Re: Networking in DOSEMU

I want to ask something not concerning python.
Here, you give yourself an excellent hint that this is the wrong place to ask this question.
actualy i want to know how can i connect to my novel server thru dosemu.
Here, I am wondering if you mean Novell. Searching is facilitated by correct spelling of the thing you seek. Let's take a look... http://www.google.com/search?q=dosemu+mailing+list Strangely, not the first hit on google, but the 3rd hit looks pretty good: http://dosemu.sourceforge.net/mailinglist.html good luck! _________________________________________________________________ Don't just search. Find. Check out the new MSN Search! http://search.msn.com/

I've made some headway implementing a primitive hypertoon using VPython. I came up with the hypertoon concept some years back, per this 1996 web page: http://www.grunch.net/synergetics/hypertoon.html The basic idea is you have this set of key frames, say A-F, and a set of cartoons or "scenarios" which always start and finish with one of the key frames. In between you have some transformation that takes you smoothly from the starting key frame to the ending one. A dictionary of scenarios might look like this: scenarios = {} # Scenario( function, starting frame, ending frame) scenarios['sc0'] = Scenario(trans0, 'A', 'B') scenarios['sc1'] = Scenario(trans1, 'A', 'C') scenarios['sc2'] = Scenario(trans2, 'A', 'D') scenarios['sc3'] = Scenario(trans3, 'B', 'C') scenarios['sc4'] = Scenario(trans4, 'C', 'D') scenarios['sc5'] = Scenario(trans5, 'D', 'B') ... So you get this spaghetti ball, or network, of interconnected nodes (the key frames), all interconnected by transformations (the scenarios). At each node, you may randomize the selection process, as to where to go next. def play(self, n): node = keyframes[0] lastthing = None for i in xrange(n): nodelist = self.nodes[node] # list of candidate toons nextsc = nodelist[randint(0,len(nodelist)-1)] # random pick lastthing = self.scenarios[nextsc].play(lastthing) # play it node = self.scenarios[nextsc].finish # here's where you end up The result is a continuous smooth motion visualization, with a lot of repetition, and with a lot of random sequencing. If the network is quite large, then you might keep seeing new stuff even after quite some time. Per my interest in geometry, my theme for my first VPython hypertoon is polyhedra and their interrelationships per what Bucky Fuller called the concentric hierarchy (or cosmic hierarchy when he was being cosmic). I started out with the following key frames: # A: rh dodeca all alone # B: octahedron all alone # C: cube all alone # D: tetra all alone Later I added: # E: cubocta all alone # F: Icosa all alone Not sure what's with this "all alone" -- emphasizing that we end up with nothing else on stage I guess. The first thing I did was script scenarios among the initial four key frames. I wrote a transformation in each direction, e.g. if there's a toon from A to B, there's also one from B to A. By transformation I don't mean anything very fancy. I basically just trace out edges (they progressively elongate, so each scenario takes a few seconds -- I adjust the frame rate). For example: def trans10(thing): print "trans10: tetra -> cube" tet = rbf.Tetra() # D: tetra all alone tet.draw() if thing is not None: thing.delete() invtet = rbf.Invtetra() invtet.draw(trace=True) cb = rbf.Cube() cb.draw(trace=True) tet.delete() invtet.delete() # C: cube all alone return cb Here we're going from the tetrahedron key frame to the cube key frame. I