@Mr.Wizard @Kuba From @Mr.Wizard From both of your perspectives, do we indeed need another moderator? I'm honestly not sure if the lack of candidates is an indicator of a good or a bad health of our community. For a normal user like myself, it seems that are current moderator/high-rep users handle the site quite well. We don't have any chat abusers whatsoever, especially compared to other sites.
For the last weeks, it was mostly Kuba who was around all alone and still, at least to me, it seemed that he could handle things quite well. Beside this, we have a lot of other "community jobs" that our high rep-users do for the community for free. I'm thinking of the Paclet server, web-site creator, BTools and the service connections that @b3m2a1 crafted. Szabolcs maintains IGraphM and MaTeX. I'm responsible for the SE-Uploader, the SE-Toolbar and the IntelliJ Plugin.
These are only examples, but I'm not sure how many of these project would be there, openly accessible to everyone if it wasn't for the community here.
I believe we are indeed a good team and it doesn't matter that I don't have a diamond in front of my name. When I need super-powers, you guys are always around and never felt left alone. So for me, it is not that I don't want to step up as a candidate. I just have the impression that it is not needed which in turn says a lot about the work our current moderators do :)
Finally, I cannot tell often enough how impressed I am when I look how much editing and review work @m_goldberg is doing that mostly goes unnoticed. I value this so much, that I even consider spending 2h debugging Javascript.
"For the last weeks, it was mostly Kuba who was around all alone and still, at least to me, it seemed that he could handle things quite well." That may well be the case but it's not a fair workload, IMHO. I'm much less involved these days, J.M. is ill, and R.M hasn't been a regular contributor in a long time, though he does help with CM duties from time to time. I think there should be at least two fully-active CM's at any given time, to discuss issues etc.
@eyorble I respect that position. For what it's worth my comment wasn't intended to bully anyone into stepping forward, or an implicit request for an explanation. Simply it is my observation that compared to when the site was formed (separate from Stack Overflow) the wind seems to have gone out of our collective sails.
Intro
PaneSelector looks like a pretty idiomatic way to toggle displayed content e.g.:
PaneSelector[
{True -> progressBar, False -> button}
, Dynamic @ processing
]
Analogous If version:
Dynamic[ If[processing, progressBar, button] ]
The difference is that PaneSelector content will be co...
@HenrikSchumacher Hm, what do you mean with the comment that you can download the package only once? The comment is under your answer: mathematica.stackexchange.com/a/173933/42046
@halirutan. I have become very much dependent on your buttons, so thanks for making the new version available so quickly.
@Kuba. I am thinking about becoming a candidate for moderator. However, I don't think I can put more than three hours a day into the job, and some days not even that much. Is that enough time to be a useful assistant to you?
What algorithms exist to draw planar graphs without edge crossings in a way that they are easy to interpret by humans?
There are multiple algorithms that can handle any planar graph, such as Schnyder's algorithm or Chrobak-Payne. These typically produce a result that is not very useful.
Here's...
Note that we can't just blow this up by converting to polar coordinates and applying some non-linear transformation to the radius. That often results in intersecting edges.
@Szabolcs I had given your question some thought but could not really come up with something easy. The vertices needed to be classified into groups that share certain properties. If one had this, one could probably come up with an algorithm but the problem is so damn hard. Especially, when you don't work with graphs everyday.
I made some progress since then, and the Tutte layout seems very promising with some additional modifications.
First, we can choose the outer face. That can make it much nicer. When the graph has symmetries, it can be done based on that.
Mathematica's implementation doesn't allow choosing the outer face, but I now implemented that.
Second, the weighting helps a lot. I have not seen this done anywhere else though and I don't know why. It would be nice to prove that the layout is convex and crossing free even with the weighting ...
Finally, this is guaranteed to work only for 3-vertex connected graphs. Those are graphs which cannot be disconnected by removing any two vertices.
Since then I learned a bit more about such planar graph, e.g. that the embedding on a sphere is unique if it is 3-vertex connected, and that the faces are proper polygons when it is 2-vertex-connected.
To handle graphs that are not such, we can try to add some more edges to make ti first biconnected, then triconnected. That is not a trivial problem, but it turns out to be necessary for other planar drawing algorithms too and there are papers for it.
@m_goldberg I am putting far less than that into the site in general at this time, so it certainly would be a significant improvement. Further I don't think that I have ever spent as much or more than that on a regular basis on moderator duties, and I used to pull my weight in that department quite well I think. It's just not that time-demanding. Occasionally I'd spend most of a day retagging questions and deleting old closed questions, etc., but that was my choice and not a necessity.
i got a very strange behaviour at a ParametricPlot3D
Could someone check if Manipulate[ ParametricPlot3D[{2 x, 2 y, Norm[{x, y}]^2 - 1}/(1 + Norm[{x, y}]^2), {x, -0.5, 0.5}, {y, Sqrt[3]/2, t}, PlotRange -> Full], {t, 5, 100}] behaves the same way on your machines
@DominicMichaelis Don't think that Mathematica screwed it up. The online manual talks about the need to add additional sampling points and recursion for some functions. Also, you might want to add in Mesh -> {15, 150} or something similar to show the contour lines as $t$ changes.
@m_goldberg, I second what @Mr.Wizard said, it is not as much time consuming, and based on the impact you have on the site I think it is more than enough :)
@Kuba, @Mr.Wizard. I would probably need to put in more time than you two do because at my advanced age I have slowed down quite a lot. Everything is slower: reading, typing, and most troubling, thinking. I might be down close to 50% of the speed I had in those activities 30 or 40 years ago.
@m_goldberg i think you are great, i have always found your answers very helpful and clear. you also edited my very first post here when i did not know the basic rules .. i am grateful you are here.
Intro
PaneSelector looks like a pretty idiomatic way to toggle displayed content e.g.:
PaneSelector[
{True -> progressBar, False -> button}
, Dynamic @ processing
]
Analogous If version:
Dynamic[ If[processing, progressBar, button] ]
The difference is that PaneSelector content will be co...
What algorithms exist to draw planar graphs without edge crossings in a way that they are easy to interpret by humans?
There are multiple algorithms that can handle any planar graph, such as Schnyder's algorithm or Chrobak-Payne. These typically produce a result that is not very useful.
Here's...
