« first day (188 days earlier)      last day (2672 days later) » 
00:00 - 22:0022:00 - 00:00

12:03 AM
@Rojo not a good sign. I'm sure the last thing they want to happen is for a thriving black market to form.
 
@OleksandrR. They think they can force handle the market
I remember a couple of months ago, one of them said that "they had come to an agreement with the main exchange sites so that the "illegal" exchange rate would go down". Not in those words. Of course, he was just talking nonsense
 
Managing expectations by giving false, misleading, or confusing statements seems to be a key tool among central bankers today. Greenspan was a master at that and Bernanke seems to be quite accomplished as well. I don't regard it as a good thing.
{|class="toccolours" style="float: right; margin-left: 0.5em; margin-right: 1em; font-size: 85%; background:#c6dbf7; color:black; width:20em; max-width: 25%;" cellspacing="5" |style="text-align: left;"|Known as Fed Speak, the convoluted rhetoric has befuddled even the wisest of interpreters. |- |style="text-align: left;"|— Wealth Building Strategies in Energy, Metals and Other Markets |} In monetary policy of the United States, the term Fedspeak (also known as Greenspeak) is what Alan Blinder called "a turgid dialect of English" used by Federal Reserve Board chairmen in making intentionally...
 
R.M
Watch out folks, this might be the beginning of anonymous spam/vandalism mathematica.stackexchange.com/suggested-edits/345
 
@R.M Grrr
 
@R.M Uh oh. Well, am I glad you got your 20k so you can deal with this for us! :)
 
R.M
12:10 AM
Anyone with 5k should be able to see the suggested edit queue if I recall correctly, but thanks :)
does any bold person want to check out that link? :P
Probably it's just some random asdfgh smashings
 
12:22 AM
@R.M yes, they don't seem to be real addresses. Perhaps testing the waters. I was a moderator on a discussion forum and we had real trouble with spam until eventually I changed it so that new members' posts had to be moderator-approved. The volume of submitted spam posts was still somewhat overwhelming though.
 
acl
but what's the motivation? testing to see if spam links can be inserted?
 
R.M
This is the internet. People don't need any motivation
 
acl
a good point, but, why not post insults if you're just going to vandalise things randomly?
 
lmskdhfjlajhmlajcmlfaslvfajsdfa.com/alskdfjasldkj
 
acl
@Rojo jjjjjjjjjjjjjjjjjjjjjjjjjjjjj
jjj jjj jjjjjjj jjj?
still, easily dealt with,it seems
 
12:31 AM
@acl Mmmpppmpm fppmfpmmmfmp'fmm fmpmfpmmmfmp fmmfmfpfmpfmppffmmmppmpm fmpppf ppmmppmmmppp?
I could've used MMA but I'm eating cheese with one hand
That "but" made no sense
I need to go to the toilet but my ex is doing a final exam right now
That one either
 
acl
the second less than the first
 
Yes
However, I feel there is some relationship between my delaying going to the toilet and the exam. Anyway, moving on
 
acl
@Rojo could be
the universe is ever mysterious
 
12:56 AM
@acl Realizing how mysterious the universe is by not going to the toilet sounds like a drunkard's path to Zen
 
I went already
Couldn't take it
 
acl
@belisarius a random walk eventually covers the plane, and clearly Zen lies on the plane. QED
 
@acl The dimension of Zen is irrational, as any beginner can testify
 
1:16 AM
@acl a surprising number of spammers are not bots, but actual people (being paid a pittance for a huge volume of spam). They may be checking to see how much effort they need to go to, to get a post full of nonsense published, with the intention of coming back later and upping their game--or, their efforts, thus compensation, might be measured by some rather stupid software that can't tell these from real spam links.
One of their favorite tactics, and one that works well, is copying questions from other forums with a similar subject. Fortunately I think our topic is narrow enough that this approach is unlikely to be feasible.
 
Hey guys, anyone knows an easy and intuitive way to make these figures? mathworld.wolfram.com/images/eps-gif/HopfAlgebra2_1000.gif
 
@jmlopez File not found? wrong link?
 
@jmlopez looks like a job for LevelScheme. I haven't used it, personally, but the general layout is perhaps not dissimilar from an energy level diagram.
 
@OleksandrR., has anyone ask this before? I've been looking for something like, function map figures, but I haven't found anything related
Looking here: mathworld.wolfram.com/HopfAlgebra.html makes me wonder how they made those pictures.
Is there a name for these type of figures?
 
