Hungry penguins have inspired a novel way of making sure computer code in smart cars does not crash. Tools based on the way the birds co-operatively hunt for fish are being developed to test different ways of organising in-car software. The tools look for safe ways to organise code in the same way that penguins seek food sources in the open ocean. Experts said such testing systems would be vital as cars get more connected.
Engineers have often turned to nature for good solutions to tricky problems, said Prof Yiannis Papadopoulos, a computer scientist at the University of Hull who, together with Dr Youcef Gheraibia from Algeria, developed the penguin-inspired testing system. The way ants pass messages among nest-mates has helped telecoms firms keep telephone networks running, and many robots get around using methods of locomotion based on the ways animals move.
Penguins were another candidate, said Prof Papadopoulos, because millions of years of evolution has helped them develop very efficient hunting strategies. This was useful behaviour to copy, he said, because it showed that penguins had solved a tricky optimisation problem - how to ensure as many penguins as possible get enough to eat. [...] "There must be something special about their hunting strategy," he said, adding that an inefficient strategy would mean many birds starved.
Tux was not involved.
(Score: 2) by Dr Spin on Tuesday January 31 2017, @08:02AM
I look forward to my next car being better at finding fish!
OTOH, if the use of Penguins demonstrates that Wince is incompatible with in car entertainment, I shall be greatly relieved.
Warning: Opening your mouth may invalidate your brain!
(Score: 2) by bob_super on Tuesday January 31 2017, @06:17PM
> I look forward to my next car being better at finding fish!
Given their target demographics, how long before NASCAR jumps at the opportunity to include fishing in their car races?
I'm already quite surprised that they haven't included guns and explosives yet.
(Score: 0) by Anonymous Coward on Tuesday January 31 2017, @10:24AM
From the article:
This sounds as if they want to automate the job of the software engineer.
(Score: 0) by Anonymous Coward on Wednesday February 01 2017, @02:14AM
This already exists and it's called genetic programming (basically you run a genetic algorithm on the code structure itself, which is easiest in Lisp, or on command sequences that act as code). There's nothing to see here except a new name so someone can clam their search heuristic is something new.
It sounds like they don't know what fuzzing is. With a software program you can automatically target every input path. You don't need to have your testing code flock to the area which has triggered the most bugs (I think that's what they're trying to describe? The article is completely lacking in real technical details, it's mostly hand-waving marketing bullshit.). Testing buggy areas harder generally means you're wasting time finding ways to re-exploit bugs you've already discovered. If a negative number causes a crash, there's no value in retesting that feature with 200 different negative numbers.
(Score: 3, Interesting) by butthurt on Tuesday January 31 2017, @12:35PM
Millions of virtual monkeys have almost typed out the entire works of Shakespeare by bashing random keys on simulated typewriters.
The virtual monkeys, created by an American programmer, have already typed up the whole of the poem A Lover's Complaint and are 99.99 per cent of the way through the Bard's complete works.
The experiment attempts to prove the theory that an infinite number of monkeys sitting at an infinite number of typewriters would eventually reproduce the works of Shakespeare by chance.
-- http://www.telegraph.co.uk/technology/news/8789894/Monkeys-at-typewriters-close-to-reproducing-Shakespeare.html [telegraph.co.uk]
https://en.wikipedia.org/wiki/Infinite_monkey_theorem [wikipedia.org]
(Score: 0) by Anonymous Coward on Tuesday January 31 2017, @01:23PM
Of course that experiment does nothing of the sort, as those are not real monkeys. Probably the programmer didn't even determine the probability distribution of characters typed by a real monkey on a real typewriter.
Probably the problem as stated is ill-defined anyway since it doesn't say which species of monkey and which sort of typewriter (mechanical? electric? Which keyboard layout?), or how much training the monkey got (does a monkey even start proper typing when given a typewriter, but not trained to type on it? I'd expect it to just hit the keyboard pressing several characters at once, jam the letters and then quickly lose interest as the typewriter doesn't seem to do something interesting).
(Score: 2) by bob_super on Tuesday January 31 2017, @06:15PM
The beauty of infinity is that it doesn't actually matter what the probabilities are, as long as they're not zero.
(Score: 2) by maxwell demon on Tuesday January 31 2017, @07:47PM
A monkey that quickly loses interest surely has zero probability of typing the works of Shakespeare.
The Tao of math: The numbers you can count are not the real numbers.
(Score: 2) by bob_super on Tuesday January 31 2017, @07:56PM
If you have a monkey with an infinite life, the probability that he will get back to typing is not zero.
(Score: 0) by Anonymous Coward on Tuesday January 31 2017, @08:56PM
You can't assume that. I'm still alive, but I have yet to take a botany course or show any signs of interest in botany. Even if I lived forever, I'd probably still not take any botany courses.
Just because it's infinity doesn't mean you can just assume that a non-zero probability will be large enough for the result to occur.
(Score: 2) by bob_super on Tuesday January 31 2017, @09:31PM
In a sense, you can. The math sense.
If the probability of anything doesn't decrease faster than the inverse of time, then mathematically, it will happen at some point over an infinite duration.
