The Whiteboard

user55340

12:22 AM

@GlenH7 that prefix/suffix problem is kind of interesting to look at for a little bit. You make the sets for all the prefix and suffix forms (and identify the set equivalencies - number of ways to do this depending on language as the sets are being built - things like java.util.concurrent.CopyOnWriteArraySet could be fun to play with there).

user55340

And then just go through and identify the pairs. The neat part with that copy on write set would be (I think) that foo.add(1); foo.add(1); won't create a new set on the second call. Thus, you can blindly add and assign... man, you could do some wonderfuly odd code with that.

user55340

The set itself as the key to another set (where the sets are the same)...

user55340

For [3, 5, 7, 3, 3, 5], You could have P:{3,5} => {3,4} and P: {3,5,7} => {0,1,2}.

user55340

Hmm, I wonder if hashcode() for the two is the same... I bet it is... could even use the same structor for both... and get your answer from it.

user55340

(I had the P and S mixed up previously...)

user55340

12:30 AM

{3,5} => { P => {0,1}, S => {3,4,5}} ... and then for each key in the hash, you multiply the size of the P set by the size of the S set and sum it up and you've got your answer.

user55340

Ok... that was kind of fun. Time for some food.

user55340

Thinking about it even more before I get food, you don't need the copy on write set if you are reasonably confident on the hashcode in Java on the HashSet to start with (also store the number of values in the set and it should be resolved very hard to get a collision). Because then you wouldn't need to store the the set itself, just its hashcode and you wouldn't need to make copies of the structure for each distinct add.

user55340

(for any wondering what I'm babbling about... )

user55340

0

Q: help in theory regarding prefix_suffix_set

I found this question at codility.com, but I don't understand the question. Can you help me figure out what do they want? I am more interested in the theory than a code. A non-empty zero-indexed array A consisting of N integers is given. A prefix_suffix_set is a pair of indices (P, S) such that ...

algorithms

user20683

it's an interesting problem to be sure from a toy standpoint

user20683

12:38 AM

curious as to the real world benefits

user55340

1:46 AM

@WorldEngineer Probably few, its a coder challenge. There may be some similar problem in genetics or various forms of substring matching... but in general, its a mental "can you figure out the problem that you may not have seen before, and design a solution that scales.

user20683

@MichaelT It shows up heavily in Automata proofs as I recall

user55340

The neat part with the hashcode approach is that it can be done in O(3N).

user55340

One pass for prefix, one pass for suffix, one pass through the resulting structure.

user55340

Something similar in the NDFA - DFA conversion if I recall correctly (which then has application for writing a regex to DFA compiler).

user20683

2:01 AM

something like that

user20683

I seem to recall it has to do with how to use lambda to prove that there is a midpoint where you can determine palindromes or some such

user20683

being critical to determining if your braces are balanced

user55340

2:27 AM

user55340

Had to do that before someone did something like upvote me before I got such a beautiful number...

user55340

I mean, its not like I upvoted someone from 1337 rep to 1347... oh wait, I did.

user20683

so you went from Leet to the Plague Year

user55340

Its a bit before I get to 31337 rep.

user20683

yeah

user20683

2:29 AM

14159

user20683

that's the next geek one

user55340

-9

Q: Lock your visible score

I just noticed that my P.SE score is 7337. While not as awesome as 1337, it got me to thinking that it would be nice to be able to lock your P.SE score at a certain value. Either discarding upvotes or having a private score that you can set public once you get over the novelty of being "1337". I ...

feature-request status-declined

user55340

So, the best I can do is take a screen shot of it.

user20683

aye

user55340

There are times I want a "Close - Off topic - WTF" reason.

user55340

2:40 AM

15

Q: Does good physical shape make peers underestimate you as a programmer?

It's no secret that programming is a profession that involves few physical activities. Brute force as well is not used (normally) by programmers to solve problems they encounter at work. No wonder that some programmers disregard keeping their bodies in shape. And no wonder that treatment of fa...

office social

user20683

@MichaelT Nuked as Off-topic: Career Advice

Jimmy Hoffa

Least we don't see colorful trolls like

Jim Gay

1

breaker-box brick electrical floor flooring foundation level

user55340

And a flagging I go...

user55340

And you do need more than a 70 amp pannel.

Jimmy Hoffa

I know, but it is more, my house inspector got that part wrong

Which doesn't inspire confidence

user55340

2:50 AM

Installing a subpanel... thats were the fun is.

Jimmy Hoffa

I'm just hoping the current panel can be reorganized to make it ok enough for the AC

user55340

Looking at it... I'd go with a subpanel.

user55340

Are you talking about central air or a window unit?

