« first day (191 days earlier)      last day (4281 days later) » 

12:00 AM
@R.M nice! but I'm not going to follow the logic :)
it is neat, I'm not making any critic. It's just that I am lazy right now :)
 
R.M
heh :)
 
I remember Yaro did a whole mess with SatisfiabilityInstances that ended up saving 98% of the computation time. But I can't remember more than that
 
R.M
It might be useful here too... I just haven't fully understood how to use it
 
@R.M Also, I suspect that question was HW
The Bresenham line algorithm is an algorithm which determines which points in an n-dimensional raster should be plotted in order to form a close approximation to a straight line between two given points. It is commonly used to draw lines on a computer screen, as it uses only integer addition, subtraction and bit shifting, all of which are very cheap operations in standard computer architectures. It is one of the earliest algorithms developed in the field of computer graphics. A minor extension to the original algorithm also deals with drawing circles. While algorithms such as Wu's algorith...
 
R.M
I don't know about that... his explanation for the industry application seemed to fit with his earlier question about a month ago
If he's doing some petroleum engg courses, homework will hardly be of the type where you'd have to find line intersections
I'm guessing he's more a beginner engineer at an oil company (or possibly an intern) who has been given a simple problem
 
12:20 AM
@R.M hehe I remember how cozy it was to solve simple problems with only a handful of variables. Perhaps you are right. I made no comment about it because I am not sure enough. But think of it: If you can set the final grid coarseness , I think you can use the algo I linked above, which is too well known
 
R.M
I first thought that was his intention, but it wasn't. I thought it was a dupe of this one — I think Heike used the above algo or a variant of it (actually, I didn't follow the logic fully)
 
@R.M I see no substantial difference. The adaptive grid size?
 
R.M
Yes
 
12:43 AM
@belisarius Closed as dup of the Q you migrated. No merge, since there aren't any answers.
 
@SjoerdC.deVries Ok!
 
Bresenham is for fixed scale pixels so isn't applicable in this case
 
@SjoerdC.deVries working on it
 
I once programmed this algorithm so I could run the same code on my Atari as on our mainframe.
 
R.M
I guess you can now run your mainframe programs of the past on your iphone
 
12:46 AM
It worked OK, but the native Atari call was faster.
 
R.M
I call Bresenham. — stevenvh yesterday
Now that makes sense... I didn't know the name of this algorithm
 
@R.M I still run some mainframe assembler code here
he
 
R.M
I thought it was some weird form of "I call shenanigans" in some language
 
I guess so. One of my early mma mistakes was trying to invert an integer 100x300 matrix
It broke out of all batch queues I could place it on except for the low priority week-lasting queue.
 
@SjoerdC.deVries hehe "to Siberia"
 
12:50 AM
Somehow I missed the difference between integer and real calculations in MMa.
The resulting matrix was 20 MB. My mac had 4 MB at the time.
 
@SjoerdC.deVries Integers are more compact, faster and ... oh my!
 
All the matrix elements in the result were rationals with hundreds of digits
 
@SjoerdC.deVries We had some spare time of a PerkinElmer at a nuclear facility. I'd better not tell in public what I did
 
You switched off temperature controls in order for your calculations to start?
 
@SjoerdC.deVries No, fortunately we hadn't access to the real time subsystem. They knew us
You know, we were young and single. A real danger for the society.
 
12:56 AM
;-)
There's a chat cast in the teacher's lounge in two minutes. Anything I should bring up there?
 
@SjoerdC.deVries About what?
 
I think it's open, no specific topic today
 
@SjoerdC.deVries Could you pass on a link?
I love bullying teachers
 
It's mod only
 
ohh
no bullying
snif
 
1:04 AM
I'm sorry. Would you like to become a mod?
 
mmm. How much?
 
General admiration
 
Good song.
 
yesssss
 
 
2 hours later…
3:32 AM
Re this off-topic non-answer - should we in addition suggest to him that it could be self-answered question on StackOverflow, just not here?
 
@Verbeia I flagged it
 
4:07 AM
@Verbeia He linked or mentioned his site address twice in his post. I think he is almost spamming
 
4:48 AM
@belisarius Deleted
 
ok!
How was your ivory tower chat?
@SjoerdC.deVries BTW, take a look at this answer
2
A: What does "EquationalLogic`FindCounterexample[]" do?

Matt W-DPretty cool; EquationalLogic`FindCounterexample is still pretty mysterious (at least to me), but I managed to get it to do something more interesting than spit back numbers. In[1]:= EquationalLogic`FindCounterexample[f[a, i] == a, f[a, a] == a] Out[1]:= {{f -> ({{1, 2}, {1, 1}}[[##1]] &...

 
@R.M Done.
 
@Rojo You too
 
@belisarius Messy. Half of it was in a live feed on google+, which didn't work on my iPad. Not much talking. A bit about welcoming new users and some about rudeness on certain sites.
 
R.M
@belisarius Does my code snippet work well for you?
 
