
“DESIRE” by Granino A. Korn, author of Electronic Analog Computers book
that
Bill mentioned. There is the modern, free and open source SciLab
recommend
by control engineers.

I can’t say I’m a fan. They are certainly useful to some people, but for
me,
there’s too many options and elements to choose from, they are big and
cluttered. The biggest pain is that they lack connection to some kind of
a
‘world’. If you want to simulate something with realistic physics, you
really need to understand the physics of it. That is one reason I like
robotics - the physics part takes care of it self, I just program the
control part.

In recent years there was some amazing work on physics engines, mostly
driven by game developers, and I’m hoping to find a way way of
connecting
control systems design with 2D and 3D environments with relatively
realistic
physics, and all that in the browser, so people could see it and feel
what
happens instantly.

There are great tools to do that naively on a desktop, such as the free
Microsoft Robotics Simulator or V-Rep studio:
http://www.coppeliarobotics.com/ Wonderful thing, very fun, but also too
many options and models and scripts and ways to control the robot. I
made an
inverse pendulum in V-Rep, it kinda worked, but never as stable as I
wanted
it.

Adam


 --
 Richard S. Marken PhD
 [www.mindreadings.com](http://www.mindreadings.com)

 It is difficult to get a man to understand something, when his salary
 depends upon his not understanding it.  -- Upton Sinclair

–
Sent from my Android phone with K-9 Mail. Please excuse my brevity.

MartinT · April 1, 2014, 6:38pm

[Martin Taylor 2014.04.01.10.52]

This is an unfortunate date to be suggesting new ideas, isn't it?

But this is a good old idea, and one I totally support. In fact,
such a forum exists at (Exploration of Complex
Adapting Control Systems), though it is rather moribund, the last
message on it having been posted at the end of July 2011. It was
started in February 2004. ECACS was started for two main reasons,
firstly because of the very limited half-life of discussion threads
on CSGnet, which meant that the wheel had to be reinvented many
times, and secondly because discussion of the complexity of
real-world perceptual control seemed to be crowded out on CSGnet in
favour of one-loop and two-level control systems acting in a
homogeneous environment.
Here is the current state of the ECACS site.
On the ECACS home page there are links to “Links” (to other
PCT-relevant sites), “Archival Documents”, “References”, and
“Forum”. All of these are now out of date, but could be updated as
users see fit. In the Forum, there are five main areas, built around
the metaphor of a virtual “Explorer’s Club”. The areas are:
“Who’s who and what’s what” for self-description by those embers
who want to make their interests and skills known
“Chart Room” for stable hypotheses and experimental results
“Pub and Grill” for “what’s cooking now” – ideas, speculations,
debatable issues and hypotheses, etc.
“Behind the Club” – stuff related to club management, policies,
polls and so forth
“Ye Ol’ Haunts” Intended for things that should be remembered but
are not of current interest. Actually used only for discussions of
the organization of the site.
There isn’t much in the “Chart Room” – Four posts on “Theory” and
eight on “references and Bibliographies”. There is, however, a
predefined topic area for “Software Repository” for links to and
descriptions of software relevant to complex PCT systems. Other
predefined topic areas with no content are Experiment and
Application, which also appear in the Pub and Grill.
The Pub and Grill had most of the action during the active life of
the Forum. It has seven major topic areas, entitled “Theory”, “Blue
Yonder”, “Roughneck Rooms”, “Application”, “Walkabouts”, “Mapping
and Charting”, and “Experiment”. On them there are, in total, 1270
posts, some of them quite extended dissertations. Most (765) are on
the Theory topic, which currently has 18 subtopics, the most active
of which are “What is PCT for?” (217 posts under 5 sub-subtopics)
and “Mathematical dynamics” (171 posts under 4 sub-subtopics). Other
realtively popular “Theory” subtopics were “Sociological Issues” (80
posts in 5 sub-subtopics) and “Emotions and affect” (88 posts in 2
subtopics).
The “Blue Yonder” topic has 11 subtopics, of which the most popular
were “Beliefs and attitudes” (45 posts), “PCT and the Persistence of
Belief” (38 posts), and “Development of monotheism” (21 posts). Blue
Yonder posts often have no obvious base in PCT, but deal with
observations that eventually should be subject either to denial of
the observation or explanation in a PCT analysis.
“Roughneck Rooms” are the place for personal comment, particularly
of the insulting kind. Posts in the other areas that seemed to
contain flames or flame-bait were arbitrarily moved there by the
moderators.Today scanning a few of the other old threads, I see
several that perhaps should have been moved here
“Application” has three subtopics “Economics” with 8 sub-subtopics,
several of which have over 30 posts, “RTP” and “Organized Conflict”.
“Walkabouts” is for topics that have no obvious other home
“Mapping and Charting” is supposed to be an area for developing
“Maps and Charts” before putting the completed versions into the
Chart Room – sort of crowd-sourcing things that can later be used
for further development of PCT in complex environments. At the
moment, it is empty.
“Experiment” likewise has little content, other than a misplaced
thread that is partly about the use of real data in economics.
All these threads are still open for continued comment, unlike
e-mail threads, which tend to vanish into the mists of time within
weeks after the last message on them.
I guess that the forum software should be updated, but it still
seems to work. I posted a test message to it today, and received an
email message telling me (and I hope other ECACS members who had set
their profile to do that and who had not changed their addresses
since 2011) that the message had been posted.
The ECACS forum can be configured to allow users to choose to be
able to post by e-mail, but I haven’t checked whether it actually is
set up that way at the moment. And I don’t know whether the existing
ECACS forum could mass import all the messages from CSGnet, but it
would be a useful thing to do, especially if it could go back into
the archives. However, even if the Forum coud automatically import all the CSGnet
messages, I think it would require quite a bit of human curatorial
management in choosing the best place for the hundreds of different
threads that have accumulated over the years, many of which
recapitulate old threads under new names, and many of which split
into different threads under the same subject line. An especially
time-consuming bit of curation would be required to split off the
personal spats into the “Roughneck Rooms”. ECACS did try, not always
successfully, to ensure that the threads elsewhere in the forum were
conducted with a reasonable degree of courtesy (not a word much used
nowadays in respect of internet discussions). Furthermore, Dag
Forssell has an archive of CSGnet going back to who-knows-when, so
simply to import the messages would have no added value unless the
actual discussion topics were split and combined from the different
subject-line threads.
I just discovered that the discusware site seems to have gone out of
business, so the ECACS site software can’t be updated. Whatever I
tried today just worked, so it is functional. If anyone now wants to
post to the existing ECACS forum, they can do so, but first they
have to be registered as a member. That’s quite an easy thing to do,
but it requires a moderator’s approval, which is usually automatic,
at least for names known from CSGnet.
When I googled for “discusware” there were several hits describing
successful migration to other supporting software, including a not
too expensive commercial service to migrate it to phpBB. If there is
interest, I might try that.
Martin

