Thursday, 18 August 2011

Popular misconceptions about Science and Scientists

The popular press, and people in general, seem to misunderstand what scientists are, what they do, and what their goals are. This article clarifies these matters.

1. Scientists do not all work in laboratories, do not all wear lab coats, and they do not all perform wacky experiments with chemicals. The only scientists who work in laboratories are experimental physicists, experimental chemists, and experimental biologists. Sometimes they are assisted by lab technicians. Lab technicians are not full scientists; that is, they are usually low-level employees with training in cleaning bottles and other equipment. Mathematicians and theoretical physicists do not work in labs.

2. Sometimes, lab technicians are higher-level students paying off a bursary or study loan. In that case, they may be postgraduate students. A postgraduate student is one who has already got her undergraduate or bachelor’s degree in her area of study, and is now specialising. Her supervisor, or the person to whom she reports, will usually be a scientist or professor. A professor is a university employee who has published a lot of research papers and has been awarded the title ‘professor’. This is because he ‘professes’ to know his area well. Not all professors are scientists. You can get a professor of Engineering, for example; some Engineers do not consider themselves scientists; ‘scientists’ are more theoreticians, whereas Engineers are more about implementation. Most professors have PhD or doctorate degrees. Your general practitioner (GP) or medical doctor usually just has a Master’s degree - one level lower. There are different degrees in different countries, but the full list, in the order of amount of study involved, is: Diploma (in some countries), Bachelor’s Degree, Honours Degree (in some countries), Master’s Degree, Doctorate, Post-doctorate. A professor can have anything from a Master’s to a Post-doc; ‘professor’ is an honorary title given on the basis of research work. Not all lecturers at university are professors; often each department or subject area only has one full professor and several adjunct or associate professors.

3. A scientist is usually a university employee, and usually a lecturer. But some scientists work for parastatal entities or private research labs, pharmaceutical companies, weapons companies, or government-funded research labs. An example of this is CERN - the European Nuclear Research Centre in Switzerland. They have a large particle accelerator underground spanning many kilometers, and study the particles that make up matter. They’re scientists who work for an entity that is not a university.

4. The term ‘scientist’ usually applies to someone studying a science, rather than, say, economics, education, humanities, architecture, and other disciplines. Mathematics, chemistry, biology, physics, are considered ‘hard’ sciences. ‘Soft’ sciences might include the social sciences, such as population studies, politics, etc. One might consider these to be sciences because they make use of statistics - a form of mathematics - in their research. Statistics is not a form of guesswork. It is a strict form of mathematics with strict formulae and rules.

Importantly: not all scientists work in medicine. You often see the remark online that "why are scientists researching this useless thing when they could be finding a cure for cancer?" - well, that just means that the commenter doesn't understand that science has different research areas. It's like saying, "Why do biologists study living organisms when they could be designing rockets to get to Mars?". Scientists have specialist areas of study. They only study the area that interests them. Few of them are interested in medicine. There are a wide range of research areas. Here's a short list: Mathematics (Applied and theoretical), Computer Science, Statistics, Physics (Particle, Newtonian, Astro, Geo, Relativistic, etc.), Chemistry (Applied, Biological, Industrial, Physical, Inorganic, Organic, etc.), Biology (Botany, Zoology, Evolutionary, Palaeo, Human, Micro), Medicine, Psychology (Neuro, Clinical, Industrial etc - some debate about how much of psychology is a science), Sociology (debatable), Geography, Geology, Palaeontology, Archaeology, Anthropology, etc etc. Only one of these deals with medicine.

5. Mathematics and the hard sciences differ in that mathematics is not experimental, and assumes the truth of its tenets, such as 1 + 1 = 2. The hard sciences - physics, biology and chemistry - experiment with observable evidence, and make up ‘theories’ to explain the evidence. So they use experiments and evidence, whereas mathematics does not. Mathematics has a number of branches, such as computational and applied mathematics, which overlaps with Engineering, Physics, Computer Science and Astrophysics, and it also has the branch of Statistics. Some subject areas, therefore, are a bit fuzzy in terms of which areas they fall under. So, for example, you can study computer circuit design in Computer Science as well as Electrical Engineering.

6. The process used in the hard sciences to ‘discover’ something or come up with a theory or law more or less goes like this. A scientist or layperson will make an observation. In other words, he or she will see something happen or that just exists, and want an explanation. The scientist will then use her existing theoretical knowledge to come up with a ‘theory’. A theory is a rigorous, strict, mathematical model of what could explain the observation. The theory includes a predictive phase, in which it claims that if it is true, certain things that have not been observed yet, will be observed under certain conditions. An experiment will be performed to test the theory. If the experiment succeeds, the scientist will repeat the experiment, and then write a paper on it, which will get reviewed. So, for example, we may want an explanation of what water is. A scientist will theorise that water is a combination of Hydrogen and Oxygen, in a 2:1 ratio, formed by heat. She will perform an experiment to test this theory. She will have a ‘null hypotheses’ which assumes that the theory is false, and she will have a ‘control study’ which tests something that is partly unrelated to see if it produces any water, too. So, the null hypothesis will say that water is _not_ 2H2 + O2. And the control will be something else, say; just keeping the Hydrogen in a gas cylinder without heating it in the atmosphere, and seeing if water just appears inside it.

