« first day (426 days earlier)      last day (4048 days later) » 

12:45 AM
Is this on topic? I mean: software design for something big is a nontrivial thing and needs probably more thought and details than can be put on a Q&A site, or?
 
acl
1:00 AM
@RolfMertig it seems to be too vague to get a useful answer, anyway
 
@acl and not really much effort taken yet
 
acl
@RolfMertig I think that even if it gets clarified, the question it seems to be asking (how do I design nontrivial software, how do I choose data structures) needs a book, not a Q&A site. as you said
 
@acl or a consultant ...
time to sleep. bye
 
acl
night
 
 
1 hour later…
2:34 AM
@RolfMertig a highly paid consultant ... :)
 
 
9 hours later…
12:02 PM
@MichaelE2 Can you join chat?
 
12:28 PM
@rm-rf @Mr.Wizard @J.M. @Verbeia anyone care to re-open the blog chat? I got a question from SE concerning that and think the chat would be a good place to put it to...
"that" meaning the blog facilities as such :-)
 
12:46 PM
@halirutan Are you still here?
 
12:59 PM
@gpap you need 3k rep for close votes...
 
@MichaelE2 Yep
@MichaelE2 Although your answer is very detailed, it was already clear to me how Dynamic works.
Now, when I read my question I'm not so sure I pinned down what I really wanted to ask.
Let me give another example:
@MichaelE2 This is the example from the Refresh page:
Manipulate[{x, Refresh[y, TrackedSymbols :> {x}]}, {x, 0,  1}, {y, 0, 1}]
So good so far, lets add one small detail which breaks the behavior:
Manipulate[{x, y, Refresh[y, TrackedSymbols :> {x}]}, {x, 0, 1}, {y, 0, 1}]
@MichaelE2 Can you explain this in more detail? The only sentence in your answer about this is "There is only one Dynamic, so everything is updated or nothing is."
 
1:20 PM
@halirutan I figured you knew how Dynamic worked, but for the sake of others and to introduce the idea of a branch, I had to say something....
The expr in Manipulate becomes
Dynamic[Refresh[{{x, y, Refresh[y, TrackedSymbols :> {x}]},..]]
Only expressions inside Dynamic are updated, and if they are the whole expression is reevaluated....
The Refresh you put inside the Manipulate makes that y expression depend only on x. But there's no x in the expression. It does not force y to depend on`x`.
 
@MichaelE2 Yes, I see this but in the first example there is only one Dynamic too and 2 Refresh functions.
 
You mean this:
Manipulate[{x, Refresh[y, TrackedSymbols :> {x}]}, {x, 0, 1}, {y, 0, 1}]
 
Yes.
 
The whole expression depends only on x because:
x depends on x and Refresh[y, TrackedSymbols:>{x} depends only on x (only is the sense of at most).
So changing y does not cause an update. Does that make sense?
The confusing thing (it was for me once) is that TrackedSymbols only limits what the expression depends on. It doesn't expand it.
Try Manipulate[Refresh[x; y, TrackedSymbols :> {x}], {x, 0, 1}, {y, 0, 1}] and see if it makes sense. (x; y depends on x and y, but TrackedSymbols will limit the dependence to x only).
 
@MichaelE2 So with Refresh and TrackedSymbols I cannot prevent the expression inside Refresh to update, I can only tell what triggers an updated for this small part?
 
1:34 PM
Yes.
 
@MichaelE2 Ok, now this all makes sense.
 
I'm in the middle of editing my answer. How much of this should I add?
 
@MichaelE2 Make it clear, so that others understand it ;-)
 
OK, I'll do my best.
 
@MichaelE2 Thank's for your patience.
 
1:37 PM
@halirutan No problem. By the way, is there an easier way to respond to a message on chat? I have to (remember to) click on the previous message, and my aging hands would rather use a keystroke or something
 
Hello @halirutan. What question are you talking about?
@MichaelE2, salut
 
@MichaelE2 I remember I have seen a browser-plugin which eases the handling inside chat. Btw, this exists for the SE page too and it is pretty awesome. You can navigate, answer, up and downvote with keystrokes only.
 
