The underlying data here comes from the Ptolemy instrument, a gas chromatograph mass-spectrometer (ion trap, in this case) which can measure ratios of molecules in a range of 14 to 140 Da. (It might also be able to identify somewhat larger species by inducing fragmentation and measuring the resulting pattern of smaller molecules within the effective mass range, but I haven't found if this capability was included).
The article very clearly indicates the data is from COSAC, an instrument on the lander. It also indicates that the sampling came from the "atmosphere" of the comet rather than drilling.
Indeed, good catch - that wasn't in the originally-linked article and I completely overlooked COSAC when looking up the instruments. So here is some info on COSAC:
It is also a GC-MS, but rather than an ion trap, it is a time-of-flight analyzer with a much higher mass range (up to 1500 Da in the widest mode). It is neat to have two mass specs on a single relatively small probe. They are intended to be complementary, as discussed here (briefly, PTOLEMY is designed to measure isotope ratios and the COSAC TOF is designed for organics as it can measure chirality with dual GC columns, and also with a much higher mass range):
[edit: for anyone else interested in mass-spec, I found some slides with an overview of MS instruments on previous probes - I was familiar with several, but there have been many more than I realized: http://jfsm2012.sciencesconf.org/conference/jfsm2012/pages/C...]
Personally I think the idea that life/prebiotic soup came to earth on a comet is suspect, for the reasons explained in these blog posts and the links therein:
Two problems with the 'organic asteroid' theory are the 'racemization problem' (how to explain why life only uses L isomers rather than both the L and D isomers present in meteorites), and the fact that the amount of organic matter in meteors is tiny. Compare that to the opportunities for life to arise on earth in its diverse chemical environments.
See the calculations by Jeffrey Bada discussed in the second link.
It's possible that there's no reason life chose L or D isomers, but choosing one and sticking to it was advantageous (I use the word "choose" loosely here).
Abiogenesis is one of those things where there is way more speculation than evidence. Some speculation says that it is because the molecular machinery maintenance to handle both would be twice as taxing for organisms.
Of course, having the reverse isomer molecules would be advantageous against predators (who couldn't digest it) and a boon to natural selection.
> Of course, having the reverse isomer molecules would be advantageous against predators (who couldn't digest it) and a boon to natural selection.
This, of course, doesn't really mean much. Something being advantageous doesn't imply that it must show up as a product of natural selection nor does it imply that such an adaptation must survive into perpetuity.
This is the kind of change you really don't expect to have an evolutionary pathway. Random genetic changes to genes won't suddenly start a full system of reverted isonomers out of the blue, and there's way to slowly get there (it's negative all the way through than a little positive afterwards).
I gave it as an example of there being too much speculation.
There is absolutely no evidence for it and it is not meant to be taken as as serious as some of you are taking it. If anything, it proves there is too much speculation -_-
It renders predation on you maladaptive, because the predator doesn't get back the energy expended in hunting, so over time it does actually make you less likely to get killed.
You're correct, and I didn't mean to make it seem as though it was an intentional process.
But historically, when one aspect of the ecosystem goes unchecked, something usually comes along to take advantage of plentiful biomass, one notable exception being humans.
Don't forget domesticated cats. It's still a wild animal, but one that has adapted itself to live off humans in quite clever ways (essentially by mimicking human children).
Yes, except even herbivores occasionally show evidence of eating meat. Deers are occasionally seen eating birds, similar has been seen in Cows (IIRC Lal the cow was documented eating its owners chickens from the coop) and in fact all Bovid species occasionally show eating birds.
If there's a world of untapped food, this exceptionally rare behaviour would rapidly become advantageous, especially when the prey isn't adapted to the unusual predator.
Evolution has no guide, but all evidence shows it evolves into almost every available niche and quite specifically that any large available resource will be tapped.
So what if some herbivores will occasionally eat meat? You're assuming that the organism should look, act, and behave strictly within the boundaries of an English description.
Again, evolution is still unguided and to assume that something will evolve into an open niche is a gambler's fallacy.