do this by drawing an inverted tetrahedron (black), intersecting the first one's mid edges, then connecting the resulting 8 points by tracing out a (green) cube. Then I delete the two tetrahedral, leaving only the cube. There's one subtlety here: each scenario returns the final polyhedron as an object, and this gets passed in to the next scenario as an argument. The next scenario then superimposes an identical shape, before deleting the one that got passed in. The purpose is to create overlap, so there's no point at which the stage is entirely empty. The fanciest I've gotten with the transformations so far is between E and F (the icosahedron and the cuboctahedron). For those of you familiar with Bucky's geometry (actually a philosophy), there's this thing called the jitterbug transformation that relates the two. I found this somewhat difficult to implement in VPython and ended up with a kind of double-buffering, where two polyhedra progressively draw and delete themselves as the vertices shift position. I'm posting the source code to my web site, in case you want to eyeball it in any detail (or even run it). I'll be changing it of course, as time permits. This version requires Python 2.4, meaning I'm using the VPython 2.4 experimental. However, it'd take only minor tweaks to make it run in 2.3 (I treat 'set' as a built-in and use 'sorted' is all). hypertoons.py -- the main module wherein all the scenarios are defined and from whence the hypertoon is launched rbf.py -- a library of concentric hierarchy shapes, repurposed from POV-Ray work to support VPython vectors instead. The repurposing is incomplete and fuzzy around the edges. coords.py -- where I've rolled my own vector classes, except now I hardly need 'em cuz I'm using VPython's native class. See: http://www.4dsolutions.net/ocn/python/hypertoons/ The only reason coords.py is still in the picture is I get the whole show rolling by defining a large set of vertices as globals within rbf.py. These are the key "points of interest" within the concentric hierarchy, out of which all the polyhedra are built. For historical reasons, these key points start out being defined in this weird coordinate system defined by four rays emanating from the center of a tetrahedron to its four corners, labeled (1,0,0,0), (0,1,0,0), (0,0,1,0), (0,0,0,1) -- a coordinate system that requires no negative numbers. It so happens that our points of interest are very easily expressed as vector sums (linear combinations) of these four. I immediately convert these weirdo vectors to traditional XYZ vectors in rbf.py, but I'm still using the old coords.py to start with with "Qvectors" (Q for Quadrays)[1]. Here's what that looks like in more detail (defining A-Z): ORIGIN = vector(0,0,0) A = vector(Qvector((1,0,0,0)).xyz) # center to corner of tetrahedron B = vector(Qvector((0,1,0,0)).xyz) # " C = vector(Qvector((0,0,1,0)).xyz) # " D = vector(Qvector((0,0,0,1)).xyz) # " # tetrahedron's dual (also a tetrahedron i.e. inverted tet) E,F,G,H = B+C+D,A+C+D,A+B+D,A+B+C # tetrahedron + dual (inverted tet) = duo-tet cube # octahedron vertices from pairs of tetrahedron radials I,J,K,L,M,N = A+B, A+C, A+D, B+C, B+D, C+D # octahedron + dual (cube) = rhombic dodecahedron # cuboctahedron vertices from pairs of octahedron radials O,P,Q,R,S,T = I+J, I+K, I+L, I+M, N+J, N+K U,V,W,X,Y,Z = N+L, N+M, J+L, L+M, M+K, K+J Kirby PS: Something I've discovered about VPython: apparently copy.deepcopy doesn't work reliably to copy vector objects. [1] http://www.grunch.net/synergetics/quadintro.html

