« first day (179 days earlier)      last day (4302 days later) » 
00:00 - 05:0005:00 - 20:00

5:15 AM
@Rojo I'm back now, briefly; are you here?
 
Hey
Was just here
 
Do you need me to run something?
 
Well, haven't tested this more than very little, and I'm suuuure it must be very buggy, but check it out and let me know. Code is ugly
I paste it here?
@MrWizard
 
darn, sorry, looked away
Yes, please.
 
Oka
AppendTo[$ContextPath, "Internal`"];
I got lazy to type the context
ClearAll[private`count];SetAttributes[private`count, HoldRest];
private`count[i_Integer, expr_, count_Symbol]/;(expr;++count;False):=Null
private`count[i_Integer, expr_, count_Symbol]/;count===i:=(Remove[count];RemovePeriodical[private`count[i, expr, count]])
private`count[___]:=Null;
ClearAll[createScheduledTask, startScheduledTask,
stopScheduledTask, $scheduledTask, removeScheduledTask,
scheduledTasks, scheduledTaskObject, runScheduledTask];
SetAttributes[{createScheduledTask, runScheduledTask}, HoldFirst];
SetAttributes[scheduledTaskObject, HoldAll];
SetAttributes[{stopScheduledTask, removeScheduledTask}, Listable];
createScheduledTask[expr_, {time_, count_: 1}, start_: Automatic] :=
1 + Length[scheduledTasks[]] /. c_ :>
scheduledTaskObject[c, expr, {time, count}, start, False] /.
private`STQ[scheduledTaskObject[id_, expr_, time_, start_, _]]:=Length[Cases[scheduledTasks[], scheduledTaskObject[id, expr, time, start, _]]]>=1
private`STQ[___]:=False;
the "idea" was for it to be similar to what you can find in v8 docs
Just, starting with lowcase
 
5:20 AM
Certainly a lot of work went into this. Thank you.
Is there a simple case you suggest I test this with?
 
No problem. I'm not proud of the code however, could be way more elegant
Sure
runScheduledTask[Print@"hello", {2, 5}]
That should print "hello" 5 times every 2 seconds and then stop
 
Confirmed. :-)
 
Good
It probably doesn't update the task object to inactive in the end and other minor things, but, in time
Let me know tomorrow or another day of bugs and stuff
 
I need to sleep soon. I'm going to putter around the site for a few minutes, but I don't have enough time to work through this. Tomorrow I will try to get back to you and test it further.
 
Buhbye, leaving the chat too
 
5:26 AM
Thank you again. If this works reasonably well I'll get you the bounty promised. Bye Rojo!
 
@belisarius arrivederchi
 
@Rojo Auguri e figli maschi
 
 
3 hours later…
8:47 AM
Not bad for a 2D graphics if I say so myself.
 
9:43 AM
Is there official news that we can use the logo? (I saw this ) I somehow missed that, can someone point me to the relevant discussion?
 
@Szabolcs I think the discussion started here
 
@Heike Thanks! But it's so disappointing ...
 
@Szabolcs I agree.
 
10:04 AM
@Jin Don't bother with vector formats for shaded 3D Mathematica graphics. We can send a very high resolution bitmap instead, PNG or TIFF if you like (speaking from experience making figures for publication)
@belisarius I like this one.
@Heike Did you colour it yourself, or is it lifted from one of the previous know posts?
Should I add (optional) auto-update to the image uploader palette?
 
10:29 AM
@Szabolcs I used ParametricPlot with a ColorFunction based on the direction of the normal to the 3D surface.
 
I'd love to keep the pentagonal and dodecahedral symmetries if possible.
I just found this, by complete accident:
 
@Szabolcs I made a chain with three links from a solid block of wood once, but that was child's play compared to that.
 
Then there are V Bulatov's sculptures (belisarius posted them several times): bulatov.org/metal/index.html bulatov.org/wooden/index.html
I'm wondering if we could take inspiration from Penrose tilings for the logo as well
Or make the logo completely 3D instead of 2D (something that stands out and jumps out of the page as 3D)? That would be too great of a departure from the current design maybe ..
 
@Szabolcs I was thinking of that. It would be a shame to have to change that as well to make it match with the new logo.
 
@Heike Penrose tilings can be made round. I can make a rounded version of the same dart-kite pattern by drawing the green-red arcs from here: upload.wikimedia.org/wikipedia/en/d/d6/Kite_Dart.svg
Or any other curved shape, in this spirit:
 
10:39 AM
@Szabolcs I wouldn't have recognised that as a Penrose tiling
 
@Heike It's not, I just tried to illustrate how I imagined to make the Penrose tiling have curves. I didn't find an image yet when I posted that, but here's one now:
 
10:50 AM
flowery
 
11:01 AM
I'm trying to figure out if I can incorporate knots into the tiling pattern without making it look too busy (and still keep it compatible with how I made it periodic)
 
@Szabolcs I like that better than the original actually.
 
