« first day (2660 days earlier)      last day (715 days later) » 

12:00 AM
That might be me being picky about such details, but I know that designers also care about these things.
2 hours later…
1:53 AM
@b3m2a1 OK, enough fiddling.. One of the easiest solutions I have found is to skip the edges all together and make them zero thickness. Then, simply making an heptagon behind the whole drawing which has a mean/similar color to all colors reduces the artifacts and doesn't introduce any edge-clutter
Small like on the site
500% zoom
2:15 AM
@halirutan that’s basically what I did, although I did some minor adjustments of the edges
Hides most issues quite well
@b3m2a1 And now I rehacked the whole logo to not replace triangles, but keep the larger ones in the background.
Lets see how this works out
2:59 AM
Alright, complete Bullsh#$t. We need a new logo. I suggest the following
Has zero rendering artifacts. Good night
1 hour later…
4:17 AM
Alright the second. Still not in bed, but I think I have it. Find the svg here
2 hours later…
6:32 AM
Here is what we have to decide between based on what people have suggested:
* With or without "the"
* "and", "&", "/" and no conjunction, "Powered by Wolfram Language"
* centered, right-aligned
* smaller or larger font size
6:59 AM
@C.E. Yes, and I'm a bit lost when I think about how we want to vote for all the possible permutations. One point though: Fontsize won't be up for much discussion. Depending on what comes in the second line, there isn't much room when we want to have a nice proportion of space between the parts.
@halirutan Personally I'm not sure it's necessary to have the community vote exactly about how the text should be displayed. Make your best judgement and if people are unhappy they will ask for a change. No need to design by committee on this one :)
@CarlLange What the native speakers among us could decide is if "the" sounds better in combination with "Wolfram Language". My feeling tells me yes but I'm not sure how much difference it makes.
@halirutan My belief is that it's referred to by SW and thus WRI as "Wolfram Language", no "the".
Well, looking for written proof to back that up gives me nothing. Just in the livestreams I suppose.
@CarlLange Hmm..
@halirutan In any case, I think it's personal preference whether you prefer it with a "the" or not. Seems fine either way :)
7:27 AM
@halirutan if you want an & or Powered by I’d say you probably want it. If you go with / that reads as Mathematica “slash” Wolfram Language which somehow feels more palatable
On the other hand it’s the kind of thing that’s not worth obsessing over given how little anyone is really apt to be bothered by whatever the community goes with
7:51 AM
What I think looks best is having "the" when there is no conjunction and not having "the" when there is a conjunction.
I feel the same about right-aligned or not. I prefer right-aligned except when there is no conjunction.
"/" and "Powered by Wolfram Language" I don't think convey the right thing. "/" makes it sounds like Mathematica and Wolfram Language are synonymous and "Powered by Wolfram Language" makes it sound like Mathematica is built in Wolfram Language.
"and" makes it sound like they are two separate things, which I think is right.
@C.E. I agree. I would like to have it right aligned with "and" and without "the". I don't like the symbols "&" or "/" in this. It feels like a playground then. "Powered by" sounds awesome but it just doesn't feel right.
 StringJoin /@
  Tuples[{{"Mathematica\n"}, {"and ", "& ", "/",
     "Powered by "}, {"the ", ""}, {"Wolfram Language"}}], "\n\n"]
8:09 AM
Arnoud originally asked if "& Wolfram Language" could be added to the image, in case it's helpful
There is also precedence for using & on Stack Exchange: Unix & Linux, Programming Puzzles & Code Golf, Music: Practice & Theory, Motor Vehicle Maintenance & Repair, Software Quality Assurance & Testing, Anime & Manga, Board & Card Games, Psychology & Neuroscience, Gardening & Landscaping, Veganism & Vegetarianism Stack Exchange
@C.E. That might be a valid argument.
Still don't like it, but it is something objective.
8:38 AM
Vote for your favorite and say if I should add another variation
Q: Site Logo Text: Mathematica and XXX. Which wording exactly?

halirutanIt's safe to say that after quite a discussion, the general consensus is that we should add "Wolfram Language" as sub-text to our header image. Now, the question is what exactly should we put there. I prepared 3 versions of things I like the most, where "Mathematica and Wolfram Language" is my fa...