4:55 AM
@SjoerdC.deVries We should try to qualify. It will be an honor to be considered in those chats
 
R.M
For me I keep getting an error that the notebook is already open even though I close it :O
but it works nevertheless
 
@R.M Yep! pretty well! a bit of screen flashing, that's all
 
R.M
@SjoerdC.deVries Thanks
@belisarius Yeah, I'm trying to fix that. Just need to set Visible -> False before creating the notebook, but I'll have to insert that into the raw notebook expression. Not sure if it's a good idea
Trying to see if there's another way
 
@R.M It gave me an error the first time I tried. Closed Mma, restarted, and presto!
@R.M I hate making errors while manipulating big expressions (images or whatever) in the output cells.
@R.M We have a new very nice prospect user. See my link above
 
R.M
I saw that answer — very nice!
 
4:59 AM
yep
@R.M errr wait
please evaluate
EquationalLogic`FindCounterexample[f[a, i] == a, f[a, a] == a]
his first piece of code
What does it return?
@SjoerdC.deVries Could you try it?
 
R.M
@belisarius his answer
 
@R.M Damn! It is a version issue
returns 0 here
 
R.M
EquationalLogic`Prove[ForAll[x, Composition[s, k]@k@x == x],
 ForAll[{x, y}, Composition[k, x]@y == x] &&
  ForAll[{f, g, x},
   Composition[s, f]@g@x == Composition[f, g]@Composition[g, x]]]
Code for the second
 
yep I did it already
returns True
But the first is what I want to get to
my version 8.0.0.0.0.0 is outdated
 
R.M
hehe
 
5:14 AM
well, enough for today. Nights!
 
R.M
aha! Setting the definition to $Pre works without errors.
Not fully sure why $PreRead gave errors, but oh well...
 
@R.M Haaaaaaaaa I told you :p
Nights, now for real
 
R.M
bye
 
 
2 hours later…
7:38 AM
This is a duplicate.
 
 
2 hours later…
9:59 AM
Just playing.
 
10:23 AM
@MrW do we really need as a tag? It doesn't seem like it could see much use.
 
11:00 AM
@Szabolcs looks like a chip :)
 
 
1 hour later…
12:15 PM
@Szabolcs Indeed it does look like a computer chip. Are you going to share the code or make us reverse engineer it?
@Szabolcs I'm not sure yet. :^)
 
12:30 PM
Hi @MrW
 
@belisarius unfortunately I can't test at the moment. No mma on my iPad, remember?
 
@Szabolcs Darn, I didn't get the ping sound. I just posted an answer which I think will be helpful. Please take a look. Also, do you want the buttons aligned only right, or right and bottom, to stay in the lower corner?
 
So Row does take the ImageSize option! Why did I think it didn't?
 
@Szabolcs I don't know -- probably because Column doesn't?
 
I only wanted them aligned to the right for now.
Maybe that's the reason, and I was always using Column. Actually I just made something very similar to your solution, but using Pane instead of Row
 
12:41 PM
Is my answer a solution in that case, or does it break in your application?
 
It is a solution, I already upvoted it.
 
Okay.
 
I don't usually accept until a few hours have passed, but it is a solution.
 
@Szabolcs Wouldn't expect you to! I'd give it 24 hours myself.
 
R.M
1:07 PM
@Mr.Wizard What does envelope mean in that context?
 
@R.M It means "label".
 
R.M
no, it does not... even if it were applicable here (and would be an abuse of the word), it would have to cover the jar completely
 
@R.M In graphic design this kind of transform is known as an "Envelope Warp" or "Envelope Distort." I was seeking to make this Q&A easier to find by searching.
 
R.M
Umm... not quite the same, really. Merely adding won't help — see for yourself. I don't think the tag is useful... Maybe you can use the term 'envelope warp' in the body, if you feel that strongly about it
 
@R.M I don't feel strongly about it.
I'll get rid of from the jar label question, but I actually think it should be on this one now that you direct my attention to it. What do you think?
 
R.M
1:27 PM
It might be better on that, but still am a little skeptical about getting too specific with tags. Envelope by itself doesn't mean much and is not a "topic", so to say. It is also uncommon enough that a simple search for the term gives you all related answers
Even sites that have some envelope related questions (math, dsp) don't have a tag for it.
 
@R.M What is the reasoning behind a minimum number of questions for a tag? It seems to me that even if there are only a handful of questions the tag still serves to organize, and it doesn't really get in the way that I can see.
 
R.M
I think a tag vaporizes after 3 months or so if it has less than X questions in it... I think X is 5
 
@R.M In that case I don't see the problem with these rare tags: if they fail to reach critical mass in that time frame the will be removed automatically.
 
R.M
It also increases the chances of arbitrary tagging... Every new tag is an opportunity for someone to use only that tag. If people begin browsing by tag, like we did on SO, it is easy to miss the small ones
I don't think that should be a justification for using it... I mean, do you think is a useful tag for Szabolcs' recent question? Sure it fits the question, but what purpose does it serve our site?
 
I think that would be taking it to the absurd extreme, but just to argue the point, suppose we did. If in three months no other hedcut questions are posted it cleans itself, but if for some odd reason there are five questions posted nobody has to go back and find this one to tag it, or have it get lost or whatever.
Of course one must draw the line somewhere but "envelope" describes a number of things I believe, including e.g. convex hulls.
 
