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


Forgot your password?
Note: You can take 10% off all Slashdot Deals with coupon code "slashdot10off." ×

Comment Re:Comparison? (Score 1) 235

While many CS papers deal with the math up front, many don't. Most AI papers, for example, don't have rigorous formal math, but rather describe what was done and what the results were.

Exactly. And there are plenty of CS papers in other subfields that take an empirical approach to CS research, too. E.g., analyzing the performances of competing algorithms by running experiments with data (either "real" or simulated) is not uncommon. To say that "CompSci is pure math" and nothing more is simply incorrect.

Comment Re:I wish I could buy GMO seeds (Score 2) 295

We count on them to outcompete native plants (if corn (which was actually from Central America I believe) can't outproduce native prairie grasses in Iowa and Nebraska then we won't have any corn).

I definitely see your larger point, but I think you are confusing "outcompete" with "outproduce".

With regards to invasive species, "outcompete" usually means that the invasive species is able to displace native species from their natural habitats. Kudzu and tamarisk are good examples of invasive species that can outcompete native species. We certainly don't count on our crop plants to be able to do that, and most of them can't. For example, you generally don't see corn or wheat displacing native vegetation in the U.S. without a lot of help from humans. In fact, many crop plants can easily be outcompeted by native weeds without human intervention (e.g., tillage or herbicides).

Comment Re:Actually there is a 34% CHANCE... (Score 1) 385

You are correct that the temperature observations in the four datasets are not all statistically independent. There are a finite number of weather observation stations in the world, so of course there will be overlap in the raw data used to generate the datasets. That's why I described them as "methodologically independent datasets, all derived from raw temperature data from land and ocean surface temperature observations." The State of the Climate report uses similar language. In other words, they might share input data, but the methods used to generate the final datasets (e.g., how to perform data quality control, how to interpolate missing data, etc.) are independent. That was all I meant. So yes, you are correct that the data themselves are not all independent. The semantics are messy and annoying, so I am sorry if what I meant wasn't clear.

ALL the datasets they use in this study are directly controlled by the NOAA. They are each adjusted and calibrated... by the NOAA...

But that's objectively not true. Look at the methods used to generate the datasets. Yes, all four datasets use GHCN, but some of them use other data sources in addition to GHCN, and they don't all use ERSST. They take very different approaches to deciding which stations to include, how to correct for missing data, and so on. So yes, there is overlap in source data, which is probably inevitable if you are trying to compile a global dataset, but the final products certainly do not "come from the NOAA" or even rely exclusively on NOAA data. (Caveat: This is based on my non-expert readings of dataset summaries and descriptions.)

Look, I definitely see your point about the datasets not being statistically independent. That is absolutely correct. But claiming that they are all "directly controlled by the NOAA" and "adjusted and calibrated by the NOAA" comes across as disingenuous. It's probably best just to say that they're not statistically independent and leave it at that.

my bias corrects from 35 to 30. The figures based on the math alone were showing something around 35 to 38 percent. But given that we've had corrections to the models and the figures going on for years and they always correct them DOWN... I personally decide to read the numbers as being slightly lower than cited if only in anticipation of the next correction.

Whatever works for you, I guess, but applying an arbitrary 5% downward adjustment because your gut tells you the numbers might be biased is not very defensible. Unless you have actual evidence that the station readings are biased upwards, or that the datasets are fudging the anomalies upwards, you really have no idea whether a correction is needed, let alone how large of a correction to apply. You could be right -- I don't know, and neither do you. Arbitrarily changing the study results because of a hunch is sketchy, at best. Consider the opposite: Some have argued that the JMA dataset underestimates the true extent of climate warming, but I doubt you'd accept an arbitrary 5% upward correction as a result, and neither would I.

Regardless, I suspect we can both agree that, in the end, the precise probability that 2014 was actually the warmest year isn't all that important. The general trend probably matters more, and no one is disputing that 2014 was one of the hottest five or ten years on record.

Anyway, thanks for the interesting (and civil) discussion.

Comment Re:Actually there is a 34% CHANCE... (Score 1) 385

Independent presumes that they're not all under the same organization's control. ALL the datasets they use in this study are directly controlled by the NOAA. They are each adjusted and calibrated... by the NOAA...