 
4 hours later…
2:40 PM
@RM, hey
 
2:51 PM
Stole your shape
 
R.M
3:01 PM
nice :)
 
@RM I chose a nice letter
Check this out
Grrr I and W
 
3:20 PM
 
R.M
@Rojo What font is that and why isn't it centered?
it is centered in mine by default
 
@RM Bulatov on tilings
@Szabolcs Check that link
 
3:37 PM
@belisarius I've actually seen some of the original Escher tilings in real life.
 
R.M
@belisarius very impressive! I spent some time looking through his other art and some of the slides... I wish there was a notebook I could download and play
@belisarius I've wanted to ask you if he was related to Yaroslav
 
yep
Yaro is his son
 
R.M
I didn't know he was at WRI though... I thought he was still at Google
 
mmm I guess we should not speak about private lifes here in public
@RM Oh perhaps I tangled his work with another guy. Let me check
wait
 
R.M
Yaroslav Bulatov, Corvallis, OR
133 5
 
3:44 PM
@RM Yep he is still at google. It was another person who went from Google to WRI
Sorry about the mistake
 
@RM well, I wanted the size of the text to be fixed relative to the rectangle
so I did something heike suggested, from the docs
I'd like it to be centered, of course
 
R.M
what's your code for the tile?
 
@Rojo How did you rescale the coordinates of the letters?
 
This is what I'm doing
@Heike horribly manually
ClearAll[letter];
i:letter[s_String]:=i={{LightYellow,EdgeForm[{Thick,Darker@Yellow}],Rectangle[{0., 0.}-1/2,{0., 0.}+1/2,RoundingRadius->.1]},{Black,First[First[ImportString[ExportString[Style[s,Bold,FontFamily->"Helvetica",FontSize->1.5],"PDF"],"TextMode"->"Outlines"]]]/.{x_Real, y_Real}:>{x-0.42, y-11.308}}}
Scan[letter, "A"~CharacterRange~"Z"];
Partition[
Graphics[#, PlotRange -> {{-0.6, 0.6}, {-0.6, 0.6}}] & /@
letter /@ "A"~CharacterRange~"Z", 5, 5,
1, {}]~GraphicsGrid~{Frame -> All}
 
@RM Still, we could ask Yaro if he can get a notebook for playing around with that beatiful tilings
 
R.M
3:48 PM
yep
 
@Heike I can think of taking all the coordinate points of the joined curves of all the letters. Find the x min and max to get the scaling. Then the y median to get the y translation. and the x translation, on a letter by letter basis, the median of the x values... Let me try that
 
R.M
Perhaps something like:
ClearAll[letter];
i : letter[s_String] :=
 i = Graphics[{{LightYellow, EdgeForm[{Thick, Darker@Yellow}],
     Rectangle[{0., 0.} - 1/2, {0., 0.} + 1/2, RoundingRadius -> .1]},
     Inset[Graphics@{Black,
       First[First[
         ImportString[
          ExportString[
           Style[s, Bold, FontFamily -> "Helvetica", FontSize -> 1.5],
            "PDF"], "TextMode" -> "Outlines"]]]}, {Center, Center},
     Center, Scaled[1]]}, PlotRange -> {{-0.6, 0.6}, {-0.6, 0.6}}]
Partition[letter /@ "A"~CharacterRange~"Z", 5, 5, 1, {}] // Grid
 
@RM Niiice
so I will stop trying what I just said, thanks
 
@Rojo That's what I was thinking of. This is what I got:
ClearAll[letter];
i : letter[s_String] := i =
With[{fc =
First[First[
ImportString[
ExportString[
Style[s, Bold, FontFamily -> "Helvetica", FontSize -> 1.5],
"PDF"], "TextMode" -> "Outlines"]]]},
{{LightYellow, EdgeForm[{Thick, Darker@Yellow}],
Rectangle[{0., 0.} - 1/2, {0., 0.} + 1/2,
RoundingRadius -> .1]}, {Black,
fc /. FilledCurve[a_, b_, c___] :>
With[{mean =
Flatten[Reap[
Cases[b,
p : {__?NumericQ} :> (Sow[p[[1]], "x"];
Sow[p[[2]], "y"]), Infinity];, {"x",
"y"}, (Max[#2] + Min[#2])/2 &][[2]]]},
Partition[
Graphics[#, PlotRange -> {{-0.6, 0.6}, {-0.6, 0.6}}] & /@
letter /@ "A"~CharacterRange~"Z", 5, 5,
1, {}]~GraphicsGrid~{Frame -> All}
 
@RM Oh, but I think I'd rather have them as primitives more than as graphics... Feels more efficient but perhaps I'm talking nonsense?
 
