Since when are physicians a type of scientist? They can do science, but that's not what the qualification is primarily for.
This is the position I'm in. But I was paid about $25,000 with good health insurance to be a graduate student, now I'm paid about $38,000 with good health insurance and extremely flexible hours in my current dead-end job, so it could be worse. The only college/high school classmates I can compare myself to and think "I would be better off if I had done that" are the doctors, computer programmers and certain engineers.
Goat farm it is, then, with seasonal off-farm labor as a coder. (That sounds sort of tolerable, actually, except that I'm getting old.)
There's a new Slate 'initiative' about how to get more scientists and the comments are, rightly, full of scientists/ex-scientists complaining that there are no jobs or jobs don't pay enough. My field is so dependent on gov't funding that the recent cut backs have nearly wiped out jobs. I want to go back to Canada and that is currently looking like 4+ years and a change of gov't before there might be jobs.
But as I keep reminding myself, there was never really a HIGH demand for ecologists.
Ned's position is very common which is exactly why the wife and I switched to teaching and the cop thing. Generally the dead end stuff we were seeing our cohorts end up in didn't even have the upside of a pension. Fuck that.
The part about people going into debt for a PhD is really weird. Is that a biomedical thing?
6: Yes, I know some people who became teachers (not in states like Pennsylvania whose leaders are abandoning the concept of public education) and are pretty happy.
Since when are physicians a type of scientist? They can do science, but that's not what the qualification is primarily for.
In education policy circles, it all seems to get lumped together under "STEM", along with IT and other things very unrelated to traditional research science. Which seems to be the main problem: the goals of most science education and the nature of the STEM jobs opening up don't seem very well-aligned. An education in a traditional research science isn't necessarily going to help you data mine for large corporations' marketing departments, which seems to be the kind of "STEM" field where there actually is a lot of job growth.
I briefly thought of going into teaching (most of my extended family are teachers) but, because I was on the grad school track, I would have had to do at least 2 more years of undergrad.
And, as things have turned out, my family members are having a really hard time getting full-time permanent teaching jobs. Only my cousin who teaches in French (and whose father is a principal) had an easy time of it. My aunt and other cousin who teach science/math had to work as subs before getting a job. Well my cousin gave up and moved to the north where she's starting again.
It's always a great challenge to go where the puck is going to be, rather than where it is. Most of us can't manage it, or advise our children well.
This situation has gotten so bad in Germany that quantum chemists are resorting to becoming Chancellor as an alternate career path.
At no point in my career have I been surrounded by trainees who all want their own lab. There are a few people who want their own lab and they have a good chance of getting it. Most people see that the faculty spend most of their time writing grant applications that have a 10% chance of being funded, under pressure of having to shut down their lab if they run out of grant funding, and either say "That sounds horrible" or "My grant applications would never get funded so why bother". People say they'd like to do something less tenuous and work for a drug company, or the EPA or CDC or something. Problems seem to arise from this, though.
- Drug companies aren't hiring people in the US much anymore
- The US government isn't hiring people in the US much anymore
I think people are fully prepated for "postdoc" to be a career instead of a supposedly temporary position. People who stay in the same lab for years and years represent a ton of institutional memory. But at the moment, the only people who stay in the same lab for years and years are "technicians" who don't have advanced degrees.
The part about people going into debt for a PhD is really weird. Is that a biomedical thing?
No. I can't speak for more "field biology" type areas such as ecology or evolutionary biology, but if you go for a pH.D. in any of the biochemistry, molecular/cellular biology, biophysics & etc. type fields without being provided with full tuition + stipend then you're doing something seriously wrong.
Masters level scientists with an actual dissertation are slightly more employable, and chemists have more flexibility than biologists.
5 years ago Bill Gates spoke at the Harvard Commencement. There was a woman next to me who had a Ph.D in a science field who said, "I hope he doesn't talk about how we need to train more scientists. A bunch of my friends just got laid off by [insert name of Pharma company]. There aren;t any jobs.
||
Did anybody here go to Dartmouth who would be willing to send me an e-mail? I have a question about their alumni e-mail system.
|>
I think CCarp gets it right, and I need to stop assuming I can give my students any career advice that isn't simply cautionary (e.g. "Don't go to law school").
A friend of the family has a Ph.D in biology from Berkel/ey, did a postdoc at an Ivy League school in the same circadian rhythm stuff he was working on in hamsters and went to a sleep lab for people for another post doc. He applied for jobs and got none. He can stay on at the sleep lab on some kind of cheaper trainee program that usually goes to foreigners. He has a wife and two kids. She's a doctor but not qualified to work in the U.S. and is 10 years his senior. He's loath to move for a non assistant professor job, because he's living in the apartment attached to his parents' place rent-free and has some help with childcare.
I think he's considering getting a teaching credential. And he has a lot of publications.
7: no, that's just nuts. If you're going into debt to get a bio PhD, ur department is doin it rong. Obviously you may go into debt while doing a bio PhD, if you have to handle medical crises or support a family or whatever. But you should never ever have to pay for school itself, and you should get adequate stipend support from your advisor's grants.
18: I think that the friend I mentioned above kind of ran out of savings during his postdoc even with a subsidized apartment. Manhattan is so expensive, and even with cheap dance things, supporting a family was a lot. But he didn't actually go into debt.
It's always a great challenge to go where the puck is going to be, rather than where it is.
Definitely.
Things have worked out OK for me but that's because I did a postdoc using an obscure technique that was just about to take off and become very popular (got to learn form the technique's inventor).
If I'd gone for one of the hot topics of the day I'd probably be out of science right now.
The thing is this wasn't a case of superior judgement on my part. It was basically pure luck. Unfortunately, as advice goes "get lucky" is about as useful as "choose your parents wisely".
Ph.d.s in mining engineering seem to do okay. I don't know whether that counts as science if you're doing research. There are industry jobs available.
I think people are fully prepated for "postdoc" to be a career instead of a supposedly temporary position
Really? Wow, not IME. My cohort wanted either to postdoc for the necessary few years before trying for TT jobs, or to get the hell out and go to industry.
I have a friend who works for JPL and NASA for a few grants in a row, then for oil exploration firms long enough to refill the family coffers. It's not that the nominal JPL salaries are particularly low, but the work is so sporadic and unpredictable that the actual income is tiny.
There's a real Ponzi-scheme aspect to today's "sectors with job growth". Both education and health care are expensive services much of whose value is not real but taken as given, especially to the extent that they're growing. And like finance and consulting, they're both essentially support services that presuppose a healthy rest-of-the-economy that can afford it all.
I dislike the term meta-economy because that conflates such things with art, primary education, and other intangibles that are perfectly legitimate, but something's wrong in that vein.