Really? The conclusions of the State of the Climate report are based on analyses of these four datasets: HadCRUT4, JMA, NOAA/NCDC, and NASA/GISS. You are seriously asserting that NOAA directly controls all of these?

As to citations of where I'm getting my information from, I read some stuff from this guy:

That's fine, but it still doesn't support your claim that the actual probability is 34% or that there are "estimations as low as 30 percent". The guy in your link estimated the probability at 47% using the NOAA/NCDC dataset, and at 39% using the NASA/GISS dataset. So again, where does your 34% estimate (or 30% estimate) come from? I am not claiming you are wrong; I'd just like to be able to evaluate these alternative studies for myself.

In the mean time, I also reread the relevant portion of the State of the Climate report, and it appears that even though their overall conclusions are based on all four datasets, the 48% estimate might have been based on the NOAA/NCDC data alone. And it certainly is possible that particular dataset gives a higher estimate than any of the others. Having not examined the data myself, I don't know. That makes me even more curious to read the details behind the 34% and 30% numbers that you cite. I suspect that the JMA dataset might lead to a lower estimate, since it appears to show a less steep warming trend over the last few years than the other datasets.

That being the case... I just have to take their predictions with a grain of salt and correct them down a little bit on the assumption that figures are going to be a little inflated.

Wait a minute -- so now you are saying your probabilities are a result of an arbitrary "correction" you applied because you assume that the reported numbers are "a little inflated"?

Comment Re:Actually there is a 34% CHANCE... (Score 1) 385

The probabilities are based on which dataset you use. The dataset cited outputs that number.

What do you mean by "the dataset cited"? According to the methods description, the conclusions of the State of the Climate report are based on analyses of four methodologically independent datasets, all derived from raw temperature data from land and ocean surface temperature observations. So the 48% calculation comes from a combined analysis of four datasets.

There are a couple others and some of them have estimations as low as 30 percent. I was splitting the difference given that the number cited by the NOAA the outlier.

What are the other datasets you are using that the State of the Climate report didn't account for? Since you are suggesting that the 48% result is due to cherry picking (you call it "the outlier") and not representative of the "true" probability, I'd like to see the study (or studies) that arrived at an estimate of 30% so I can evaluate it (or them) for myself. As I said before, of the six links you provided, five are not relevant and the one that is relevant gives the same conclusion as the State of the Climate report and does not agree with your much lower estimate.

Comment Re:Actually there is a 34% CHANCE... (Score 1) 385

Thanks for the links, but you didn't answer my question. I understand perfectly well that there is estimation uncertainty in the "hottest year" rankings. That wasn't my question.

To remind you, here's what I asked: "That's not what the report says, so I'd be interested to know how you arrived at that conclusion. (I.e., how you calculated the probabilities.)"

The report states that the probability that 2014 was the hottest year on record is 48%. You said the probability is 34%. Therefore, you are arguing that the probability estimate in the report is substantially wrong. I assume (or at least hope) that means you either have a reputable source for your claim or you calculated the alternative probability yourself. Either way, I'd like to know more.

Unfortunately, the first five links you provided don't say anything about this, and the last link actually says that you are wrong -- it also gives the probability as 48%. So your own source disagrees with you.

So... maybe a 30 percent chance of being the hottest year in the last 20~30 years?... sounds about right.

Well, according to the report and your own reference, that doesn't "sound about right." They both say it's about a 50% chance that 2014 was the hottest year of the last 135 years. So again, can you provide any more insight into how you calculated the probabilities? I am genuinely curious.

Comment really helpful (Score 1) 263

From TFA:

Python is a general-purpose language, which means it isn’t used for just one purpose such as Web development.

Oh, so that is what "general-purpose" means! I'm still not sure I understand, though. Can you give me some examples?

For example, if you’re hired to write apps that interact with operating systems and monitor devices, you might not need to know how to use the Python modules for scientific and numerical programming. In a similar fashion, if you’re hired to write Python code that interacts with a MySQL database, then you won’t need to master how it works with CouchDB.

Got it. So with Python, I don't need to spend time learning things that I don't need to know. Python does sound like quite a useful language!