Jimmy Hoffa

@MichaelT central air. a subpanel would be limited to 70 amps and still gives me no extra amperage over the 125. I probably need to shell out the $1400 and get upgraded to a 150 panel.

user55340

Upgrading the panel is probably a good thing. The subpanel I thought was 100 amps... and its not going to change the service to the house, but rather where you can put a new circuit.

user55340

3:05 AM

Central air will draw between 15 and 20 amps (guestimate based on other things I've seen).

user55340

So if you put it on one of the other 20 amp circuits, you'll be tripping it all the time with other things.

user55340

And the 15 amp probably won't do it.

user55340

I'm not sure what you've got on the 50s or the 30... but they probably wouldn't take too kindly to putting an AC uint on there.

user55340

At which point it becomes things like "Do you want to put take a bedroom and put it on the same circuit as another bedroom where they both might fit to free up a 20... though I'm not sure how practical that would be.

Jimmy Hoffa

@MichaelT they'd be adding an additional 20 breaker for the AC, the question being is there enough juice to support it... at least asking that on DIY made me feel comfortable about one thing, I won't overamp my box because while I'm lacking a main breaker the way it's wired right now the 50s and 30 are all acting as individual main breakers supplying the spaces below them. Another 20 breaker beneath any of those 3 spots may involve me tripping it and needing a new panel, but it won't overamp

well..*almost* won't overamp it, it's a 125 amp panel that's apparently pulling up to 130amps before all the breakers will flip.

but most likely one of the 50s or the 30 would flip independently before all 3 of them maxed and floated in that 5 amp overage space not flipping

The guy told me there's enough room for it, so I figure I'll let them do that and when I find the breakers snapping I'll just upgrade the panel, and if it turns out they're fine, no big deal. It's not a huge house and doesn't run much electricity anyway. Gas range, gas hot water heater, gas furnace, the only big electric sinks are the oven (not stove), the TV, and? Wife's hair dryer I guess? Otherwise electricity goes to laptop (that's pretty low amp, no?) and lights (CFLs)

user20683

3:23 AM

@MichaelT @JimmyHoffa Here is the perfect board game to go along with your discussion:

user20683

bullypulpitgames.com/games/fiasco

Jimmy Hoffa

@WorldEngineer Good point, I'm buying a fire extinguisher...

user41796

@WorldEngineer And you haven't deleted that question why?!?!

user20683

@GlenH7 because talking with girlfriend and forgot

user41796

ha! then I get a flag for it! :-)

user41796

3:33 AM

better yet, decline the flag so I can create an epic MSO rant and complain bitterly about how the diamonds are out to get me and it better. stop. now. dangnabbit.

user20683

@GlenH7 I was nice and didn't decline

user41796

Thanks for not calling my bluff since I don't have the creative juices to come up with a rant that good.

Jimmy Hoffa

@GlenH7 Just to be clear, the diamonds are out to get you.

user41796

@JimmyHoffa some, not all.

Jimmy Hoffa

(Why do you think Yannis has been missing for a while? @WorldEngineer is just acting decoy while Yannis implements the plan)

user41796

3:35 AM

(don't burst my bubble)

user20683

@JimmyHoffa Decoy?

user41796

@WorldEngineer - he's not in on the story. Doesn't realize your account is really a sock puppet for Yannis

user41796

What I never got is why you wanted a sock poppet that was also a diamond. Never quite made sense to me.

user41796

minor rant about audit reviews. I'm a bit torqued that I had 3 or 4 audit reviews tonight while going through the close queue. I maxed out with 5 left to be reviewed still in the queu.

user20683

@GlenH7 I am not a sock.

user20683

3:40 AM

I am a shoe.

user41796

I apologize if I have offended you. Not sure I've ever heard of a shoe puppet though.

Jimmy Hoffa

@GlenH7 They were obsoleted by the invention of socks.

gnat

7:38 AM

24

A: Let moderators navigate the review queues at will

I don't think this needs to be so convenient as to require buttons on each review item, but a list - similar to the existing review history list - of pending review tasks per-queue would probably come in handy for moderators now and then. After considerable discussion with Anna (who championed ...

new (not very new but still) toy for moderators and 10Kers, wow

CV reviews history: programmers.stackexchange.com/review/close/history

LQ reviews history programmers.stackexchange.com/review/low-quality-posts/history

...and the first WTF along the way:

LQ review: programmers.stackexchange.com/review/low-quality-posts/38725

5 voters recommend deletion, 1 says looks good, post is kept staying, WTF?

makes this queue look pretty useless

another one, 4:1, keeps staying, until likely flagged for moderator to delete: programmers.stackexchange.com/review/low-quality-posts/36820

teethless

user41796

11:00 AM

@gnat Shows as deleted on Sept 1.

user41796

@gnat It's not quite as bad as the review would have it appear. That was the FGITW answer on that question, and it's a common response to many in a troubleshooting scenario. I think the "looks good" reviewer is trying to explain how to improve that answer as opposed to just deleting it. LQP really should have another option to mark as bad but not delete worthy. Disclosure: I also put an answer on that question.

rwong

11:58 AM

opencv.org/survey.html

To maintain its high quality, OpenCV is looking into crowdfunding its development. Please, fill out that short survey to help us know what you guys need. Click on !

user20683

12:36 PM

@rwong might go leave that in DSP's chat

Transcript for