6 ms·

Having worked with a lot of researchers, in general, my experience has been that if something isn't the latest research meme/trend/buzzword, they don't give a f

Having worked with a lot of researchers, in general, my experience has been that if something isn't the latest research meme/trend/buzzword, they don't give a fuck. "Something that we did 10 years ago" means they already got a publication about this idea, and that was that, they've long since moved on. Actually deploying something into the real world, effecting real change, is the hard part, but most researchers are focused on getting the next publications accepted, and only that. Not every researcher is like that, but it's a problem. The researchers I worked with, also, unfortunately, had this idea that engineering is somehow beneath science. That's just worrying about the details. They think that the idea itself is the real advancement. They think of themselves as smarter and more important than engineers, and they don't like when engineers press them on how they would actually implement their ideas into something real. Sorry if I sound cynical, it's because I am.

BurningFrog 6y ago
The rule in academia is "publish or perish". So the publication focused scientists are only doing what they have to.
- tachyonbeam 6y ago
  Yeah, but I think the problem is that you get a lot of "scientists" who are only optimizing to game the system and maximize the publication count. It leads to most of the published work following very safe and boring research directions that everybody else is already working on. There is less scientific progress as a result. Contributing something that is useful and novel is often not rewarded. Risking yourself exploring a new branch of your field is very dangerous for your career. In my own biased opinion, about 90-95% of published papers are contributing nothing new to the field.
  - BurningFrog 6y ago
    Just to be clear, I agree with all that. Science, in its current institutionalized form, is starting to look pretty broken to me. It's time for something new.
zarkov99 6y ago
They probably are smarter than Engineers, on average, but that does not make them more valuable to society at large, though it might in their academic habitat.
- lumost 6y ago
  “Smarter” is a really vague term. At least in CS/Software it’s incredibly common for people to sound much more knowledgeable than they are, or for promising junior engineers to get stuck in a hole. Given the selection bias inherent to graduate school pathways* in the US, I wouldn’t presuppose any innate skill gap between Fresh Phds and baccalaureates. * A top BSc candidate with a comp eco background would need to forgo ~800k in compensation for a 4 year PhD. While in grad school, they must be able to survive on ~20k per year in direct income. Fresh from a BS they may have 100k in student debt, or have been fortunate to have wealthy family or studied internationally where the cost of education is lower. People willing and able to go for grad school do not represent an unbiased sample of the top students.
ISL 6y ago
Every part of the innovation chain has understanding of the difficulty of their own link. Without every link, it doesn't work. Consider the SR-71: Is it the pilots, the crews, the administrators who secure funding, the voters, the integrating engineers, the test pilots, the supply chain, the metallurgists, the engine designers, the manufacturers of test equipment, the standards bureau, the aircraft architect, the people who specified the initial requirements, the tire manufacturers, the refueling crews, the avionics manufacturers, the chemical suppliers for the bespoke starting fuel, the aerospace engineers who did the fundamental research, Chuck Yeager, the Wright Brothers, Bernoulli, the thousands of spouses who sacrificed to support their partners, or the fundamental researchers who ascertained the physics necessary to make the aircraft fly and evade detection? Every one of them, and more, can lay claim to the statement: If we hadn't done what we did, the SR-71 wouldn't have flown. I'm a fundamental physicist who has spent the entirety of this century trying to get at the bottom of things because it is the foundation upon which our understanding rests. When we make a measurement, however, even those experiments rest upon work and resources provided by a vast array of engineers, suppliers, manufacturers, administrators, support staff, physical plant, custodial support, funding agencies, and taxpayers.
- ouid 6y ago
  Did the SR-71 evade detection? I was under the (perhaps wrongful) impression that it just outran the consequences of detection.
  - the-dude 6y ago
    I was under the impression it flew too high to be reached by surface to air missiles.
  - eximius 6y ago
    It's stealth was fairly advanced for the time. It was a bit of both.
  - classichasclass 6y ago
    That's pretty much it. The standard countermeasure for launch detection was simply to accelerate and outrun the missile. It had some early stealth attributes but could still be detected by the systems of the time.
xenocyon 6y ago
I'm not sure which field you're in, but at least in electrical engineering, the situation is very different from your description. EE academics are deeply enmeshed in real-world problems, and justifying real-world impact in publications is considered more or less essential. This is also enforced by the bottomline: a lot of research is sponsored by industrial consortia or even individual companies (Intel, Microsoft, etc). Furthermore, many though not all academics at some point spin off some part of their research into commercial ventures, and for some that even becomes their primary concern.
- tachyonbeam 6y ago
  I'm on the CS side. Was surrounded by people who love math but seem to believe that programming code is somehow dirty and less worthy than equations. The difference is probably that EE academics have an engineering background (are engineers themselves)?
  - nitrogen 6y ago
    A more apt comparison might be CS and physics?
- rjsw 6y ago
  I'm currently seeing mechanical and materials engineers who have just discovered computers and are planning to do fundamental research on data modelling. They ignore existing international standards that deal with the area since they are not new enough. I think that EE has always been more connected to industry.
  - lumost 6y ago
    I think the publish/perish culture has pushed academics closer to industrial research horizons than they were in the past. This leaves a few fields in an awkward state where academics are just credentialed junior engineers/practitioners struggling to differentiate their skills and activities from every one else.
gameswithgo 6y ago
I think your take on it seems fine. The researchers did the poc, later it goes mainstream. The researcher should not care or be interested in the mainstream practical application anymore, they should be on to the next thing.
- tachyonbeam 6y ago
  It's not just that though. It's also risk aversion. If you're optimizing for numbers of papers published, and you're optimizing for your confidence that your next paper is going to be accepted, it leads to redundant and uninteresting research. The research world, just like every other social circle, is full of people who are just parroting each other, repeating to each other what everyone already believes, and patting themselves on the back. Maybe, once in a while, one person takes a risk, gets a cool result, and then 100 others rehash the same idea with small variations. At some point, the field collectively decides what the next cool research meme is going to be, and everybody follows along. So yeah, structural battery packs stopped being cool at some point, I guess. Except that battery tech 10 years ago wasn't the same as what we have today, so maybe the results don't directly transfer. The topic hasn't been studied in a contemporary context, maybe the old results are out of date, maybe they would really be worth investigating again... But who cares, that topic has been done, it isn't cool anymore, it was 10 years ago. All the cool kids are doing solid state batteries nowadays.
  - solraph 6y ago
    > The research world, just like every other social circle, is full of people who are just parroting each other, repeating to each other what everyone already believes, and patting themselves on the back. I'd be curious to know what your experience is that leads you to this opinion. A more charitable explanation would be that game changing discoveries are by their nature rare, and even when they do arrive, they only outline a new space for exploration, which means that by definition, most research is going to be filling the gaps in that new space, both for positive results, and the depressingly large chunk of work that is discovering what doesn't work. The results of this negative outcome often look the same.
- deleted 6y ago
PeterisP 6y ago
That's kind of the whole point of specialization - a good framework to look at this is the Technology Readiness Level principles; and different institutions are specialized to work with different maturity levels of a tech and that's ok. The requirements to work on core principles that might have an application (and fund&organize that work) are fundamentally different from finding product/market fit for something that's known to be possible and are fundamentally different from optimizing a process for making that thing 1% cheaper than last year, so those are generally done by different institutions and different people. So yes, from the point of someone doing fundamental research, once a proof of concept exists, their part of the work is done - there's a lot of further steps to get that to a commercially viable product, but those steps should be done by someone else who's better at it and whose organization is structured completely differently to facilitate this completely different process. And meanwhile they'll working on some other tech that's ready for that engineering stage yet. Theoretical physics isn't applied physics isn't prototype engineering isn't process optimization engineering. You need people and organizations working on every TRL step, and they aren't interchangeable, you need the preceding step well-funded to make your work possible and you need to move the results to the following steps since those will harvest the actual end-user value in the end.

