You can subscribe to this list here.
2003 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
(37) |
Jul
(109) |
Aug
(117) |
Sep
|
Oct
|
Nov
|
Dec
|
---|---|---|---|---|---|---|---|---|---|---|---|---|
2004 |
Jan
|
Feb
|
Mar
|
Apr
|
May
(1) |
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
From: <ben...@id...> - 2004-05-22 12:30:55
|
Dear Open Source developer I am doing a research project on "Fun and Software Development" in which I kindly invite you to participate. You will find the online survey under http://fasd.ethz.ch/qsf/. The questionnaire consists of 53 questions and you will need about 15 minutes to complete it. With the FASD project (Fun and Software Development) we want to define the motivational significance of fun when software developers decide to engage in Open Source projects. What is special about our research project is that a similar survey is planned with software developers in commercial firms. This procedure allows the immediate comparison between the involved individuals and the conditions of production of these two development models. Thus we hope to obtain substantial new insights to the phenomenon of Open Source Development. With many thanks for your participation, Benno Luthiger PS: The results of the survey will be published under http://www.isu.unizh.ch/fuehrung/blprojects/FASD/. We have set up the mailing list fa...@we... for this study. Please see http://fasd.ethz.ch/qsf/mailinglist_en.html for registration to this mailing list. _______________________________________________________________________ Benno Luthiger Swiss Federal Institute of Technology Zurich 8092 Zurich Mail: benno.luthiger(at)id.ethz.ch _______________________________________________________________________ |
From: John A. T. <joh...@po...> - 2003-08-23 03:42:31
|
http://pyx.sourceforge.net/ if only it did pdf in addition to crappy PS... (though maybe PS->pdf conversion would work...) -J |
From: Robert F. <fe...@di...> - 2003-08-14 23:54:59
|
I added cmdline.py to the Tools directory. It's opensource from somewhere, can't recall exactly. I like it for input to scripts. I only put it in Tools so I wouldn't have to track it down each time I need it. -r |
From: John A. T. <tu...@la...> - 2003-08-13 16:51:58
|
lots of stuff: o split "ortho" and LSLR results into separate data files, o renamed data files, o made two separate Gnuplot scripts, one for "ortho" and one for LSLR, o corresponding mods to Makefile |
From: Austin M. <ami...@la...> - 2003-08-13 16:11:47
|
all the problem 6's. only LSLR two_dx includes all 128s. run on : ORTHO, LSLR[all methods] regular two_dx[not fuzzed] fuzzed, two_mat, two_dx, swapped fuzzed, swapped -> this should be compared to fuzzed NOT swapped LSLR kershaw, problem 1 completed. -> here only problem 1 is different, up to 128 now. austin |
From: John A. T. <tu...@la...> - 2003-08-13 03:38:09
|
Austin Minnich wrote: > no-don't know what happened. > this should have everything. yep, that one's got it - the problem 1 file has a boat-load of extra garbage in it, but it doesn't matter - already had those results anyway... -J |
From: Austin M. <ami...@la...> - 2003-08-13 02:52:49
|
sorry, with all these checked out copies I am getting really confused. one small change indicated below. > just so we're all on the same page: > > ORTHO[all methods]: > > DONE: > problems 1-6 > fuzz 0.0, fuzz 0.1, fuzz 0.4 > > problems 1-5 > kershaw > > IN PROGRESS: > problem 6: >>>>> two_dx, two_mat_mesh, fuzz=0.4 > REALLY: two_dx [before we did two_mat_mesh and fuzz on two_dx mesh] > problem 6: > two_dx, two_mat_mesh, fuzz=0.4, SWAP D's, dx's > > problem 6: > fuzzed, two_mat_mesh, fuzz=0.4 SWAP D's, dx's > > > LSLR[all methods]: > > DONE: > problems 1-6 > fuzz 0.0, 0.1, 0.4 > > IN PROGRESS: > problems 1-5: > Kershaw[partial results sent out, and may not converge for 128 or finish in time] > > problem 6: >>>>>>>>>>>>>> two_dx, two_mat_mesh, fuzz=0.4 > REALLY: two_dx > problem 6: > two_dx, two_mat_mesh, fuzz=0.4, SWAP D's, dx's > > problem 6: > fuzzed, two_mat_mesh, fuzz=0.4 SWAP D's, dx's > > failed: > problem 6, D ratio = 1e6 > [but ortho didn't] > reason = fail to converge in 20,000 iterations. > > maybe done by tomorrow?... > > austin > > > > ------------------------------------------------------- > This SF.Net email sponsored by: Free pre-built ASP.NET sites including > Data Reports, E-commerce, Portals, and Forums are available now. > Download today and enter to win an XBOX or Visual Studio .NET. > http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01 > _______________________________________________ > Chama-developers mailing list > Cha...