There was actually a wife of a semiconductor engineer who asked Obama this question. His response was essentially "But they tell me there's an engineer shortage!" Well of course they do. They also tell you that taxes are too high and regulations are too heavy. Obama offered to pass his resume around. Full service president.
As a generally ignorant person, I would have thought that for a majority of students, there's a diminishing marginal utility in going for the PhD rather than sticking with a masters, for most non-academic applications of science. Does an oil exporation company really want the guy with a PhD in geology other than as a head of the geology division (a position that turns over rarely, and is as likely to be filled from within by someone with an MS and 15 years experience)?
Comment 12 isn't that far from the truth. You have at least one very-long-time lurker who was a lab scientist back in Ogg's days on the blog, and has since left the rat race of science to become a university president....
27: Master's degrees in the sciences are kind of in a weird limbo. A PhD is a qualitatively different thing from an undergraduate degree: it means you're capable of doing science on your own, instead of just sitting in classes and learning about it. Master's degrees at US institutions in (at least some) sciences usually mean you started on the PhD track and stopped somewhere along the way, and thus might not be much more than glorified undergraduate degrees meaning a person got through a year or two of extra coursework and a qualifying exam. It's different if it's meant to be a terminal Master's program, I guess, or in some of the European systems, where the Master's degree is evidence of doing research.
27: I'd bet that's true for the majority of students, but there are niche fields where companies very strongly prefer PhDs. They seem to be fields that are so obscure (or, if you prefer to be positive, cutting edge) that the only way to get good lab or project experience is with a PhD.
When I was shopping for a specialization, I remember talking to professors optics (stuff like ultrafast or non-linear, not classical), RF MEMS, and FV (automated theorem proving), and getting warned that it would be basically impossible to get a job without a PhD.
You can make a lot more with a B.S. in EE if you're willing to take a job doing software, so the monetary returns are not only diminishing but negative. But, some people like optics more than programming, just like some people like bio more than programming, so there aren't necessarily diminishing returns to utility or personal satisfaction or whatever you want to call it.
30.1 -- That's what I meant.
30.2 -- And you still have the puck problem because as a field looks like it presents opportunities, people are rushing in, so the pipeline gets crowded (I would guess this happens much faster now than it did a generation ago) and you always run the risk that needs change enough a decade down the line that your hyperspecialty no longer commands a premium. Shorter: as AL noted above, there's a lot of luck involved.
i knew a korean grad student, female, that's important, who would stay in the lab until 2-3am to collect data for another grad student, male white, who would leave flexibly whenever, cz they worked for the same PI and the PIs decide (in their lab anyway) which data goes to whose paper, it's fast disillusioning how all the science works, grants writing and their allocations procedures, everything goes through connections, and the same jungle rules work like everywhere, the more powerful exploiting the less powerful, so it's all exploitation of ideas and labor, and foreign postdocs slaving all around, if all the scientific immigrants magically were to disappear one day almost all the research labs would become almost empty, perhaps
the job ads are as if like so revealing, people need this or other assistant only, not many real positions are out there, looks like nobody likes to work just prefer to hire an assistant who would do all the hard work for the minimum payment and will be just mentioned in the presentations as s/he has golden hands to collect the data
presented, no much other credits, except publications of course, no real perspective to grow as an independent researcher too, if you are female and a foreign science degree holder
but what to complain, sure, just that the US is meritocracy seems like a myth, unless one is a real genius, perhaps, not your ordinary researcher
Being in (academic) Computer Science myself, I probably have a very different perspective from that of people in Bio or Chemistry. Our PhD graduates have little problem getting decent-paying and relevant jobs. But a few points:
- as others have pointed out, nobody should finance their own PhD education. If a school admits you without tuition support and stipend, DO NOT GO. Take the hint. Admission without support in my field means "yeah, we sort of like you but not enough to give you support, so go somewhere else".
- having just 14% of grads achieve faculty positions is not the problem. The problem is not having a suitable job market for the other 86%.
- seems to me that this might be another example where the need to shape a simple message for public consumption interferes with accuracy. It is so much easier to say "we need more scientists and engineers" than "we need more people in A and B but not in C or D".
Kenneth Arrow (and Armen Alchian and Bill Capron) wrote an economics paper for RAND about the "shortage" of scientists and engineers in America. This was in 1958.
how can you make such racist comment as 32?
7: The part about people going into debt for a PhD is really weird.
I have no idea about the sciences, but it's not uncommon in the humanities. Which is not what this thread is about.
Yeah, the length the pipeline is a problem, since it's impossible to tell where the puck is going. Worse yet, a lot of students don't even know where the puck is right now.
The optics job market fell apart* over a decade ago, and there's still a huge number of students coming out with optics PhDs. The cause seems to be that universities hired a lot of optics folks back when optics was hot. Those professors aren't going to stop doing research just because the job market sucks, and they need students, so they recruit students. I'd say that 80% of the grad students I talked to had no idea what they wanted to do, so they just did whatever their professor told them to. Most didn't really shop around, and went with whoever offered to fund them first.
I suppose you can blame students for not researching things themselves, but I think the institution should take some of the blame for absolutely never talking about this sort of thing at all. But, if you try to bring it up, you're told that "we're not a trade school", as if even thinking about money or employability is beneath students.
I purposely chose a specialization that isn't the most lucrative thing around, and I keep turning down offers to move into a 'better' field, so I can understand that some people would prefer to take less money to do work that's personally more interesting. I think that should be an informed choice, though, and it seems to me that schools are purposely keeping students uninformed.
* it's not like people graduating with optics PhDs are destitute, but people I know who are good, but not brilliant, are pulling in 70-80k out of school. Good, but not great, people can make $100k+ with just a BS if they're willing to take the right kind of job.
but people I know who are good, but not brilliant, are pulling in 70-80k out of school
Am I reading you wrong, or are you saying this is evidence that the job market "fell apart"? That doesn't look like a job market falling apart to me.
That doesn't look like a job market falling apart to me.
In fact, that looks like the job market I should have gone into.
29: Canadian Masters (M.Sc) require a research project which is a fully defended dissertation. I know that at Ottawa you can start straight in biochem in a Ph.D. program but that a masters with a research project is often done before admission to the Ph.D. program.
44: Right, Canada is different, although from talking to the head of graduate studies at one Canadian university not that long ago, it sounds like at least in their department they're gradually transitioning to the US model (in that students are typically doing the MSc and PhD under the same advisor, and thus effectively combining the two into one PhD program).
Of course, I know a hack who plagiarized a lot of his PhD thesis who's making $200k+ out of school, but that's Wall Street for you.
I'm not heebie.