@jmlopez I really don't know. Making figures is far outside my usual domain; I usually avoid it as far as possible. All I know is LevelScheme is highly regarded and can make energy level diagrams amongst others. What these diagrams are called I don't know either: usually they are just referred to as "diagram"...
 
R.M
1:27 AM
@jmlopez you can do it with TikZ, for sure
Hold on, I remember a question that might be close enough...
You can do something like that using the graph objects... with the vertices being H\[CircleTimes]H, the edge labels being Δ, etc.
 
@R.M, yeah, that's what I have in mind
I can certainly do it but I was just wondering if anyone had made some sort of function to make it easier.
Or maybe find out if the person who wrote mathworld.wolfram.com/HopfAlgebra.html used Mathematica to make the figures.
 
R.M
probably not, unless they create these diagrams a lot in Mathematica... if I were writing a publication and needed to make something like this, I'd probably just use TikZ
 
I think I'll just stick to Mathematica for the moment. It shouldn't be too hard to make functions to tell it how to display a map from a set to another.
 
R.M
Anyway, this question was what I was thinking of... You just need to change the vertices, colour the edges, etc. which shouldn't be too hard. Also, that uses the older functions because Mr.W doesn't have v8. The newer graph objects might be more convenient.
A simpler approach using graphics primitives would actually be better I think.
You can use Jens' post on circuit drawing as a framework for creating the basic blocks that will lay out your set connections
 
2:00 AM
@R.M, thanks, the first question is exactly what I want. I like how the vertices can be moved in the second piece of code that Mr.W provided.
 
2:18 AM
@jmlopez It's called category theory, and those diagrams are the hall mark of it. But, that is as much as I know of it.
Category theory is an area of study in mathematics that examines in an abstract way the properties of particular mathematical concepts, by formalising them as collections of objects and arrows (also called morphisms, although this term also has a specific, non category-theoretical meaning), where these collections satisfy some basic conditions. Many significant areas of mathematics can be formalised as categories, and the use of category theory allows many intricate and subtle mathematical results in these fields to be stated, and proved, in a much simpler way than without the use of cat...
 
2:37 AM
@rcollyer That was Algebra 4xx when I was at uni
 
@belisarius we never touched upon categories, only groups and rings.
 
@rcollyer Well, I didn't take that course either
 
@belisarius :)
I find I'm using a lot of group theory as it is very useful in quantum mechanics.
 
@rcollyer Some friends said it was really interesting
 
Math only messes your MMA
You end up taking the Trace of a matrix
or creating a Set[stuff, stuff2, stuff3]
 
2:48 AM
:)
 
and then finding a Partition[Set[stuff, stuff2, stuff3]]
The Range[Sin[t]]
etc etc
 
@rcollyer Most physical theories are based on some kind of symmetry, yes
 
@belisarius It is stuff that you'd be able to figure out on your own if you just thought about it for a bit. Obviously that only applies to the first couple of weeks, but that is a good enough hook.
@belisarius and all conservation laws.
Noether's (first) theorem states that any differentiable symmetry of the action of a physical system has a corresponding conservation law. The theorem was proved by German mathematician Emmy Noether in 1915 and published in 1918. The action of a physical system is the integral over time of a Lagrangian function (which may or may not be an integral over space of a Lagrangian density function), from which the system's behavior can be determined by the principle of least action. Noether's theorem has become a fundamental tool of modern theoretical physics and the calculus of variations. A g...
 
@rcollyer Yep. That was Mech 101 :)
 
@belisarius we didn't get to that until 3rd year (or were supposed to at any rate).
Did they teach Lagrangian mechanics that early?
 
2:51 AM
@rcollyer Yep Mech 101 was in third year :)
 
@belisarius Ah. I saw the 101 and assumed that it was a first year course.
 
@rcollyer BTW Did you see the The Feynman Videos for Dummys on the MS site?
 
@belisarius you pointed me at them. I only looked at the one.
 
He explains cons. laws for lawyers. Wonderful
 
That's sounds really funny.
 
2:54 AM
@rcollyer RF was a funny and happy person, yes
 
@belisarius don't forget certifiably brilliant. Although, he always said his toolbox just contained a different set of tools.
 
@rcollyer What I admire the most of him is the huge area of expertise.
 
@belisarius he never stopped playing with what he learned. Constantly pushed himself to go farther. A lot of times it was just to show off, the safe cracking, etc. But, other times, it was just to learn something new.
 
