@kirma I think Mathematica has been used for the design of simulation games before. The main use case wouldn't be implementing the game. Take something like this (which I chose because I know it's Unity based). Imagine that you can get any sort of data in real time, as the game is running, and analyse or plot it in Mathematica. Then you could modify the game state programmatically from Mathematica.

@kirma Also see here: en.wikipedia.org/wiki/Luc_Barthelet I believe he used Mathematica during the development of Sim City at EA. Now he's at W|A and I believe he's behind this project. So this time there must have been something prior first hand experience that lead to the conclusion that this project is worthwhile.

I agree with most of what you guys said, just pointing out that IMO the Unity integration is not useless.

@PlatoManiac Mathematica, as a language, is rather complex and messy though and has many dark corners and unpredictable interactions. Julia is a much simpler language that was designed to be JITtable from the start.

@Szabolcs @PlatoManiac have you guys ever had any experience with Clojure? Would be curious on your 2 cents if you have

@rm-rf you sound a bit fed up with Mma ;)

@MikeHoneychurch No, I don't. Learning new languages, especially interesting or intellectually challenging ones, is a lot of fun. Clojure is LISP, right? Unfortunately I almost never find any use for these languages and I can afford less and less time to learn something that I would never use.

I think I went through most interesting paradigms already, so at this point there's little that could be really exciting in a new language. I'd rather challenge myself with some math instead ;-) Or maybe I'm just getting old!

Another thing is that as time passes, libraries get more and more important and languages less important. Much of the utility of Mathematica now comes from the large number of useful functions it provides. It's no longer possible in 2014 to choose any language we find beautiful because we no longer implement all the basic routines ourselves. We use libraries.

I'm not really a Fortran fan and I do certain types of scientific programming that Fortran really isn't meant for ... e.g. graph/network manipulations.

So at this point it's either C++ or high level stuff that I use. (C lacks too much and provides no advantages over C++ for scientific programming. One can always choose to use "simple C++" only but still use the convenient stuff such as STL containers.)

Right now I use C++ and Mathematica 90% of the time, I even call C++ code from Mathematica directly through LibraryLink. I use other high level languages when necessary, R/MATLAB/Python. I'm courting Python more and more because I feel that it's unsafe to rely on Mathematica too much ... And then there's Julia on the list of things to learn one day.

One day I might lose my Mma license or Mma may decline in quality so much that it fails ... at that point it wouldn't be good if Mma were the only high level language I know really well.

@Szabolcs thanks for your 2 cents. interesting

@MikeHoneychurch If I had 3 spare months with a computer I'd probably learn something new and interesting that promises absolutely nothing useful ... and it might turn out to be useful in the future despite everything. Unfortunately, as time passes, we can afford such luxury less and less ... Missing the golden undergrad years with all that spare time for experimentation ...

Uhuuu!

@Murta Congrats! What was the award for?

Oh, I see

Congrats again

Pic of @Murta plus.google.com/…

Picture of some members of the stack team. @Kuba @Fonseca @StephenW? @RolfMertig @PatoCriollo

@PatoCriollo What a join!

@Murta is the first innovation award of the night

Cool :)

@Szabolcs I do understand there ar real use cases for prototyping, but that earlier blurb muddled the waters intentionally, in my opinion. It should be more clearly labeled as research, instead of development or even implying deployment.

This question involving non-distributive integrals seems like a bug to me, but it didn't get much attention so I don't know if I'm thinking this straight. We will benefit from a deeper insight. Can anyone check it out please?

@murta Hey Rodrigo, congrats!

@YvesKlett tks!

How was the conference?

@belisarius it was the Wolfram Innovation Award

@YvesKlett I liked it. But not big news on V11. No beta this time.

but a great time!

Looks like it. Enjoy your award!

Tks! It was cool.

Even better for an active MMA.SE member :D

Gotta run - have fun and safe travels!

@Murta Nice! Now I know where to shop when in SP

Hi all, a quick question if anyone's around :-)

Suppose I have a list of integer like so {1, 4, 3, 2}, and I want to make it to a list of pairs like this: {{1,4},{4,3},{3,2}}

What's a clean way of doing it?

Alternatively, you can see the list of integers as vertices in a path graph, and the desired output is its edge list

@Juho Partition[{1, 4, 3, 2}, 2, 1]

Thanks man, I'll give it a try

EdgeList[PathGraph[{1, 4, 3, 2}]] this works almost too :-)

but I get <-> edges instead of commas

@MichaelHale Yep, that does it nicely. Thanks!

I miss the days where the starred messages weren't links

@Rojo please say something interesting for me to star

@Rojo excellent