Naked Science Forum
General Science => General Science => Topic started by: neilep on 21/07/2007 18:11:30
-
Pick a Science....any science !!...look at it.....put it back in the pack !
What do ewe think are the next bigggest breakthroughs required !?
Either cite one from one particular science or list a few...!
-
Required, or likely to occur?
Required would be to discover a way to, as Marcus Brigstocke put it on Radio 4 this week, "get militant islamists to chill out and have a cup of tea instead of blowing people up. (meanwhile, can we have our planet back)"
Chris
-
If you can class socio-economics as a science, then finding a way to get billions of people free from poverty & starvation.
Failing that, a way to make humans as intelligent as beavers! [:D]
-
Pick a Science....any science !!...look at it.....put it back in the pack !
What do ewe think are the next bigggest breakthroughs required !?
Either cite one from one particular science or list a few...!
I imagine genetic science coupled with nanotechnology may have a chance of making it big time in the next decade. There is, of course, the issue of ethics at play here that needs to be addressed before people feel comfortable about adopting bioengineering on a large scale.
Another area could be the alternative energy research, although this is more of a hunch than any solid evidence to back up that prediction. I only talked with a friend who does biofuel research to get some rough idea about it. However, this may take more than scientific research, as the willingness of the established corporate players to adapt to the new alternative energy is as influential as any to make this happen sooner.
I think there's a revolution underway to revolutionize the home entertainment industry with the advances in computer science and network communications(this is probably not categorized quite as a science, more of a corporate thing). For example, TV may be replaced with more interactive programs between broadcasters and the viewer (kind of like what Julie Christie did in Fahrenheit 451), and more tailored towards the individual viewer's tastes (like TiVo, but more sophisticated).
On a more idealistic thought, I'd like to see the transmission of information breaching the speed of light and the Shannon capacity in the field of communications. I list this as a possible breakthrough that will have some significant impact. However, this may not be happening anytime soon.
-
Find a way to prevent the Arctic Current flowing from the Arctic Ocean through the Greenland - Nova Scotia and Greenland - Iceland gaps from shutting of as Arctic water temperatures rise. If this current stops flowing, then the world climatology will tank, and there will be a very messy period while the winds and rain sort out a new pattern for the rain. Starvation over vast areas will most likely occur.
Malthus strikes again.
-
When do you think that something like a dedicated "citizen" universal card or chip will be introduced so that is serves as credit card and card for storing personal data for home use and for instance for internet login (identity) verification like voting in various elections and buying things.
And where first?
-
When do you think that something like a dedicated "citizen" universal card or chip will be introduced so that is serves as credit card and card for storing personal data for home use and for instance for internet login (identity) verification like voting in various elections and buying things.
And where first?
Perhaps some security/privacy issues must be addressed first (either through technologcial means such as better encryption algorithms, or some
-
When do you think that something like a dedicated "citizen" universal card or chip will be introduced so that is serves as credit card and card for storing personal data for home use and for instance for internet login (identity) verification like voting in various elections and buying things.
And where first?
Perhaps some security/privacy issues must be addressed first (through e.g., technological means such as better encryption algorithms) before such a universal card/chip is accepted by the public. The technology of making such a card is available and pretty mature at that. It's really an issue of public's trust on ensuring privacy of information. In other words, its wide acceptance is more driven by culture than a technological push.
-
OK things to help humanity are nice. biotechnology has been going for a long time and will continue to improve. How about some fundamental science?
My suggestion for a real fundamental breakthrough in science is a greater understanding of the "laws" of complexity. This is where many individual entities interact to produce a common object that is more than just the sum of its elements. There are a great many ways in which simple physical laws lead to complex objects. These go all the way from the emergence of the physical laws we now see out of the uncertainty of the big bang through the origin and evolution of life on this planet (or other planets maybe with very different characteristics) right to the behaviour of human societies.
These laws are even more fundamental than the laws of physics and chemistry because they would operate in any system whatever the physical laws happened to be.
-
Chemometrics is another type of research that will be important in the near furture.
Basically, it involves programming computers to learn to recognize the normally undetectable differences in gasses, solutions or solids.
Some of my colleagues are developing an IR spectrophotometer that can detect different types of infections/abnormalities in blood samples. A normal IR spectrum of a blood sample would tell you nothing, but by running several thousand known blood samples, the computer can "learn" to recognize the minute differences, and be "taught" to diagnose diseases.
It uses a crystal, so the IR light doesn't even really pass through the blood. The light interacts with the blood on the surface of the crystal. This instrument, which could be contained inside a briefcase, would revolutionize blood tests as we know it. Normal blood tests are expensive and time consuming. An IR spectrum only takes ~2 minutes. This could greatly decrease the cost of doctor visits, as well as improve efficiency of medical aid to third world countries.
Its a long way from production, mainly because it takes thousands if not millions of known samples to program the computer. The more samples it analyzes the more statistically accurate it becomes. In several years this may become the normal way to do hundreds of analytical tests.
There are a vast array of other possible applications for this research, but the concept of computers that "learn" seems to scare the mainstream public. (Too many evil computer movies... I guess??)
-
There are a vast array of other possible applications for this research, but the concept of computers that "learn" seems to scare the mainstream public.
As someone who works with computers, it scares me too, but for different reasons.
The problem with learning computers is: how do you debug them?
A computer that is learning is substantially non-deterministic. You cannot really say what it has learnt (if you could say that, then you can program it that way in the first place).
This is not to say that learning computers do not have a very valuable role to play in research, but it would frighten me silly if I knew that the computer controlling the flight of the latest passenger airliner had only 'learnt' how to fly (i.e. that nobody had actually been able to prove that it can do its job in all foreseeable events).
OK, I do agree that the problem is that humans still have that non-deterministic element, and computers that depend on human programmers are vulnerable to human omissions, but to my mind, having a learning computer and a learning human only compounds the risks.
As I say, learning computers can be very valuable research tools, or possibly useful in non-critical applications (e.g. voice recognition software); but it critical software (e.g. medical software, or worse still, in transport control systems), I would rather have the learning systems to learn a pattern of behaviour, then have a researcher understand the basis for what has been learnt, and then program another computer with a deterministic program based on what has been learnt from the self learning computer.
-
Required, or likely to occur?
Required would be to discover a way to, as Marcus Brigstocke put it on Radio 4 this week, "get militant islamists to chill out and have a cup of tea instead of blowing people up. (meanwhile, can we have our planet back)"
Chris
Politically, this may be valuable; but in the longer term context, I actually believe that "terrorists" (whether Islamist or other) are as necessary a part of wider human development as anything else. Ultimately, all political power (that includes the official State) is enforced by violence and threat of violence (which is why they are so terrified of violence that is politically motivated being used against them). Every government of every State believes it should have a monopoly on violence, but is monopoly ever really a good thing in the longer term?
-
Violence does not work. The only solution to terrorism is to have benign laws alolowing maximum personal freedom. Minimise the reasons why people should become terrorists. Minimise the holding of weapons. Be watchful. and accept a small attrition rate from the occasional real looney without going wild. There is no such thing as a zero risk environment.