@li... > https://lists.sourceforge.net/lists/listinfo/chama-developers > |
From: Austin M. <ami...@la...> - 2003-08-13 02:41:13
|
no-don't know what happened. this should have everything. > Austin Minnich wrote: > > > sorry for all the files, this one finished earlier than expected. > > This has kershaw mesh, all orthos, in correct format, problems 1-5. > > prob_5_mesh_kershaw.gplt.good still seems to be missing the results > for the finest mesh - is that expected? > > -JT > > > > > ------------------------------------------------------- > This SF.Net email sponsored by: Free pre-built ASP.NET sites including > Data Reports, E-commerce, Portals, and Forums are available now. > Download today and enter to win an XBOX or Visual Studio .NET. > http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01 > _______________________________________________ > Chama-developers mailing list > Cha...@li... > https://lists.sourceforge.net/lists/listinfo/chama-developers > |
From: John A. T. <tu...@la...> - 2003-08-13 02:31:30
|
so I happened to be talking with Doug tonight as he was leaving, and I mentioned the fact that we were finding that the method for computing the face conductivities in T was pretty bad, and he responded that it was surprising because it was just a harmonic average, right? well, no - so I showed him where we were looking in the code to find that it was using what we're calling the "gamma" scheme: k1*k2 (V1 + V2) / V2 V1 \ -1 --------------- = (V1 + V2) | -- + -- | k1*V1 + k2*V2 \ D1 D2 / and he said "well, looks like a bug - and it's probably mine" interesting! that's been in there for a while! he pointed out that we have some two-material conduction stuff in the test suite that they thought was working, so I'll look into that... -JT |
From: John A. T. <tu...@la...> - 2003-08-13 00:29:51
|
Austin Minnich wrote: > sorry for all the files, this one finished earlier than expected. > This has kershaw mesh, all orthos, in correct format, problems 1-5. prob_5_mesh_kershaw.gplt.good still seems to be missing the results for the finest mesh - is that expected? -JT |
From: Austin M. <ami...@la...> - 2003-08-12 23:24:02
|
just so we're all on the same page: ORTHO[all methods]: DONE: problems 1-6 fuzz 0.0, fuzz 0.1, fuzz 0.4 problems 1-5 kershaw IN PROGRESS: problem 6: two_dx, two_mat_mesh, fuzz=0.4 problem 6: two_dx, two_mat_mesh, fuzz=0.4, SWAP D's, dx's problem 6: fuzzed, two_mat_mesh, fuzz=0.4 SWAP D's, dx's LSLR[all methods]: DONE: problems 1-6 fuzz 0.0, 0.1, 0.4 IN PROGRESS: problems 1-5: Kershaw[partial results sent out] problem 6: two_dx, two_mat_mesh, fuzz=0.4 problem 6: two_dx, two_mat_mesh, fuzz=0.4, SWAP D's, dx's problem 6: fuzzed, two_mat_mesh, fuzz=0.4 SWAP D's, dx's failed: problem 6, D ratio = 1e6 [but ortho didn't] reason = fail to converge in 20,000 iterations. maybe done by tomorrow?... austin |
From: Austin M. <ami...@la...> - 2003-08-12 22:17:45
|
probably worthy to note that all LSLR was done with weight = 1 / d^2 austin |
From: John A. T. <tu...@la...> - 2003-08-12 20:17:04
|
attached is one of the most interesting plots so far, and it may warrant some additional runs o though LSLR beta and delta remain identical, now there are diffs (small) between them and gamma - alpha remains much worse o most interestingly, it appears that the 3 good LSLR's may be degrading on the more-refined meshes - so what I'm thinking is that it might be worthwhile to run at least one of the 3 "good" LSLR's (but preferrably all 3) on: 1. one more finer mesh just to see whether it does fall apart, and 2. run all the meshes with a higher D1/D2 - these were D1=1 and D2 = 0.001 - maybe another with D1=1000 and D2=0.001 Austin, can you fire those off? -J |
