I think that's not in opposition to what the article says. The main point is that computer science largely dismisses social sciences, hitting realizations and creating problems which social scientists have long considered and solved. To fix this, more exchange of knowledge must happen between computer science and social sciences. In order for social sciences to embrace computing, computing must realize that social science is a necessary part of their work.
What social scientists consider "solved" does not usually jive with reality. That attitude in itself shows the field still needs to mature. They are also notorious for being more "enlightened" from my experience interacting with them at University.
Don't you think connecting the field more closely with computer science would help it mature, though? Both sides could benefit from this: Both get methods, challenges and insight from each other. It might bring both a bit closer down to earth - computer science gets a better idea of possible social challenges their work could face or produce; social science gains new perspectives from more technical and practical students.
I find some of what makes it into CACM is just absolute dreck. To be honest I think they are having trouble finding authors for articles. I can't tell you how many times I've seen articles from Kate Matsudaira prattling on about managing a software team (which, based on conversations with my wife, who is an actual director of software development for a large team, seem to convey naive entry-level management insights at best) or endless articles from Ivar Jacobson trying to evangelize the Next Big Thing to Replace Agile for software engineering.
At the same time there are usually 2-4 articles per issue that are very good. CACM also presents curated lists of interesting research articles, so (as practitioners) we do not need to sift through every article from every conference proceeding or journal publication talking about how some squeezed an extra tiny incremental fraction of x out of an O(n^x) algorithm or incremental papers discussing how compiler type inference algorithms could be improved this tiny little bit with this one weird trick.
And we occasionally get really insightful practitioner articles as well; for example, some interesting bits of how Google manages their code base, or what an incredibly difficult problems Coverity faces when statically-analyzing (supposedly)-compliant C/C++ code.
Translation: 'We are horrified that Computer Science students - some of whom might go on to be the next tech billionaires - are managing to get through university without being completely brainwashed by leftist demoralisation!'
Don't you think that, even from a strictly scientific perspective, more exchange of knowledge between computer science and social sciences would be good? Isn't it a shame that, for example, social media platforms don't benefit from the decades of research in social science to create a platform that's more beneficial for society? They hit roadblocks and cause problems which could have been easily prevented with this knowledge.
I think it's similar to the relationship between computer science and statistics - both can live without the other, but both have a huge benefit of connecting their knowledge, working together, and even merging their curricula (into data science).
If you haven't read the whole article, I really recommend doing so - I was put off by the title as well, but it's an interesting analysis and makes some good recommendations.
1. The courses taught are very low level, below the level of building platforms. There really isn't a social science perspective to algos, OS, CPU arch.
2. This is already the case, 50% of classes must be from outside the department with more rules about what is required
3. This is already the case
The author seems misinformed about the nature of CS programs and how departments already work.
My point about ACM pushing garbage content shall remain
> My point about ACM pushing garbage content shall remain
Sadly, I think you're right. Not sure if it's an attempt to remain relevant or if the group itself is just waning, but the quality over the past number of years has diminished. It's like they're now just trying to meet minimum word requirements.
> Don't you think that, even from a strictly scientific perspective, more exchange of knowledge between computer science and social sciences would be good?
Hmm. Not necessarily. If you're speaking strictly from the point of view of sharing knowledge, well, there's nothing really stopping that now. But if we're talking about moving computer science departments into social science departments under the pretext of sharing knowledge, I'm not sure that's necessarily a good thing. It might be, but I'm not sure.
What would the effect be of putting physics in the social science departments? Or mechanical engineering? I think if you can answer that question, you can answer it for computer science, and by extension you could just say you could introduce all of these programs into the social science departments to share knowledge... in which case you could just have a university.
I agree that the unfortunate title of the publication makes this unclear, but that's pretty consistent with what the article says:
> Despite the title of this essay, I'm not actually advocating for the institutionalized transfer of computing departments into social science faculties—such a move is no doubt highly impractical and implausible—but rather for a change in mentality, a recognition that the field now and in the future will have more affinities with the concerns of the academic social sciences, and fewer with the natural sciences or engineering.
