Quoting the original post:

Is is popular in regards to the vertical axis (stackoverflow.com) but it does poor in regards to the horizontal axis (github.com

The vertical axis doesn't include questions that appear on Wolfram Community. I wonder how much difference it would make to include them. However, ff other languages have non-Stack-Exchange communities, this would inconsequentlal.

If the authors of those projects could put the source of these demonstrations in GitHub, it would really help improve the perceived popularity of the language. And for new demonstrations, I would modify the "upload a new demonstration" webpage in order to encourage authors to send a GitHub link for the files.

Do you think this is a good idea?

I think this is a good idea, but would I think it would require more comments in the demonstrations' code. This may be quite an effort; crowdsourcing would help.

Before my little rant I need to preface things by saying Mathematica is the right choice for me.

From the perspective of some one who chose Mathematica // Wolfram Language Standard Version to be the engine driving my personal Continuing Educational Endeavors I can give an example, not at all untypical, of my Mathematica Documentation experiences that came up in my today's self study sessions, e.g. ......?Set...Shift Enter.......in WL does not at all include this syntax:

Set [ lhs , rhs ]

This particular syntax context which is conspicuously absent, is both the Functions primary defining context and at the same time is absolutely necessary for understanding the language used to explain the Set Functions concept of it's HoldFirst Attribute.

I had to learn this Syntax ( Set [ lhs , rhs ] ) from Professor Richard Gaylord's second of three You Tube videos on Wolfram Language Fundamentals.

Firstly. I am sure Mathematica is the only choice for me, warts and all, but this sort of anecdote tends to provide ammunition for those around the water cooler wanting to take scoffing pot shots at Mathematica and Wolfram Language.

John Soper

It seems common that a few explanatory variables account for a substantial part of the behavior of a complex system.

It might be amusing and interesting if Mathematica could be used to discover and informatively display the contribution of each of those variables for a language in that diagram. That might also show which of those variables has the greatest effect on the ranking of that language and even hint at where to most efficiently spend time and energy to change the ranking. Such an interactive display might even draw attention to Mathematica, in a good way.

I think that's a great idea. In addition, I'd like to see a large collection of recipes for common mathematical/programming tasks, organized and indexed on Github. For example, it would have been awesome if I didn't have to write the code below (to generate a list of values with a 'goal seek' target). It seems like a lot of the documentation and posts for Mathematica are more about how to use the language and less about these sorts of tasks.

x =.
begValue = 1;
endValue = 0.25;
listItems = 40;


f[1] = begValue;
f[n_] := f[n - 1]*x

x = Last[
   First[
    Flatten[
     Solve[f[listItems] == endValue && x > 0, x]
     ]
    ]
   ];

list =
 Drop[
  Flatten[
   Reap[
    Do[
     Sow[x^i]
     , {i, 0, listItems - 1}]]
   ]
  , 1]

x =.