From: Austin M. <ami...@la...> - 2003-08-12 20:15:54
|
sorry for all the files, this one finished earlier than expected. This has kershaw mesh, all orthos, in correct format, problems 1-5. LSLR kershaw is still going... about to start runs on the two_dx mesh for LSLR and ortho. austin |
From: Michael L. H. <Ha...@la...> - 2003-08-12 19:14:02
|
I just committed the write-up to the repo. -Mike Added the write-up and associated file structure from Caesar. Also, several changes from earlier versions. ---------------------------------------------------------------------- Added Files: Tag: BRANCH_DIFFUSION Examples/LoboHeat2D/diffusion/results/paper/.cvsignore Examples/LoboHeat2D/diffusion/results/paper/README Examples/LoboHeat2D/diffusion/results/paper/images/harmonic_avg.fig Examples/LoboHeat2D/diffusion/results/paper/images/harmonic_avg.ps Examples/LoboHeat2D/diffusion/results/paper/images/Internet/backgrounds/hallstone.jpg Examples/LoboHeat2D/diffusion/results/paper/source/GNUmakefile Examples/LoboHeat2D/diffusion/results/paper/source/diffusion.tex Examples/LoboHeat2D/diffusion/results/paper/source/make/commands.gm Examples/LoboHeat2D/diffusion/results/paper/source/make/environment.gm Examples/LoboHeat2D/diffusion/results/paper/source/make/latex.gm Examples/LoboHeat2D/diffusion/results/paper/styles/MLHcommands.tex Examples/LoboHeat2D/diffusion/results/paper/styles/MLHfloatrule2e.sty Examples/LoboHeat2D/diffusion/results/paper/styles/MLHlatex2html.tex Examples/LoboHeat2D/diffusion/results/paper/styles/MLHmargins.tex ---------------------------------------------------------------------- |
From: John A. T. <tu...@la...> - 2003-08-12 16:51:16
|
I've got all the result datafiles set up, and the plots are working, but they're getting a bit cluttered now, so I'm working on better presentation/labelling at the moment -J |
From: Michael L. H. <Ha...@la...> - 2003-08-12 16:27:02
|
Austin Minnich writes: > in particular, it seems to be the alpha choice of LSLR that is not > converging....something's funny with that one. Yes, it is different. How, you say? Well, for a constant D, the choices for harmonic average D simplify a bit. Choices beta, gamma and delta all reduce to the constant D value, regardless of the geometry (look at the definitions and assume D1=D2=D). So, in other words, beta, gamma and delta LSLR are all equivalent for constant D problems. In fact, we can skip running two of them on all the constant D problems (1-5). Choice alpha, however, does not simplify to D for constant D. It simplifies to: |\vec{r}_1 - \vec{r}_2| D_{12}^h = D --------------------------------------------- |\vec{\Delta r}_{1f}| + |\vec{\Delta r}_{2f}| On an orthogonal mesh with constant D, this does simplify to D. But for a general mesh, it doesn't. Choice alpha has some other good properties, mainly in conjunction with choices for DeltaR in the Skewed-Ortho methods. But it doesn't seem to be good for LSLR on constant D problems... -Mike |
From: Austin M. <ami...@la...> - 2003-08-12 16:06:51
|
all four LSLR methods on probs 1-5, shape 8-64(except where alpha failed to converge). also, 128s are not converging for any method. austin |
From: John A. T. <tu...@la...> - 2003-08-12 15:58:31
|
Robert Ferrell wrote: > I'll try to attend the meeting, but why don't you cover the Chama stuff > in your intro. I did my preliminary song and dance already, so you might > as well use the time to cover you guys' work. ok - I'll do up a slide or two on it, and run it by you... |
From: Robert F. <fe...@di...> - 2003-08-12 15:50:35
|