@rcollyer He did brilliantly so many things that I could only be ashamed of my performance :)
from safe cracking, yes, to playing bongos
not to mention physics
 
2:59 AM
Ha
nice!
He was also such a honest person!
 
I wish I could say I came up with it, but it was on a blog I read called Study Hacks which, among other things, advocates a different way of approaching careers and learning. Worth the read.
 
@rcollyer Seems a nice site. Will take a look. Thanks!
 
@belisarius the author is somewhat of a heretic in his views. But, there is little denying his approach works: 2 or 3 books while working on his Ph.D.
 
@rcollyer That depends on how long he took to get his PhD :)
 
@belisarius I think within a 6 - 7 year time frame, so normal.
 
3:07 AM
@rcollyer Not bad for 3 books
 
@belisarius 4, the dissertation.
 
Given that conditions, in your dissertation you stand up and say "I wrote 3 books already, Thanks"
need to leave
nights all!
 
Well, they weren't exactly on his dissertation topic, but I bet the snake wrestling portion of the defense went quite easily.
 
Nite
 
Night!
I should go, also. Night @Rojo!
 
3:11 AM
Niite @rcollyer
 
 
4 hours later…
8:10 AM
@Szabolcs Thank you for your link, I found the notification right after opening stackexchange
 
 
1 hour later…
9:23 AM
@Szabolcs I just used your SEUploader palette to upload category graphs. it works very well so far. :-)
 
@magma I'm glad :-) Let me know if you have any problems with it!
So you're going to post the answer now?
 
yes within 15 minutes
 
9:57 AM
@MrW I also voted to close, but what I don't like about the upvoted answer on this question is that it suggests automatically modifying init.m. It's unsafe, things go wrong easily, it's not a good advice for a beginner question.
 
10:35 AM
@Szabolcs Well, he makes a backup and only appends, so there doesn't seem to be that much harm.
 
 
2 hours later…
CHM
12:27 PM
@belisarius just saw this posted on HN.
2
 
@Szabolcs Posted the answer. It took a bit more than 15 min. I hope it is informative
 
 
5 hours later…
acl
5:09 PM
@rcollyer 7 years is normal??
 
5:26 PM
@acl hehe ... I saw more than that too
 
@acl The US median is 6, but pushing 7 years, now.
Note, this is directly from a Bachelor's degree program, so it includes Master's level work.
 
acl
@rcollyer if I had taken 3 years to do my first degree and 7 for the phd, I'd have gotten my phd last year!
bloody hell
 
The market in the US is also flooded, so it is harder to get a job. Hence, the longer time spent as a grunt (grad-student), multiple post-docs, and then maybe, just maybe a faculty position opens up.
 
@acl Think of it. Now you are one of the small group of really well payed scientists that also understands life, universe and beyond.
 
@acl what's it run where you are?
 
acl
5:31 PM
@rcollyer I did it in the UK where it's 4 (on average). most of my contemporaries took 4 or 5 years. I was stupid and did it faster
here I think it's 3, after 2 for masters and 3 or 4 for the first degree (I am not sure, as there are a number of paths)
 
@acl 4 is getting very rare here. But, some friends did it in 4.
 
acl
@rcollyer I did it in 2... not a good move
 
@acl So, 5 - 6 then which puts it in a little below us.
@acl why?
 
@acl That depends on your postdoc prospects
 
acl
@rcollyer because if you do a bsc in 3 years and a phd in 2, you get your phd 5 or 6 after you start. you then have to compete with people who get their phds 10 years after they start, and of course it's not that easy if nobody looks at anything except year of phd
@belisarius yes well I barely survived (I switched subjects immediately after I finished, which didn't help)
 
5:34 PM
@acl Ah. Smaller publication record.
 
acl
@rcollyer right. switching subjects hurts even more
 
@acl yes, yes it does. Hopefully, I won't have to go through that.
 