hypertoons.py rbf.py ... coords.py... See: http://www.4dsolutions.net/ocn/python/hypertoons/
There was another dependency I spaced mentioning: colors.py -- just a mapping between color words and associated RGB strings, used in various programs I've got. I've added that to my site. Kirby

Kirby, Thanks for the update. I had tried to run this earlier and noticed the missing dependency, but had not yet tracked down what was up. I've got it running now (with a couple tweaks for Python 2.3). This is a nifty application for demonstrating stereo visualization. Just set scene2.stereo = 'redblue' and scene2.stereodepth = 1.5, slip on the red-blue glasses, and enjoy the show. Awesome. It's even better with active or passive stereo, but I know must of you don't have that option. Thanks loads for the posting. --John ps. For stereo, it generally works a bit better to set the background to medium gray (.5, .5, .5) rather than black or white. Kirby Urner wrote:
hypertoons.py rbf.py ... coords.py... See: http://www.4dsolutions.net/ocn/python/hypertoons/
There was another dependency I spaced mentioning:
colors.py -- just a mapping between color words and associated RGB strings, used in various programs I've got.
I've added that to my site.
Kirby
_______________________________________________ Edu-sig mailing list Edu-sig@python.org http://mail.python.org/mailman/listinfo/edu-sig

Thank you for these pointers John. I'm hungry to play this in stereo, primitive as it is, but a source of frustration in my life is I've lost that whole stack of paper 3d glasses I handed out (and collected back) at OSCON 2004. So this project to view in stereo is on hold. I'll search for the glasses again soon (box?), and/or invest in another stack (I want to at least have them by April, for use in class). On another front, of course I've fantasized about firing off two Hypertoon classes, each within the same domain of scenarios, but each running on its own thread, traveling the network solo (with threading essentially giving equal time to each). Once you've got two threads, you can get any number (processor limits make too many impractical). Given the way these toons are designed (they all traverse within what Fuller called the concentric hierarchy in synergetics), multiple play heads would play well together, adding to the intricacy of the effect. Plus I need more scenarios to begin with (growing and shrinking balls, stuff unfolding/refolding, tetrahedron inside-outing, more complete jitterbug...). Hopefully this Hypertoon idea goes open source in a big way. I've been trying to get it going since 1996, but people only really start to get it when they have working source code, i.e. obviously what I'm talking about is highly doable. And not just in Python (but hey, what a wonderful world in which to prototype!). Kirby
-----Original Message----- From: John Zelle [mailto:john.zelle@wartburg.edu] Sent: Saturday, March 12, 2005 2:38 PM To: Kirby Urner Cc: edu-sig@python.org Subject: Re: [Edu-sig] Hypertoons!
Kirby,
Thanks for the update. I had tried to run this earlier and noticed the missing dependency, but had not yet tracked down what was up. I've got it running now (with a couple tweaks for Python 2.3).
This is a nifty application for demonstrating stereo visualization. Just set scene2.stereo = 'redblue' and scene2.stereodepth = 1.5, slip on the red-blue glasses, and enjoy the show. Awesome. It's even better with active or passive stereo, but I know must of you don't have that option.
Thanks loads for the posting.
--John
ps. For stereo, it generally works a bit better to set the background to medium gray (.5, .5, .5) rather than black or white.

Kirby Urner Sent: Sunday, March 13, 2005 8:01 PM Cc: edu-sig@python.org Subject: RE: [Edu-sig] Hypertoons!
<<SNIP>>
On another front, of course I've fantasized about firing off two Hypertoon classes, each within the same domain of scenarios, but each running on its own thread, traveling the network solo (with threading essentially giving equal time to each). Once you've got two threads, you can get any number (processor limits make too many impractical).
OK, one last set of tweaks for the day, and I'll give it a rest for awhile. Got to work on some other projects. But (drum roll), I *did* get the two hypertoons to run in their own threads, sharing the same display. It's pretty cool to watch. Had to change the goofy way I was making so many key vertices global -- better to work with copies of the key vertices, and stuff them in a class (changes to rbf.py -- also made the icosahedron cyan). http://www.4dsolutions.net/ocn/python/hypertoons/ My proposal to speak at OSCON this year was accepted. This Hypertoon concept will be a part of my talk, along with Elastic Interval Geometry and some other stuff. Kirby

As might be expected, my hypertoon has generated some interest on Synergeo, one of the eGroups frequented by Fuller Schoolers. One user found something interesting, and to me surprising. Some of the smooth motion "jitterbugging" I incorporate, where rods smoothly change position (the whole rod does, versus the progressive elongation of individual rod tips -- another kind of animation I employ) ONLY seems to work well in Python 2.4 with the experimental VPython 2.4. There's a thread on this: http://groups.yahoo.com/group/synergeo/message/21263 (is one part of it). He's responding to my: http://groups.yahoo.com/group/synergeo/message/21262 I thought it was a problem with frame rate and his Celeron 2.2, or video card. But it turns out that as soon as he changed to the experimental VPython, everything worked as I'd described. Certain kinds of motion that hadn't been apparent before, now were. Interesting. I haven't done the necessary experiments on my end to verify this is the problem. I've got everything to my satisfaction on the newest VPython on my faster computer, and tried Python 2.3 + VPython on the slower laptop. But I never tried Python 2.4 + Experimental on the laptop. Kirby