@Rojo Hi.
 
@Rojo Hi btw.
 
@halirutan Thanks, I'll look for it.
 
1:41 PM
@MichaelE2 This one for the main page:
66
Q: Official keyboard shortcuts

balpha Screenshot About This is the official keyboard shortcuts script for the Stack Exchange sites, as announced on Meta Stack Overflow. It is still in active development, and will eventually move from being a user script to being a full-blown (optional) feature when it is ready for that. Do...

 
@halirutan Thanks!
 
@halirutan Question read
 
@MichaelE2 And I think this one for chat
29
Q: SE Chat Modifications -- Keyboard navigation and commands for chat

Tim Stone Screenshot Use /command shortcuts to perform common chat tasks: See message history inline: Easily preview replied-to messages: And much, much more... About Legends tell of a prolific Meta Stack Overflow chatter who despised using their mouse above all things. In an effort to keep t...

 
Now, time to read the answers to see if it's already there
Wow loooooooong
 
@Rojo Sorry...
 
1:44 PM
Haha, it's great
@halirutan My answer would be an average of Michael's and andre's
I think andre's said the most important thing to say, but without enough explanation
 
@Rojo Yes, that's what I thought.
 
@MichaelE2 @Rojo Even m_goldberg had the right example. So three accepts I guess ;-)
 
@halirutan Everybody's seems to want that
 
@Rojo Hehe, yes thats nice.
 
1:49 PM
@Rojo Once upon a time, I was tracking two deer, x and y....
 
Haha
 
2:00 PM
@Rojo Nice summary. You forgot: And they lived happily ever after.
 
@MichaelE2 Edited
 
@Rojo The moral is a good addition, too.
@Rojo Now you've got it all :)
 
@halirutan I used to use this, but now I use vimium for Chrome, so I can use the keyboard for all sites
 
@MichaelE2 Thanks. I honestly think it's worth it to imagine it "in order"
because it does make a difference
 
@rm-rf The chat script seems not to work anyway in my chrome here.
At least I don't know how to use it.
 
2:17 PM
@halirutan it works in chrome here. You type those commands into the text box
for example, type /help and it should open up a popup showing the list of commands
 
Hmm, yes, you are right. Works.
Is there a manual?
 
@halirutan Use ctrl-up to change focus on the messages (you should see a green highlight) and then move up/down and press R to reply, Q to quote, S to star, E to edit etc.
@halirutan I think the readme on github has a good list
 
@rm-rf Works like a charm ;-)
really nice
 
@Rojo Stories are good for that (to put things "in order"). As I've grown, my imagination has become more and more visual. Hence that darned graphic -- I hope some find it helpful.
 
@halirutan it also makes it easy for you to reply to your own messages (to keep the chain logical)
(someone asked about it a while ago... maybe Brett)
 
2:22 PM
@rm-rf yes, I see. This was one thing that bothered me a while ago.
Especially when you write long monologues.
@MichaelE2 You have to try this chat-script. It's really nice.
5 mins ago, by rm -rf
for example, type /help and it should open up a popup showing the list of commands
4 mins ago, by rm -rf
@halirutan Use ctrl-up to change focus on the messages (you should see a green highlight) and then move up/down and press R to reply, Q to quote, S to star, E to edit etc.
 
/help
 
you need to install the script first :P
 
@Rojo you have to install the scrip first ;-)
 
(:
 
ups
@rm-rf when I use a command like /star, is there any way to type the message id faster?
Or do I have to type all 8 digits?
 