It's true that nothing is stopping the exchange of knowledge now, but the minor, often bad ethics course in CS doesn't give students the impression that social sciences is a field they can and should learn from. That's why students should be sensitized as to which problems they may face with the software they write, which are directly related to and possibly already examined by social science.
In my opinion, this would be similar to the statistics courses I had in my CS studies: Scratch the surface enough to get a feeling for what's realistically correctly solvable and implementable by a programmer with no deep statistical knowledge. That way, no time is wasted with half-correct implementations caused by not knowing that there's a deeper, complex statistical problem underneath.
Physicists and mechanical engineers simply do not wield the same amount of power that computer scientists/software engineers do. FYI the article does not suggest literally putting CS in the social science department. TFA is just arguing that in light of how CS is applied in society, computer scientists need to be more broadly educated. I was a cryptography PhD student for awhile and I can tell you by-and-large CS academics are so far up the ivory tower it's a wonder they don't die of asphyxiation. To ignore the reality that algorithms, and those that design them, hold an enormous, and growing, amount of sociopolitical power is not just willfully moronic, it's arguably leading to the death of liberal democracy.
On the contrary it's easy to make this argument for physics.
Would physicists have developed the atomic bomb if they had been required to take a few extra courses in social sciences and if physics departments had an extra multi-disciplinary professorship? I don't believe this would have made any difference.
The suggestions are at the same time radical and pointless. Arguably he is simply trying to colonize other fields, as he accuses CS of doing.
Historically that may have been true; I won't argue that point.
The difference between the atom bomb and CS is that the latter has a ubiquity of access and reproducibility that subject it to completely different ethical and philosophical concerns. The atom bomb doesn't change or influence our understanding of how political discourse operates in modern democracies. Things like social media and GPT-3 do.
I'd like to add that, when building an atom bomb, you are well aware of what you're doing - why you're still doing it is an entirely different topic (an interesting one which, in fact, falls right into the social sciences). Software engineers at facebook, on the other hand, are likely not aware of the full impact of their work. Their seemingly harmless and objective algorithms literally, although indirectly, cause deaths, and the engineers working on them should be aware of that.
The number of post-hoc internal memos and studies clearly shows that there are many consequences that FB engineers did not anticipate (though FB is not unique in this regard; consider how YouTube's recommendation algorithm feeds misinformation & outrage).
The whole point of TFA is that pure intelligence (particularly "computational thinking" type of intelligence) is no substitute for the knowledge and methodology already explored by various humanities and social science fields.
> The whole point of TFA is that pure intelligence (particularly "computational thinking" type of intelligence) is no substitute for the knowledge and methodology already explored by various humanities and social science fields.
This is such a false dichotomy. There exists many, many technical programs that go to field and do that type of "social" research, using the precisely the same methodologies, knowledge of which end up with UX designers, product designers, product managers etc. This TFA readily admits. In other words, anything that social sciences can offer is already being incorporated into those programs. (I am talking about likes of human computer interaction, user experience, information processing science, network science, and tons of other hybrid programs)
What deaths? I would tend to disagree that the algorithms caused deaths in any event, unless they were maybe physically brought about by the algorithm in some way... You might as well say electricity or the internet itself caused those deaths.
I don't think the issue with big social networks (and other big tech actors) is one of ignorance. It seems to me many of the bad things they do are very deliberate.
Sometimes the things they don't care about are also very deliberate. Google is perhaps the champion of causing harm by not caring about things. This is sometimes ascribed as being caused by their "engineering-centric culture", but I fail to see how engineering excellence can mesh with the real world experience of using their products. A broken clock is still broken even if all its gears are beautiful.
More to the point, I really wonder how exposure to the social sciences may help fix any of this.
Strongly agreed. Exposure to social sciences won't fix problems caused by your business model, because market competition is stronger than individual morals. Even if you decide not to proceed in harmful way, there will be a competitor who's less moral and will corner the market in your place.
The more I participate in the "tech won't fix social problems" kinds of discussions, it dawns on me that it's not the tech that's the problem, but the business models behind that tech's development and deployment. Sure, in small companies, the "tech person" and the "business person" are sometimes the same individual, but I still can't think of a single example where harmful technology was irresponsibly deployed "because it would be elegant/useful/cool" - the problems happen when the motivation is, "because it's easy money".