In 32 "jungle rules" is clearly racist.
Highly location dependent- we have trouble getting non-PhD science people because we're competing with all the pharmas moving to Boston. We have three open positions and we're going to be opening about 4 more. PhD level is tighter just because of higher salaries and fewer openings. Growth area that I see is bioinformatics, and my general belief is you're more successful if you're chemist and learn biology on the job, or a chemist or biologist and learn informatics, than to start from the other end (computer science person learning biology, or biologist learning chemistry.)
The assumption in the article that a PhD = academic track is silly. I know PhD chemists who have done consulting, patent law, finance/investing, teaching, pharma, as well as traditional academic, and not just as a backup, but as careers where a PhD is required.
Ooh, are we playing 20 questions? Is read more racist than a breadbox?
sorry for the lack of caps. my phone seems to fail at capitalization in forms.
if spending an additional 7 years in school beyond a b.s. reduces your earning power by $30k a year, that doesn't seem like a very productive degree
45: of course totally field and school dependent. U of Toronto/Alberta/BC definitely encourage going right into a PhD or rolling over into one. But in the other schools, it's less common (all of this is in my field and related biologies). In Canada though there's never been the academic jobs of the US and there's the further confusion of either hiring foreigners or Canadians who have done some schooling out of Canada.
Anyway, most of the gov't jobs I know either require a PhD (research scientist) or are open to BScs with experience. Of course now they're hiring PhDs because why not? (sigh). I even know some people with high school deplomas being replaced with MSc
7 years? Yikes. But still, presumably one doesn't go get a PhD just because they want to make money.
28.--Drinks are on Thomas Jefferson!
28.--Drinks Jobs Drinks are on Thomas Jefferson!
Starting from the same report, Frank Pasquale has a decent framing of this at Balkinization.
53: true, and there's nothing wrong with that if people know what they're getting into, but the impression I get from most people is that they expect the degree to increase their earning power if they go into industry. they also expect the degree to take 4 years, which is wildly optimistic, based on the actual statistics.
I've been ABD since Clinton's first term.
To be fair, it's possible that I've merely dropped out.
I think we need to invent a Mineshaft Honorary Doctorate and give one to Moby.
Even better, you could maybe write a nice letter to my committee explaining that I probably should have formally dropped out.
The MhD for MH. A little like Major Major Major Major.
59: Really? Most people I know doing PhD hope it will take 4 years but know it isn't likely and figure their employment potential will be different though not necessarily better with a PhD. I think incoming PhDs are fairly realistic and enter because getting into a PhD program puts off the job search for another however many years. And it's a reason to move somewhere new with a little bit of stability. Finishing a PhD may not be the best career move but starting on one can be.
|?
Long term job opportunities for scientists in
...with the correct forms of gov't of course, if we're lucky. Or brave.
We ain't got a freaking generation. We ain't got five years. AGW is hitting now, will not hit uniformly (Dallas is ok right now, folks, barely above average. And wet.), and the outliers and extremes will be disastrous for everybody, because we are all connected.
For whom does the asphalt melt? It melts for all.
|>
||
Anyone with any interest at all in the media should read the excellent post at SCOTUSblog by Tom Goldstein giving a very detailed review of the initial reporting of the ACA decision. Excellent stuff and I found it kind of riveting even though I knew how it came out in the end.
Via LGM.
|>
65: Well, that's just my impression. I'm subject to my own biases, and I certainly haven't talked to people at a random sample of schools, and I don't even have a random sample of people at the two schools where I've known a lot of grad students.
I think part of my experience may come from my age. I finished undergrad in 2003, and there have been a lot articles written about why people shouldn't go to grad school since then. Back then, phds.org had a few articles about how the alleged STEM PhD shortage was a crock, but my recollection is that most people dismissed anyone who brought that up.
But, I still meet people who don't know what they're getting into, even now. A friend of a friend just dropped out after three and a half years, after his advisor suggested that an academic career wasn't going to happen with him only publishing a couple papers have three years. He was devastated; he was really shocked that a PhD wasn't an automatic path to an academic career. I don't know how common it is to be that naive now, but there I still meet people who are totally clueless about the EE job market.
Why am I getting the sense that pursuing a Ph.D. in the humanities as well as the sciences is not something people should be interested in doing any more?
That is, I know that financially it's not, and the job market for pretty much everything sucks, but it's beginning to sound as though the actual knowledge that pursuit of a Ph.D. provides is considered shrug-worthy. Which is strange.
We wouldn't really want that as a society, would we?
@70
We were (grimly) chuckling over what bullshit the famous "imminent shortage of scientists" study from the 80s was when I was a grad student in 1995.
I guess the word didn't get out as widely as it should have.
71: I think these are positive statements folks are making, not normative ones, parsimon.
71 came out sounding naive. I mean to say: in the humanities it's long been known that the additional knowledge value of a Ph.D. is considered useless by most of society. I just didn't realize that it was becoming the same in the sciences.
69: Even as someone intently following the news at the time, I don't understand the plaudits that article's getting. We already knew CNN and Fox were so anxious to be the first-to-break they misread the decision, and others weren't. All I see in the article is predictable details being filled in, with a bit of color. OK, CNN weren't in reality the idiots they made themselves look like, but so what?
48: My BF applied for an MSc job at your place maybe 5 years ago. The pay was significantly lower than a comparable pharma job, and he felt like a Masters candidate got more respect at pharma, that you wound up being more of a glorified technician and less of a researcher.
74
71 came out sounding naive. I mean to say: in the humanities it's long been known that the additional knowledge value of a Ph.D. is considered useless by most of society. I just didn't realize that it was becoming the same in the sciences.
The knowledge value of a PhD in the sciences is often not all that great. You can spend a lot of time on your thesis work which is on some esoteric unimportant problem which no one has bothered to solve before. If you don't pursue an academic career it likely won't be of much use. You generally do not spend your time in graduate school obtaining a broad (but shallow) overview of the entire field which would be of more value in later life.
The value of the knowledge of a Phd is tied up in there actually being a job available that uses that knowlege. A scientist isn't someone who knows a lot about science. Ascientist is someone who does science.
And yet people who do get jobs are a bunch of incompetent hacks. I might have to stop reading the arxiv; a listing like tonight's is going to give me an aneurysm. My offhand comments on facebook constitute roughly the same amount of thought that's going into published papers. Christ.
Yeah, we're lower paying than industry, but our prestige is edible.
"A scientist isn't someone who knows a lot about science. Ascientist is someone who does science."
Ha. Guess I'm halfway to being a scientist.
76: The preparations, the role of SCOTUSblog, who was where at the White House, how various outlets reacted, who gets a press pass etc., The CNN and Fox narratives are mere side stories.
