« first day (1828 days earlier)      last day (2662 days later) » 

12:10 AM
@R.M. So you are saying we missed our own birthday?? Sounds exactly like it could have been me.
 
 
10 hours later…
10:31 AM
I have just discovered that the presentations of the WTC2016 (Wolfram Technology Conference, Champaign) are available.
 
 
3 hours later…
1:19 PM
Gee, I missed when they fixed MatrixForm:
IdentityMatrix[3] // MatrixForm
%.{1, 2, 3}
 
1:34 PM
I'm going to put here in the chat room a block of code of 40 lines. I have done some tests in the sandbox chat room, so It should work. It's rather tiedous and slow. Hope it will not disturb anyone. This block of code will probably be use for further discussions. Thanks.
Let's go ...
`` <<NDSolveFEM` ``
sh=2/10;sh2=2/100;sw=3/10;
coordinatesList={{0,0},{1,0},{1,sh},{1,1},{0,1},{0,sh+sh2},{sw,sh+sh2},{sw,sh},‌​{0,sh}};
bmesh=ToBoundaryMesh["Coordinates"-> coordinatesList,"BoundaryElements"->{LineElement[{{1,2},{2,3},{3,4},{4,5},{5,6},‌​{6,7},{7,8},{8,9},{9,1},{3,8}}]}];
bmeshRotated=ToBoundaryMesh["Coordinates"->((RotationMatrix[45 Degree] . #)& /@ coordinatesList),"BoundaryElements"->{LineElement[{{1,2},{2,3},{3,4},{4,5},{5,6}‌​,{6,7},{7,8},{8,9},{9,1},{3,8}}]}];
Row[{bmesh["Wireframe"],bmeshRotated["Wireframe"]}]
{mesh,meshRotated}=ToElementMesh /@ {bmesh,bmeshRotated};
{gr01,gr02}=Through[{mesh,meshRotated} @ "Wireframe"]
\[Epsilon]r=If[y<=sh,{{11.7,0.},{0.,11.7}},{{1.,0},{0.,1.}}];
\[Epsilon]rModified=If[(-x+y)/Sqrt[2]<=sh,{{11.7,0.},{0.,11.7}},{{1.,0},{0.,1.}‌​}];
{ContourPlot[Norm[\[Epsilon]r],{x,-0.1,1.1},{y,-0.1,1.1},PlotLegends->Automatic‌​,Epilog->First[gr01]],ContourPlot[Norm[\[Epsilon]rModified],{x,-1,1},{y,0.,2.},Pl‌​otLegends->Automatic,Epilog->First[gr02]]} //Row
\[Rho]c =10^-8.;
\[Epsilon]0=8.86*^-12;
op=Inactive[Div][-\[Epsilon]r.Inactive[Grad][u[x,y],{x,y}],{x,y}]-\[Rho]c/\[Eps‌​ilon]0;
opModified=Inactive[Div][-\[Epsilon]rModified.Inactive[Grad][u[x,y],{x,y}],{x,y‌​}]-\[Rho]c/\[Epsilon]0;
\[CapitalGamma]D={DirichletCondition[u[x,y]==0,x==1||y==1||y==0],DirichletCondi‌​tion[u[x,y]==10^3,0<=x<=sw&&sh<=y<=sh+sh2]};
\[CapitalGamma]DModified={DirichletCondition[u[x,y]==0,y==-x+Sqrt[2]||y==x+Sqrt‌​[2]||y==x],DirichletCondition[u[x,y]==10^3,0<=(x+y)/Sqrt[2]<=sw&&sh<=(y-x)/Sqrt[2‌​]<=sh+sh2]};
ufun=NDSolveValue[{op==0,\[CapitalGamma]D},u,{x,y}\[Element]mesh];
ufunModifed=NDSolveValue[{opModified==0,\[CapitalGamma]DModified},u,{x,y}\[Elem‌​ent]meshRotated];
{Show[ContourPlot[ufun[x,y],{x,y}\[Element]mesh, ColorFunction->"Temperature"],bmesh["Wireframe"]],Show[ContourPlot[ufunModifed[x‌​,y],{x,y}\[Element]meshRotated, ColorFunction->"Temperature"],bmeshRotated["Wireframe"]]} //Row // Style[#,ImageSizeMultipliers->{1, 1}]&
I have finish to write my block of code. Sorry for the interruption.
Here is the result of the code above :
 
2:02 PM
!! Apparently, my the copy-paste procedure of my code above has failed.
I give up.
For those intrested, the question why : how can NDSolve solve this problem, which involve that :
1) the tangential component of E at the dielectric interface must be continous
2) the orthogonal component of Epsilon[x,y]*E must be continous (which imply that orthogonal component of E alone is discontinuous)
though these informations are stated nowhere in the problem description (= in the code)?
 
2:49 PM
Here are links to the WTC 2016 youtube playlist and to the videos on wolfram.com, where you can also find the slides: youtube link, wolfram.com link
2
Easy enough to find I suppose, but maybe it saves somebody from searching for wtc and getting conspiracy theories.
I guess it can't hurt to remind @Kuba about the presentation "Building Maintainable Dynamic Interfaces" (haven't watched it myself yet) (youtube link)
 
something something jet fuel can't MapThread steal beams?
 
I guess it can't hurt to remind @Kuba about the presentation "Building Maintainable Dynamic Interfaces" (haven't watched it myself yet) (youtube link)
MapThread[f, stealBeam[stealBeam[1, 2], stealBeam[3, 4]]]
(*MapThread::list:List expected at position 2 in \
MapThread[f,stealBeam[stealBeam[1,2],stealBeam[3,4]]].*)
 
@JacobAkkerboom thanks :) was there and seen it live. many good tips.
 
@Kuba Ah nice :). Ok I should watch it then
 
3:05 PM
@JacobAkkerboom unfortunately it does not address scaling and modularity in reasonable extent
k, have to go, have a nice day.
 
3:16 PM
D'oh, too sleepy this morning. My comment above is nothing new. Nothing has changed.
Hmm, still not ready for chat... :)
 
 
4 hours later…
7:30 PM
@JacobAkkerboom Of course, the one I'm most interested in checking out is notably absent: "A Writer's Workshop with Stephen Wolfram: Covering the Nitty-Gritty about the Book-Writing Process and 'Toolkit'". sigh
 

« first day (1828 days earlier)      last day (2662 days later) »