<quote who="John A. Turner"> > fe...@di... wrote: > >> Does look pretty good. Is there any way to see the local error in the >> taylor series reconstruction? > > probably - I'll think about that... > >> Telluride stuff showed that to be "not >> good" (tm), and it would be interesting to know if that's a bug, or >> something different is being measured. > > which is definitely something we hoped to achieve with this project - > finding out what the method should be capable of (modulo various > assumptions about how results on structured 2D extrapolate to > unstructured 3D...) The T stuff has so many unknowns in it, and errors are the most likely explanation for discrepancy. > btw, Robt., we've volunteered to do some show and tell at the T team > mtg. on Thurs. - we have 30 minutes - want to say a few words about > Chama, or shall I just include it in my intro section? I'll try to attend the meeting, but why don't you cover the Chama stuff in your intro. I did my preliminary song and dance already, so you might as well use the time to cover you guys' work. -r |
From: Austin M. <ami...@la...> - 2003-08-12 15:40:26
|
in particular, it seems to be the alpha choice of LSLR that is not converging....something's funny with that one. austin |
From: Austin M. <ami...@la...> - 2003-08-12 15:38:02
|
On Tue, 12 Aug 2003, Michael L. Hall wrote: > John A. Turner writes: > > so this is pretty good news for LSLR - although it's a constant D > > problem, it's a fairly nonortho mesh with a nonlinear solution... > > Actually, this is pretty much what I would expect, after some > thought. The LSLR method makes a local linearity assumption (first > order) which gets plugged into the divergence discretization (plus > one order, according to my hunch). In my experience, this gives a > second-order method. > > The problem, of course, is that the local linearity assumption it > makes is reasonable with constant D, but wrong with variable D. The > question is how bad the assumption will be on variable D problems. any luck on problem 7? that would be a good one for this. attached is kershaw stuff for ortho methods-all methods, probs 1-5, almost all shapes(problem 5 only has to 64, but I thought I'd get it out since it probably won't be done till end of today.) we are getting convergence problems with kershaw and LSLR-won't converge in 20000 iterations. shape is usually 128, but have seen 64(problem 2 w/ 64 iterations failed to converge). maybe use preconditioner? austin |
From: John A. T. <tu...@la...> - 2003-08-12 15:35:20
|
fe...@di... wrote: > Does look pretty good. Is there any way to see the local error in the > taylor series reconstruction? probably - I'll think about that... > Telluride stuff showed that to be "not > good" (tm), and it would be interesting to know if that's a bug, or > something different is being measured. which is definitely something we hoped to achieve with this project - finding out what the method should be capable of (modulo various assumptions about how results on structured 2D extrapolate to unstructured 3D...) btw, Robt., we've volunteered to do some show and tell at the T team mtg. on Thurs. - we have 30 minutes - want to say a few words about Chama, or shall I just include it in my intro section? -J |
From: Austin M. <ami...@la...> - 2003-08-12 15:02:36
|
> two quite different curves for the 4 LSLR variants - not sure > which lines obscure the LSLR beta and LSLR gamma results yet, > but interesting that LSLR alpha appears to be 0-th order, while > LSLR delta sure looks 2nd-order I noticed that LSLR alpha didn't look too good, we might want to take a look at the again. everything seemed okay when I checked it over, but the results are so different from the others.... austin |
From: Michael L. H. <Ha...@la...> - 2003-08-12 06:35:58
|
John A. Turner writes: > so this is pretty good news for LSLR - although it's a constant D > problem, it's a fairly nonortho mesh with a nonlinear solution... Actually, this is pretty much what I would expect, after some thought. The LSLR method makes a local linearity assumption (first order) which gets plugged into the divergence discretization (plus one order, according to my hunch). In my experience, this gives a second-order method. The problem, of course, is that the local linearity assumption it makes is reasonable with constant D, but wrong with variable D. The question is how bad the assumption will be on variable D problems. -Mike |