@halirutan to my eyes the & just looks more correct than "and" here, especially since we're not putting the Wolfram Language on the same footing. One possible variant that could be even nicer would be to put the conjunction in yet a different color. Like you could have a lightish-gray & there
@b3m2a1 yeah, yeah.. after preparing I like it too. I admit it.
That's how Unix & Linux do it, although in their logo the two are treated as equals
The benefit of graying out the conjunction is that it is clearly distinct from the Wolfram Language in a way that it's not entirely here. On the other hand it might feel wrong given the columnar layout.
And right-alignment
@b3m2a1 And we don't have a gray in the rest of the logo. The colors for the fonts are coming from the graphics and the red one is our site-red.
Gray does kinda fit in everywhere, though.
I'm still using the old logo though so I probably have no leg to stand on here in terms of making suggestions :)
8:48 AM
@b3m2a1 have you managed to put dynamic module variable's controller in an attached cell? even with InheritScope -> True it does not seem to work. It works for global variables though
@Kuba Nope. It was causing me grief at one point so I started storing state in the FE using FrontEnd`SetValue.
Only works if you got a small amount of stuff to store though
@b3m2a1 ehh
AttachedCell is annoyingly buggy in terms of standard Dynamic
It actually works quite well with the other FEPrivate and FrontEnd constructs but not the documented ones
@halirutan @b3m2a1 and others thanks for taking care about Wolfram Language header thing. If you need any help from my side let me know, I was not interfering because I don't have strong opinion on this subject and you are doing well anyway.
@b3m2a1 I'm not sure. We already have a lot of grayish things going on in the background
8:56 AM
Eh looking at it I'm actually kinda partial to it over the mono-color but I don't really mind either way.
When I finally update my browser plugin to not use the old logo I'm probably gonna hide the "Wolfram Language" part because I hate the name.
So do not work hard to accommodate me
9:27 AM
@b3m2a1 a follow up to AC problem: with global variables this complains:
RawBoxes@DynamicBox@ FEPrivate`If[x[[1]], DiskBox[{0, 0}], CircleBox[{0, 0}]]
Those primitives can be large so I don't want to transfer them each time
So I don't want to use system IF
I was wondering about posting a question: 'general graphics pane selector'
10:28 AM
@halirutan Could we also have a site preview for the other & alternative? It would be easier make it easier to compare them.
2 hours later…
12:35 PM
@C.E. Oh boy, what have I gotten my self into..
@halirutan You're doing a great job, it's almost over now. Thank you!
@C.E. At least I learned a lot more about the Chrome dev tools and local assets. And of course b3m and me have some more experience with SVG now.
1 hour later…
2:01 PM
@halirutan What about this for the logo? Dropbox link. Looks like this on different zoom-levels:
The idea is to first merge everything with the same color into one shape, and to "merge-accumulate" them with the previous one, going from the inside out. This way, there are no edges between shapes, just overlapping shapes, so you get away with no edge colors at all
@LukasLang Visually, there doesn't seem to be much difference to my latest version
Except that the outer red in my version appears to be a bit darker.
@LukasLang Can you share your implementation?
@LukasLang However, since I drew additional lines on the edges and you only merged existing shapes, your version is smaller in size and I would prefer this.
Very nice job.
1 hour later…
3:36 PM
Q: How to use SPARQLExecute for SPARQLSelect on Entities?

EdmundWhen using SPARQLSelect on Entity Types the entire entity store is downloaded to the local cache. For example, SPARQLSelect[ { RDFTriple[SPARQLVariable["city"], EntityProperty["City", "Name"], SPARQLVariable["cityName"]] } /; SPARQLEvaluation["CONTAINS"][SPARQLVariable[...

More SPARQL fun. These functions feel very experimental; which is why they are.
Still, though, it seems odd not to have an easily accessible SPARQL endpoint for entity queries considering SPARQL on entities have been showcased numerous times leading up to the release and within the release videos.
3:59 PM
I have to admit, I am having trouble getting my mind around this SPARQL stuff
Is there any effective way to connect it with the already existing Entity framework (or does that not even make sense?)
@chuy The SPARQL functions can take entities (and list of entities) as the graph store.
It seems okay for operations on a single entity graph. However, because it is downloading the values in the query for all of the entities supplied to the query you can quickly run out of memory when the Cartesian product of the entity count of the entities in the query are created
4:31 PM
What would be the advantage then? I guess I haven't seen an example where this seems useful (yet).
5:24 PM
@chuy If it didn't download everything locally then the advantage is being able to perform entity queries that span entities. Like the one in my post that queries both "City" and "Airport" entities.
5:38 PM
I'd like some help in identifying a function that works will work similarly to this, but w/o the periodicity (so there is one "peak" so to say") `Manipulate[
Plot3D[Sin[x + w]^2 Sin[y + z]^2, {x, 0, Pi}, {y, 0, Pi}], {w, -1,
1}, {z, -1, 1}]` My limited mathematical function literacy is not up to the job.
6:04 PM
@bobthechemist What about a normal distribution?
  PDF[MultinormalDistribution[{x0, y0}, 5 IdentityMatrix[2]], {x,
  {x, -10, 10}, {y, -10, 10},
  PlotRange -> Full
 {x0, -10, 10},
 {y0, -10, 10}
@C.E. Yep, that's exactly what I wanted; I was trying to get it to work with NormalDistribution and forgot about the multiple variate version. Thanks.
6:16 PM
That is a SymbolDependencyGraph function
6:49 PM
@C.E. With your help, now I can do really important things like make my LED periodic table do sunrises:
@bobthechemist Nice :)
Beats grading lab reports...
I'll bet
7:30 PM
I have to say these ILD live-streams leave me a little bit cold
It's all just adding little unnecessary flourishes to the language and not focusing on the real stuff that would make it better.
7:57 PM
...and from Mathematica array plot prototype to display on the actual periodic table...
user image
2 hours later…
11:04 PM
One thing I will never get about Mathematica is how the language is seemingly designed to stunt independence in its user base. Like it seems designed to be a toy language with functions that do lots of cool things but rarely generalize well which means that you could know how to use all the fancy built-in functions but still not be particularly good at composing them into targeted use cases for what you really care about...

« first day (2660 days earlier)      last day (715 days later) »