Wolfram Mathematica

« first day (96 days earlier) ← previous day next day → last day (4389 days later) »

1:38 AM

@CHM I'd like to think that the V9 release is the event that accelerates interest in the stack so that it turns the corner and gets out of beta.

rcollyer

1:51 AM

@BrettChampion could be interesting. Any hints on what's going to be in it? :)

yoda

2:21 AM

@rcollyer there's probably a truck load of NDAs that he'll be violating if he fed our curiosity ;)

He can anonymously snail mail us though... with letters cutout from old newspapers :P

rcollyer

2:39 AM

@yoda I figured. I was mostly being facetious. But, he does have my email, and I won't talk, much ...

1 hour later…

Brett Champion

3:51 AM

@yoda Just the one NDA. :-) I think my favorite new features are *********** and ************.

8 hours later…

F'x

11:34 AM

so, we had election results yestarday in France, and the statistical charts generated on the fly always gives some truly wonderful results… I couldn't take snapshots, but it reminded me of this excellent chart:

acl

@Fx that's right; let's put the numbers on screen, along with some nice coloured bars. no need for them to mean anything

F'x

11:51 AM

@acl yeah, I don't even begin to understand how you can produce that kind of chart

plus, it really changes the perception: if you put "economy" at its true 62% size, the others then rightly seem so unimportant

rcollyer

12:28 PM

@BrettChampion Looking forward to testing them out! If I notice them at all. I have this habit of not noticing new features for a release or three.

belisarius

12:40 PM

@acl it is a Log[] chart !

2 hours later…

tkott

2:39 PM

@acl @belisarius reminds me of xkcd.com/558

belisarius

@tkott :D

acl

hah, yes, that's brilliant

rcollyer

Wow, I like the 30 secs v. all night comparison below it. :)

2 hours later…

belisarius

4:17 PM

l = {9, 9, 10, 62};
p = l /. x_ -> 100 + x - x/ Log@x;
posTxt = Text[
Style[#[[1]], FontFamily -> "Helvetica", 15, Bold,
Background -> Black, White,
LineSpacing -> {1, 1}], {#[[2]], #[[3]]}] &;

s = BarChart[p, BarOrigin -> Left,
ChartStyle -> {EdgeForm[None],
Darker /@ {Blue, Brown, Red, Green}}, BarSpacing -> Large];
t = BarChart[180 {1, 1, 1, 1}, BarOrigin -> Left,
ChartStyle -> EdgeForm[None], BarSpacing -> Large];
q = BarChart[220 {1, 1, 1, 1}, BarOrigin -> Left,
ChartStyle -> {EdgeForm[None], Lighter /@ {LightBlue, LightBlue}},

Eli Lansey

@belisarius Ha!

belisarius

They are not cheating. Just rescaling :)

rcollyer

@belisarius Well, that is cheating if you don't tell people as everyone will assume that you're using a linear scale, not logarithmic.

belisarius

@rcollyer I am joking, of course :)

Heike

@rcollyer It could still be linear if the origin is at -200% or so

Great, we just lost our government

tkott

4:43 PM

folks, this answer by @Jens (http://mathematica.stackexchange.com/a/4028/360), is awesome except for failing with autoLegend[ListPlot[{1, 2, 3, 4, 5}, Joined -> True], {"bla"}]. The issue is the style extractor with the type of data there is from the ListPlot: {Hue[0.67, 0.6, 0.6],
Line[{{1., 1.}, {2., 2.}, {3., 3.}, {4., 4.}, {5., 5.}}]}. The current pattern matcher: Cases[lines, {c__, Line[__], ___} :>
Flatten[Directive @@ Cases[{c}, Except[_Line]]], Infinity], can't find that type of data. Is there a way for `extractStyles` to play nice with the current setup and deal with the 'bug'?

belisarius

@tkott You should post a question!

tkott

@belisarius recently i've been posting questions that have one liner answers, so now I just throw the question by the chat :)

belisarius

@tkott Ha! But by doing so, the answers are not usable by anyone except you!

