@whatihear's banner p

whatihear


				

				

				
0 followers   follows 0 users  
joined 2022 September 07 03:01:39 UTC

				

User ID: 917

whatihear


				
				
				

				
0 followers   follows 0 users   joined 2022 September 07 03:01:39 UTC

					

No bio...


					

User ID: 917

Where are all the scientists?

After some experiences doing research during a bachelors and masters degree, I had had enough of an experience with academia writ large and academic research in particular to know that while I enjoyed some aspects of it, it wasn't for me. Getting a PhD is a grueling experience that I don't have the work ethic, and, in the field I was most interested in, possibly the intelligence to make it through. Since I've never spent any time as a practicing scientist, I've long felt like it wasn't my place to criticize the institution of science or scientists. Now, a few years later, I've gotten more experience with academic research second hand through a close relationship with a graduate student who is experiencing all of the trials and tribulations that I bailed on in favor of a cushy job at one of our economies massive digital rentiers.

Writing my master's thesis was the closest I got to actual academic research, and the experience left me decidedly soured on the whole endeavor. I did all of the work. My adviser knew little about the subfield I was focused on, and barely had the time to read enough papers to keep up with her own subfield much less learn one that a student who would only be there for a year was interested in. I didn't really expect anything different given the nature of her incentives and my prior experiences. I was however a bit surprised when I mentioned off-hand that I was doing all the work and she took exception, pointing out that she gave advice and suggestions during out weekly meetings. These meetings consisted of me explaining what I had done and her giving mostly terrible and out of touch advice that would have crippled my project had I heeded it. I believe she made one genuinely helpful contribution to the research over the course of the year. Again, this is all perfectly reasonable and I hold no resentment about it, but I did find it bizarre that she didn't realize just how little she had contributed to the actual work. This was not a relationship between two researchers working on a problem together, it was a relationship between one (somewhat bumbling and lazy) researcher and a mildly interested administrator whose job was mostly to make sure I had done sufficient work to graduate.

Since that experience, I've gotten to witness that dynamic second hand through my graduate student friend and her extended network of graduate student friends. Her advisor is much more hands on than mine was, but we know PhD students with advisors just as out of touch with their work as mine was with my thesis work. Even my friend's blessedly engaged advisor never does any actual research. He has not touched any of the lab equipment in anger in years (exceptions are of course made for photo ops). He does not write any of the scripts to analyze the data his students collect. The one area where he is intimately involved is in the production of the true output product of academic research: technical writing about the research. Specifically, he edits the technical writing of his students furiously. His other main contribution towards research is securing and retaining funding, a task which is done by the production of yet more technical writing.

Though our society calls my friend's advisor a scientist, he isn't one. Calling him a science manager would be more accurate, but I think that that too is not quite right. I'm a programmer by trade, and another analogy that springs to mind is the technical team leads who get to the point where they stop writing code and are wholly occupied by ancillary activates such as writing design documents, coordinating with other teams, and reviewing the code of more junior programmers. If this goes on long enough, these people's skills can dull. I'm reminded of the time that I interviewed a fellow who had previously spent a 10 year stint at Netflix and, if his resume was the be believed, lead some fairly impressive projects. I expected him to breeze through the interview, but he couldn't code his way out of a paper bag. I think professors engaged in research are generally some combination of a science manager and one of these not-so-technical team leads.

By the time anyone gets tenure, they have been forced by the inexorable Moloch to have spent the last several years acting as one of these science coordinators. If they try to do any research themselves they will get out-published by someone willing to hire 15 grad students and keep 6-10 balls in the air at once. At this point they will have spent around 5 years being an actual if initially poorly trained and little supported scientist (a grad student), then probably a few more years on top of that (as a postdoc on multiple postings). On top of that, they must be unusually intelligent and driven to have gotten so far. This means that they still likely have their edge when they get tenure. At this point, they even have the option of of becoming a real scientist again. They could stop hiring grad students to cut their lab down to a reasonable size and actually start spending time in the lab themselves. Few choose this option when it means giving up on becoming at the "top of their field" and when their compensation is tied to the amount of grant money they can bring in.

Upon achieving tenure by being an effective science coordinator, the ambitious academic continues in much the same pattern that got them to where they are. They may retain their edge, after all these are our best and brightest, but I suspect that many of them on some level go the same way as the poor fellow from Netflix. I imagine most don't fall quite so far as he evidently had, but as time goes on, the skills that they learned as graduate students and postdocs will dull and fade. They are required to teach classes, so their theoretical fundamentals remain very strong, but their ability as actual practitioners falls off. That friend of mine frequently complains that the senior "researchers" she works with ask her to do things that are just clearly not going to work from the perspective of someone who is in the trenches day in and day out.