2:30 PM
@halirutan You will need to type all the digits, but the script easily shows you the id of the message on the message itself (see to the right, next to the time stamp)
 
@MichaelE2 @Rojo Here is the link to the chat script again: stackapps.com/questions/2105/…
And here you can find the ReadMe which shows how to use it: github.com/rchern/StackExchangeScripts/blob/master/README.md
 
/help
There we go
Thanks
 
2:47 PM
@halirutan Thanks
 
@rm-rf on a different topic, I tried to SENotebook an answer containing a Graphics3D[] with Spheres in it and that did not go down so well, the graphics were posted as textual expressions
Not sure if that is just a quirk of my setup, though.
 
@YvesKlett hmm... can you do a quick check to see if it works for a different Graphics3D (perhaps a plot3d or something similar)?
Or even better — if you ctrl-shift-e (show expression) the cell with the 3d graphics, does the structure look like Cell[BoxData[GraphicsBox[... or is it something else other than a GraphicsBox (perhaps GraphicsBox3D?)
 
3:03 PM
will do... lots of GraphicsBox3D and similar
ugh, now my system is messed up copy/paste-wise. Will get back once I got time and a reboot (my graphics driver setup is not really happy right now, too).
 
@YvesKlett ah, ok. It's an easy fix. I only considered GraphicsBox.
I'll fix this a little later and push it
 
3:38 PM
@rm-rf and things like SphereBox and CuboidBox feature prominently, too.
 
 
1 hour later…
4:59 PM
@YvesKlett I think only Graphics3DBox is necessary here, as it is the outer wrapper even for SphereBox et al. I've pushed a small change. Please pull it and try again
 
@rm-rf would you mind posting the link once more?
 
btw, do you use git or download each time?
 
me, just downloading :P
 
@YvesKlett here :)
 
Yay!
Will report asap (but don´t hold your breath) :-)
@rm -rf Oh, and I think I encountered another thing while using a Texture, but cannot reproduce right now.
 
5:05 PM
@YvesKlett Might have been related to the same thing... It doesn't matter what the contents of the graphics 3d are, it's handled the same way the image uploader palette does. I just wasn't handling any graphics 3d. I think textures should also work fine now, but if it doesn't, please let me know. Hopefully in the next few weeks, I'll get some time to improve it and show a progress bar so that you know when the conversion is complete
 
beautiful!
 
 
3 hours later…
7:49 PM
This question does not look like a duplicate of the linked post to me:
2
Q: Exporting large numbers to a text file

user6463I need to export large numbers to an ASCII text file, but as simple digits without powers of ten, e.g. I need to write 123456.789 to the file and not 1.23456789E10^5. It seems like such a simple thing, but I can't figure it out without some serious kludge. Is there a neat way of doing this? Than...

Would someone please take another look? It seems as though this new one is about formatting and output, whereas the old one was about building numbers from parts.
 
@Mr.Wizard Six of one, half dozen of the other. But, the OP finds that the solution to the other problem works for him:
Thanks cormullion, Chris Degnan's longform[] does exactly this — user6463 5 hours ago
 
@rcollyer Well I have an answer for the new question I'd like to post that I don't feel answers the old one. Now what?
 
hmmm ... how different is it?
 
It's not complete yet but I expect it to be shorter than longform. I also think the old question is confusingly written.
Sorry, I'll be right back.
 
@Mr.Wizard considering the author, that may be kind
ping me when you get back.
 
8:06 PM
@rcollyer I'm here. I took another look at the old post and I guess it does cover this. I certainly think the new question needs to remain visible to the search engine as I expect it will be more easily found, or at least understood.
 
@Mr.Wizard okay. I won't vote to delete it.
Does your method apply to the original question?
 
It was string based for export to a text file, so probably not directly. The new user's problem is solved however and that's what matters.
 
works then.
 
@rcollyer Don't vote to delete any duplicates unless it is utterly crap
 
See you around. :-)
 
8:11 PM
@Mr.Wizard bye.
 
(there are quite a few with pending delete votes in the queue)
 
@rm-rf I have yet to vote to delete since getting the ability to do so (again). Something about a waiting period ...
 