But that line of thinking reduces the issue to the usual problems of society run by market economy, which is already a well-trodden ground, so you can't find new scapegoats there.
I agree that the market economy is the underlying problem of it all. However, I wouldn't completely dismiss computer scientists' lack of knowledge in social science as part of the reason. I'm hopeful enough to think that, knowing the incredible harm the software they are writing could inflict on other people or society as a whole, some computer scientists would be less willing to participate in it.
"Google is the champion of circumventing long established norms, such as the prohibition on wiretapping, and evading responsibility for doing so" says a lot more than "causing harm".
Blaming Google for that is some bizzare history retconning bullshit. You want the DHS, FBI, CIA, and NSA for that.
Calling Google wiretapping would be like calling your bank a wiretapper because they log every transaction. Those previously listed bad actors with badges agencies doing so would be business as usual.
It is utterly exasperating to see "Big Tech" magically become the magnet for blame of things they don't even remotely do and see people believe any ole bullshit that confirms their preconceptions.
Your hostility is a little weird. Maybe you think *.gov are worse and should be mentioned before Google. That is a legitimate position to have.
On the other hand, Google drove surveillance vans all over the world, hoovered up people's home network wifi frames, and promoted the engineer who "accidentally" introduced that "bug". If that isn't circumventing a norm against wiretapping, I'm not sure what is.
You're joking right? You're proving the article's point. The social sciences have informed policymakers for decades. The US was founded by social scientists (as TFA defines them, the social sciences include philosophy and so on).
Isn't that more worthy of damnation than praise? Superpredator broken windows, and all other sorts of convenient bullshit. Now policymakers are all about very selective hearing for demagoguery to be fair. We know this well enough because the absolute dumbasses still think cryptography backdoors are a good idea in spite of every credible expert screaming at them for decades "No you fucking morons! It would weaken our own security while not even helping for actual bad actors!"
Definitely, social sciences have gotten a lot of things wrong. And the pipeline from science (social or otherwise) to policy is pretty broken, to put it mildly. Isn't it all that much more important that computer science absorbs that history, so as not to make the same missteps?
Ah, well, if TFA defines the Founders as social scientists, then I can call today's AR-15 a muskeet. Precisely one of the issues of today social sciences is the absolute perversion of language (on purpose).
You can make everything a social science with a bit of imagination.
My, admittedly limited, understanding of current social science theory seems to claim more and more that boundaries of race, class, gender etc are real boundaries that preclude identification and understanding. I identify with men, because I am a man, I can't possibly understand a woman's point of view for the same reason. We are encouraged to believe in privileged positions where knowledge is spesific to this position and irrelevant in others.
Computing requires a common language and a common understanding, as does any discipline where you are dealing with the real world. If you extend social theory into computing you WILL end up with people claiming it is discriminatory to privilege any ideas within the field. Sounds like a nightmare to me.
You seem to be thinking of gender studies? I don't know much about that, especially in the US. It may be right to criticize that field and its methodology. Social science, however, also includes fields like psychology, history and law, all of which I assume you'd agree are hugely important for society.
Also, as I've mentioned in another comment, the article is not advocating for computer science and social sciences to 'become the same', but to improve communication and mutual appreciation and knowledge. (The title is a bit unfortunate.)
IDK - I don't think too many students get into the computing field because they are power hungry. If they gain a lot of power because they "release the next big X" then that's a byproduct and not the motivation.
Ha, I was going to make this exact argument. The flip side of this is that it cost Zuck almost nothing to get Facebook up and running with users using it. To build any type of nuclear weapon I assume that would cost millions to billions, thereby making the barrier to entry rather high.
That is not an endorsement of the authors proposal however. I would like to see it go the other way: treat it more like an engineering discipline and as part of that students would be required to attend the engineer ethics course that (almost) all engineering curriculums have.
But even admittedly that's not seeing the forest for the trees. The questionable systems that are built that are the result of many people working willingly together across a variety of disciplines (to the author's suggestion even). I'm not so sure making computing more multi-disciplinary is the right answer because in fact and in practice it's already that way.