http://www.4dsolutions.net/ocn/python/hypertoons/
My proposal to speak at OSCON this year was accepted. This Hypertoon concept will be a part of my talk, along with Elastic Interval Geometry and some other stuff.
Kirby
On the Hypertoons front, I thought we were done with the drafts for Pyzine, as I'd sent several, and only got back suggestions the problem was on that end, with parsing software. I even went to the trouble to convert it to restructured text, which is an unnecessary step if we're not yet in final formatting. Now it turns out the article is unacceptable as it stands, is in need of a substantial rewrite. News to me. I'm tired of working with Pyzine and will no longer give them the option to publish my article. Kirby

[ Kirby Urner ] ------------------------------------------------------------ | Now it turns out the article is unacceptable as it stands, is in need of a | substantial rewrite. News to me. I'm tired of working with Pyzine and will | no longer give them the option to publish my article. That is sad news. Nevertheless, I'm convinced that many of us nourish hope of seeing your article surface somewhere else. By the way, even though [1] makes reference to [2] and therefore to Hypertoons, maybe mentioning Hypertoons directly in [1] would be appropriate ? [1] http://www.python.org/sigs/edu-sig/ [2] http://www.4dsolutions.net/ocn/cp4e.html best regards, Rod Senra -- Rodrigo Senra <rsenra |at| acm.org> ------------------------------------------------ GPr Sistemas http://www.gpr.com.br Blog http://rodsenra.blogspot.com IC - Unicamp http://www.ic.unicamp.br/~921234 ------------------------------------------------

-----Original Message----- From: edu-sig-bounces+urnerk=qwest.net@python.org [mailto:edu-sig- bounces+urnerk=qwest.net@python.org] On Behalf Of Rodrigo Dias Arruda Senra Sent: Sunday, July 24, 2005 2:13 PM To: edu-sig@python.org Subject: Re: [Edu-sig] Hypertoons!
[ Kirby Urner ] ------------------------------------------------------------ | Now it turns out the article is unacceptable as it stands, is in need of a | substantial rewrite. News to me. I'm tired of working with Pyzine and will | no longer give them the option to publish my article.
That is sad news. Nevertheless, I'm convinced that many of us nourish hope of seeing your article surface somewhere else.
I just slapped it up on my cp4e.html, where it could have been all along. But since an editor of Pyzine approached me at Pycon, having sat through my open space session (people could book them at will), and suggested I submit an article to his magazine, I was giving Pyzine first dibs. That commitment stretched through many weeks and months (Pycon was last March), but was never infinite. I'll just self-publish, charge nothing, and enjoy whatever circulation the reading deserves. Yes, it's possible that reworking with some Pyzine editor could have substantially improved it, but the process wasn't working, as I'd already been asked to reformat in restructured text, whereas Word PDFs were fine for advancing copy or proofs.
By the way, even though [1] makes reference to [2] and therefore to Hypertoons, maybe mentioning Hypertoons directly in [1] would be appropriate ?
[1] http://www.python.org/sigs/edu-sig/ [2] http://www.4dsolutions.net/ocn/cp4e.html
best regards, Rod Senra
Now that Hypertoons! has a link from the cp4e page, I think we've got a clean linking architecture. A goal with sigs/edu-sig (the web page) is not to promote my own projects too much. Kirby

This is a nifty application for demonstrating stereo visualization. Just set scene2.stereo = 'redblue' and scene2.stereodepth = 1.5, slip on the red-blue glasses, and enjoy the show. Awesome. It's even better with active or passive stereo, but I know must of you don't have that option.
Thanks loads for the posting.
--John
OK, found my 3d glasses. My daughter and I were just blissing out for awhile. Way cool, that feature. Kirby
participants (4)
-
John Zelle
-
Kirby Urner
-
Lee Harr
-
Rodrigo Dias Arruda Senra