« first day (874 days earlier)      last day (3600 days later) » 

3:05 AM
@Szabolcs Red is better ;)
 
 
3 hours later…
5:46 AM
@halirutan Yep, fully agree. I think this is generally the reason why Java is not so much used in heavy computations / number crunching. Paul Graham wrote some 10 years ago that he considered Java to be an evolutionary dead end, and I don't know of anything that would contradict that statement. But it seems that the memory issue is more general, and is there at the level of JVM, so other JVM languages are likely suffering from it too.
@Rojo That reminds me of one line of Wesley Snipes's character in Passenger 57 movie: "Always bet on black".
 
 
5 hours later…
10:43 AM
@halirutan I second that. One reason Excel imports fail so often for large files is that the Excel importer is Java based.
@rm-rf The OP of this question deleted his question directly after receiving answers in comments. Not a big problem, as the question has appeared in many disguises here (in fact I was already looking for duplicates), ...
... but I really hate this behaviour (keeping the answer for yourself). I suggest we undelete it and then close as duplicate. What do you think?
 
 
1 hour later…
11:54 AM
@SjoerdC.deVries agreed!
 
12:06 PM
I posted something on unix.se about a Mathematica related issue, if any of you have any idea please let me know :)

http://unix.stackexchange.com/questions/136396/clipboard-behaviour-when-copy-paste-from-mathematica-to-gooble-chrome
 
12:17 PM
@Öskå You should have asked here first :-)
@Öskå I guess you are using Chrome (no matter whether beta or stable) and not Chromium?
 
Yes I'm using Chrome indeed
@halirutan and I still can Copy/Paste on here.. :) But last time I had a clipboard related issue you advised me to ask there.. :D
 
@Öskå It depends on the whether in Oslo..
@Öskå OK, here we go:
 
I didn't get the Oslo joke :P
I can ask a proper question on mma.se if you like :)
I'm just going to copy/paste the same question
 
@Öskå I could have said it depends on some water level in China.. I just mean I know the answer in this special case and what you did (asking on Unix) was the correct way in general
 
That's why I asked on the chat here as well :)
 
12:23 PM
@Öskå I suspect the issue is related to the java one I reported here. This all comes from Mathematica which rapes the system clipboard by not ending the copied text with a zero byte.
 
Oh lord, now I remember you giving me that link last time.. :)
 
@Öskå After the last update, Chrome unfortunately shows the same behavior. I haven't found another way than using Chromium instead of Chrome.
@Öskå If you have time, you could post an issue on the official Chrome bugtracker.
 
mhm, I might do that indeed, it's really annoying. But when doing xclip -out I get Error: target STRING not available after copying something from Mathematica.
So is it Chrome-related or Mathematica/XClip-related?
 
@Öskå This is what I get after copying your code in Mathematica
xclip -o -verbose -selection clipboard | hexdump -C
Connected to X server.
Using selection: XA_CLIPBOARD
Using UTF8_STRING.
00000000  61 62 63 64 65 66 67 68  69 6a 6b 00 00 00 00 00  |abcdefghijk.....|
00000010  00 00 00 00 00 00                                 |......|
00000016
 
If I copy: StringJoin @@ CharacterRange["a", "z"][[;; 14 - 2]] I get:

Using selection: XA_CLIPBOARD
Using UTF8_STRING.
Type is STRING.
00000000 53 74 72 69 6e 67 4a 6f 69 6e 20 40 40 20 43 68 |StringJoin @@ Ch|
00000010 61 72 61 63 74 65 72 52 61 6e 67 65 5b 22 61 22 |aracterRange["a"|
00000020 2c 20 22 7a 22 5d 5b 5b 3b 3b 20 31 34 20 2d 20 |, "z"][[;; 14 - |
00000030 32 5d 5d 00 00 00 00 00 31 00 00 00 00 00 00 00 |2]].....1.......|
00000040 b0 eb eb 00 00 00 00 00 20 67 76 05 00 00 00 00 |........ gv.....|
 
12:39 PM
@Öskå Now copy the input not from within Mathematica but from the chat-message above and do the same xclip command in a terminal.
You will see that there are no additional bytes after the 2]]
 
Indeed, so I guess that Mathematica is f*cking things up and Chrome can't handle it? I have no clue how I could report that tbh :)
 
@Öskå It needs to be fixed at Wolfram. I already did this for the current beta but I wouldn't hope for much..
 
