Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!


Forgot your password?
DEAL: For $25 - Add A Second Phone Number To Your Smartphone for life! Use promo code SLASHDOT25. Also, Slashdot's Facebook page has a chat bot now. Message it for stories and more. Check out the new SourceForge HTML5 Internet speed test! ×

Comment Re:Likelihood by race (Score 1) 209

Playing with percentages can get easily confused which is what I suspect happened to you because your math is way off.


Percentages are handy for comparison, but can be easily confused or used to conflate facts.


... a white man is several times more likely to be shot by a black man, than the other way around

Thus, 1/1,046,032 whites was an interracial murderer. A black person statistically has to meet over 1M whites to run into an interracial murderer (whitey is pretty safe these days, apparently). ...1/92,176 blacks was an interracial murderer. A white person has to meet 92K random blacks to statistically meet an interracial murderer. ... 1/32,608 black men was an interracial murderer. Thus, any random young black man is 240% more likely to murder a white than any random white person is to murder another white person and any random young black man is 20,900% more likely to murder a black person than any random white person is to murder a black person.

You are simply reading the GPs text differently. It hinges on the meaning of the phrase "likely to be shot by a black man", where one can read that as either "likely, in the course of daily life, to be shot by a black man" (my interpretation) or "likely, upon seeing a young black man, to be shot by him" (your interpretation, and possibly the GPs).

Comment Likelihood by race (Score 1) 209

By the way, a white man is several times more likely to be shot by a black man, than the other way around. It's not some 20-30% difference. It's several hundred percent more likely.

This is getting really off-topic, but that statement is ridiculously far off. Murder is quite well traced by the FBI, so let's take it as a proxy for shootings. The FBI has this nice table of 2013 statistics (other years would be broadly similar).

From the table we can see that there were 409 murders of whites by blacks. With a white population of roughly 200 million, that makes about 2 parts per million. We also see 189 murders of blacks by whites. With a black population of about 40 million, that makes just under 5 parts per million.

From this we see that a black person is about 2.5x more likely to be murdered by a white person than the other way around. That's the opposite of what you said.

Now, it is true that if you restrict to the cohort of actual crime victims, things look different. For example, given that a white person is one of the 3,000 murder victims , chances are about 14% that the murderer was black. In comparison, given that a black person was one of the 2,500 victims, chances are about 7.5% that the murderer was white.

Even viewing the statistics this way, we do not reach "several hundred percent", but rather 80%.

Comment 2010 book by Kim Young-Cheol (Score 1) 38

There's a book titled (roughly) "Think Samsung" that was published in 2010 (link). It's said to give a disturbing picture of Samsung's corruption, and was even reviewed in The New York Times. It was written by Samsung's former chief legal counsel.

In his book, Mr. Kim depicts Mr. Lee and “vassal” executives at Samsung as bribing thieves who “lord over” the country, its government and media. He portrays prosecutors as opportunists who are ruthless to those they regard as “dead” powers, like a former president, but subservient to and afraid of Samsung, which he calls the “power that never dies.”

Comment Re:Why is Slashdot anti-trade? (Score 4, Interesting) 158

Don't forget the $50 Billion in job losses to offset the $25 B in gains. We tried this crap with NAFTA etc and it only benefits the rich.

People who actually have studied this and know something about it disagree with you.

I don't blame you, it is an easy mistake to make because benefits are diffuse while costs are concentrated and easy to identify, especially due to the inadequacy (in the USA) of the trade-adjustment assistance program.

Comment Re:Pretty interesting (Score 1) 412

As the others have said, Wikileaks isn't out of play at all.

You're right, they are not necessarily out of play on this. The out-of-play scenario I have in mind (with zero evidence for it, mind you) is that the organization with all this data has been sending it to Assange (or making it available online) encrypted with his public key.

With Assange unable to handle things, they would have to find another part of Wikileaks they trust to share their publication goals. Perhaps that's not so easy.

Comment Pretty interesting (Score 1) 412

With WikiLeaks (likely) out of play, whoever has been sending WikiLeaks the Democrats' data will either have to find another channel for release, or stop releasing. In the former case, that may give intelligence agencies a better idea of their target.

I wonder if there was a back-channel conversation with Ecuador -- something like "Whoever is behind this, Ecuador is effectively acting as an accessory to some outside party attempting to alter the US presidential election. Is that *really* how you want us to treat this?"

Comment Re:pixel (Score 1) 212

Except that in the case of platforms, you may have a lot of money tied into either the "App Store" or "Google Play" for apps/music/video/etc

In addition, there's the pain of migrating calendars and the like. There are export options and tools to assist in that, but it could still get painful I fear. For music, Spotify customers would be fine but I am not sure about Play or iTunes, even where it is just a matter of playlists.

It's hard to tell of course how many people this matters for. In effect the article is telling us fewer than 50% of Note 7 owners are both clueful and worry about migration problems.

Comment Re:Man = 1000 (Score 1) 162

Infinite places does not mean everywhere. This is a common misconception when dealing with infinite sets.

Suppose you have infinite many places, as many as the natural numbers.
You may have infinitely many places numbered by even numbers, while still not have the other, infinitely many, places with odd numbers.

So a more correct translation would be "in many places". But then again, if you are talking about infinite sets, the concept of "many" is also tricky, and leads into questions of set cardinality, aleph numbers, etc. which fortunately is a lot more interesting than the usual Netflix soap operas.

Uh, thanks Lars. Where would we be without iconoclastic pedantry on Slashdot?

Since we're being pedantic, you may wish to learn how to parse the phrase "loosely meaning".

Comment Man = 1000 (Score 1) 162

The Korean "man", which incidentally is actually pronounced with a long "a" to rhyme with the English word "on", is the same as 1,000 but can be translated as "infinite" in many situations. For example, a fountain pen is translated into Korean as a "1,000 year pen" or "man-youn-pil" (see here).

The "bang", which is also pronounced with a long "a" to rhyme with the English word "on", means a "place" or a "room", as others here have noticed.

Thus this word (as with many Korean words) is a portmanteau, in this case loosely meaning "infinite places", which makes the translation to the English word "everywhere" fairly reasonable.

All that said, like most of the posters here, I think this choice of branding is truly hilarious.

Comment Entropy Canary (Score 1) 102

I have considered keeping a "Ransomware canary" around. I'm thinking of, say, a Word .doc file on a network drive. A process on some separate computer then checks its entropy on a regular basis, or on file change notification if available, to make sure file entropy has not grown huge.

The idea fails for local files because (as I recall) the more sophisticated ransomware inserts itself as a filesystem driver. That's a likely problem for some of these researchers' heuristics as well.

(Expanding on something I wrote a while ago)

Slashdot Top Deals

Thus mathematics may be defined as the subject in which we never know what we are talking about, nor whether what we are saying is true. -- Bertrand Russell