Catch up on stories from the past week (and beyond) at the Slashdot story archive


Forgot your password?

Comment Re:The algorithm isn't clever, but scales well. (Score 2) 81

Especially for DNA comparisation, this shortcut won't work. Moreso, it is antithetic to the task at hand. We want to know the edit distance, or in DNA speak, we want to know how many mutation we need to get from DNA1 to DNA2, for instance to determine relation. The shortcut that we just look if all letters are present and begin and end are ok, and don't care about their sequence in the middle of the word doesn't give us the edit distance, as it hides exactly the answer to the question we asked to begin with. It's ok for determining which words are there, and it speeds up reading, but only if all sequences of letters are actually words we know perfectly well. It would slow down the reading to a crawl if our task was for instance to read a text in a foreign language and translate it with the help of a dictionary. If we then stumble upon scrambled words, it will take ages until we figure out the right sequence of letters to look the word up in the dictionary.

Comment Re:In the comments below the interview... (Score 1) 90

It works in this case, because everyone knows what you are modelling: You are using the english language analog to the language of Westeros, which makes sense for english listeners, because they already associate a scotch accent with "north", and a spanish accent with "south". As the whole series is in english anyway, this works. With this choice, you can control the context, because what you are suggesting with the variation of the accent is equivalent to the setting in the fantasy world.

But using a dying language in a movie just because it's a dying language and you want to preserve it makes no sense in most cases, except you find a language environment, that is a) somehow historically or locally connected to english as the main language used in the movies and b) carries with it all the connotations you want to use. You could for instance have the people of the Southwest of Westeros talk Cornish (Kernowek), if they should have their own language and if it should be distinct from the main language. But using for instance one of the languages once spoken in Tierra del Fuego does not make sense, as the connections to English are so different from anything we get told about Westeros.

Comment In the comments below the interview... (Score 5, Insightful) 90

In the comments below the interview, there were several comments along the line that Hollywood should not pay someone to invent a new language, but rather revive one of the many languages on the verge of extinction. One answer of one commenter was that exactly those people who invent languages for fun and for a living are also exactly those linguists who preserve those languages destined for extinction.

I would add a second thought: First, it doesn't make sense to have an invented place speak a real language in lieu of an invented one. It just creates a confusing context. Lets say the people of the eastern regions in Game of Thrones would speak a language like the Mansi language. It would somehow place Westernesse into the Ural mountains as Mansi is spoken east of the Ural. People from West Siberia, who might not speak Mansi, but recognize the sound of it would always be somehow reminded of their home land instead of being immersed in a phantasy world, and the Mansi people then would then wonder if the people of Westeros should somehow be identified with the Russians, and why there is no Khanty language (a neighboring language both locally and linguistically) in the series.

Chosing a language always sets a context, and if you want to control the context, you can't chose languages at will.

Comment Re: No Way ! (Score 1) 110

I have an issue with the notion that something should be allowed because it is possible. Yes, it is possible to collect the data about my buying habits. But should it be allowed? It is also possible for me to ram a knife into your chest, but should that be allowed?

For a long time, we didn't think about the consequences of wholesale collection of all available data of people, because the sheer amount of data meant, that it wasn't done for all people in the most complete manner. There were specialized professions which collected as much data as possible, but only for a certain subset of people: tabloid journalists for celebrities, spies for high profile persons in politics, investigators for people accused of a crime and private eyes for targets the paying customer named. And we thought, that those limits somehow made sense, as most of us are neither celebrities, high profile persons in politics, accused of crimes or in an enduring conflict that makes one side willing to pay large sums of money to private investigators. And we somehow felt, that those persons are special, and thus permitted special care, and we told ourselves, that those persons deserved it because of their life choices.

But with the availability of data storing and processing, we all are now in the role of celebrities, accused, politicians or people in deep conflicts. Suddenly it's not a choice anymore to get your privacy constantly violated. It happens to all of us. And now we see that the old idea, that if something is in the public view, it belongs to the public, and it should be allowed to indiscriminitely record, store and process it, is working against us.

About 300 years ago, we understood that for special types of data, it shouldn't be the case. Works of Art were in the public view, but it was explicitely forbidden in the Statute of Anne in 1710 to record, store and process them without explicit permission, because mostly everyone understood that those works somehow incorporate a value.Now with the advent of big data, we also see that other data, which are not Works of Art, still have an intrinsic value, and if everyone is allowed to siphon it from us, we lose, and someone else who is not us, gains, albeit it's us who created the value in the first place.