You might realize, after spending 20000000 years watching porn, playing video games and reading Confucius a few thousand times, that maybe you could dedicate some time to knowing something about those odd plants around you.
Unless you're shipped to outer space without any plants nor study material and spend eternity avoiding all planets with plants (qualifies as decreasing faster than time).
(Score: 2) by maxwell demon on Tuesday January 31 2017, @09:39PM
It's infinitely many monkeys, but not monkeys with infinite lifetime. Although infinite life time won't help either if the typewriter rusts away before the monkey gets interested again.
The Tao of math: The numbers you can count are not the real numbers.
(Score: 2) by bob_super on Tuesday January 31 2017, @10:07PM
Isn't an infinite number of monkeys, and their equally infinite amount of typewriters, equivalent to a single eternal monkey with an infinitely-working typewriter, for the purpose of assessing the probability of something being typed?
It's a good thing that the original posting was so uninformative, considering how far off-topic we've drifted.
(Score: 2) by maxwell demon on Wednesday February 01 2017, @04:26AM
No, it isn't, because the monkeys, and thus their typing-probabilities, are not time-independent. Fore example, what does sound more credible, a single monkey getting bored after 1000 minutes, or 1000 monkeys getting bored after 1 minute?
The Tao of math: The numbers you can count are not the real numbers.
(Score: 2) by bob_super on Wednesday February 01 2017, @05:03PM
If you put a monkey in a featureless room for eternity, with only a typewriter... He's gonna type.
(Score: 2) by maxwell demon on Wednesday February 01 2017, @06:37PM
Are you sure he's not rather going to examine the typewriter, ruining it in the process?
The Tao of math: The numbers you can count are not the real numbers.
(Score: 2) by bob_super on Wednesday February 01 2017, @07:01PM
A few posts ago in this giant digression, I did specify an infinitely working typewriter.
Because even an infinite amount of monkeys with an equally infinite number of typewriters would not be able to type the whole works of the Bard on a single one, if you don't provide them with paper or ink.
/thread at this point. Beaten that eternal monkey to death enough.
(Score: 0) by Anonymous Coward on Tuesday January 31 2017, @09:03PM
The beauty of infinity is that it doesn't actually matter what the probabilities are, as long as they're not zero.
I'm not sure what I think about this. From an abstract mathematical perspective you are (probably) right... but are you really?
For the sake of argument, let's say there are 2^10 particles in the universe, and the universe is 2^10 picoseconds old. Now imagine probability of randomly typing out Shakespeare is 1 out of 2^2^1000000000000000000. So really... it is effectively impossible in the universe.
So does it merely matter that it is a non-zero probability, because I somehow feel like you need more than that, even if I don't know what or how.
(Score: 1, Informative) by Anonymous Coward on Tuesday January 31 2017, @08:59PM
That project is as useless and as deceptive as the "___ passes the Turing Test" headlines we occasionally hear.
In my mind, the relevant statement is, "If the nine-letter sequence appears anywhere in one of Shakespeare's writings, it is matched against the relevant passage in a copy of the Bard's complete works, and is checked off the list."
That's like saying, "aa, at, ag, ac, ta, tt, tg, tc, ga, gt, gg, gc, ca, ct, cg, cc." There, I've sequenced the entire Human genome, as every DNA strand is some combination of those DNA pairs!
The project could potentially be interesting, but it is certainly nothing like the infinite monkey theorem. If anything, I'd say that it lends credence to exactly how unlikely it is to create the complete works of Shakespear at random, as simply producing the 9-letter combinations is taking this long, and the works are millions (?) of letters long and thus something like 2^1000000 more difficult (back-of-the-envelope math here).
(Score: 2) by hendrikboom on Tuesday January 31 2017, @02:05PM
Is it possible to mod an article down instead of a comment? While amusing to talk of penguins and programming, there is no technical information in this article at all. It is pure fluff. Not even a hint as to how they are going to reorganize the software process.
(Score: 3, Funny) by KritonK on Tuesday January 31 2017, @02:10PM
Perhaps they simply mean that, by using Linux instead of Windows, they achieved a significant decrease in the amount of software crashes.
(Score: 0) by Anonymous Coward on Tuesday January 31 2017, @08:59PM
The real question here is why this is better than the Monte Carlo algorithm that's typically used for testing plane software. You can't necessarily foresee and test all possibile conditions, but if you test enough combinations of conditions you can be sure that the car is safe to whatever level of safety you need. You'll probably still find situations that don't work right, but they're rare enough that they can be fixed and hopefully, it takes long enough for that to happen that people don't get a bad attitude towards the technology.
(Score: 2) by hendrikboom on Saturday February 04 2017, @03:36PM
My guess is that finding problems in one area of the software might be a clue that there would be other bugs in the same area. But that's a well-known enough fact that it hardly needs to be inspired by hungry penguins.
(Score: 2) by requerdanos on Tuesday January 31 2017, @04:10PM
The plot of the book "Prey" by Michael Crighton [wikipedia.org] is driven by some software called PREDPREY, a fictional set of artificial intelligence algorithms modeled after prey-seeking animals. I don't recall whether penguins or fish were mentioned. As you might predict, the algorithm learns... a little too well.
I assume it's called PREDPREY because of an eight-character filename restriction.