tkott

5:02 PM

@belisarius Maybe I just have a higher threshold of 'interesting' or 'usable' questions

in this case, the answer is that I did not read the docs well enough --- I need to make sure the default option PlotMarkers->none is set...

belisarius

Oh well, one shoud Read The Fabulous Manuals :)

tkott

image_doctor

@Heike Not to worry, I'm sure a very similar one will be along shortly

Heike

@image_doctor That's what I'm worried about.

belisarius

5:18 PM

►

acl

@Heike you could aim for the belgian's record

image_doctor

@Heike Wonders what would happen if citizens all got an email each morning asking for their votes on national decisions ... far too much like true democracy for politicians to imlpement

belisarius

@image_doctor Hacking emails would turn out to be a very well payed job

image_doctor

@belisarius lol , yes it would - maybe those finger print readers compaq/hp kept fitting to laptops might come in handy :)

belisarius

@image_doctor I have one of those. It scares me to death to trust my work to their software

image_doctor

5:34 PM

@belisarius if I think too much about it, most software sacres me ;)

acl

6:33 PM

All, I am having trouble opening a particular tiff file with mma

I'd appreciate it if people could try downloading and opening it on their machine

sblom

@acl Like Import[] just refuses to work?

acl

no it imports it but doing things like ImageCrop etc causes all sorts of errors and eventually crashes mma

I am trying to upload it somewhere but it's 60MB

(additionally, I have noticed that if I Import a tiff with a preview, I get a list of the image and the preview image; ImageCrop on that list crashes the whole thing, too)

well, this is taking too long; I'll do it later tonight

Szabolcs

7:01 PM

@acl Use ge.tt or if you have Dropbox, use that.

@acl You can mail me if you need a quick test.

CHM

@BrettChampion I reckon ************ is the new Make[] function, as in Make["Coffee", "Espresso", "Guatemala", "DarkRoast"] or Make["Bagel", "Toast"->3, "CreamCheese"]?

For some unknown but extremely annoying reason, I can't access Nature Chemistry articles with my uni's VPN - only Nature articles.

Szabolcs

7:32 PM

@CHM They're different subscriptions---same here.

CHM

@Szabolcs but my department pays for both, apparently.

Maybe it's just a problem with the proxy not being configured to route traffic for Nature Chemistry

CHM

7:45 PM

@Szabolcs seems the proxy isn't properly configured, I'm using a VPN now and it works fine.

Sjoerd C. de Vries

8:39 PM

It's amazing this question actually gets answers. I've been reading it a couple of times and can't find anything that resembles a question. I'm tempted to close it as NARQ.

Q: developing Domain Specific Language and CKEs

We'd like to develop Health IT "active document" functionality that includes (1) interpreting users' natural-language (2) performs automatic markup to tag, link and journal clinical documentation and (3) communicate orders and queries to web services such as terminology servers and Computational ...

Eli Lansey

@SjoerdCdeVries not only an answer, but a Leonid answer, no less

Sjoerd C. de Vries

alancalvitti is the user with the most closed questions, and he doesn't seem to learn

R.M

> I understand this is very vague and underspecified, but in one of Leonid Shifrin's excellent summaries, he mentioned that MMA capabilities include developing Domain Specific Languages and am curious where to start in investigating the capabilities.

Seems valid — boils down to "where do I start looking at developing something like a mini W|A"

R.M

9:02 PM

@SjoerdCdeVries This one seems valid, but I voted to delete some of the other closed ones... a lot of them have two votes and needs just one more (if anyone has the time to vote, there's more here)

rcollyer

9:50 PM

@CHM No, that's been left out as incomplete as someone during testing entered make["Universe"] and nearly killed us all. We have to wait until v. 10.

@RM and my suggestion amounted to why develop it yourself when you can use WRI's.

Hi @Verbeia

I'd stay to talk but got to run.

acl

10:20 PM

@Szabolcs thanks: dropbox is what I've tried but it was too slow

I've now found the problem and it's not related to the specific file (this is a fancy way of saying I was being stupid).

however, mma really does crash if I do the aforementioned stupid thing

« first day (96 days earlier) ← previous day next day → last day (4389 days later) »

all rooms