I agree that there should be room for ethics but the recommendations in the article seem extreme and lacking.
Especially the "Web development" example in section 2 seems to be quite off. In fact the counterargument for section 2 seems to be very convincing!
CS/IT is such a broad, rapidly progressing field that you cannot be prepared enough on technical subjects.
As someone who has no formal CS education I'm impacted immensely by this. I don't only need to keep up with current technology (mostly in my free time) but also often had to catch up on topics that my peers learned in University.
I'm not complaining though. I love it. But to say that there is _easily_ enough room for additional, non-technical/mathematical subjects seems to be quite unrealistic! Especially if we look at the hiring practices of our industry. Apparently having a degree is not nearly enough even from a technical standpoint.
I majored in CS in college, and I would say there is actually easily enough room for additional, non-technical/mathematical subjects in the curriculum.
As part of my major, I was required to take: two calculus courses, two physics (or chemistry–but almost nobody did) courses, linear algebra, and two discrete math courses. I can confidently say that aside from some very basic physics from the first of the two classes, I have used absolutely none of the knowledge in any of those courses in my life, and if we are just talking about my career (mostly working on Etsy and Trello), then I haven't even used the physics. The rest, I have effectively forgotten completely.
Some of my CS electives also proved not to be that useful for much; for example, my courses on computer graphics, compilers, and operating systems. I retain some basic knowledge from those courses which sometimes come in handy for quickly understanding tangentially-related concepts, but definitely nothing to spend entire semesters on.
Of course, I'm not arguing that nobody who studies computer science should learn the things I've listed, but it seems silly to me to make them required. I took both my calculus and physics courses during my first year of college, and by my third year, I had probably forgotten almost 100% of it. In terms of my career trajectory, and even in terms of the value of knowledge/education for the sake of being knowledgable/educated, it was a complete waste of my time and money.
Often times, I wish I was able to learn a wider variety of things in college.
I think the bigger issue here is that computer science is so tightly coupled with software engineering. I hope that as time goes on, we get better at separating the disciplines in the same way that we've separated other sciences from engineering (e.g. chemistry vs chemical engineering, physics vs mechanical/civil engineering, etc.).
By the same argument, chemical engineering belongs with biology, because the chemicals produced (and their byproducts) impact biological systems. Which is true, but it still doesn't mean that the two disciplines should be lumped together.
Computing may need more exposure to social sciences, but it absolutely does not belong within the social sciences. (More accurately, product management or program management needs more exposure to the social sciences - those who decide what programs are going to do and how they're going to interact with people and society. Computing in general - algorithms, programming techniques, and so on - doesn't even need that.)
I like the idea that we should make the fact that it isn't real science concrete so people know what they're getting into.
Allying it with the bit of universities that have abandoned any idea of the value of ideas being linked to actual evidence though is a terrible idea.
Make it part of the engineering department, those guys get the meeting place between a technical discipline and something that has to actually exist in the real world.
Do CS students not already take social sciences classes as part of the general degree requirements at their university?
At MIT, every undergraduate is required to eight courses in arts, humanities, or social sciences. That works out to taking one arts, humanities, or social science class per semester. Specifically, you have to take one from art, one from humanities, and one from social science. You also have to pick a particular field and take a series of courses in that field.
Caltech is similar, but not quite as structure. Every undergraduate must take 36 units of humanities (Caltech has a different unit scale than most others--one term of most classes earns you 9 units, and there are 3 terms per academic year) , 36 units of social science, and 36 units of humanities or social sciences.
For someone who took the minimum number of units possible to get a CS degree at Caltech, 22% of their coursework would be in humanities or social sciences. 20% for someone who took the average CS course load.
I'd just assumed that it was like that at most other schools too.
> Shouldn't engineers who design social media algorithms have some basic knowledge of social dynamics and psychology?
Firstly, let’s not reduce decision making to something only software engineers do because typically product managers have huge, even the most, responsibility on product decisions.
Secondly, what makes you think they (incl. engineers) don’t have that basic knowledge? Most software people I know are autodidacts at heart and are interested in social domains just as much. These people are also fed the warm data from their social software systems and can robustly hypothesise on them (they have to) because features based on those hypotheses is applied back to real life next quarter.