4:01 PM
 
R.M
This looks nicer than mine
@Rojo Oh well... I had to use Graphics earlier to inset it
 
@RM But my code still a bit muddy.
 
R.M
not the outer one though.
@Heike gets the job done.. .that's what matters :)
 
@Heike I'm trying to see what to tweak in yours to make the letters slightly smaller for the W to fit
 
R.M
A slightly smaller font size? 1.4 seems to make the W fit
 
4:05 PM
Yeah
1.3 or 1.2, because in real use the pieces will be smaller so the edges will end up thicker
unless I made them fixed, which I should also ask or browse the docs
 
R.M
make a giant arcade game
 
EdgeForm[{Thickness[0.05], Darker@Yellow}], nice
Haha
You think pieces should fall continuously or in steps like in tetris?
 
R.M
letter rain!
I think steps would be nice... gives the user some time to think
 
Haha
Many options
Several pieces could fall at the same time
Ok, steps
If steps then maybe its better to use a grid instead of a full graphics? Opinions?
Nah, graphics
 
4:20 PM
You could consider using tiles with Qu instead of Q. Otherwise it might become too tricky.
Slightly more transparent version (maybe):
ClearAll[letter];
i : letter[s_String] := i =
With[{fc =
First[First[
ImportString[
ExportString[
Style[s, Bold, FontFamily -> "Helvetica", FontSize -> 1.5],
"PDF"], "TextMode" -> "Outlines"]]]},
{{LightYellow, EdgeForm[{Thick, Darker@Yellow}],
Rectangle[{0., 0.} - 1/2, {0., 0.} + 1/2,
RoundingRadius -> .1]}, {Black,
fc /. FilledCurve[a_, b_, c___] :>
With[{mean = (Max[#] + Min[#])/2 & /@
Transpose[Flatten[b, Depth[b] - 3]]},
FilledCurve[a, b /. {p__?NumericQ} :> {p} - mean, c]]}}]
sc = Scan[letter, "A"~CharacterRange~"Z"]
 
@Heike You mean write "Qu" inside the piece?
I think it will do more harm than good when they try to visualize fast what words they can build up
 
@Rojo Yes.
 
I'd rather edit it manually to make it very Qish and less Oish
or what you are saying is that in English Q always comes with an U next
Is that true?
 
@Rojo A word game I play on my phone does this. It's because a Q is nearly always followed by a U.
 
Then it's a good idea :)
 
4:27 PM
@Rojo Yes, that's what I meant.
 
No, I'll let it be tricky, because
given that DictionaryLookup works with many languages
its nicer to make it more general
You just better not put a Q in a bottom corner
bottom right
 
R.M
and a z and x next to it :)
 
Get all the words in a language, tally the characters they use, and make them drop with that relative frequency
Now, putting the letters that sholud go near near, is part of a good player's game ;D
 
@Rojo Scrabble already optimized the chars freq
you could steal their tables
 
@belisarius as long as they are available for the langauges in DictionaryLookup
and not just English
However, even then, this should be didactic more than mega-optimal, so I think it's better to do it MMAlly
 
4:36 PM
Many editions of the word board game Scrabble vary in the letter distribution of the tiles, because the frequency of each letter of the alphabet is different for every language. As a general rule, the rarer the letter the more points it is worth. Many languages use sets of 102 tiles, since the original distribution of one hundred tiles was later augmented with two blank tiles. __NOTOC__ {| id="toc" |- ! colspan="3" | Table of contents |- | style="width:33%; vertical-align:top; padding-left: 8px; padding-right:15px" | *English *Afrikaans *Arabic *Bulgarian *Catalan *Croatian *Czech *Danis...
 
Woaa
 
hehe
 
The source of the wiki article gtoal.com/wordgames/langdist.html
 
Stepping on a giant's shoulder
 
4:55 PM
@belisarius I just posted the code of the scheduled tasks as an answer mathematica.stackexchange.com/questions/2806/… , care to test it?
 
@Rojo In v8?
 