@acl sounds like you need a nice advisor
with strong relations
 
acl
@belisarius I had, I just ignored them. but yes, strong connections is the only thing that counts most of the time
 
@acl Oh yeah. That's life
 
acl
5:36 PM
well it worked out and I can't claim I didn't enjoy it.
@belisarius well now I advise my people, and of course I tell them to do exactly the opposite of what I did. let's see if they listen
(but I also made sure they got good connections, better than I have in fact, so it should be OK either way)
 
@acl That sounds good
 
acl
@belisarius yes: if all else fails, one of them will hire me eventually
 
@acl nepotism
 
@belisarius :)
 
acl
@belisarius right, if you can't beat'em, join'em
 
5:40 PM
brb
 
acl
i just need to graduate/supervise enough people now to have a network :)
 
@acl You just need to get enough money to run a group as large as Oppenheimer's!
 
acl
@rcollyer exactly
 
R.M
The # of years to graduate also depends on the field... in Engg, most people want to go for a job, so they aren't interested in building a publication record or attending conferences beyond what's necessary (and for vacations), so they graduate in 4. In sciences, where most people tend to go for a post-doc, the average is 5.5-6. Humanities and social sciences are much worse and average 7-8
 
acl
@R.M but is funding really available for 8 years?
 
5:45 PM
@acl define "available."
\begin{rant}I despise with a deep passion the program suite I am working with. F-ing thing seg faulted and I don't know why. \end{rant}
 
acl
@rcollyer I mean, do you typically stop getting paid after n years? or is money found to pay you until the end?
@rcollyer the spice of life!
 
@acl I think it is a mixed bag, and never, ever pays well.
 
acl
@rcollyer that is a given of course
(except in switzerland apparently)
 
@acl seriously? even for the humanities?
 
R.M
@acl not really... well, if you manage to find a TA position every year, then you can get by. The upper limit for university funding is 8 years here. Not a penny after that
 
5:48 PM
Gotta run and figure out this bloody mess. Bye all.
 
acl
@R.M 8 years for a phd is not that unreasonable as an upper limit!
 
acl
this is good
who prepared the image?
 
R.M
I did
 
acl
@R.M how did you produce the torrent of messages?
 
R.M
5:52 PM
On[] ;)
followed by Import["error", "HTML"], I think
 
acl
it's still printing messages here...
it produced a 4.5MB notebook file!
brilliant
 
R.M
Heh, I wonder if the developers ever use On[] (i.e., all messages) for any kind of debugging... There's just way too much info to be useful, I think
 
acl
@R.M the same is true for Trace but, with a bit of massaging, it can be very useful
(massaging of the output, that is)
I think that was one of the most educational answers I've seen here or on SO
 
R.M
I agree, that answer really showed me what can be done with Trace. Although, to be fair, I seldom (almost never) use Trace and friends... I guess it's high time I started poking into uncharted territories (for me)
 
acl
@R.M if you use traceView as described there, Trace can become very useful
OK so if I write an answer saying "Read the manual" for this, am I mean? he does ask "what should I do?"...
 
R.M
6:07 PM
I don't see a harm in asking if he has read the docs for NMinimize (although, I wouldn't put it as rtfm)
 
@R.M Y[x] = y[x] /. sol
This is definitely someone that need to read the manual.
 
R.M
@OleksandrR. haha! I didn't actually read the question, just the title
 
Considering whether to vote to close as too localized, or as a duplicate of this user's previous question.
 
acl
@OleksandrR. I'd go for "too localized", since it seems unlikely people confused about how to define functions would find it
 
R.M
I agree
All he had to do was use JM's answer to his previous question: dsaad[t_] = Q[t] /. First @... he didn't even copy correctly
 
6:15 PM
@R.M yes, exactly. Actually on that account I decided to downvote as well as voting to close.
 
I thought the Epic badge was gold, and now it turns out it's silver :(((
 
@OleksandrR. Could you please explain your downvote to the OP?
 
@R.M What eactly did you edit in the blog post draft yesterday?
 
@Szabolcs I think you need 150 times to get gold
 
@belisarius I'm not that obsessive
 
R.M
6:17 PM
@Szabolcs nothing
 
@R.M WP said you edited it. Maybe just whitespace.
@R.M I'll put it in an etherpad and convert it to MarkDown. I want to do this because Google Docs doesn't seem to like plain text, and it's just too difficult to convert between formatted Google Docs and something Wordpress compatible. Etherpad supports plain text and tracks the edit history. What do you think?
 
R.M
I had it open in edit mode, but did nothing and closed it after a few hours when I left for the day. It might have autosaved my name
I was going to make some edits to the code and organization. I'll not touch it if you're working on it
I won't be able to get to it until a few hours later anyway
 
@Szabolcs The Obsessiveness level is always measured with a very subjective meter
 
R.M
@Szabolcs Yes, I like etherpad
 
I wanted to start editing, but it really bothers me that the edit history is not tracked. That's why I'm in a hurry to move it to another editor.
 
