Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[kaian-a-coel]

In other news, whereas the Curiosity rover had previously not detected methane, it now apparently has: http://www.bbc.com/news/science-environment-30456664

It shows up from time to time, blown in from some source in the rim of the crater to the north. Which opens up the potential to determine whether it is caused by life if they can get a decent sample of it:

One way to investigate whether the methane on Mars has a biological or a geological origin would be to study the types, or isotopes, of carbon atom in the gas.

On Earth, life favours a lighter version of the element (carbon-12), over a heavier one (carbon-13).

A high C-12 to C-13 ratio in ancient Earth rocks has been interpreted as evidence that biological activity existed on our world as much as four billion years ago.

If scientists could find similar evidence on Mars, it would be startling. But, sadly, the volumes of methane detected by Curiosity are simply too small to run this kind of experiment.

"If we had enriched our sample during one of the peaks, we might have had a shot at looking at these isotopes," explained Dr Mahaffy.

"I think there is still some hope. If the methane comes back, and we can enrich it, we'll certainly be trying."

It might have to do with the time (0035), but my mind is blown that we could potentially proove the existence of life on mars by smelling faint whiffs of methane and counting the number of neutrons the carbons in said methane have. With a robot. On mars.

[MetalSlimeHunt]

It's hardly amazing proof, but it is a sign that is otherwise difficult to explain. Another issue is the presence of methane on Mars at all. It should be rapidly destroyed, but remains.

[ChairmanPoo]

It might have to do with the time (0035), but my mind is blown that we could potentially proove the existence of life on mars by smelling faint whiffs of methane and counting the number of neutrons the carbons in said methane have. With a robot. On mars.

There are other explanations for methane formation though. IIRC it abounds in gas giants and comets.

[alway]

Yes. Which is precisely why it states it is looking for specific isotopes.

In unrelated news: http://newsoffice.mit.edu/2014/mathematical-relationship-in-superconductors-1216

MIT researchers have discovered a new mathematical relationship — between material thickness, temperature, and electrical resistance — that appears to hold in all superconductors.

The interesting note is that this was based on experimental results; so between existing theory and these results we can assume there are some new theories to be found which may shed more light on superconductivity. And these results may help shed light on that.

[Evil Knievel]

By Stephen Hawking , Stuart Russell , Max Tegmark , Frank Wilczek:

Success in creating AI would be the biggest event in human history. Unfortunately, it might also be the last, unless we learn how to avoid the risks.

[Putnam]

Well, yeah, when it comes to existential risk an uncaring/hostile AI would be the biggest one by far. I mean, there may be something like like random-ass gamma ray bursts from some perfectly-aligned supernova in the large magellanic cloud or something, but that's not preventable at all except by leaving, so it's not really something that should be considered (at least, not directly; it's pretty well an accessory to the standard "spread away from Earth" goal of mitigating existential risk which is overall probably the most effective method)

[i2amroy]

Personally I think the spook is happening a bit soon. We are very, very likely to have limited AI's that are able to act sort of like humans and sort of not for quite a while to work out all the bugs and limits before we ever get an AI that is able to exceed humans in every way. In my opinion that article is kinda like an article that was claiming that we weren't thinking enough about astronaut safety... when we still hadn't even got a successful unmanned rocket off of the launch pad. Like any scientific developments, it's very likely that the field of AI will have AI's that can learn in some areas and be constrained in others, our unmanned rockets in the analogy, that we will be able to use to work out the biggest bugs before they actually risk anything serious. We'll probably still eventually have some sketchy moments (just as we did in the space program), but hopefully by that point in time we'll be able to avoid/fix them for the most part.

[ChairmanPoo]

Pfizer is going to relaunch Mylotarg  http://www.pmlive.com/pharma_news/ash_pfizer_eyes_re-launch_of_mylotarg_622985

[alway]

Interesting thoughts of the effects of using GPS in the construction of massive building projects: http://bldgblog.blogspot.com/2014/12/spacetime-glitches-frozen-into-built.html

Edge to edge, the GPS constellation can apparently be considered something of a single device, a massive super-detector whose "time glitches" could be analyzed for signs of dark matter.
*snip*
The temporal distortion—a kind of spacetime wave—would propagate across the constellation, taking as long as 170 seconds to pass from one side to the other, leaving forensically visible traces in GPS's navigational timestamps.
*snip*
Here, he specifically mentioned the risk of space weather affecting the accuracy of GPS—that is, things like solar flares and other solar magnetic events. These can throw-off the artificial stars of the GPS constellation, leading to temporarily inaccurate location data—which can then mislead our construction equipment here on Earth, even if only by a factor of millimeters.