In all seriousness, the article doesn't even have its facts straight. Consider:

Any Python newbie needs to know which types are immutable, which means an object of that type can’t be changed (answer: tuples and strings).

No, that's not the correct answer. Numeric types are also immutable, and that includes integers, floats, complex numbers, and Booleans. Frozen sets are immutable. (To be fair, frozen sets are a relatively obscure type unlikely to be used by beginners.) There are probably others I'm not thinking of off the top of my head.

Comment chances are good? (Score 1) 108

From the summary and TFA:

Chances seem good that Amazon will ask future teams to build machines that are even smarter and faster.

The chances that Amazon will want future warehouse robots to be "even smarter and faster" are "good"?? Okay, I suppose the probability that they will want future robots to be dumber and slower is technically non-zero, but I'd at least revise "good" to "almost certain".

Comment Re:inbreeding beneficial? (Score 1) 111

All good points. I would suggest a few language corrections to make your statement a bit more precise, though (my changes in bold):

The problem with inbreeding is that you can get two copies of a single deleterious, recessive allele quite easily, and rare genetic diseases that appear only when the same allele is present on both chromosomes in a pair suddenly start popping up more often.

The issue with inbreeding depression is not getting two identical copies of a chromosome (which, because of chromosomal crossover during meiosis, is extremely unlikely to happen), it's getting two copies of an allele (or set of alleles) that causes a recessive genetic disease to be expressed.

Comment inbreeding beneficial? (Score 3, Informative) 111

From TFA summary:

Dr Warren Booth, an evolutionary biologist at the University of Tulsa, who previously discovered an instance of parthenogenesis in snakes, said: "This is basically a very extreme form of inbreeding. Most people think of inbreeding as bad, but it could be helpful in purging deleterious mutations from a population."

Most people think of inbreeding as bad, because it almost always is bad. Inbreeding depression is a very well documented, and well understood, phenomenon that can increase the extinction risk of critically endangered species. The idea that inbreeding can somehow be "helpful in purging deleterious mutations" has been discussed before, but a recent study found that even if small (e.g., endangered) populations are actively managed to control both inbreeding and outbreeding, the negative effects of inbreeding depression generally outweigh the benefits of removing harmful alleles. And that is a best case scenario, with reproduction carefully controlled to produce an optimal genetic outcome, which obviously does not happen naturally.

For these sawfish, asexual reproduction is most likely a desperation strategy used when the population has gotten so small that it is difficult or impossible to find mates. It is extremely unlikely that it will somehow improve the population's genetic fitness; more likely, it will lead to further decreases in genetic diversity and a corresponding loss of overall fitness.

Comment summary is inaccurate (Score 1) 385

According to the summary (and linked article): "What job is hardest for a robot to do? Mental health and substance abuse social workers (found under community and social services)."

If you bother to read the actual research paper, the authors concluded that "recreational therapists" were the least likely jobs to be computerized, with a probability of 0.0028 (0.28%). Plus, there are two other jobs ("first-line supervisors of mechanics, installers, and repairers" and "emergency management directors") that also have lower probabilities of computerization than "mental health and substance abuse social workers". For an article that contains barely more than 10 sentences, one would think that they could have at least bothered to get their main point correct.

Comment Hard to predict how this will turn out (Score 5, Insightful) 135

It's hard to predict what the end result of this will be.

On the one hand, I can imagine that letting the mass spying provisions expire, and forcing the bulk data collection to stop, could actually be a win for privacy in the long run. After all, inertia is powerful, especially in politics. It is much easier and less controversial to say, "let's continue with our existing domestic spying program" than it is to say, "now that we stopped spying on everyone for a while, let's start spying on everyone again."

On the other hand, letting everything expire could create an environment where it becomes easy for fear to rule the day (or, easier than usual). We'll no doubt have politicians eager to scare us with stories of how letting bulk domestic surveillance expire makes us unsafe and vulnerable to terrorists, and so "we need to do something now before we die!" Then, new spying legislation could be hastily pushed through that is no better (or worse, depending on your perspective) than what we have now.

As I said, I think it is hard to predict the ultimate outcome, but if the recent past is any indicator, I sadly suspect that fear will win.

"More software projects have gone awry for lack of calendar time than for all other causes combined." -- Fred Brooks, Jr., _The Mythical Man Month_