My manager is a very good programmer, but he does not view himself as a programmer, he views himself as a manager. As such, he does not make an effort to tell me how to do my job, though he surely has opinions on it. He understands that because he not working on the code as part of his day to day work, he doesn't have the right context to make technical decisions, and I do.

Science coordinators do not possess the graciousness of my manager. Since they view themselves as researches rather than coordinators of researchers, they are quite willing to hold forth on the right way to do things. For this reason, bad decisions are made in the pursuit of research when science coordinators tell senior grad students how to do their job. A complicating factor here is the fact that junior grad students do need to be told what to do, as our system does almost nothing to train them to be scientists before they are expected to be generating data in the lab (they are pretty smart though, so in time they will learn through osmosis from the senior grad students).

Another problem with the fact that science coordinators do not view themselves as managers is the fact that they tend to make no effort to actually learn or apply the art of management. When one of my older relatives made the transition from being an individual contributing engineer to an engineering manager, he spent about a year poking through management books. I'm generally pretty skeptical of the MBAification of things, but the field is not entirely without merit and I do think managers should at least take the time to think deeply about what it means to lead people and to hold power over them. At the very least, they should recognize that they are managers and they have some new, uniquely people focused responsibilities.

Micromanagement and mismanagement in general makes the lives of grad students hell, and maybe even leave some productivity on the table (though given the brutal competition of academia I tend to think that the professors that make it are the ones who have figured out how to wring every last drop from their grad students). Beyond that though, there are societal impacts. Whenever a "scientist" wades into the public discourse, they are inevitably a science coordinator rather than a practicing scientist. They are likely well suited to that role since the one remaining technical activity they engage in is technical communication, but the public is still deceived by this inaccurate title.

I think the larger harm done by this system is the utter waste of human capital. As I've mentioned a few times so far, grad students come in as untrained neophytes, so don't become productive for a year or two. Even once these young researchers become competent and effective practitioners, they are still inexperienced. They reach their most experienced period as scientists when they are postdocs, but postdoc postings are not long enough to delve deeply into a field. Then, at the height of their powers, they become mere coordinators. There are no graybeard scientists. This, more than anything seems like a tragedy. Brilliant professors should be doing science, not writing grant proposal after grant proposal and copy-editing their students work.

Much ink has been spilled about the fact that technological progress is getting slower and slower per researcher hour as we push the technological frontier further and further out. Scott shows some good data that illustrate this point in Is Science Slowing Down. Like Scott, I tend to think that the low hanging fruit theory explains what is going on here, but I wonder if missing scientists might be another factor.

Didn't people get murdered in CHOP/CHAZ? It may not have lasted long, but there was quite a bit of damage.

In the western legal tradition, violent crime is not a tort, or at least not exclusively a tort. If Brown murders Jones, the case is The State of Maryland vs Brown not Jones vs Brown. This is because violent crime causes lasting damage to the social fabric. I would argue that just the establishment of a law-free zone without any violent crime does a ton of damage to the social fabric. The knock on effects of delegitimizing the state's agents of violence are where most of the damage was.

I have one friend who is a great guy, quite smart and funny, and quite willing to help out any of his friends at the drop of a hat. Unfortunately, he has the misfortune of not being all that attractive. He's in his early thirties and single, and I don't think there is anything wrong with him. Some guys just have it hard.

I think the fact that they are a smaller country is one reason. China is so big and economically muscular that they can throw their weight around. Smaller countries must integrate into an international system if they want to do well. Also, china is not the only datapoint. South Korea started off a military dictator ship and transformed into a (somewhat shaky) liberal democracy. Singapore is known for being the model authoritarian state, but they have been very slowly loosening the ratchet. Various peripheral European states have been influenced to clean up their act in various ways by the economic power of the EU.

I wrote a wall of text, so maybe you missed the bit where I said "He does not write any of the scripts to analyze the data his students collect." It's not that the grad students are the experimentalist and he is doing all of the experimental design and analysis, they do all of the analysis. My understanding is that the experimental design process is somewhat collaborative between the PIs and grad students, and I would say that participating in these experiment design meetings is doing science, but doing a bit of science on the side does not a scientist make.

The analogy with programming is not importing modules, it is writing design docs. You need to be a good programmer to write good design docs, but if all you do is administrative management tasks plus collaborating with actual programmers to write design docs, I'm sorry, you're not a programmer.