R.M
6:20 PM
Also, etherpad is open source and can be installed on custom domains/servers. I think it would be nice to bring this up with SE and as if we get get a mma.blog/sandbox or something like that were registered blog contributors can collaborate
 
@OleksandrR. Thanks!
 
@belisarius sorry, I probably should have commented first and then downvoted to avoid any confusion rather than doing it the other way around.
 
@R.M THIS is THE solution IMO :-)
 
@OleksandrR. No no, mine was just a reminder b/c he is new in the site.
 
R.M
@Szabolcs Yeah, I liked that suggestion. Will take time before they implement it though...
 
6:29 PM
Okay, so, Etherpad. Which is the best one? PiratePad?
 
R.M
That works... that's what was used for the tag brainstorming sessions, no? There's also beta.primarypad.com/p/Ta2gPpHxBi
 
@Szabolcs the gold one should be called Honorificabilitudinitatibus
 
R.M
I think Mr.Wizard is half way there
 
7:18 PM
@Jin You should drop that message here in the main room :-)
 
Jin
"I'm going to purchase Mathematica today!" - Jin
4
although i'm a bit confused about the licensing. I only plan to use it to create artworks, e.g. geometric patterns, like the awesome ones you guys made the for the site design
should I buy the commercial version $2k+ or the personal one?
the commercial one is a bit overkill for what i'll be using it for..
 
acl
ba
ab
ah so it works :) froze up earlier
jjj
ok I'll stop spamming and restart the browser
 
R.M
@Jin hmm... tough to say. I'd say that a home license is sufficient, because you aren't using Mathematica primarily for your professional work — only for artwork, possibly as a hobby. On the other hand if you consistently use work created with it in your design work, then I think you might have to get the pro license
 
I really don't know the answer to that, Jin
 
R.M
It might be best to write to WRI's support/sales contact and find out
 
acl
7:28 PM
@Jin ah, how about playing with the trial for a few weeks first? it's a pretty strange language if you're used to python, C and the like
(not to rain on any parade or anything :) )
 
Jin
@acl I have downloaded the trial. I'll give it a try once I get some free time. (got another site launch soon)
 
The trial can't Export, probably an essential feature when evaluating it for such a purpose ...
 
acl
@Szabolcs really? I guess that complicates things...
 
Yes, it can a couple of annoying limitations. I don't remember exactly what. I think it was version 6.0 when I tried the trial.
@Jin Do you use other programming systems which are made for artists, such as Processing or NodeBox?
 
R.M
Or what programming languages are you familiar with?
 
Jin
7:32 PM
@Szabolcs no.
But I used to program quite a bit. so I'm comfortable picking up new syntax 'n stuff
I did a lot of ray tracing back in the day.
 
Looking forward to your questions, Jin :-)
 
Jin
@Szabolcs ha. I'll try to figure out as much as I can, on my own first.
I really miss Autocad.
 
@Jin We don't have many art-related questions yet, but check out this tag
 
Artlandia has a Mathematica package. I think they experimented in Mma before they made the illustrator plugin.
 
R.M
7:36 PM
@Jin Wait till you see this :)
 
Jin
can Mathematica do recursive bitmapping?
here's what i mean, from my blog post: 8164.org/ps-tut-minihendrix
 
What is recursive bitmapping?
 
Jin
in my blog, it's only 1 lvl deep. i wonder with Mathematica, you can define "n" level
 
@Jin Making it ...
 
Jin
of course the original photo probably needs to be high res, and has good contrast
 
R.M
7:39 PM
@Szabolcs There was a mathematica oneliner on this... by Yves, I think
 
@Szabolcs I already saw it done ... if I only remember where
 
R.M
No, Zdeněk
 
You found it?
 
R.M
Go down to "recursive image collage"
 
Okay, I'll go watch Doctor Who then :P
That's why I can't make a demonstration ... can't find a topic which is not implemented yet.
 
Jin
7:42 PM
btw Jeff tweeted about Mathematica.se today
did you see that http://mathematica.stackexchange.com/ is now outta beta with a beautiful new @jzy design? Fantastic technical community.
@codinghorror *well actually,* the logo and patterns were designed by the community, using Mathematica! http://meta.mathematica.stackexchange.com/a/555/96
 
R.M
@Jin If you have Mathematica, evaluate this:
With[{im =
   ImageTake[
    ColorConvert[
     Import["http://www.8164.org/wp-content/uploads/2008/09/jimmy_\
normal.jpg"], "Grayscale"], {1, 350}, {30, 380}]},
 ImageAssemble[
  Round[Rescale[
      ImageData[i = Nest[Darker, ImageResize[im, 50], 3]]] 10] /.
   n_Integer :> Nest[Lighter, i, n]]
 ]
 
I hope import works in the trial version ...
 