R.M
1:39 PM
Ok, I just checked and X=1 in my comment above, not 5. Also it is 6 months
I wish I could point to some of the arbitrary and over specific tags that people add and I remove them in my editing. But there is no way to search for them and I'm sure those tags don't exist anymore, so nothing to see. But yes, it used to happen a lot. Maybe a little less now that the threshold has slightly increased
 
@R.M In that case orphan tags with only one question are a problem. What would you suggest as a reasonable minimum? I think three questions is reasonable, if there is plausible anticipation of growth.
 
R.M
Well, I tend to think of tags as being for some reasonable, but not over specific grouping. In most cases, tags should stand alone, because if some chap comes by and uses that tag alone and not one of the more common tags, it should still be clear what the intent is (of course, someone will most probably fix it by then).
It's hard to keep track of tags if we get too specific... in the end, each one creates a specific tag his way. Like for example, you'd use for convex hull and someone else would use , yet you can't make them synonyms because envelope has other meanings
In any case, I'm done on this topic for now... gotta run. Do what you think is best.
In other news, temporary #3! The last 24 hours has been a crazy round of musical chairs between Szabolcs, Leonid and myself
I'm coming for you @Mr.Wizard!
 
2:07 PM
@R.M I don't doubt it. You've been quite busy!
 
 
5 hours later…
7:08 PM
This one is a prequel of my blog post
1
A: parameter optimization

belisariusFor solving ODEs like yours for a variable number of functions, you could do something like: dims = 3; k = RandomInteger[{-2, 2}, {dims, dims}]; DSolve[Table[ D[f[i, x], x] == Table[f[j, x], {j, dims}].k[[i]] && (D[f[i, x], x] /. x ->...

@SjoerdC.deVries Hi!
 
@belisarius When do you expect you'll be ready?
 
@Szabolcs Don't know. Still polishing the code
 
I don't want to publish anything for two weeks, maybe longer.
But if you are ready, or near ready, please let me know. I need to know how many posts to expect.
I would like to keep a (semi-)regular posting schedule.
 
@Szabolcs Sorry, I am not able to take compromises. My workload is chaotic
 
@belisarius That's why I want to publish at regular intervals, and not right away. I don't want to push anyone to finish by a deadline. Instead I want to put finished post in a queue and publish on schedule. Originally I thought of starting with a one-month interval, to make sure that there will be enough posts, but people thought that's too long. 2 weeks is an absolute lower bound though---I'm not sure we'll have enough posts in the long term to keep a 2 week schedule.
 
7:19 PM
@Szabolcs We could also keep a think tank for themes, so anyone wanting to write, but lacking inspiration could grab one from there
 
@belisarius Yes, good idea. Feel free to post ideas in the meta thread even if you don't plan to write them yourself. I am hoping to do a series on making pretty pictures, like the ones Jin linked to
 
@Szabolcs Yep., Those are nice
@Szabolcs Will we have tagging for a minimal recommended reader level on the posts?
In the blog posts, I mean
 
7:44 PM
I'm not sure it's necessary, but if people find it convenient, we could do that
Let's have a few more posts first to see
I need to log off now, getting married tomorrow.
11
 
@Szabolcs WOW ... big moment! Congrats!
 
 
1 hour later…
9:17 PM
@Szabolcs This really is a year of change for you. Congratulations. hope you have a nice day tomorrow (and beyond that of course).
 
acl
@Szabolcs congratulations!
 
10:05 PM
@SjoerdC.deVries Pin that starred saying from @Sza!
 
R.M
10:29 PM
@Szabolcs congrats!
 
@R.M No toads at the party
What could be a great question to celebrate Szabolc's marriage? I thought of a few, but none of them will survive any reasonable good taste filter
 
R.M
Is there mma at the wedding?
 
How do you think he designed the cake?
 
R.M
211
Q: What code would you have on your wedding cake?

SembianceI'm getting married on April 4th, 2009. My future wife has agreed to let me have actual code on top of the wedding cake. Woot! There isn't very much room, four, maybe five lines (it's only a 6" cake because the main cake is a bunch of cupcakes). Any language is okay, Java, JavaScript, XML. A l...

 
@R.M That would be a nice question. With a variant. Draw a wed cake in Mma. One that Sza deserves
 
R.M
10:36 PM
Vote to close!
 
of course
Only After Heike's answer
Or Verbeia's
Girls are talented
 
R.M
I can't eat it, so no deal.
There's also this: stackoverflow.com/questions/2420689, although it is for a birthday cake...
 
I answered a question about a present on SO. Let me search for it
Ohhh
I was lazy
solved it without programing
6
A: A code that is a word forwards & a different word backwards

belisariusI tried to transform FESTUNGDRESDEN into MARIA. I found a possible encoding that does not satisfy all the specified conditions, since one of the letters needs more than 10 symbols. A "manual" procedure: As both words share only one letter (the "R") I broke up both words as follows --------...

Won the bounty anyway
I wonder if the site's design is cake-compatible
 

« first day (191 days earlier)      last day (4281 days later) »