81: Have you tried reading real journals?
85 made me laugh, which is mean. I have no idea if real science journals are any better than arxiv.
The knowledge value of a PhD in the sciences is often not all that great. ...You generally do not spend your time in graduate school obtaining a broad (but shallow) overview of the entire field which would be of more value in later life.
The knowledge value of a PhD is identical with the the value you give the subject and the role of the subfield within the subject.
I did my PhD on Wittgenstein's late philosophy of mathematics, a topic I haven't touched since. I still think it is a good topic worth knowing about. On the other hand, I gather Heebie doesn't care a whit about the math she wrote about although it was a most serious endeavor.
85: then I would be reading the same crap months later.
At least Facebook gives me a way to semi-publicly shame them.
87
The knowledge value of a PhD is identical with the the value you give the subject and the role of the subfield within the subject.
This is all very well but the topic under discussion was the value given the knowledge by "most of society".
Are STEM* workers in demand in other parts of the world? (Cue Slashdot argument about the vast number of meaningless BRIC certifications, and the tiny fraction of those -- still a vast number -- given to superb thinkers.)
*This is short for 'you have to pass calculus', yo. Hence the inclusion of MDs.
88: essear, does physics have a problem with people spitting out papers in 10 days (or however much)? Why is that?
I mean to say that one would never ever write and attempt to publish a paper in literature or philosophy or history in under, oh, three months, for the truly ambitious, and more like six months to a year for the thoughtful. Or [grimace] maybe that's just me.
82: Right. I think that the gap between industry and your place is greater at the Masters level than at the Ph.D. level.
He also found the attitude of some of the Ph.Ds. condescending.
93: right now it's because the Higgs was just discovered and people are scrambling to grab credit and citation counts by vomiting papers I could have written better in half an hour onto the internet.
No one should go to grad school or professional school, because there are no jobs when you finish. A bachelor's degree is essentially useless, and these days you have to take out too much debt. High school is a waste of time, middle school is awful, and elementary school does not inculcate the skills that today's employers are looking for. And I admit I have rarely used anything I learned in kindergarten at any job I've had.
All I Really Need to Know I Learned on the Veldt.
I'd be in blog love with Bave if he weren't already taken.
96: You do sort of wonder if the people giving the advice aren't actively trying to discourage kids from trying anything. Maybe the daytime TV people figured they needed a bigger audience.
It is a problem when Obama is promoting science education and there are no science jobs. If he was promoting legal education, that would be a problem too. He should do something about promoting more jobs.
But banks are getting money for free from the government! How are we supposed to create more jobs, other than by giving banks extra money that theoretically they could then loan to nonexistent entities that if they existed would employ people?
We had a huge recession, the economy shrunk, and labor was particularly hit hard. This affects every job in the country, including science jobs.
102
We had a huge recession, the economy shrunk, and labor was particularly hit hard. This affects every job in the country, including science jobs.
Obviously the recession made things worse but there was and is an underlying problem. Despite self-interested predictions to the contrary by industry groups there is no general shortage of science PhDs. Making obtaining one a dubious proposition particularly if you aren't top ranked. And if you aren't top ranked going into grad school (which means being admitted to a top program) the odds are you won't be top ranked coming out. So encouraging marginal students to obtain science PhDs is likely doing them a disservice.
99 - the daytime TV industry: also declining.
101:QOTD, from Brenda Rosser
"Insolvent central banks are lending money to insolvent banks who buy government debt from insolvent governments who lend money to the IMF which then lends it to insolvent governments to pay back insolvent banks."
105 is wrong in one very important respect.
You do sort of wonder if the people giving the advice aren't actively trying to discourage kids from trying anything.
"Lisa, you tried and failed, but you learned a very important lesson - never try."
You generally do not spend your time in graduate school obtaining a broad (but shallow) overview of the entire field which would be of more value in later life.
I can't speak for the sciences, but my own undergraduate and graduate education in philosophy did _exactly_ that. In fact, you couldn't have gotten in to the doctoral program I did without evidence of a decent amount of taught graduate study, and a substantial undergraduate education in the subject. I taught a few people who had undergrad degrees in philosophy who were nevertheless being required to take a few extra courses in philosophy -- because the department judged they hadn't done enough -- before they'd admit them to the 'masters',* which was in turn a prerequisite for the doctorate.
* i.e. http://en.wikipedia.org/wiki/Bachelor_of_Philosophy#University_of_Oxford
106.1:Rosser laughs at that in comments
Krugmanophilia. Techocractic idiocy.
"Central banks can tax, cut spending, and print unlimited amounts. Nations with sovereign currency can't go broke."
"And if they won't? If they don't?"
"Umm. Central banks can tax, cut spending, and print unlimited amounts. Nations or sovereign currency areas can't go broke."
"Tell to the Greeks, Irish and Spaniards. Tell it to the 100,000 American teachers fired last year. Tell it to Obama, who said America was broke."
The difference between "insolvency" and "willful default" isn't helping the long-term unemployed whose UI is ending or those who are losing their food stamps.
108
... is that fair?
"Life is unfair" as JFK famously observed not too long before some madman blew his head off.
"There is always inequity in life," he mused. "Some men are killed in a war and some men are wounded, and some men never leave the country, and some men are stationed in the Antarctic and some are stationed in San Francisco. It's very hard in military or in personal life to assure complete equality. Life is unfair."
109.1: no, that's not the wrongness I was referring to.
Tell to the Greeks, Irish and Spaniards.
OMG, bob hasn't heard of the euro!
Next:homo economicus says:"Default? We aren't in default."
Well, it's funny what kinds of social obligations are binding, because both 10 year treasuries and food stamps are social obligations. We can only "default" on the former?
(Anybody know why DeLong went off on Graeber yesterday? Anybody care?)
||
Even FDL hasn't gotten the memo.
"Obama Adviser Says President Will Not Extend Bush Tax Cuts, Even Temporarily: '100% Committed'"
New Plan, locked and loaded, temporal order
1) Election. Doesn't matter (for this purpose)
2) Tax cuts expire, spending cuts = over fiscal cliff, Shock Capitalism? Oh noes? So scarey!
3) New Congress, New or repeat President, balance much doesn't matter
4) Something somwhere between Simpson-Bowles and Ryan Plan
5) Whichever President, numbers in Congress will play a part in where we land on the line S/B to R, but since S/B is to the right of my worst nightmares, I really can't give a fuck.
6) Four more years, worse than the last four, but not quite bad enough.
|>
109: bob, I'm surprised that you, of all people, are falling for the "insolvency" rhetoric. In today's environment, insolvency is code for class war raged by the rich on everyone else. This article about Richard Koo talking to German leaders makes it as clear as it possibly can be, without saying it outright.