R.M
It should... it's what goes out that matters :)
 
Jin
@R.M "n_Integer :> Nest[Lighter, i, n]" does "n" determine how "deep" it goes?
for example, if i want the small faces to be made up by smaller faces, then n =2?
 
acl
8:04 PM
no, that just darkens the small copies n consecutive times
 
R.M
@Jin No, not really. The first i=Nest[Darker,...,3] darkens the image (in the middle of the code) — this is the darkest it will be. The final n_Integer:>... part repeatedly lightens this image according to the brightness value in this darkened image
The size of the face is determined by the ImageResize part. There it is resized to 50px wide. Changing that will give you smaller/larger faces
 
acl
well, this code is not written in a very extensible way
 
R.M
True, it was meant to be for a "oneliner" competition :)
 
acl
8:32 PM
@R.M oh right!
 
Jin
this is fantastic. i'm excited to learn
 
@Silvia 欢迎聊天
 
@Szabolcs 谢谢:) just come in for curious :)
 
Jin
你們講中文?
 
@Jin 我是中国人, @Szabolcs 好像是学过中文的 :)
 
Jin
8:46 PM
@Silvia i am too!
 
对,我在学习中文 :-) 可是看不懂繁体字 :-(
It's bedtime for me though
 
@Jin Nice to see another Chinese :D
 
I always thought he was 韩国人
Ohh ... what do we do with this one?
0
Q: Request for Text on Mobius Strip

SaraI've been looking for a way to create an image of my own text in the shape of a mobius strip. I know nothing about Mathematica, but noticed some images from the forum under "Rotating an image along a Mobius strip" that were similar to what I had in mind. Could someone create this for me and jus...

I hate to vote to close on new user questions.
 
@Szabolcs Is it because of his avatar or his name? I have the same illusion because his name Jin.
 
I thought I read it on some SE blog, but it seems I was wrong.
 
8:54 PM
我能阅读繁体字,不过大部分不记得如何正确书写……
 
Jin
@Szabolcs actually my name can be a Korean name too.
 
acl
@Jin a quick and dirty modification to that code gave me the above result
 
@Jin I'm thinking maybe I know you in some Chinese site like Douban?
 
acl
png here
 
@Szabolcs I voted to close because she was not planing to use a Mma driven bot to make the tatoo
 
R.M
9:00 PM
She has to wait for v9, to do Tattoo[image, Location -> "Arm"]
 
Jin
@acl interesting. my jimmi hendrix photo is rather tiny, so the end result is hard to see. how about you try it on this one: presstletter.com/public/Image/Maria%20Elena/Maarten%20Baas/…
 
acl
@Jin actually that shouldn't be a problem, you could just replace each pixel by an image of jimmi. i just slapped the thing together to see if it would work, and chose random parameters
I am now working on the (more appropriate) image of js bach :)
 
Jin
can you post your code, I'm going to steal it :)
 
@acl JavaScript Bach?
 
acl
@Jin hang on it's running on js now
it's hard to make generic
 
9:05 PM
I'd like to see one which doesn't lighten or darken the small images but Binarizes them with the appropriate threshold.
 
acl
I probably should ask a question on main and see what the others come up with
 
Unfortunately I managed to crash the Front End at the last step (out of memory...) so I lost my code :(
 
acl
@Szabolcs you mean you'd like to dither with mini-copies instead of changing their brightness?
 