Yeah
Just start the symbols in lowercase
runScheduledTask, etc
 
ok. let me finish what I am doing now, and I will give it a try
 
Thanks
 
$10
 
4:58 PM
I'll pay you in a month, when it's not enough for a banana
 
$10 BLUE U$D
 
Haha
I've already found that it isn't working fine with delayed starts
 
@Rojo for delayed starts you need to schedule a task that schedule a task
 
That's what I supposedly did, but clearly not well
AHHHH
I missed a damn "`"
Haha
(ok, 3 of them)
 
@Rojo did you implemented ResetScheduledTask?
 
5:09 PM
No, didn't konw it existed
 
hehe
 
I'll edit in a minute, and look that at the docs
in the docs
 
@Rojo and StopScheduledTask?
 
Yeah
 
good
@Rojo and ScheduledTaskObject?
 
5:11 PM
Edited
Yeah
Lowercase
 
you did a nice work!
 
Only if it works
 
@Rojo I believe in Murphy, I believe in God ...
 
No messages however
and I am still not sure what ResetScheduledTask does, hum,m
It's like a StopScheduledTask with the option to change the time specs on the fly? Humm
 
@Rojo Create and Run both have time parms. Reset goes back to the Create ones
 
5:20 PM
@belisarius I assumed that Run creates a whole new one
just like Create
but starts it already
 
sorry, I mean Start, not run
 
If Start takes time parameters, they are undocumentes
or not easily seen at the top of the docs
 
let me check
ResetScheduledTask stops the task if it was running.
When the task is reset, the time until the next activation is set to the full time interval.
and it takes time parms
 
Ok, I'll asume its like stop, with the chance of changing the time pars. Thanks!
 
Hi, Rojo and belisarius.
May I ask you a question?
 
5:26 PM
@Andrew hi
go ahead
 
I don't think this is a good enough question to post on the list, but I still would appreciate any help you can please give me.
 
Hello @Andrew
Just say it already ;)
That's what belisarius is here for
 
spit it out
 
In Windows, one often specifies a color by R, G, and B values.
 
puke it down
 
5:27 PM
Can I do this in Mathematica?
 
yep
 
There is RGBColor[]
 
paste some code
 
I would like to make a peach colored circle
 
@Andrew, evaluate "Red"
and see the output :)
 
5:28 PM
I tried Graphics[{RGBColor[255, 169, 128], Circle[]}]
 
{Graphics[{Red, Circle[]}], Graphics[{Thick, Circle[]}],
Graphics[{Dashed, Circle[]}],
Graphics[{Red, Thick, Dashed, Circle[]}]}
 
But it does not give me a peach colored circle.
 
The numbers go from 0 to 1 only I think
Is that right @belisarius?
 
arguments from 0 to 1
 
So how should I normalize?
 
5:29 PM
yep
divide by 255
 
divide by 255?
OK.
It works great!
Thanks!
Talk to you later!
 
another happy user
 
:-)
 
:D
We should have an auto-message like "all our operators are too drunk to answer, call later"
 
Haha
Do we still have an operator?
If we dialled 100 something
?
 
5:33 PM
@Rojo No idea
 
5:46 PM
 
How cute
 
@Rojo shame on you
 
R.M
Hello? What did you say? The hyperbolic pentagon is trademarked? Ok, so how about 10 of them?
 
@RM No hyperboles. They are Beziers
and .. there are 11
 
R.M
This guy writes his posts terribly and I've told him about 3 times so far to not use alternating capitals, unnecessary ellipsis, lowercase i, arbitrary code blocks, etc. and he show no signs of making an effort of improving/correcting
This is insane
 
6:02 PM
@RM well, if I was 100% lazy I wouldn't even use code blocks
 
R.M
It's almost as if he's trying to be annoying... there is no particular pattern there. No txt speak, no slang... it's mostly correct, but annoyingly capitalized and highlighted.
 
@RM Probably not a native speaker
 
R.M
I edit a lot and I come across several common patterns among people... this guy is plain annoying. I'm very patient with grammar/spelling/english mistakes and will readily fix them, but this seems blatant, especially when I've corrected his previous ones and commented every time
The dude's not even German, as it doesn't make sense for German capitalization style
 
@RM OK. Cool down. Do not edit the post this time. Leave a comment and downvote if he doesn't do the job or ask for help
 
R.M
Well, I'm not doing anything this time... I just left it as is
I don't want to downvote, as that would be petty and be downvoting for the wrong reasons
 
6:11 PM
If you ping a low rep user that can't come to chat
what happens?
 
R.M
You can't ping anyone that hasn't been/is in chat... only mods can do that
 
I see
 
R.M
A simple test is to write @ and the first few letters... if it autocompletes, you can ping them.
 
