Obama Knows His Computer Science

Obama fields a question intended for a Google interviewee and gives a surprisingly knowledgeable answer.
8266says...

If its not a total duplicate why even complain about it ? A short clip of an interview is totally differant then the whole thing. It obviously highlights something very interesting. Boo the dupe fairys.

MarineGunrocksays...

Sorry. I didn't take the time to check the dates.

>> ^DrPawn:
If its not a total duplicate why even complain about it ? A short clip of an interview is totally differant then the whole thing. It obviously highlights something very interesting. Boo the dupe fairys.


Because a dupe is a dupe is a fucking dupe. Just because it's a shorter clip doesn't make the other content disappear.

jonnysays...

DrPawn, bamdrew -

from the FAQ [emphasis added]:

Duplicates will be *discarded. A duplicate video is one which contains content already on VideoSift in a published, queued, personal queued, or dead video submission. Minor changes in content, like a few additional insignificant seconds of video or alternate background music, will still be considered dupes.
If you want the rule changed to make an exception for short bits from existing long posts, then post to Sift Talk to get some discussion going. But please don't flaunt this rule, as it will inevitably lead to this kind of fiasco.

hixsonjsays...

Sorry guys. Didn't mean to post this as a "fucking" dupe. Just thought it was a funny clip. Some people might not want to watch a 20 minute interview and would just rather see this funny bit.

Kruposays...

Flagging this for an administrative un-discard command. I would just burn a promote on it but I don't have one available right now.

Video is not a dupe because it was published on Feb 15th (the other vid was published on Feb 16th).

Goldies, please be careful when firing off discards, thanks.

jonnysays...

My bad folks - I did not intend to discard this. I didn't realize a blockquoted invocation would get carried out, nor did I even particularly notice the '*' in the quote. I'll promote this back tomorrow when I have the power point available again.

Krupo - if that is the case, then it needs to be added to the FAQ.

[edit] bamdrew, re: your example - full episodes of conan shouldn't be getting posted anyway. The general point is well taken though.

8266says...

Fine dupe fairys are all over the place. I dont agree this is a dupe as defined by the rule.

I watched both vidoes. And the first is about a 15 minute bore fest.

Dupe fairys, I respectfully disagree with why this video has to be deleted. I see your point that it violates the rules. I just differ in the interpretation of duplicate content when this video clearly brings something "new" even if the new is only it highlights Obamas computer literacy.

MarineGunrocksays...

Pawn, you're missing the point: It's not a dupe because it was posted first.

And don't be immature and get all YouTube on us by calling us "dupe fairies" just because we like to adhere to the rules. It's what sets this site above the rest.

lertadsays...

Bubble sort is a pretty basic form of algorithm, so it shouldn't really be too surprising he knows about it as long as he's shown at least some kind of interest on the related subjects.

Basically what I'm trying to say is it's impressive but not the so-impressive-he-must-have-been-told-ahead-of-time impressive.

Kruposays...

>> ^jonny:
My bad folks - I did not intend to discard this. I didn't realize a blockquoted invocation would get carried out, nor did I even particularly notice the ' ' in the quote. I'll promote this back tomorrow when I have the power point available again.
Krupo - if that is the case, then it needs to be added to the FAQ.
[edit] bamdrew, re: your example - full episodes of conan shouldn't be getting posted anyway. The general point is well taken though.


LOL, I didn't realize the discard was in the quote. Yeah, treat the * like a live bullet around all words at all times. Sorry for harshing on you.

Regarding the exception, it's a Sift Convention. As we're all aware, laws and rules can either be handed down as a formal written policy, or in something more akin to a tradition.

The Unitard Convention or more of the latter, and the reasoning behind it is buried in the Sift's comment database. We can always google is out if we really need it.

Thanks for the fix dag.

dgandhisays...

It is a cute answer, but not indicative of much computer knowledge.

What do you all think would be the best answer?

Off the top, without knowing the answer to "efficient how?" or "sorted how?" I vote R-B tree.

On the other hand if "efficient" means minimum memory usage, you would probably have to go bubble sort.

HaricotVertsays...

There is never a good reason to do a Bubble Sort. The only reason anyone ever learns it is because it is used as an "easy" introduction to writing algorithms in Computer Science 101 courses. Bubble Sort just happens to be a brute-force method that mimics the way some people would manually sort a random list of integers given unlimited time and paper. At the very least, for a O(n^2) algorithm, you can go with Insertion Sort which is guaranteed to perform faster or equal given the same data set, with the same amount of memory usage.

As for the actual algorithm that would be used, Quicksort is probably the most obvious choice, although they did not specify what their sorting goal was, so "It depends." is probably a fine answer.

dgandhisays...

>> ^HaricotVert:
There is never a good reason to do a Bubble Sort. The only reason anyone ever learns it is because it is used as an "easy" introduction to writing algorithms in Computer Science 101 courses.


Bubblesort is not always the wrong answer. For instance if "efficient" means using only 32Million bits RAM, two 32bit registers (plus the hacky use of an execution counter), and a few dozen lines of assembly, you can't really implement anything else. The answer is based on the constraints.

That being said, now that I think about it, loading and dumping an R-B tree is logically equivalent to quicksort, which does seem like the best general solution.

8266says...

The question about how to sort a million 3 digit integers is interesting and misleading.

The best way is not to sort them at all but count them!

So you could store the entire sorted data set in one object oriented variable, lets call it dataset.

dataset.999=25
dataset.998=22
dataset.997=18
....

This means that there were 25 3 digit intergers that were equal to 999 and 22 that were equal to 998.

The dataset would require only a 1000 integers to store.

rychansays...

DrPawn is right. The big O for sorting algorithms is typically calculated for real numbers. If you're using integers, especially a small range of integers, just count them in linear time.

dgandhisays...

>> ^DrPawn:
The question about how to sort a million 3 digit integers is interesting and misleading.


I think he said 32bit integers, not 3 digit integers. Radix sort, or in the absence of associated data, simple counting, as you suggest, would be very bigO efficient in that case.

vairetubesays...

i read all this stuff and links about duping, then i finally watched this video... it's just a set-up joke and typically executed.

grin @ "dupe fairies" and related funniness.

Send this Article to a Friend



Separate multiple emails with a comma (,); limit 5 recipients






Your email has been sent successfully!

Manage this Video in Your Playlists




notify when someone comments
X

This website uses cookies.

This website uses cookies to improve user experience. By using this website you consent to all cookies in accordance with our Privacy Policy.

I agree
  
Learn More