I've long had a feeling of disquiet about just how far removed from the necessities imposed on us by the fact of our physical existence we are in the modern era. For me, this manifested most strongly as a revulsion towards a career that just pushes paper or people around, and made me interested in STEMy stuff, but I don't think I'm alone in this. I think this sense of disquiet undergirds a lot of strange behavior you can observe in left wing hippies and right wing homesteader types. For some people it isn't the disconnection from physical reality (by this I mean mostly agriculture and manufacturing), but instead from what one might call the societal production function. My grandmother once remarked that as a girl she thought the only jobs worth doing were being a soldier, a teacher, or a farmer, and I think this sentiment was coming from a similar place as my feeling that you have to be getting your hands at least a little dirty.

I think you're a bit like my grandmother in that you feel that there are certain social truths that will remain inescapable so long as we are a bunch of jumped up apes. As I get older, I've started to come around to this view more even as I loosen my grip on my old feeling that work is fake unless you're building a little. Pinker really convinced me that Hobbes was right, and watching the world devolve into chaos as US hegemony fades is only strengthening that view.

"has an advantage" != "is better". Two different competitors will always have a multitude of different advantages and disadvantages relative to one another. It just happens that the results of male puberty have a very strong advantage over the results of female puberty for almost all sports.

I don't think your local interlocutors have advocated for stripping women of the right to vote (though maybe someone has elsewhere in the thread). I've read them more as hoping for a shift in power at the level of culture and norms, but basically disparaging that this is possible.

Personally, I'm somewhat optimistic that we will find a better balance in the long term, but believe the short term will be bumpy. We've only had gender equality in the west for maybe half a century, and that equality is not perfect because some aspects of patriarchy have inertia and most feminine privilege remains unexamined due to the nature of the feminist movement. I think feminism is mostly downstream of the industrial revolution, but cultural evolution can take awhile and I think we still have a ways to go before we have adjusted to the economic substraight of the information economy.

A software engineer, on the other hand--or better yet, a software architect--need not necessarily do any programming. They can offload the tasks that require that specific technical skill to programmers.

There is no meaninfful distinction between programmers and software engineers. I consider myself a programmer because I feel like it captures what I do more accurately, and refer to myself as a software engineer in situations where it is financially beneficial. Software engineering is programming plus bureaucracy. Lots of things involve bureaucracy. When it comes to software engineering, programming is the main bit. If you take out the programming, it's not software engineering anymore. I have no patience for someone who thinks they are contributing technically by building a pie-in-the-sky UML diagram and demanding that actual programmers implement their out of touch vision.

I suspect that this is at the root of the contention between your perspective and mine. Do you regard doing science as a set of technical skills? Or do you regard doing science as making progress on our ability to predict and manipulate the physical world?

I think you're right about that this is where we disagree. If we take doing science as "making progress on our ability to predict and manipulate the physical world", well that applies to the electron microscope salesmen, academic departmental secretaries, directors of corporate research orgs, plumbers who install chilled water systems in labs, the maintainers of python and r, and any number of other people who contribute in some small way to the broad economic activity of advancing science. You my protest that since science coordinators work a bit closer to the main body of the academic work than the directors of a corporate lab, they are scientists, but both of those roles are mostly about coordinating the technical work.

So if I run a bio-chem lab (the Hooser Lab at Stanbridge) and my goal is to progress what we know about what causes aging and what may halt the process in mammals, then my main job is to make sure that my lab can actually make useful progress in my goal. I need to break down what my lab needs to do, what resources it needs to do that, and how I can get those resources. Then I get those resources, and oversee the process. And as much as I enjoyed writing scripts to analyze data when I was a postdoc at Whatihear Lab at Oxbridge, maybe my time would be better spent on reviewing drafts for publications (because I have the breadth of knowledge to connect that esoteric result to broader field, or to suggest in the discussion multiple probable interesting consequences), and speaking with grant-giving foundations (because I have built my reputation as a serious scientist and they will take me seriously), while a postdoc in my lab oversees the data analysis.

Within our current system, that's what you need to do to push research forward. It doesn't mean you would be a scientist in that situation.

I'm not blaming PIs for the current state of affairs. They are operating within a system of constraints and incentives that they had no role in building. I'm just pointing out that they are not scientists, despite being the best trained people to fulfill such a role.

...they did.

When? If you are talking about the resettlement of the German tribes inside imperial borders, they only did that after smashing them in battle and disarming them. Sometimes they would skip the smashing in battle bit, but there would always at least be a negotiated settlement before the tribes were allowed to move into the empire uncontested. Towards the end of the empire they stopped disarming the Germans, but that was definitely not by choice, it was because the empire was falling apart.