BurningFrog 6y ago

The rule in academia is "publish or perish". So the publication focused scientists are only doing what they have to.

tachyonbeam 6y ago
Yeah, but I think the problem is that you get a lot of "scientists" who are only optimizing to game the system and maximize the publication count. It leads to most of the published work following very safe and boring research directions that everybody else is already working on. There is less scientific progress as a result. Contributing something that is useful and novel is often not rewarded. Risking yourself exploring a new branch of your field is very dangerous for your career. In my own biased opinion, about 90-95% of published papers are contributing nothing new to the field.
- BurningFrog 6y ago
  Just to be clear, I agree with all that. Science, in its current institutionalized form, is starting to look pretty broken to me. It's time for something new.

zarkov99 6y ago

They probably are smarter than Engineers, on average, but that does not make them more valuable to society at large, though it might in their academic habitat.

lumost 6y ago
“Smarter” is a really vague term. At least in CS/Software it’s incredibly common for people to sound much more knowledgeable than they are, or for promising junior engineers to get stuck in a hole. Given the selection bias inherent to graduate school pathways* in the US, I wouldn’t presuppose any innate skill gap between Fresh Phds and baccalaureates. * A top BSc candidate with a comp eco background would need to forgo ~800k in compensation for a 4 year PhD. While in grad school, they must be able to survive on ~20k per year in direct income. Fresh from a BS they may have 100k in student debt, or have been fortunate to have wealthy family or studied internationally where the cost of education is lower. People willing and able to go for grad school do not represent an unbiased sample of the top students.