@rcollyer Just not instantly... you can, 2 days later. You can see the list here
 
@rm-rf right, and thus far, they have been gone prior to me being able to exercise my new power.
@rm-rf Used it, once. I can now live the rest of my life in peace. :)
Maybe it should be integrated into the standard review page ...
 
@rcollyer It is under review...
I mean, the feature request is being reviewed, not that it is currently in /review...
 
8:19 PM
@rm-rf it did make me chuckle.
 
why, do you want to use your super powers?
 
@rcollyer I don't need anyone's votes for that... just that 3 people have commented, yet no flags/close votes.
 
we're being polite, for now. I considered it, but I thought the newbie should be given a chance before the lions descend.
 
@rm-rf sorry. Usually I do but this time I forgot.
 
8:22 PM
I'll be happy to eat him alive once he demonstrates that he is a Darwin candidate.
 
@rcollyer That's not a newbie question... that's a lazy question. Also, not his first
 
@rm-rf six of one ... and, as far as his user page is concerned, it is the first.
don't worry. we'll get to the crunchies and munchies soon.
 
Hi, I am going to scan the Wagner book early next week. The file is going to be rather large and I wonder what you guys think when I split it in say 5 files. Each part of the book in its own PDF file.
 
@rm-rf oh look, the blood bath has begun. lunch time. :D
(I wish I could tell you where this is coming from ...)
 
@Matariki Works for me... As long as I can get my hands on that book, you can split it how many ever ways you want :D
@rcollyer Are you still in NY or IL?
 
8:26 PM
@rm-rf NY.
 
Working from home working out well for you?
 
It has its good and bad moments. No commute to speak of. Difficulty getting into the office on time. The usual.
I get to have lunch with my wife and kid. Bonus.
 
@rm-rf OK. I'll intend to go for better quality scan then and split the file in 5 or so.
 
@rcollyer Yeah, that must be nice. In my case, I have to ward off a cat that wants to snuggle/lay on my laptop
*cats
@Matariki So is Todd in charge of the distribution? How does it work? We all email Todd and he sends it to us?
I hope it doesn't become too much for him to handle...
 
@rm-rf 2 year old who wants to watch Ormi the pig. Or, muppets. Lots and lots of muppets. Better than Barnie.
 
8:35 PM
@rm-rf I haven't talked to Todd about the distribution details yet. I am not sure if Todd will be in charge as he did the dealings with McGraw. I'll ask him today how he envisions the distribution.
 
8:46 PM
@Matariki Ah, ok. I misread his answer... MGH might be the one handling distribution (it wouldn't make sense for them to not do so). I thought the email he had listed was for such purposes, but it was mainly to get in touch with him re: scanning.
 
@rm-rf Yes. I just contacted him. Lets see how the scanning goes.
@rm-rf Do you know of any other copies floating around here (of the book I mean) in case there are bad pages in my copy?
 
@Matariki No, sorry. I've not tried to get hold of any illegal copies.
 
No, I don't mean illegal copies. Just a paper version
 
Oh, you mean folks with the book? Yeah, halirutan has one, so does Leonid.
Also, I think mgoldberg has it and someone else mentioned a few days ago that they managed to get hold of one. Szabolcs got one on loan from a library, but that was a while ago and I'm not sure if he still has it
 
@rm-rf. Ok, that's good just in case a page or two are of bad quality
 
8:53 PM
So yeah, there are folks here with the book in case you need a backup copy for bad scans
 
@Matariki I have the book too in case..
 
9:30 PM
@halirutan Thanks. I have not come across any really bad pages but a few that have writing on it. Those can however be removed from the scanned pages easily.
 
@Matariki "How can I remove writing from a scanned page using Mathematica?" :D
 
9:48 PM
@rm-rf You just use the inbuild function RemoveWritingFromScannedPage[page_] ;-)
 
10:25 PM
@OleksandrR. @rcollyer There we go: it's a keygen license lol
 
acl
is that what the last comment is saying? I've been staring at it for a couple of minutes but can't make sense of it...
@Matariki if you want, I can OCR the PDF, add the text as a layer, and send it back to you (if you do not have suitable software)
(if you send me the PDF, that is!)
 