113: Well, neither will I fall into the Krugmanism
"Well, the central bankers are well-intentioned people who, umm, are misguidedmisinformedgodwescrewedup macro. They can print as much as they need! Here, I've proved it beyond doubt!
So. Print!
Now Soon Any minute I'm waiting. I'll make another speech. You can print now. What's going on?"
It's ain't a class war until both sides have declared it and started fighting. I'm also waiting. On Krugman.
I'm late to this post, but I'd like to echo the points made by 48. There are many valuable careers for people with science PhD's that are not in academia or research. We hire a lot of people with advanced degrees to help with patent prosecution and the other non-research company that we compete with for talent is Gold/man Sachs. Now, Gold/man might only be hiring people with a PhD in EE or CS, but we hire PhDs in fields like bio-engineering as long as they have coding experience.
48 and 115: But that's just so weird. It sounds like you're both looking for programmers and you're settling for people who take up programming as a hobby or as an interest in addition to their major interest. I guess that will get you people who are very interested in programming.
It seems like 'We hire PhDs (to do something other than their field)' isn't really an answer to the job shortage.
I went to a top-ranked science/tech program, so I guess the fact that almost all of my cohort got fine jobs (though damned few in academia and several in things like consulting, finance, and government) is not worth noting?
116- Well, obviously some of those non-research jobs are still relying on the science training (interpreting Markush structures in chemical patents, consulting for area-relevant companies.) But in general a lot of PhD training is also guiding your own experiments, record keeping, knowing how to do research using the literature, and specific things like coding in certain languages. Some of it is credentialism too, of course.
116: I should also say that my firm does 98% software/hardware patents so we're not a perfect match for non-CS/EE people. Lots of patent firms do biotech and pharmaceutical work where people would be better placed. My general point is that I think we should still be training people in the sciences. I have a BS in chemistry and while I do often wish I had been chemE or chem/CS, I'm glad I focused on the sciences.
I disagree with the extreme gloom. Within-field stable lifetime employment is unlikely for most science PhDs, but so what? Getting to do research science for a few years is a fantastic experience, intellectual mountain climbing, another wasteful pursuit that nobody should do ever.
The people I know who are succeeding in-field either walk on water or chose a very practical subfield. I am a physics PhD 1996 from a good but not top-tier department. My classmates seem mostly content with their jobs-- some teach, some do computer work, most have found other fields that use part of their training.
115: It's interesting that you bring up both patent prosecution and Gold/man Sachs when talking about "valuable careers", because there's at least a colorable argument to be made that the growth in both the propertization of ideas and the financialization of the broader economy are, in fact, Bad Things; they are certainly part and parcel of the broader shift in the political economy of the US over the past generation and a half (which most folks here treat skeptically). So while it's true that these are certainly socially respected and remunerative careers, this only highlights the broader dilemma the country seems to be in. Are we producing folks with the skills to produce real wealth, or are we shuffling around the chairs for who gets to claim their outsized share of a not-really-growing pie?
In other words, what Minivet said at 24: "And like finance and consulting, they're both essentially support services that presuppose a healthy rest-of-the-economy that can afford it all."
121: Fancy forms of guard labor.
What Ttam says in 107 about philosophy applies in many ways to my experience in history, only I still don't have the doctorate. The obvious problem for the history phd as it was when I got started is you didn't learn nearly as many identifiable* skills that crossover into other fields of work as you do in the sciences and social sciences. Kids these days who get themselves into programs with some kind of digital component are probably going to be better off, but mostly still not research professors.
*I think identifiability is key. People say that they value research and writing skills, but those are harder to show to a potential employer who doesn't know you and doesn't have the time to get to know you than skills like coding, quantitative methods, etc.
I've been reading Acemoglou and Robinson's blog, and I've been thinking about their overarching idea that what stalls economic growth in poor and middle-income countries is "extractive elites" who rig the country's rules in their own favor. It sorta fits what's been happening in the US.
It doesn't seem possible to me that writing well in English is harder to demonstrate than writing well in a computer language. I can believe that many employers act as though it is harder to demonstrate.
My favorite group blog! covering one of my favorite subjects! It's like an awesome sandwich.
116/119:
Folks seem to think that there's less demand for PhD chemists (esp. organic chemists) to go into IP/patent work. True, untrue?
Read. What do you want? Do you want someone to say that "all that jazz" is racist? No one will do that, because it's not racist. It's an idiom, a turn of phrase, that generally means "all that stuff". The word "jazz" came to be used because jazz (music) is often improvisational and operates as variations on a theme (a musical theme). To say "and all that jazz" is to say "and things that are variations on that theme".
So please relax.
I think that deletion will continue until you stop shouting that "all that jazz" is racist, and making other insulting comments. Lots of people here have written comments that no one else agreed with. If people don't agree with you, and don't respond to your comments, usually the thing to do is find some other thing to do, rather than insist repeatedly, with insults, that people listen and reply.
"All that jazz" can be used to describe something right that's occurring as well. So please, just let it go.
Parsi, the best thing to do -- other than getting read the help that she clearly needs, but which we're in no position to provide for her -- is to disengage.
Read, I think that your comments are being deleted because you're insulting people.
I'm out.
124: Walt, am I off base in thinking that fundamentally we're dealing with the fallout from an immense supply of new, cheap labor? This is used to undercut domestic wages (and politically as leverage to dismantle public services and labor laws), and, for a while, the consequent drop in demand was papered over by supplying more and more cheap debt to consumers nd some asset bubbles. Eventually, being unsustainable, this stopped, and now were dealing with the aftermath of inequality and deficient demand.
And I'm sure there are other scams going on, and the elites are taking advantage with shock doctrine power grabs.
And, to avoid sounding too protectionist, I should acknowledge that free trade has likely lifted a lot of people out of poverty. I'm just not convinced unrestricted free trade was the best way.
92: Are STEM* workers in demand in other parts of the world?
I don't know the numbers on this, but anecdotally, yes. Singapore has been recruiting a lot of researchers in biology. (Not a lot in absolute terms, but a lot for such a small country.) China also has pretty massive funding for research, although the work environment is not so great in other respects.
It doesn't seem possible to me that writing well in English is harder to demonstrate than writing well in a computer language.
Actually, that kind of rings true, because to demonstrate that your code is good, you just have to compile it and run it and either it crashes or it runs slowly or it runs fast. But demonstrating what good English is is a subjective and context-dependent thing. Is Charles Bukowski a good writer of English? He is? OK, so shall we hire him to write user documentation for MRI machines?
shall we hire him to write user documentation for MRI machines
YES.
shall we hire him to write user documentation for MRI machines
That's one for the "Paperback Writer" thread. Alas, I have far too much work to do this morning or I'd give it a go.