I'm not even going to attempt to argue this, because I sense I'm simply going to be banging my head against the wall because you're going to be locked into semantics and obsessing over "unguided".
Convergent evolution, and history argue against you. C4 photosynthesis evolved independently 60 times. There's no guarantee it will, and I'm not arguing it will. However, the weight of evidence argues that if there's a readily available niche it is a matter of time.
If you argue I won't win the jackpot, you're likely right - in fact I'd put money on my probability of losing. However, if you argue no one will win the jackpot in the next 100 years you're an idiot.
You're the one trying to use the semantics of "herbivores" to make a point in biology. By assuming that there is even a "jackpot" to inevitably be won is making the gambler's fallacy. There is no "jackpot".
Quit projecting and just stop already, electromagnetic. You can't even bow out and admit you're wrong gracefully, so you skip straight to name calling.
You don't even know what the Gambler's Fallacy is! Gambler's Fallacy is betting according to a limited collection of statistics (red came up 5 times in a row, bet on black!) and not to true probability.
I didn't intend to refer to you personally as an idiot, but turing a phrase however that appears very lost on you.
I'm certain you're just intentionally being a troll, because you're unwilling to make a single basis of an argument or even support your reasoning and can't even use your terms correctly, and now you're attempting to bait me with condescension.
I believe what they're getting at is that it's possible to do things to other species which leaves them digestible when they weren't initially. For example, human cooking, bird gizzards, cows chewing their cud, or rabbits eating their feces to give a second pass of digestion.
You talk about cooking meat, but we still have enzymes in our stomachs (e.g. pepsinogen) that only digest L-peptide chains. As a predator, you do not digest D-peptide chains.
- It is analagous to eating raw grass.
- It is not analgous to eating processed/cooked grains, hence it is non-sequitur.
- It is a matter of biochemistry, not culinary arts.
I gave the original speculation as an example of there being too much speculation around abiogenesis and evolution. This discussion proves it.
I confess ignorance about this theory, but I see it a lot and I don't understand. It just seems bizarre to think that life began on a comet. Sure, we want to know the origin of life. What does pushing it off the Earth solve? I guess my question is, What are the positive reasons for thinking life arrived from a comet? I can't think of any. As a layman, the theory is so strange I'm not even sure why it needs to be refuted. Can someone explain why this idea is popular?
One reason for its appeal is that the earliest fossils / signs we have of life on earth are pretty close to the ground not longer being molten lava. An origin of life in comets would give more time.
Well, the first life that existed was maybe some kind of rna-like molecule with very little to no protection. Not even a cell.
I do not think any of these could leave any traces, even a fossil.
I see it more along the lines of being a hypothesis. We do not have nearly enough evidence to put it off the table, even though it is not the most likely explanation.
> Once the rechargeable secondary battery has been warmed by sunlight again, Philae will restart and the DLR LCC team will take their places at the control consoles again.
This is the first I've heard that Philae can wake back up. I was under the impression that it had no solar capabilities whatsoever, and that the battery was the only power it had available.
Yeah the probe is covered in panels. The main issue is that it did not land where they intended and it is mostly in shadow. I am not a rocket scientist but... it looks like it will be getting closer to the sun as time goes on here so maybe the light it can get later will be more intense and enough to power it?
The main issue I read was that even though it gets some light (something like 1 hour of light per 12 hours) the battery can't recharge when it's cold. So Philae needs light to heat up the battery and only then it can start charging it so it needs a longer exposure to sunlight or it needs to be closer to the sun so it heats up faster.
Most electronics are more efficient when "cold", but when we say that we mean "cold" by our earth standards, i.e. something like 15 degrees celsius. I think I read that Philae's temperature after landing was less than -160 degrees celsius. Getting anything to work at those temperatures is hard...
I've understood it has solar panels. But what would be the reason they didn't put a nuclear battery on board? That would work for years.
If the direct output power of the nuclear battery was too low, it could still be used to charge the rechargeable battery until it can make the lander do work again for a few hours/days.