(this is from the blog)
though I realize that darken first, binarize later might lead to the same effect
 
Hello
 
9:08 PM
so figuring out the binarization thresholds as a function of the desired intensity might have been a waste of time
 
acl
@belisarius a [combination] (schillmania.com/projects/soundmanager2)
 
Hi @Vitaliy
 
acl
hi
 
How do I invite someone to chat?
 
acl
(just the man for the job, by the way)
 
9:09 PM
@VitaliyKaurov With care
 
@belisarius ;-) i can guarantee that
 
Perhaps just send him a link?
Does he have enough rep to talk here? If not, I'll try to grant the permission once he joined.
@Vitaliy I think the invitation feature only works if the person is already present in another chatroom. Otherwise, just send him a link to this room.
 
acl
@Szabolcs I am not sure you need to binarize. one approach to dithering is to replace each pixel with a block of pixels, say 20 by 20, black but with a number of white dots proportional to the brightness of the original pixel and randomly located inside the block
but, I'm done playing for the day and anyway whatever I write will be done better, faster by an army of others :)
 
I'm done too
 
+1 for all in the loupe question
 
9:17 PM
Finally :-)
 
I have a problem about CellFrameLabels, I can't find a simple way to align the boundary of cells with different length of CellFrameLabels. Somebody has a hint?
As I'd like to use some background color, I can see the position of right boundaries of the main contents vary with different CellFrameLabels length. looks ugly :(
 
@Szabolcs thanks. grats on epic
 
A simplified style definition:
Cell[StyleData["teststyle"],
 CellFrameLabels->{{None,
    Cell[
     TextData[
      StyleBox[{
        CounterBox[
        "teststyle",CounterFunction -> (
          Part[{"I", "II", "III", "IV", "V", "VI", "VII", "VIII", "IX", "X", "XI", "XII", "XIII", "XIV", "XV", "XVI", "XVII",
            "XVIII", "XIX", "XX", "XXI", "XXII", "XXIII", "XXIV", "XXV", "XXVI", "XXVII", "XXVIII", "XXIX", "XXX", "XXXI", "XXXII",
            "XXXIII", "XXXIV", "XXXV", "XXXVI", "XXXVII", "XXXVIII", "XXXIX", "XL", "XLI", "XLII", "XLIII", "XLIV", "XLV", "XLVI",
 
9:33 PM
Hey @Silvia
Nice magnifier
 
Thanks :)
I once did a similar magnifier, so the code is handy.
 
Nice one. I am not too good at image processing and visual stuff, so those eye candy questions elicit my admiration :P
May not be elegant, but why don't you just pad your roman nubmers?
Btw, you can use RomanNumeral instead of the explicit list
but they will be lowercase, humm
 
R.M
@Rojo ToUpperCase?
 
I sometimes need to visualize data to pretty (as pretty as my boss feel happy :P) graphics.
I believe some function can not be used in stylesheet :(
 
R.M
17
Q: Defining functions in stylesheets

jmlopezIs there a way to define functions in a stylesheet? So far I have managed to write my stylesheets without the need to use my own custom functions or definitions. For instance, maybe I have some color that I wish to use throughout my stylesheet. myColor = RGBColor[.5,.5,.5]; The reason I would ...

 
9:47 PM
I think the stylesheet can only handle functions which can be evaluated in front-end, so render the nb do not require start a kernel.
@Rojo Great! Padding them is indeed a solution :D
 
2
A: Artistic image vectorization

faleichikThe following approach was suggested by cormullion in his answer, namely it is "spiral engraving". The idea is to approximate the image by short lines arranged in a spiral. The main issue is the lines lengths: they should depend on the spiral curvature and on the image details. I decided to imple...

 
@R.M Oh yes I know this post, but just feel it's not as "simple" as I expected. :P
 
Jin
now i'm curious if Mathematica can do hedcut arts!
Hedcut is a term referring to a style of drawing, associated with The Wall Street Journal half-column portrait illustrations. They use the stipple method of many small dots and the hatching method of small lines to create an image, and are designed to emulate the look of woodcuts from old-style newspapers, and engravings on certificates and currency. The phonetic spelling of "hed" may be based on newspapers' use of the term hed for "headline." The Wall Street Journal adopted the current form of this portraiture in 1979 when freelance artist Kevin Sprouls approached the paper with some in...
 
I mean "as I expecting"... sorry for bad English.
 
R.M
@Jin looks pretty similar to halftone
 
Jin
9:52 PM
@R.M yes, but i think the placement of the dots are more systematic than random
 
@Jin Surely this is the first Stack Exchange site launch that managed to sell Stack Exchange a software license.
Glad to have you on board as a fellow M user!
 
You need to find the lines/curves/flow in the image and follow it. Looks difficult to do it automatically.
 
R.M
Oh yeah, get SE to buy your Mathematica pro-license @jin ;)
 
I think first generate a contour plot or a field plot, then convert structured lines to dots.
 
00:00 - 22:0022:00 - 00:00

« first day (188 days earlier)      last day (2672 days later) »