@acl It means he is using a pirated version of mma and using a key generator to illegally generate a (possibly) license number and activation ID. And he was perplexed as to why that didn't work
 
acl
@rm-rf I mean, he says: "i did not expect to have this problem in case i am using a keygen licence", and I had trouble parsing this (I am familiar with the meaning of keygen :) )
but OK now that it's pointed out, he seems to indeed be saying he's using a keygen. ha
 
@acl aaah, ok that makes sense :) (I was surprised to think you didn't know what a keygen was)
 
@rm-rf good grief! Unbelievable.
 
10:45 PM
@acl Thanks, for the offer. I might do some image processing and in harder cases put it through Tesseract. Most of it are just pencil marks/writings so the use of an eraser in document preconditioning phase :D will do.
 
acl
@Matariki does tesseract insert the recognized text back into the pdf as a text layer? making the PDF searchable?
 
@acl No, its just an OCR I am calling from Mathematica. No such fancy OCR package at my disposal. What are you using if I may ask?
 
acl
@Matariki I thought Acrobat Pro, but just realized my personal laptop does not have it installed (nor do I have a personal license for it).
I do have this and have succesfully used it for PDFs up to ~100 pages.
If it does not work, I still have access to Acrobat Pro, too. I find it very useful to have long PDFs searchable
 
@acl I'll have a look at PDFpen. Yes, searchability is also a major point for me. My, admittedly ambitious, plan was to turn the book into a set of notebooks. But due to time constrains I'll get the scan done first and then see how to improve that. Maybe a community effort to get the pages OCRed would work.
 
acl
@Matariki it would have to be very accurate OCR, or else someone with lots of time and patience would need to work on it...
 
10:58 PM
@acl true
 
acl
@Matariki I think once it's scanned in some form, and in people's hands, it will be easier to organize a conversion into notebooks if we want (eg, find out how many people are willing to work on this, then split the task)
 
@acl Yes, you are right it would only make sense if enough people would be interested. However, there would the still be the license to be sorted as we have right now only permission to distribute copied/scanned pages.
 
acl
11:14 PM
@Matariki oh, I did not know there was such a restriction
 
hhh
Hello
What could cause that I cannot see most of the plots?
Like this

inQ = Abs[FractionalPart[N[#, 16]]] < 0.1 &;
check[0 | 0., 0 | 0.] := 0;
check[a_, b_] :=
With[{p = (a + b)/(a^2 + b^2), q = (a - b)/(a^2 + b^2)},
Sum[Boole[inQ[c p] && inQ[c q]], {c, 100}]];

Image[Table[(0.01 #)^(1/4) &@check[a, b], {a, -1, 1, 0.0025}, {b, -1,
1, 0.0025}]]
I have tested things on Thinkpad x60s and Macbook Mid-2012 -- does Mathematica have some hardware requirement to plot things?
(none of my computer is able to plot things :(
Wait -- it appeared now ?!
It took many minutes to plot this fractal
I cannot understand why -- the activity monitor shows that Mathematica is not taking much CPU or RAM
 
acl
try
inQ = Abs[FractionalPart[N[#, 16]]] < 0.1 &;
check[0 | 0., 0 | 0.] := 0;
check[a_, b_] :=
  With[{p = (a + b)/(a^2 + b^2), q = (a - b)/(a^2 + b^2)},
   Sum[Boole[inQ[c p] && inQ[c q]], {c, 100}]];

Monitor[
 img = Table[(0.01 #)^(1/4) &@check[a, b], {a, -1, 1, 0.0025}, {b, -1,
      1, 0.0025}];,
 a]
to monitor progress
(then Image[img])
(last post was for @hhh)
 
hhh
@acl nice one! Now I see that it is at least doing something...
Thank you, stil new...
Is there a way to stop it?
Kill kernel?
 