In the course of running a research lab, I've worked with a variety of smart people, generally with undergraduate degrees from good universities.
From that vantage point, my experience is that people who can write - and by "write", I really mean write coherent paragraphs in English. Forget about style - are rarer than hen's teeth.
They typically take whatever information they are trying to communicate and throw it on the page in no particular order, with no differentiation between the big important points and the trivial details.
When it comes to "transferable skills' I'm surprised that decent writing isn't valued more. Surely an organization will function more efficiently if you can actually explain to people what's going on and what needs to be done in a clear, coherent fashion.
re: 147
Maybe you need to hire people with a humanities background to do your writing? [cough]
...no differentiation between the big important points and the trivial details.
Aaaargh! I have a colleague who's like this in all his communications: "The gaffer's tape is in Bay2, that capacitor is at 30 kilovolts, I'm hungry."
143: I don't think that's sufficient to show that code is good. Code compilation is a pretty low standard; having it not immediately crash nor run too slow is only a bit better. There are other failure cases, like crashing on a weird edge case or subtly corrupting data. The code could even be perfectly correct but written in a style that makes it hard to extend.
I think in general, the difficulty of demonstrating skill in English versus a programming language really depends on the environment you're writing in. In a research lab, code reuse (and readability?) is less important than getting the data, so the code might not be held to a very high standard. On the other hand, there are subtleties to writing a good grant proposal or research paper that might be difficult to judge at a glance.
The opposite situation exists in a software shop (my personal experience): code is held to a high standard of being clear, modular, and reusable, while written English generally only needs to be understood by those who already have an idea of the underlying system. And so my communication skills have atrophied to a point where I mainly use grunts and hand gestures.
147.4: I think it's common for people in STEM fields--especially but not exclusively younger people--to underestimate the difficulty and importance of good writing. The suggestion in 148 isn't bad, but it doesn't solve the common problem of wanting the person who knows the details and the person who is communicating them to be the same.
...no differentiation between the big important points and the trivial details.
And so my communication skills have atrophied to a point where I mainly use grunts and hand gestures.
Generally speaking I'm a better writer of English than I am of code (which may be a scary thought) but I find that it's really difficult to write or speak clearly about projects that I'm currently involved in -- precisely because it's hard to shift gears from whatever trivial detail I'm working on at the moment to the big picture which somebody else will be interested it.
If I don't make an active effort to switch frames of reference my natural tendency is to say things like, "this software will cure cancer, bring a tear of joy to your eye and <voice rises excitedly> in the latest version we've fixed a long-standing bug in which changes to font selection made while running in compressed mode didn't last beyond the current session." (or whatever problem I've been working on).
It's always a little disconcerting when I catch myself doing that, because it's such an easy pattern to fall into. To write something useful I have to really step away from what I'm working on and often it can take an interval of time and a couple of drafts to do that.
138: I think there's something to that, but it's not the whole story. The effect of cheap labor is muted by the fact that the cheap labor is (currently) less productive per hour than American labor.
I think it's really the end of Communism that did it, though. I think extreme inequality is the "natural order" of capitalism, just because you can only work so many hours in a day, but you can own unlimited amounts of property, so the money naturally flows up the socioeconomic scale. The elites were less rapacious when Communism was around to show them they could do much worse, but now that that's gone we're reverting to the status quo.
to demonstrate that your code is good, you just have to compile it and run it and either it crashes or it runs slowly or it runs fast.
This seems about as much a proof of good code as not having any squiggly underlines in Word is proof of good prose. What was it supposed to do? Does it? How sure can we be?
Several lectures at the Cosma Shalizi Shout-out Singalong this June touched on actual programming now being more-or-less automatable, once you have acceptance/fitness tests written. Coders gonna get automated away. Testers and requirements-finder-outers last employees standing.
requirements-finder-outers last employees standing
"Well-well look. I already told you: I deal with the god damn customers so the engineers don't have to. I have people skills; I am good at dealing with people. Can't you understand that? What the hell is wrong with you people? "
153.3: What? Programming is not now more-or-less automatable.
153: Cosma is surely smarter and better informed than I am, but that sounds unbelievable to me. Was it restricted certain problem domains (e.g., CRUD)? That general claim has been put forward every year for decades, and it seems less true now than ever before, in terms of really existing development processes.
I work in one of the few problem domains where people actually take testing seriously, and despite having a pretty complete set of tests at the beginning of each project*, projects are taking longer than ever despite more re-use (not just at my company, but at every company I've heard of).
*well, compared to the level of testing in 99% of the rest of the industry. Compared to the total state space, we test an infinitesimally small part of our product. We also use formal verification, but, ironically, those methods don't scale well enough to apply them to the areas where we actually have serious bugs.
Cosma Shalizi Shout-out Singalong
Huh?
155: ++. You clearly would make an excellent requirements-finder-outer.
No, no, it wasn't Cosma saying this, but as far as this blog is concerned the Compleχ Systεms Summer Schοol was the, etc.
sral; genetic algorithms on the -- object tree? parse tree? of working code, against acceptance tests. Newish results. Not exactly just experimental results, because they were hunting known bugs in working code, *particularly* big annoying codebases. I'll dig references out of my notes if you're curious.
160: Oh, ok. If you're talking about automatic bug fixing and not creating a 'good' solution from tests alone, I think I already know what you're talking about. If not the exact papers, then similar work. Thanks for the offer, though.
160: I'm also interested; thank you for looking into it. Genetic algorithms seem like a great extension of verification, static analysis and test-driven development. But as a working developer, I'm skeptical that they can make programming "more-or-less automatable." (I also have my doubts in power looms and steam-driven pumps.) And of course, turning specifications into machine-readable data *is* programming. Or a part, of it, anyway. Eliminating some other parts of programming doesn't make it "automated" in a general sense any more than, say, automating the construction of stack frames at the beginning of function calls. But I'll reserve judgement until I see your references.
In general I think there's a very high correlation between people who do good science and people who can explain it clearly, although it's convolved with language barriers in ways that sometimes obscures it.
it's convolved with
Rad.
Eliminating some other parts of programming doesn't make it "automated" in a general sense any more than, say, automating the construction of stack frames at the beginning of function calls.
Apt.
In general I think there's a very high correlation between people who do good science and people who can explain it clearly,
I think this utterly falls apart in math.
Don't bore me with your coding details- I'm the product owner, laydeez.
165: duh
166: well that's because math isn't science. It's math.
I think this utterly falls apart in math.