So, next time you hear about a scientific ‘theory’, please understand that it does not mean the same thing as the term ‘theory’ that we use in the phrase ‘in theory, Man United ought to win’. It actually means something much, much stronger. A ‘theory’ is a series of mathematically corroborated facts and/or predictions, in science. It is called a ‘theory’ purely out of modesty. This is because any theory is still open to testing and verification, or further proof or disproof. But theories can _only_ be disproven by proper scientific method, described above. A scientist’s personal beliefs are irrelevant, and a layperson’s opinions are irrelevant. You cannot merely ‘disbelieve’ in the theory of gravity; you’re sticking to the ground quite tenaciously, and that is just a fact. The same applies to other theories; merely disbelieving them does not make the reality of their testability go away. In science, all theories which pass an experimental test, are considered facts. Atomic theory, the Theory of Relativity, and the Theory of Evolution - these are all well-established scientific facts.

7. Once a theory has been verified by experiment, it is usually written up in a research paper called a ‘journal article’. Scientists have their own magazines - called _journals_ - which describe the latest research that they have done. There are thousands of journals. They each specialise in a particular research area. Not all journals are in the hard sciences. For example, you can get journals of history, politics, and philosophy. A paper usually starts with a section called the ‘abstract’, which summarises the paper. When a scientist or other researcher submits a ‘paper’ to a journal, it undergoes a process known as _anonymous peer review_. That is, the people running the journal give the paper, with the author name removed, to the peers or academic equals of the scientist submitting the paper. The author name is removed to prevent the reviewers from being biased against the submitter, or in her favour. The reviewers or peers will then try to replicate or repeat her experiment, and/or they will check her mathematics and reasoning. They will also check that she has done her research properly, that is, read up existing recent research on the topic in their journal and other journals. If she has performed her experiment properly, if her mathematics are correct, if she has ‘cited’ or referred to existing proper research, and if her conclusions follow from her premises of her argument, her paper will be accepted into the journal for publication, with her name now visible to all. Subscribers to the journal, that is, other scientists who buy the journal, will then have a chance to write responses, usually criticisms. The author will then be able to write responses or work on the queries that come along. Strictly speaking, anyone who writes a response to a journal article is welcome to submit to the journal to see if their response is published. So, for example, if you truly believe that evolution is false - go ahead. Refute it in an academic journal. You’re perfectly welcome to do so. There are no entrance requirements, just that the paper is well-researched.

This is how science increases our knowledge. 

8. The more a paper is ‘cited’ or used as base research material in new research papers, the more the work of that paper is respected and ‘rated’. A highly-rated scientist is one who has produced a lot of cited research material. Such a scientist may get earmarked by her university for promotion to professor, and her theory may become accepted broadly as scientific fact.

9. Instrumentalism versus realism. Some scientists consider their theories to be physical facts. They also consider the entities that they describe in their theories - such as atoms, molecules, energy, waves, quarks, etc. - to be real things. These scientists are called ‘realists’. Other scientists, however, are only interested in whether a mathematical model or theory can generate good predictions, and, they are not concerned with whether their theoretical entities exist (i.e., they don’t care if atoms actually exist, as long as the theory works to make predictions). This type of scientist is called an ‘instrumentalist’, because they consider theories to be merely instrumental or useful, rather than fact. My experience of 23 years at a university tells me that most physicists are instrumentalists, but most chemists, biologists and astrophysicists are realists.

10. Funding. Not all scientists are funded by pharmaceutical companies looking for cures. Some are funded by government. They write a research proposal and a funding proposal, and a funding body, like the NRF in South Africa, or the government (DHET), or whatever, fund the research. Often just the university funds it. So, for example, the university research office may slice up the funding pie between different departments; physics and medicine, say. This means that some research funding goes to physics, rather than just medicine. This means that funding is split up from a budget for various projects. This means that not all funding goes to cures for cancer. But if there was no funding for anything but medicine, we’d not know anything about the world… e.g. how to make computers, how to make aircraft. et.c. It is not the business of a layperson who doesn’t understand science to say what aspect of science should or shouldn’t be funded or in what proportion. Many spinoffs come from funding apparently useless studies. So, for example, the astronauts on the moon resulted in Nike sneakers. Bulletproof vests were a result of someone tinkering with chemistry. Microwave ovens were an accidental side effect of a different experiment that wasn’t looking for ways to heat things. So unexpected benefits can come from apparently useless research. Landing on a comet, more recently, has relevance for our knowledge, e.g. of the early universe, and whether astronomical bodies are worth mining commercially.

The only funding in science which is questionable is medical and weapons funding, because sometimes it has corporate interests behind it. There’s no commercial funding for evolution, gravity, relativity, and other theories that have no obvious commercial motives. The way to check medical research for legitimacy is to look for it in journals and see if it has been replicated by people who are not funded by the pharma company. If it has been replicated by uninterested parties, then it’s not a case of corrupt research which is trying to sell people “unnatural” “non-holistic” cures for diseases. Moreover, empirical evidence of the efficacy of “allopathic” medicine makes it perfectly clear that it’s not merely corporate quackery. Most studies of homeopathic and related medical models show that they’re no better than placebo. But this is a debate for another article. Homeopathy is also big business, also a multi-billion industry. So in terms of research being dubious because it makes money - no; 'all natural' medicine is a huge industry. Money is also irrelevant to empirical results. If something works, and it’s in a journal, and it’s been replicated, that’s all you need. Whether a wealthy funder funded it becomes irrelevant at that point, just as the fact that Michelangelo’s work was great, was not disproven by the fact that he was funded by a very wealthy church.

Funding is awarded by applying for it to a funding body. You have to write a long preliminary study showing why your research should be funded, and the reasons can’t be its commercial viability. The reasons usually have to be benefit to humanity.

This is how science actually works. 

php 7 nightmare

OK so Centos 6 insists on installing php 5.3 and even if you download other RPMs and install them, they do not replace the existing 5.3 whic...