@RM I wrote a comment
 
R.M
Thanks :) It probably went straight to /dev/null in his brain :P
 
6:14 PM
@RM I'll not bet about it :D
 
6:25 PM
@RM That's an insult to Unix/Linux.
 
R.M
ooh.. only about 100ish away from Szabolcs
 
@RM And only about 1100 away from 20k.
 
R.M
yes! And then I can call myself TheHypnotoad
Not that the two need to be related
The goal was to reach it before we graduate, but that doesn't seem very likely...
 
6:41 PM
@RM I am sure you can tempt the SE team to delay the launching. Just tell them we are accusing each other of bribery.
 
Anyone knows a way for the FE to higlight defined symbols in black, in contexts other than global?
 
R.M
@belisarius I think WRI's lawyers are doing a good job at delaying us already
 
I mean
relative contexts
 
@RM Did you bribe them too?
 
such as blala=8
 
R.M
6:42 PM
@Rojo yes, hold on...
 
'bla'la=8
Oki
 
R.M
Search on the site for units package... I think the Q &A was by rcollyer and mr.wiz
 
Ok
 
@RM Maybe R.M is one of them.
 
found it
 
R.M
6:43 PM
6
Q: No Syntax Highlighting of Package Functions

rcollyerLately, my installation of Mathematica has decided not to highlight any symbols loaded from packages. The picture below is a good representation of this: where after loading both PhysicalConstants` and Units` BohrRadius, Meter, and Angstrom should all be black. Yet, they retain their "unknown ...

Maybe that?
 
@Heike Your paranoid Sundays are magnificent!
 
Intresting
 
R.M
Sometimes, I think my way of indexing things is funny, but this is how I remember things... {"rcollyer", "mr.wizard", "units"} --> "context highlighting"}
 
Hehehe
It's interesting, though not exactly what I was looking for
Do 'la=8
 
R.M
6:45 PM
Oh. Hmm... there's also one more on SO by 500 and Mr.Wiz.
 
that " ' " is a backtick
and try to get that black
 
R.M
ah, I see.
That's probably harder, because that needs the FE to be aware of the current context, which it doesn't (or does it?)
 
@belisarius Is there any mathematics behind that picture?
 
@RM I think your proably right, and perhaps I'll just stop coding with relative context
 
R.M
@Rojo For instance this:
 
6:48 PM
oor I'll post it as a question and see what comes up
 
R.M
`la = 8
Begin["test`"];
`la = 9;
End[];
 
Needs["PolyhedronOperations`"];
d = "Dodecahedron";
k = PolyhedronData[d, "SkeletonGraph"];
vc = PropertyValue[k, VertexCoordinates];
r = Show[Graphics[{BezierCurve[#,
SplineDegree -> 2]}] & /@ ((vc[[#]] & /@
List @@@ EdgeList[k]) /. {x_, y_} -> {x, .3 Mean[{x, y}], y})]
r1 = Show[
Graphics[{BezierCurve[#, SplineDegree -> 2]}] & /@ ((vc[[#]] & /@
List @@@ EdgeList[k]) /. {x_, y_} -> {x, 1.1 Mean[{x, y}], y})]
Show[r, r1]
 
R.M
I don't think the FE will know that the second `` `la`` should not be Black (if what you want is possible)
 
@RM I see your point, so I see how it wouldn't make much sense to color all `la according to the current context
but it would be nice anyway, hehe
 
@Heike Yes, code above
 
6:51 PM
@belisarius Thanks.
 
@RM However, the FE does know the $ContextPath and colors all instances according to the current context path
 
R.M
@Rojo Right, but it won't know that the `la inside test is a different context...
 
Hmm, I just heard something in the kitchen. Turns out a magpie was stealing kibble from my cat's bowl.
 
R.M
But you probably should post it as a question... I'm very weak in my FE foo, so I'm probably saying something incorrect
 
@RM nor it would if Begin was a BeginPackage (appends to $ContextPath), but it would however color it
and stop coloring it if the context is removed from the context path
 
acl
7:01 PM
@RM well, maybe he is trying to annoy you. what's more impressive is that two people upvoted that question
 
@acl Maybe they felt sorry.
 
@acl Those three are trying to annoy him
 
Nice
A top view of a circus
 
acl
@belisarius too much beer? :)
 
You all are trying to discourage me from becoming a renown artist. Know this: I am going to be rich and famous. In my autobiography I will scorn you.
 
acl
7:14 PM
@belisarius so what are you going to call it? Mein Kampf?
 
@acl No. "Memoirs of a Russian princess"
 
00:00 - 05:0005:00 - 20:00

« first day (179 days earlier)      last day (4302 days later) »