Proposition: Sufficiently abstract ideas are impossible to explain in more concrete terms without losing most or all of the meaning. Almost all modern research mathematics is at that level of abstraction.
What about analogy space-to-analogy space mappings?
Analogies are the lossy compression of abstraction.
||
Where's a bike thread when you need one?
Cargo bike for carrying multiple kids- xtracycle, bakfiets, bullitt, or Madsen bucket?
|>
169 seems like the wrong argument to me. It's not as if what I do is substantially less abstract than what heebie does. I think it's that the culture of mathematics (at least of certain parts of mathematics) has for some reason decided that the important thing is to state and prove theorems, but not to explain in writing the motivation for studying the questions at all. I know I took math classes in college where it wasn't until studying for the final exam that I had any idea why we had done any of the things we had done in the class; just a steady barrage of definition-theorem-proof until there was a chance to step back and look at the bigger picture.
152
... The elites were less rapacious when Communism was around to show them they could do much worse, but now that that's gone we're reverting to the status quo.
This doesn't make too much sense. The Communist countries were just as stratified (if not more so) than the capitalist countries. Of course we didn't want to get conquered by Russia but that didn't depend on Russia being Communist.
164
In general I think there's a very high correlation between people who do good science and people who can explain it clearly, ...
There is some connection because often people who can't explain something clearly don't really understand it. But science also attracts people with high math skills and low verbal skills just because you can get by without real good verbal skills.
174: I won't deny that there's a cultural issue, and that might have been what heebie was getting at. I suppose I was thinking about more in terms of public outreach--of explaining to the intelligent and curious but ignorant--and being too pithy about it. A lot of math--and (this list biased towards what I've found difficult to understand from popular science explanations) physics quantum and relativistic, and philosophy, economics, evolutionary biology--doesn't lend itself to that sort of elucidation. Explaining, say, uncountable numbers isn't too hard, but it's pretty difficult to explain much of set theory without that first step. And as you go higher and farther, it bears less resemblance to anything relatable. Explanations increasingly take the form "higher dimensional space behave that way, because they just do [or I can spend half an hour proving it to you, if you'd like]." This idea isn't as well-formed as (and much more ass-pulled than) I thought it was, but I still intuit that there's a difference in explanatory difficulty across the research boundary of various fields that correlates with abstraction.
you can get by without real very good verbal skills
Oooooh, bakfiets!
I don't know if that is the best option, but that's what I say every time I see one.
Proposition: Sufficiently abstract ideas are impossible to explain in more concrete terms without losing most or all of the meaning. Almost all modern research mathematics is at that level of abstraction.
But there seems to be absolutely no correlation between the brilliance of the mathematician and how clear their lectures are in, say, a graduate course. Or when fielding a question.
I'm actually surprised that there would be a correlation in any field? Being willing to empathize with why material is hard the first time you see it, and plan your explanation accordingly is just a very different skill from designing insightful experiments or having a profoundly deep understand of the theory.
175: The Communist countries were just as stratified (if not more so) than the capitalist countries.
Shearer: Objectively Schachtmanite.
173, 179 totes backfiets. More than two on an xtracycle is a hassle and everything else seems cobbled together. Could do bakfiets plus trail-a-bike, I guess. Dual trail-a-bikes is likely not good.
180: I'm not really thinking about teaching, but about writing. Just seems like people who are clear on why what they're doing is important, and who can explain the motivation in an accessible way, are also usually doing more important things.
In writing, I agree more. People who really understand the material will organize their paper with a birds-eye view of the material, whereas people who don't understand the material write it like they're leading you down a path and they can only see a little bit in any direction from where they're standing.
I know I took math classes in college where it wasn't until studying for the final exam that I had any idea why we had done any of the things we had done in the class; just a steady barrage of definition-theorem-proof until there was a chance to step back and look at the bigger picture.
YES. This was the single worst thing (for me) about math teaching. OK, you want me to memorize a bunch of theorems and figure out a bunch of proofs. Why? Where is this going?
180: I was thinking more of the ability to explain at the popular science level, which is different from the ability to teach a grad student audience. Sorry for misunderstanding.
Another attempt at an explanation: since math is usually cheaper than science, the gatekeepers of mathematical funding are more likely to be internal to the field than in the sciences. So, there's less pressure on mathematicians to be able to explain to non-mathematicians.
Oooh, 189 is making me mad just thinking about it. Good God did math teachers suck. Heebie, create better ones.
173: This is the family biking blog I follow, which answers some of your questions http://carfreecambridge.com/
189: I did appreciate some of Lakatos's writing on how math is presented and/or taught (and I guess Polya is somewhat similar). here he is on what he calls the "deductivist" style:
This style starts with a painstakingly stated list of axioms, lemmas and/or definitions. The axioms and definitions frequently look artificial and mystifyingly complicated. One is never told how these complications arose. The list of axioms and definitions is followed by the carefully worded theorems. These are loaded with heavy-going conditions; it seems impossible that anyone should ever have guessed them. The theorem is followed by the proof.
There are some people who are very good at math mostly because they think very clearly. Those people tend to also be very good at explaining. There are other people who are able to fight through huge complications without getting overwhelmed. Those people tend to be poor at explaining. There's a whole lot of other different types which correlate less well with ability to explain.
There are some people who figure things out because they think about it the right way, and other people who figure things out because they're clever enough and tough enough to work it out even while thinking about it the wrong way.
190.2 is an interesting hypothesis, and could be tested across various other fields.
Not sure what to say... It doesn't seem like a confusing point to me...
Maybe you're part of the group that sucks at explaining.
(I don't actually believe that. Your explanation made perfect sense to me.)
200: They're make a joke, and since they're amoral monsters, they're not going to stop until you get that they're making a joke.
If they were truly amoral monsters they wouldn't stop even then.
They're not immoral. They just don't know right from wrong.
Yeah, the problem I've heard (aside from cost) is that the Bakfiets handles like a pig and is most appropriate for small coastal countries with separated bike paths where the largest hill you're likely to encounter is a bridge over a canal. I thought the Madsen looked like a good combination of price, handling, and capacity, although the chain is about 8 feet long.
Another attempt at an explanation: since math is usually cheaper than science, the gatekeepers of mathematical funding are more likely to be internal to the field than in the sciences. So, there's less pressure on mathematicians to be able to explain to non-mathematicians.
I had a mathematician once tell me the following story: NPR called him up and asked him for an explanation about the Riemann Hypothesis. He replied that he would need about 45 minutes. They said that they were thinking closer to 1 minute. He was flabbergasted, and said something like "I'm going to have to decline. That is quite simply impossible."
He hadn't told it to deliberately showcase his naivete about news forums, but I still found it amusing.
and everything else seems cobbled together
How is the Madsen bucket cobbled together?