That was asked during the early hours of the power "uh oh" phase, and someone involved replied that Europe has no ability to manufacture RTGs for political reasons.
I've seen it a few places that they mentioned in the press conference that (mostly for political reasons) there has been almost no development of RTGs in Europe at all so it simply weren't an option.
I also heard from reddit that they didn't want to have radiation contaminate the area where Philea landed, apparently it's a bit hard to keep all of it inside of the lander.
They rotated it slightly just before the main battery ran out. They hope it might get enough light as the 'season' on the comet changes (perhaps around christmas) but no-one knows for sure.
Planetary Society blogger Emily Lakdawalla has a Tumblr that had very good coverage of what went on: http://elakdawalla.tumblr.com/ ... details of the movement, the solar charging requirements etc
This article says something different (more factual/less speculative) than the rosetta blog post.
rosetta blog post: "mission controllers sent commands to rotate the lander's main body, to which the solar panels are fixed. This may have exposed more panel area to sunlight."
article:
". The team managed to rotate the lander during the night of 14/15 November 2014, so that the largest solar panel is now aligned towards the Sun."
Ms Lakdawalla's post provided more detail, including confirmation they got feedback from the lander that its movement was successful, although it wasn't clear if they were going to actually get enough light any time soon for full operations (thanks for that hint):
" They commanded the lander legs to lift the body by 4 centimeters, hopefully raising it a bit farther above the horizon, and then rotating the lander body 35 degrees to place the largest solar panel in a position where it could hopefully receive the most sunlight.
Telemetry confirmed that the lander moved as it had been commanded to do. "
Right. They don't know where the lander is, what's around it and so on (beyond that one panoramic photo) so whether the light situation will improve with the seasons or not is impossible to know.
"Scientists are analyzing the data to see whether the organic compounds detected by Philae are simple ones—such as methane and methanol—or a more complex species such as amino acids"
I feel like there is a big difference between finding methane and finding amino acids on the comet. Surely, finding methane isn't all that interesting.
It would have interesting implications if it was produced by microorganisms. I don't know enough to comment further, but I'm guessing that's what the article was suggesting.
Organic compounds are just molecules that have carbon atoms in them. You don't need anything that is 'living' to produce them.
All life we have discovered on earth is based on carbon. So, finding these building blocks of life in space is interesting because it is solid proof you don't need to be on earth to produce them (something that might or might not seem intuitively obvious but in science you can't use intuition as proof, only as a guide). However, while making it a tiny bit more likely, it does not mean anything is alive out there.
But that does not mean all comets survive their close approach to the Sun. Some, such as the sungrazing comets[0], come quite close to the surface and can be destroyed by their encounters with the Sun.
As jnevill points out, this particular comet's orbit does not take it closer to the Sun than the Earth's distance from the Sun. The comet is on a ~6yr orbit, I believe, and so should survive for a long period of time.
Yeah, my first reaction was the same, mentioning them in the top of article while mentioning all other contributors as a note in last paragraph felt like wanting to take all credit.
But then I saw it was published on a ".de" website. You can expect it's targeted at a german audience, who naturally feels interested in knowing which part their country played into this (mentioning others just after wouldn't have hurt, though).
Thanks. Just a note to others, if you have already opened the link 'with' the paywall, even this might not work. Just open this link in another browser.
We post links to products that cost money, though, no? Journalists need to eat too. Usually it takes entire milliseconds for someone to post other links, anyway. Or you can always sneak around the paywall.
In this case I don't mind an article that has a bit more science bias rather than 'mainstream', so the DLR link is nice.
I didn't. I clicked on the German one, which I suppose I may have paid for in some tiny way, because I pay taxes in Europe. At least I'd rather think about that than people like Berlusconi and friends getting a hold of them.
I subscribe to The Economist, and occasionally buy access to the Financial Times.
Well the opinion that you "don't mind paying" for something only really means anything if you actually pay for it (or pay for something similar). If you "don't mind paying" but you're getting it free (eg via your workplace or some other way) then that opinion seems pretty void.