Comment Re: there is no (Score 4, Informative) 402

The full-coverage satellite data show continued warming in the last 20 years. I don't know which data you are looking at, but of the 20 year period you are talking about, nine out of the ten warmest years are after 2005, and 2015 might set a new all time record. If it wasn't for the extreme outlier 1998, the warming in the last 20 years would have been nearly linear. Actually, one of the Anti-AGW propaganda tricks is to take an ok sounding period like 10 years, 15 years, 20 years, in which 1998 is close to the start. Until 2008, it was always "the last decade doesn't show warming", after that it was "the last 15 years", and now, since 2013 and thus more than 15 years since 1998 are over, it's obviously 20 years. But for some reason, 2013, 2014 and probably also 2015 were pretty hot years globally, and thus the graph, that looked so convincingly anti-warming in 2008 was mildly constant until 2013, but since then, even 1998 does no longer help the argument, as even the graphs that take 1998 as their starting point have a rising trend.

Comment Re:it just makes so much sense (Score 1) 202

It's not totally false. The only reason why the companies could transfer personal data (that is data which enables one to identify an individual) to the U.S. was the Safe Harbour provision, which basicly stated that the EU Commission trusted the U.S. to have similar data protection schemes in place. And exactly this is now called in question. If the EU High Court follows the opinion of the Attorney General, then no personal data is allowed anymore to be transferred out of the protection of EU law. And that means everything, starting with e-mails sent from EU citizens to other EU citizens, from personal profiles on social media plattforms (except they get individually requested by U.S. residents, then they are allowed to be sent, but have to be erased on U.S. soil as soon as the U.S. resident is no longer looking at them), includes cloud data of European customers, up to health care data processed abroad. It means that most business plans which include storing or processing data in U.S. facilities are now called in question, as the provider of the storing and processing has to actually prove first that he follows EU law as long as he is processing EU data. Until now he just could ask the U.S. to certify him without any EU institution doubting the certification.

Comment Re:The US needs a serious spanking (Score 4, Interesting) 202

Actually, what you're saying is false. The case stems from a complaint brought by an Austrian citizen against Facebook. That constitutes bringing a claim against a US organization. The Irish Data Protection Commissioner ruled against the claim, so it was appealed to the European Court of Justice.

That might be the origin, but that's not the question the Attorney General issued an opinion about. The Irish High Court refused to hear a claim about Facebook's transfer of data and referred to the Safe Harbour treaty, and thus there was the question to the European High Court if the Irish High Court has to consider if the actual conditions of the Safe Harbour provisions were met. And now the Attorney General opinionated that the U.S. does not met the the regulations for the Safe Harbour treaty because European citizens have no legal recourse against the wholesale spying performed by the NSA and other organisations in the U.S.. The NSA spying was called unspecific and without clear goal and thus in violation of the privacy rights of European citizens. This is the first time that an European institution actually took the Snowden allegations as fact (until now most courts opinionated that the allegations have not been proven yet), and thus issued a statement that the U.S. is not adhering to the provisions and thus the European Commission can't declare the U.S. a safe harbour, thus all treaties with the U.S. about data protections are void.

Comment Re:graphical Harvard museum effort not available (Score 1) 72

The age is quite undisputed, but the actual nature of the fossils is unclear: Are they really remainings of multicellular life, are they just some strange type of bacterial colonies, ore are are they abiotic anyway and thus don't belong in the TOL at all?

I've been at the exhibition in the Naturhistorisches Museum Wien, and I've seen the fossils (or at least quite convincing replicas of them), and they looked like some type of small, round pillows with large braids. So to a non-palaeontologist like me, they looked like some type of seabed dwelling animals.

Comment Re: What? "We're sorry we got caught"? (Score 2) 301

I have a four cylinder 1.6l TDI in my car (it's a Skoda, an appearently Europe-only subdivision of VW), and I sometimes get below 4 liter/100 km, which is about 60 mpg(US). And this is normal driving to a customer site about 60 km from my company. So yes, if a sticker advertised 60 mpg (actually, the sticker advertised less than 99 g/km carbon dioxide, which is about the same), I could agree in real world driving.

And yes, it's easily possible to drive the fuel consumption up to 10 l/100 km (about 25 mpg) by inapprobriate driving.

Can anything be sadder than work left unfinished? Yes, work never begun.