Let's not lose hope :D
But @halirutan thanks for your help, at least now I understand the issue :)
 
@Öskå No problem. Glad that my wasted hours save some one else's time.
 
12:55 PM
I bet you had a better understanding of what's internally going on thanks to that, which makes it worthy.., r-r-r-ight?
 
1:34 PM
@Öskå Only if I can fix it, which I can't in this case. I still have to live with the fact that each bit code I copy from Mathematica to IDEA has trash at the end.
 
@SjoerdC.deVries I agree with you in general, but here, it seems fairly innocent and nothing sinister. OP must've had a "oh duh!" moment and deleted it. Since there are no answers and there are plenty of possible duplicates as you say, I'd suggest just leaving it alone. Perhaps if the comments required more effort (e.g. a comment saying "I tried this for a while and here's an outline of the algorithm; I'll post an answer shortly") then we could undelete it.
 
2:06 PM
@rm-rf OK, I can agree with that.
 
 
1 hour later…
3:35 PM
@Rojo Did you use it, or do you just feel close to its name? :)
 
4:31 PM
Hi guys, for some reason ParametricPlot3D isn't showing all the longitudinal mesh lines: i.imgur.com/wxZVTvA.png Any idea what could be going on?
 
@Guillochon Can you try what Mesh->Full gives you?
IMO the automatic mesh generation is messed up. Just look what ParametricPlot3D[{Cos[u] Sin[v], Sin[u] Sin[v], Cos[v]}, {u, 0, 1.9 Pi}, {v, 0, Pi}] gives you and that the missing line is exactly where the sphere closes at 2Pi.
 
4:47 PM
@Guillochon `ParametricPlot3D[{Cos[u] Sin[v], Sin[u] Sin[v], Cos[v]}, {u, 0,
2 Pi}, {v, 0, Pi}, BoundaryStyle -> Black]`
anyway its the boundary and not the mesh
 
@chuy I'll try that, waiting for some cells to finish first
 
@chuy Exactly! It's really weird this is not the default when half of the examples in the doc are closed toruses or something else where this issues shows up.
 
@chuy Splendid, that works! Thanks.
 
I can see the sense in it not being the default, but view may vary
 
5:02 PM
@Szabolcs The latter :)
 
5:43 PM
Another quickie: Is there a way to change the line thickness for BarLegend? The lines/ticks are hairline thin and difficult to see
 
@Guillochon I don't know, I can't find a way either.
@Guillochon Try digging in Options[ChartingiBarLegend]` ... TicksStyle affects colour but not thickness.
(I think)
 
 
4 hours later…
9:55 PM
@bobthechemist You asked me some time ago about Joomla. Do you have any further experience? For instance, tweaking the JCE editor so that inserting the <pre class="prettyprint"... becomes easy?
I'm not satisfied with my knowledge about Joomla. In general it is a great system, because it is easy to use, but plugin/extension/modules is kind of hard to understand at the beginning. For instance the JPrettify plugin I recommended makes nothing more than inserting in each page that the prettify.js script is loaded so that you can use the highlighting. This is kind of overkill and I would really like to know how to tweak things and do stuff by myself.
 
 
1 hour later…
11:16 PM
@halirutan I haven't had much luck with the extension, and was planning on following up with you. Regardless of the the type of code (python, bash, WL, etc) I get no highlighting but do get grey-background code. The project went to the back burner for a while. I agree that the extensions are quite cumbersome - better in Joomla 3, but still a pain. I looked at Drupal and Wordpress as alternatives, but it wasn't obvious that they made the task simpler.
 
@bobthechemist Since I looked at it today, I can give you an exact description.
@bobthechemist I write you some lines when I'm back in Linux.
 
@halirutan cool, that would be helpful and appreciated
 
11:54 PM
Another shameless plug: my WolfSpec made it into this month's edition of the MagPi. See page 22.
 

« first day (874 days earlier)      last day (3600 days later) »