ISL 6y ago

Every part of the innovation chain has understanding of the difficulty of their own link. Without every link, it doesn't work. Consider the SR-71: Is it the pilots, the crews, the administrators who secure funding, the voters, the integrating engineers, the test pilots, the supply chain, the metallurgists, the engine designers, the manufacturers of test equipment, the standards bureau, the aircraft architect, the people who specified the initial requirements, the tire manufacturers, the refueling crews, the avionics manufacturers, the chemical suppliers for the bespoke starting fuel, the aerospace engineers who did the fundamental research, Chuck Yeager, the Wright Brothers, Bernoulli, the thousands of spouses who sacrificed to support their partners, or the fundamental researchers who ascertained the physics necessary to make the aircraft fly and evade detection? Every one of them, and more, can lay claim to the statement: If we hadn't done what we did, the SR-71 wouldn't have flown. I'm a fundamental physicist who has spent the entirety of this century trying to get at the bottom of things because it is the foundation upon which our understanding rests. When we make a measurement, however, even those experiments rest upon work and resources provided by a vast array of engineers, suppliers, manufacturers, administrators, support staff, physical plant, custodial support, funding agencies, and taxpayers.

ouid 6y ago
Did the SR-71 evade detection? I was under the (perhaps wrongful) impression that it just outran the consequences of detection.
- the-dude 6y ago
  I was under the impression it flew too high to be reached by surface to air missiles.
- eximius 6y ago
  It's stealth was fairly advanced for the time. It was a bit of both.
- classichasclass 6y ago
  That's pretty much it. The standard countermeasure for launch detection was simply to accelerate and outrun the missile. It had some early stealth attributes but could still be detected by the systems of the time.

xenocyon 6y ago

I'm not sure which field you're in, but at least in electrical engineering, the situation is very different from your description. EE academics are deeply enmeshed in real-world problems, and justifying real-world impact in publications is considered more or less essential. This is also enforced by the bottomline: a lot of research is sponsored by industrial consortia or even individual companies (Intel, Microsoft, etc). Furthermore, many though not all academics at some point spin off some part of their research into commercial ventures, and for some that even becomes their primary concern.

tachyonbeam 6y ago
I'm on the CS side. Was surrounded by people who love math but seem to believe that programming code is somehow dirty and less worthy than equations. The difference is probably that EE academics have an engineering background (are engineers themselves)?
- nitrogen 6y ago
  A more apt comparison might be CS and physics?
rjsw 6y ago
I'm currently seeing mechanical and materials engineers who have just discovered computers and are planning to do fundamental research on data modelling. They ignore existing international standards that deal with the area since they are not new enough. I think that EE has always been more connected to industry.
- lumost 6y ago
  I think the publish/perish culture has pushed academics closer to industrial research horizons than they were in the past. This leaves a few fields in an awkward state where academics are just credentialed junior engineers/practitioners struggling to differentiate their skills and activities from every one else.

gameswithgo 6y ago

I think your take on it seems fine. The researchers did the poc, later it goes mainstream. The researcher should not care or be interested in the mainstream practical application anymore, they should be on to the next thing.

tachyonbeam 6y ago
It's not just that though. It's also risk aversion. If you're optimizing for numbers of papers published, and you're optimizing for your confidence that your next paper is going to be accepted, it leads to redundant and uninteresting research. The research world, just like every other social circle, is full of people who are just parroting each other, repeating to each other what everyone already believes, and patting themselves on the back. Maybe, once in a while, one person takes a risk, gets a cool result, and then 100 others rehash the same idea with small variations. At some point, the field collectively decides what the next cool research meme is going to be, and everybody follows along. So yeah, structural battery packs stopped being cool at some point, I guess. Except that battery tech 10 years ago wasn't the same as what we have today, so maybe the results don't directly transfer. The topic hasn't been studied in a contemporary context, maybe the old results are out of date, maybe they would really be worth investigating again... But who cares, that topic has been done, it isn't cool anymore, it was 10 years ago. All the cool kids are doing solid state batteries nowadays.
- solraph 6y ago
  > The research world, just like every other social circle, is full of people who are just parroting each other, repeating to each other what everyone already believes, and patting themselves on the back. I'd be curious to know what your experience is that leads you to this opinion. A more charitable explanation would be that game changing discoveries are by their nature rare, and even when they do arrive, they only outline a new space for exploration, which means that by definition, most research is going to be filling the gaps in that new space, both for positive results, and the depressingly large chunk of work that is discovering what doesn't work. The results of this negative outcome often look the same.
deleted 6y ago

PeterisP 6y ago

That's kind of the whole point of specialization - a good framework to look at this is the Technology Readiness Level principles; and different institutions are specialized to work with different maturity levels of a tech and that's ok. The requirements to work on core principles that might have an application (and fund&organize that work) are fundamentally different from finding product/market fit for something that's known to be possible and are fundamentally different from optimizing a process for making that thing 1% cheaper than last year, so those are generally done by different institutions and different people. So yes, from the point of someone doing fundamental research, once a proof of concept exists, their part of the work is done - there's a lot of further steps to get that to a commercially viable product, but those steps should be done by someone else who's better at it and whose organization is structured completely differently to facilitate this completely different process. And meanwhile they'll working on some other tech that's ready for that engineering stage yet. Theoretical physics isn't applied physics isn't prototype engineering isn't process optimization engineering. You need people and organizations working on every TRL step, and they aren't interchangeable, you need the preceding step well-funded to make your work possible and you need to move the results to the following steps since those will harvest the actual end-user value in the end.