I can believe that the bakfiets would be unwieldy. Do the others have the reputation of being better?
I had never seen a Madsen before. Probably don't listen to me on this topic.
But yeah, no, I was talking about tandem + trail-a-bike + handlebar kind of solutions.
207
I had a mathematician once tell me the following story: NPR called him up and asked him for an explanation about the Riemann Hypothesis. He replied that he would need about 45 minutes. They said that they were thinking closer to 1 minute. He was flabbergasted, and said something like "I'm going to have to decline. That is quite simply impossible."
I think it would be possible to say something useful about the RH in one minute.
Oh, I agree. It was just impossible for this guy to fathom.
207: That's funny. The Riemann Hypothesis is actually pretty easy to give a non-technical introduction to. "If you choose a really big number, what are the chances that it's prime? If we knew the Riemann Hypothesis was true, we would know the answer to that question."
I don't see how 213 is about the Riemann hypothesis and not the prime number theorem. You should be saying something like "We already know approximately the odds that a large number is prime, but we want to control how big the error in this estimate can be. The Riemann hypothesis says the error is exactly as small as you could reasonably hope for."
153, 160: I think this is the first time I've been flattered and horrified by the same speech act.
Further to 160: Are you talking about stuff like this? If so, I think "actual programming now being more-or-less automatable" is going rather far...
214: For a non-technical audience, what's the difference? If the error is smaller, then we have a better idea of how common primes are.
The difference is that one of them has been known for 100 years! I think that's a big difference! What you're saying is completely misleading even to a non-technical audience because it makes it sound like we don't know the prime number theorem.
If you want to be vague about the statement (which I don't think you have to, see my explanation), you could say something like: "2500 years ago we showed that there were infinitely many primes. 100 years ago we gave a formula for the odds that a really large number is prime. The Riemann hypothesis is an even stronger result in this direction which says that the distribution of primes is not too lumpy."
215: Batsignal! Dude, you have such a fanclub. Some of whom may even have met you.
---
Cosma, sral, Sifu et al: Yes, I was fascinated by the GenProg results (& some other work by Stephanie Forrest), and later something related from Melanie Mitchell on co-evolution. It's already a lot more useful than a baby.
Fixing bugs isn't just the 'second 80%' of software development in most cases, it's maintenance effort for the life of the code. That's a lot of programmer time, hence, cost. I do not find it at all hard to imagine half of the current programmer time being replaced by automated bugfixing. Your shop, dear reader, may have much better stats, but we know they are among the best: they hired you. Think of poor old Microsoft instead.
I also assumed that once you've sent your codebase down the GenAlg rabbithole, human beings aren't going to be fixing it, but I haven't read the (interesting-looking) linked research into whether that's true.
And the comparison to 'automating the construction of stack frames at the beginning of function calls' -- well, yeah, but that's the kind of thing that changed the number and kind of programming jobs a lot. Maybe the next change will also be good for programmers as a employed class, but... we're having a thread, yea, a decade about the replacement of skilled human jobs by automation.
And then Skynet will become conscious, and us meat citizens will all live very quietly on our CO2 fraction inheritances, playing recorder music and simulating Duff's Device to amuse ourselves. In about 500 pages, with spot glossy and a woman's jumpsuited torso on the cover... it's about time for the cyberpunk nostalgia trip, yes?
it's about time for the cyberpunk nostalgia trip, yes?
I... embody that?
218 has me thinking that someone should write a pop-science book exploring various things that look incredibly random and yet somehow are not. The Riemann hypothesis, the black hole information paradox, ...?
Jumpsuit? A little too much Tron stuff? Big Ant poseables?
I think this is the first time I've been flattered and horrified by the same speech act.
The blog has invented an entirely novel speech act.
218.1: I don't get why this bothers you so much. If we know that the error in the Prime Number Theorem is as small as possible, then we have a better idea of the probability of a randomly chosen big number of being prime. So what I said is true. If people find the topic interesting, it's not like it's hard for them to read more and find out about the Prime Number Theorem.
218.2 is a good 3 sentence explanation, though. Your initial explanation makes it sound like some book-keeping exercise that mathematicians have pursued for 100 years after solving the problem. If I were a Congressman who heard that explanation, I would vote to cut mathematics funding.
I think the best way to do it is to start with something like 213 and then backfill a bit.
If you choose a really big number, what are the chances that it's prime? If we knew the Riemann Hypothesis was true, we would know the answer to that question. Now... [218].
Make sure to mention cryptography or something, unless you want them to slash your whiteboard and wastebasket funding.
224
I don't get why this bothers you so much. If we know that the error in the Prime Number Theorem is as small as possible, then we have a better idea of the probability of a randomly chosen big number of being prime. So what I said is true. ...
Mathematicians like to get things exactly right. This is not my area but I don't think the above is exactly right. Suppose we know the probability that a coin flip will come up heads is 50%. This doesn't mean that in any particularly set of 100 flips 50 will be heads. You can study the distribution (which will depend on things like how independent the flips are) but the issue isn't the base probability of 50%.
In the case of primes talking about the probability that a large number is prime is in a sense meaningless, it either is or isn't. But you can model the distribution of primes by comparing it to the distribution you would get by randomly (and independently) labeling numbers as prime (with the probability decreasing as the size of the number increases). In such a distribution (because of statistical noise) the total number of numbers labeled prime (less than some bound) will (with probability 1) be at times be greater or less than the expected number but won't be too far away. As I understand it proving the RH would show the difference between the actual distribution of primes and the known formula behaves similarly (like statistical noise).
As I said this isn't my area so the above probably isn't exactly right either. You do have to make some adjustment for that fact that you never have for example n and n+1 both prime (for n>2) which would not be the case in the labeling model.
This is Just to Say
I have slashed
your whiteboard
and defunded
your wastebasket
both of which
you were probably
planning on
using for your work
Forgive me
it was great fun
try working
on something practical for a change
Mom
Frosted glass walls you can write on with whiteboard markers, on the other hand...
Well, they might be stupid, too. But whenever I walk around the building where all the office walls are made of those I get a bit jealous.
I suppose that instead of those I have an awesome view, so I shouldn't complain.
231: ... but sometimes you still do.
218.2 is a good 3 sentence explanation, though. Your initial explanation makes it sound like some book-keeping exercise that mathematicians have pursued for 100 years after solving the problem. If I were a Congressman who heard that explanation, I would vote to cut mathematics funding.
There's mathematics funding now? Who's responsible for this?
All of these explanations make it sound boring and pointless, they omit that the results come from analysis of a really unusual complex function. Looking at how people have managed to obtain conclusions about that thing is like listening to music from outer space or something.