acl
cmd-dot?
or ctrl-dot?
here it works fine and produces a plot
 
hhh
How long did it take for you?
 
acl
11:27 PM
I am not sure. a few minutes
 
hhh
I have "Intel HD Graphics 4000 512 MB"
 
acl
(macbook air 1.8GHz i7)
 
hhh
I have Macbook Air 2012-Mid 1.8GHz i5 8GB
 
acl
@hhh I'd have thought the "Processor" bit is more important for this
@hhh well, should not be hugely different in your case I guess
 
hhh
does this kind of MBA have support for CUDA and OpenGL?
 
acl
11:31 PM
@hhh doubt it (mine doesn't)
but can't say for sure. did you try?
 
hhh
so that is the reason many such commands do not work?!

http://mathematica.stackexchange.com/questions/21619/how-to-plot-mandelbulb-fractal-in-mathematica
(I am trying to learn fractals and complex numbers but most example plots not working...)
 
acl
@hhh most example plots from where?
anyway if you are trying to learn complex numbers, don't mix it up with learning CUDA programming
 
hhh
Mathematica's help about fractals or this site about fractals, second I wil get some
 
acl
@hhh eg these should work (Hofstadter butterfly and Mandelbrot set)
 
hhh
So you meant I cannot use commands such as OpenCLFractalRender3D?
 
acl
11:34 PM
@hhh well, I can't. If your mba doesn't support it, nor can you.
 
hhh
I don't know. I don't even know how to check.
I think it is a safe bet if yours does not support, mine also not.
 
acl
type CUDALink/ref/CUDAInformation at the top of the help browser
@hhh well anyway, not sure it's worth the effort to learn cuda programming for this
@hhh (the CUDAInformation[] thing will probably download a gigabyte of stuff and then tell you "no CUDA device" or something like that)
 
hhh
CUDA and OpenCL are supported in desktop computers?
Should I avoid all code having words such as CUDA and openCL? Cuda is for parallel so not good with MBA, what about OpenCL?
OpenCL is also some parallel thing...hmm
I would like to plot $(r,\alpha)\mapsto(r^2,2\alpha)+C=(r^2+C,2\alpha+C)$
Simple parametric plot?
 
acl
is $r$ a radius and $\alpha$ an angle?
 
hhh
I cannot fully understand this definition, it is an recursive definition but what are the initial conditions?
 
acl
11:45 PM
$C$ appears to change meaning between the left hand side and the right hand side of the equality
 
hhh
It should be simple Mandelbrot set
Suppose $C=0$ for the beginning.
 
acl
isn't mandelbrot obtained by iterating z->z^2+c?
 
hhh
$(r,\alpha)\mapsto(r^2,2\alpha)$
(this is given by my teacher -- I am trying to see what this is...)
 
acl
and you are sure nothing's mistyped?
 
hhh
Yes to the iteration question.
 
acl
11:47 PM
well there $c$ is the point you're trying to colour or not, and $z$ starts from 0 always
 
hhh
 
acl
right so it's $C$ on the LHS, $C_{x/y}$ on the RHS
I guess he's just trying to get you to understand polar notation for complex numbers
 
hhh
I cannot understand $C$ and $C_{x/y}$ here yet, trying to plot just this iterative thing...
 
@hhh I think OpenCL allows for both CPU and GPU based programming, so it depends on how the examples are implemented and whether you have the right graphics card or not
You can try Mark's code for the Mandelbrot set. I had to set block size to 1 to get it to work
 
acl
@hhh I am guessing that $C=C_x+iC_y$ is meant to be a complex number. and he's asking you to iterate a la mandelbrot set, but specifies the iteration in terms of the magnitude and angle of $z=r\exp(i\alpha)$ just to force you to think about converting between polar and cartesian form for the complex numbers.
could be completely off though
 
11:57 PM
 

« first day (426 days earlier)      last day (4048 days later) »