Something kinda fun that took quite a bit of work, but which seems to be working without any hitches:

I can take a package and drop it into another one without any modifications and it can work as a dependency -- no overwriting of the global Symbol table or any of that. And then those packages can also have dependencies. Or if you're looking to save memory if multiple dependencies depend on the same package, it's possible to include the package only at the top level and then the dependencies will load the code from there (but keep their symbols separate).

I've had something like this for a while in my BTools setup, but that required me to set flags and mess with stuff and each package had it's own hard-coded loader file so if I ever wanted to change up the loader at all (say to add an extra feature) I had to update it for every package.

Now, since the loading is centralized, none of that comes into play

Fascinating the degree to which plotting, graphics, and list-manipulation are represented amongst the top tags of the long-time users. Really gives you a perspective for what we're most useful for here.

@1010011010 Seems like a question that would be very well suited for the main site, and less for the chat.

Hi I have a small question. I have a plot and a label above a curve that goes to infinity. When I plot it, the label gets cut off halfway through. however, there is still space above (because of the axis label) and when i click and drag the picture to make it bigger the label appears (but this does not allow an export)

@KarlA Try PlotRangeClipping -> False

Not exactly Mathematica-related, but a big release of the igraph C library happened today: igraph.org

IGraph/M, the Mathematica interface, was built from the dev version, and already incorporates many of the fixes and new features in 0.8. However, recently there was a big push to get igraph into shape. IGraph/M 0.4 will be coming soon with all the additional improvements! At least as soon as Mathematica 12.1 is released so I can make sure it's fully compatible :)

@LukasLang Thanks a lot!!

cannot find words for the relief this gives me

@Szabolcs PR5 seems like it's getting pretty close, happy to do any prelim testing for you, I've got both Mac & Win.

Hello! I saw this post mathematica.stackexchange.com/questions/140199/… on stackexchange. This code is really useful to me and I want to create different operators. I searched the documentation for the "___" , but I still cannot understand the code. Could someone help me? Or how can I comment on the answer so that the specific user replies?

@Marsupilami ___ means match 0 or more arguments, __ means match one or more, _ means match just one

I thought this had an answer somewhere, but I guess the first question is would anyone feel it to be over-contrived to have a manipulate of a vector on a sphere wherein one controls directly the position of the vector on the sphere rather than moving sliders? Think something like a mouselocator tracking the position of the projection of the mouse on the surface of the sphere, perhaps showing a projection, and the click action locks in the user's choice?

@CATrevillian seems like it'd be hard to get it to cleanly...

Some strange combination of this mathematica.stackexchange.com/questions/105103/… and the way Jens does the animate here mathematica.stackexchange.com/questions/51776/…

Yeah that's what I was thinking @b3m2a1 probably would end up not being compatible enough with some slower systems

Worth a quick inquiry anyways ;)

@CATrevillian Also just moving a point on a 3D object on the 2D surface that is the computer screen isn't really well defined?

Unless you'd visually have it so that the sphere is rotating but the vector looks fixed

Ah true, hmmm, I figured if you locked in the aspect ratio of the sphere, then only letting the vector be in one quadrant, for example, it wouldn't be much different than the Locator or LocatorPane

@CATrevillian If you only have one quadrant it might work? As long as the locator is always "in front" of the sphere it'd be less confusing to work with

Yeah definitely, I agree there. I would just find it a lot more...fun? if the user would be able to directly affect the direction of a vector of a field, for example, applied to some object, like a sphere. With a static length of the vector, probably just having it be a unit vector would be least confusing

Hmmm, I seemingly need to create a new class of locator for this, maybe the idea is too large, but I do have enough other things to do that this will be fun enough to distract me, choices choices choices.... :D

I'll check back in here with my failures this evening, whether they're failures at failing, or otherwise, we shall see :}

Here's a concrete example of how one can use most of the good OOP features that Mathematica supports to define useful objects in an immutable (i.e. not memory leaky) way: mathematica.stackexchange.com/questions/213618/…

Someone basically wanted to define a nice Algebra object so I showed them how to write all the boiler plate to make that work how they wanted it

@b3m2a1 wow! That's awesome. I'll have to take a few looks at that to wrap my head around it all, but it bodes well for pushing Mathematica further onto my pure math folks.

Check this out, though, fellas, I failed at failing, but I know it can work better:

Manipulate[
 EventHandler[
  Style[Graphics3D[(*Dynamic@*){(*Dynamic@If[s,*)(*{Opacity[.5],
     Sphere[]},*)nf = Nearest[pp -> Automatic];
     {Opacity[0.5],
      First@ParametricPlot3D[{Cos[u] Sin[v], Sin[u] Sin[v],
         Cos[v]}, {v, 0, Pi}, {u, 0, 2 Pi},
        MaxRecursion -> ControlActive[1, Automatic],
        PlotPoints -> ControlActive[Automatic, 50]]}, {Red,
      Sphere[Dynamic@pp, .05]},
     Arrow[{{0., 0., 0.}, Dynamic@pp[[1]]}]}, Boxed -> False,
    PlotRange -> 1.1,
    Epilog ->

Lots of thanks to this answer: mathematica.stackexchange.com/questions/84238/…

Oops let me clean that up a bit, I know there's a lot of useless parts left over, as evidenced by my commented out portions:

Manipulate[
 EventHandler[Style[Graphics3D[{nf = Nearest[pp -> Automatic];
     {Opacity[0.5],
      First@ParametricPlot3D[{Cos[u] Sin[v], Sin[u] Sin[v],
         Cos[v]}, {v, 0, Pi}, {u, 0, 2 Pi},
        MaxRecursion -> ControlActive[1, Automatic],
        PlotPoints -> ControlActive[Automatic, 50]]}, {Red,
      Sphere[Dynamic@pp, .05]},
     Arrow[{{0., 0., 0.}, Dynamic@pp[[1]]}]}, Boxed -> False,
    PlotRange -> 1.1,
    Epilog ->
     Inset[Style["Shift+mouseover",
       FontColor ->
        Dynamic@If[CurrentValue["ShiftKey"], Black, Gray]], {Right,