What's so interesting and provocative about this is that these tiny errors created by space weather risk becoming permanently inscribed into the built environment—or fossilized there, in a sense, due to the reliance of today's construction equipment on these fragile signals from space.
*snip*
The precision required by the bridge made GPS-based location data indispensable to the construction process: "Altimetric checks by GPS ensured a precision of the order of 5mm in both X and Y directions," we read in this PDF.

But even—or perhaps especially—this level of precision was vulnerable to the distorting effects of space weather.
*snip*
The bigger the project, the more likely its GPS errors could be read or made visible—where unexpected curves, glitches, changes in height, or other minor inaccuracies are not just frustrating imperfections caused by inattentive construction engineers, but are actually evidence of spacetime itself, of all the bulging defects and distortions through which our planet must constantly pass now frozen into the built environment all around us.

[Sergarr]

Uh.

The wording of that article is kinda...

well

uh

You see, the spacetime waves

them tend to be incredibly small at best and absolutely negligible at worst.

[alway]

Space weather certainly could. As for dark matter, it would depend on the strength of the effect. If they were able to detect it with current GPS, it would need to be a temporal anomaly at least 10^-18 seconds in length, as that is the highest precision any current atomic clock (and would likely be 2-3 orders of magnitude higher or so, since the precision has been increasing over the past years, and GPS satellites likely are using some of the couple-year-old methods due to development + launch cycles).

Meanwhile, the speed of light is approximately 3 * 10^8 m/s. As such, with current GPS satellites, you likely would have a lower bound of around 10^-8 m of perturbation caused by any sort of signal messing with the timing which could be distinguished from noise.

From this: http://www.unr.edu/nevada-today/news/2014/finding-dark-matter
It states: "If the dark matter causes the clocks to go out of sync by more than a billionth of a second we should easily be able to detect such events."
Which is actually quite a large deviation by the above calculations (unsurprising, since the above lower bound doesn't account for any sort of precision anomalies originating from the other equipment in the GPS aside from the clocks). A temporal anomaly of 10^-9 seconds times the speed of light is a full 30 centimeters; a good deal greater than the 5mm talked about in the article.

[Lagslayer]

Personally I think the spook is happening a bit soon. We are very, very likely to have limited AI's that are able to act sort of like humans and sort of not for quite a while to work out all the bugs and limits before we ever get an AI that is able to exceed humans in every way. In my opinion that article is kinda like an article that was claiming that we weren't thinking enough about astronaut safety... when we still hadn't even got a successful unmanned rocket off of the launch pad. Like any scientific developments, it's very likely that the field of AI will have AI's that can learn in some areas and be constrained in others, our unmanned rockets in the analogy, that we will be able to use to work out the biggest bugs before they actually risk anything serious. We'll probably still eventually have some sketchy moments (just as we did in the space program), but hopefully by that point in time we'll be able to avoid/fix them for the most part.

Nonsense! We need to start thinking about this now! The sooner we start, the more likely we are to find an acceptable solution. You don't wait til you have an emergency before you devise a disaster plan.

edit: seems I am a few days late. Haven't been on much lately. My point remains valid!

[Frumple]

... we've been thinking about proper response to AI since, uh. Hell, at least the 50s? Over half a century now, at minimum. We've got acceptable solutions oozing out our ears, it's just a matter of picking 'em and implementing, with maybe a little adjustment for however things actually fall out instead of just the copious amounts of considering possibles we've already done.

[Sergarr]

Just implement laziness into the AI routines. Then we won't have to worry about runaway AI.

But then we would have to start worrying about how to motivate these lazy bums to do anything!

[Reelya]

Design an AI to come up with strategies to keep the other AIs in line.