One big reason people are not bringing up Tom Cruise's Scientology connection is that it is old news. For better or for worse, if a nasty fact has been out there for a long time, people don't bring it up in the discourse as much. This is how the process of un-cancellation works on an individual level (I think broad vibe shifts also have something to do with it). The Scientology thing has been litigated in the court of public opinion for quite some time.

On the other hand, assuming you're a man, you are still much more likely to be violent than much of the population. It seems to me that in order to justify your position, you have to rather arbitrarily draw a line right where it benefits you the most (you get the benefit of the doubt if you are doing something suspicious or disconcerting, but you don't have to extend the same benefit of the doubt to the group most likely to be able to harm you).

This assumes that people are in one of two states: behaving in a deranged and menacing way in public, or minding their own business. That's not really the case though, there is a pretty smooth spectrum of menacing behavior people can exhibit in public.

If observers are being good basians, they will factor in the observed behavior in addition to more contextual information about a person. A well dressed white man drunkenly throwing a single strike at someone and not following it up would seem like a much bigger deal than a similarly attired 5'0" white woman doing the same to me, partially because the woman is much less physically imposing, but also because of what I know about rates of sexed violence and my guess about the likelihood of escalation to a point I couldn't easily control.

At the same time, waving a gun around is a red-alert pretty much no matter the identity of the person doing the waving.

Argentina and Mexico are not really first world societies. Turkey and Israel both use fairly oppressive religions to control women's reproductive behavior.

I'm looking for a audiobook, series of audiobooks, or podcast that gives a historical survey of the middle east starting with the founding of Islam. Basically I'd like to have an AP Euro/AP US level of detail about the region. Normally my proceedure for this is to look for a "History of $REGION" podcast, but I found that I felt I couldn't trust the "History of Islam" podcast I found because it was done by a believer and I wasn't sure how much that distorted the narrative (I didn't get the sense he was a propagandist, just had some fairly obvious biases). This part of history seems uniquely political, so I'm afraid I would run into the same issue if I just bought a random book on the subject.

Does anyone have any recommendations? Ideally this would be by a non-believer. Even more ideally, by a non-believer from quite a while ago (somehow I feel like people might have been able to be more objective about it before 9/11).

Yeah, I think I basically agree with that unless I'm misreading you. I think "scientist" has a bunch of cachet, and we should assign that cachet to the people doing what is normally thought of as science rather than pushing paper.

The worst part was the social environment. In order to get the PhD students had to become first author on multiple papers, but the PI would assign multiple people to each research project, bringing in more people the longer it took. I'm not sure there was sabotage (I'm dumb enough to fuck things up myself, thank you), but there was definitely spying and theft of results between students. The students needed favor with the professor to buy equipment: seeking the favor of the professor resulted in schemes much like those of medieval courts. Reading The 48 Laws of Power during my PhD, the content of the book depicted the social environment of the lab quite accurately.

Jesus that sounds horrible. Fortunately, I don't think things are quite that bad in my friend's lab, but her PI is known to play favorites. There is definitely a ton of political BS.

Yeah, I saw that after I posted that response.

I think the impulse to say that the female franchise was a bad idea comes from the observation that women tend to be more prone to certain behaviors which lead to a toxic politics (women are generally the front-line enforcers of orthodoxy), which I think is actually true, but conveniently ignores the fact that men have their own set of such behaviors that are arguably worse (our violence at the personal level extends to supporting more state violence and jingoism). I think that mixed gender friend groups are generally healthier than single gender, and I tend to think that the same is true for the political realm as well as each gender provides some balance for the other.

I really don't know how things used to work, but that certainly seems plausible. Certainly as we produce more scientists the bar is probably lowered since the distribution of human intelligence is relatively fixed over time (the Flynn effect is not enough to make up for the expansion in the number of scientists, especially because it works mostly by lifting up the bottom of the distribution so we have more average people and less profoundly stupid people).

The idea of larger labs causing eroding an old apprenticeship model is interested, though in my experience smaller labs can be pretty bad learning environments because the professor does no research and there are not enough older grad students to mentor the new grad students. Maybe things would have been different if all labs were smaller and more professors actually did their own research.

100% of the trans people I know IRL do not match the Very Online Trans Person archetype at all. Admittedly, it's a very small sample size, but still. I don't think we can be sure that they all document everything online. I would expect them to be more online on average, but I'm not sure by how much.

I think I'm looking more for older history. Thanks though!