The fact that exciting scientific discoveries like this are being presented on WSJ behind a paywall is absurd. Science on the internet is going to be dramatically different in 10 years than it is right now...
It's inconvenient for us, but I wouldn't necessarily call it "absurd." The Wall Street Journal is not a 501(c). It's either this, or a flurry of intrusive ads. (Or a subscription upsell, but I digress). Different publications pick their poison differently. But someone's gotta pay for something.
Besides, paywalls aren't hard to circumvent. Someone usually posts a redirect within minutes of these articles' being linked here.
I'm not a paywall advocate, fwiw. I think it's a suboptimal monetization strategy that often directly conflicts with distribution and UX. That said, I don't think it's ipso facto ridiculous. There are better ways to monetize, but not monetizing is not an option for WSJ.
It's absurd in the context of social norms and expectations. A restaurant isn't a 501(c) either, but I would call it absurd if they charged a customer to use the restroom.
Respectfully, I'm not sure I follow the analogy. A restaurant's core business is serving food, and a restaurant charges for food. The Journal's core business is journalism, and it charges for journalism. Content isn't the "restroom" of the WSJ's business model.
As I've stated in a comment further down this tree, I don't like paywalls. I think they're an inefficient and suboptimal means of monetizing content. A paywall is basically a tax on readers who aren't savvy enough to bypass it. If you look at the economics of a paywall, it's basically monetizing the intersection of two sets of people: 1) the set of all people who really want to read the WSJ, and 2) the set of all people who can't get around a paywall when they encounter one. This intersection cohort -- call it "People who like WSJ and can't navigate paywalls" -- is probably a decent size, but it leaves a lot of would-be customers on the table. At the same time, the paywall hampers distribution and creates a bad user experience. It's an unsophisticated way to monetize content. That said, I don't find it "absurd." Bad, sure. Annoying, absolutely. "Absurd," no. Nor do I find absurdity in the premise that the WSJ should attempt to monetize its content.
I'm pretty sure that when I was in Paris, it cost $0.25 or $0.50 to use the restroom at McDonalds. Which, to be honest, made me feel better about walking in, using their restroom, and carrying on without purchasing a meal :)
> A restaurant isn't a 501(c) either, but I would call it absurd if they charged a customer to use the restroom.
The cost of the restroom visit is, in a sense, folded into the cost of the meal. That is why many restaurants only allow paying customers to use the restrooms.
Since when do social norms and expectations involve getting journalists' work for free? This basically never happened as little as twenty years ago, and everybody was fine with it. I don't think norms have changed so quickly.
> Since when do social norms and expectations involve getting journalists' work for free?
That's a false dichotomy. Websites for major news outlets have always been overwhelming "free" (as in, not behind a paywall, but they obviously try to monetize through ads).
Yes, but the web is an extremely recent phenomenon relative to journalism in general, and journalism has historically been something where you had to pay to read.
That's true, but the medium's social norms tend to be stronger than the content's social norms. After all, the vast majority of web pages are free to view, and yet all the types of content on them were traditionally paid (because, for one thing, prior to the Internet most content was much more expensive to deliver).
You think the people at the WSJ should work for free? I think that's absurd.
Have you never attempted to read original scientific papers? Most are hidden behind paywalls. Search for yourself - http://www.ncbi.nlm.nih.gov/gquery/
Not true, or at least depends on the field. I am able to find 80% of the maths, physics, comp sci papers that interest me. I think reducing all human interaction into a measure of value is absurd - but there you go, each to his own.
One suspects that there are plenty of more primary sources for this matter. (Also the WSJ is a Rupert Murdoch mouthpiece, and bypassing its paywall is a mitzvah.)
Some more info here: http://www.lpi.usra.edu/meetings/lpsc2006/pdf/1937.pdf
and: http://sci.esa.int/rosetta/31445-instruments/?fbodylongid=89...
The instrument has its own Twitter feed, of course: https://twitter.com/philae_ptolemy