···

On 2014/04/1 8:57 PM, Rupert Young
wrote:

  This is very timely as I was considering posting a
message suggesting moving to a modern forum format.

http://www.ecacs.net

  I find the current format quite awkward to read with all the
unformatted quoting. A forum format is much easier to read
especially when it comes to quoting. And we wouldn’t need to worry
about that pesky timestamp! Have a look at a forum I maintain on
www. Atheismuk.com
for an example.
  There are lots of fori around and we could add one to an existing
website. The Web host may actually provide one as an addon.
  I wonder if we could find one that allows users to still post by
email and will import all the messages from csgnet?

Matti_Kolu · April 1, 2014, 9:35pm

[From Matti Kolu (2014.04.01.2300 CET)]

Regarding importing: I did some work on the CSGnet archive files last
year. I recreated the email headers for some of the earliest messages
from 1990 and onwards, and then did some conversion work on Kennaway's
logs and Forsell's Eudora files. I parsed the results with a custom
version of mhonarc, so as to create static html (and, for some of the
header stuff, XML files) of the now-threaded messages. Then I ran the
static files through a very ugly HTML/XML parser to import them to a
database. The last step was only done for the messages for one year,
to see if it worked, which it did. Stupidly enough I seem to have left
parts of my tool chain and the rendered results on a server that is
now wiped out, but here is an example of what the (unstyled and
unformatted) results look like after they've been parsed by mhonarc.

** These files are copied from _temporary_ "working directories" that
were created as tests during the above work.**

(Scroll to the bottom and read 'upwards')

http://pct.loopgain.com/1990/threads.html
http://pct.loopgain.com/1990/date.html

http://pct.loopgain.com/1992/threads.html
http://pct.loopgain.com/1992/date.html

from a technical standpoint it is perfectly possible to feed all
messages from 1990-09 and onwards to another system. Please don't link
to the URLs above from another website. They are there as examples of
what can be done. I have added a robots.txt file to avoid any search
engines from indexing the above files.

Matti

rsmarken · April 1, 2014, 10:03pm

[From Rick Marken (2014.04.01.1500)]

···

Martin Taylor (2014.04.01.10.52)–
  On 2014/04/1 8:57 PM, Rupert Young
wrote:

  This is very timely as I was considering posting a
message suggesting moving to a modern forum format.

This is an unfortunate date to be suggesting new ideas, isn’t it?

RM: It’s no joke;-)

RM: Thanks for reminding me about ECACS. I see David just posted another possible discussion forum software. I just spoke to David and I think moving CSGNet to more “up to date” venue should be something we all work towards.