If they are following dark patterns, it is not for lack of knowledge on social dynamics or psychology, but because they would follow normativities of growth and engagement rather than social good.
Looking to social sciences, there seems to be a huge variability on how reputable and reliable their knowledge making machinery nowadays works. Hoax paper crises demonstrated that there are at least some weak spots that are very gameable.
I don’t mean this as an insult to social sciences but we could also say if engineers attended church or participated in any religious or spiritual practice to deepen their moral understandings it could be a force of good, but we can’t prescribe that exposure because it would be too ideological. I fear the same might be the motivation here.
Looks like STEMvy acting up again - trouble with jobs in sector again? My mind may be frayed cynical but this bears resemblence to the same old bullshit of previous decades of academics politics; trying to bully their way onto the gravytrain while calling it "interdisciplinary" but only adding one STEM course at best while treating them as the well of all evils in the world, something they should certainly know better than given evil long predates the accused. The "Science wars" for one. I guess it is time for shitty remakes again there too.
Seriously, there is the same pattern of vague claims to benefits, one-sided elective requirements while poisoning the well to call anyone who disagrees with them close-minded and evil. Sadly/fortunately interdisciplinary never explicitly teaches this sort of bullshitting. Sadly because it could be useful in these fights, fortunately because we are all better off with a minimium of this crap wasting our lives.
Considering that Social Sciences are in midst of a replication crisis with no end in sight I'm not that sure they have anything useful to offer, except perhaps how to best do p-hacking to get results that one wants.
The claim that computing is a social science because it has big effect on society is simply bollocks (This was explicitly claimed in the article, heck it had a subtitle "Why Computing Is a Social Science").
The effects of computing to society definitely is social science. But social sciences have exactly zero to offer on how to, as an example, define an upper bound for certain algorithm.
Just as physics has massive impact on the world the actual act of doing physics had nothing to do with social sciences. How on earth does social sciences help me to solve some particularily nasty partial differential equation? How the results are used are yet again part of social sciences as that talks about what happens to society, but the actual physics is completely out of it.
And as I said at the start, the track record of Social Sciences is not too good. I'm quite doubtful they can even say anything valid on the actual matters that fall under it.
Considering that Computer Sciences are in a software crisis since the 70ies, with no end in sight I'm not that sure they have anything useful to offer, except bloat, sidechannel attacks, cargo culting and enabling of BS-jobs, which they promised to eliminate.
I agree that computer science isn't a social science, but I think the social aspects of computer science have a lot more to do with what computer science grads do than upper bounds on algorithms. Most of what we're doing is about making products more attractive to customers.
No, much of computing belongs in the business school, communications department. Facebook, Google, Twitter, and all the ad-supported "tech" companies are part of the advertising industry. Put all the application-related courses for low-level computer science over there. Web and app development go there.
Classic hardcore CS, where people make distributed databases, new algorithms, and robot control systems, worry about proof of correctness, and study automata theory, is part of mathematics.
63 comments
[ 3.9 ms ] story [ 135 ms ] threadIf it were how the social sciences need to embrace computing, that would be a legitimate claim from my experience at University.
At the same time there are usually 2-4 articles per issue that are very good. CACM also presents curated lists of interesting research articles, so (as practitioners) we do not need to sift through every article from every conference proceeding or journal publication talking about how some squeezed an extra tiny incremental fraction of x out of an O(n^x) algorithm or incremental papers discussing how compiler type inference algorithms could be improved this tiny little bit with this one weird trick.
And we occasionally get really insightful practitioner articles as well; for example, some interesting bits of how Google manages their code base, or what an incredibly difficult problems Coverity faces when statically-analyzing (supposedly)-compliant C/C++ code.
I think it's similar to the relationship between computer science and statistics - both can live without the other, but both have a huge benefit of connecting their knowledge, working together, and even merging their curricula (into data science).
If you haven't read the whole article, I really recommend doing so - I was put off by the title as well, but it's an interesting analysis and makes some good recommendations.
1. The courses taught are very low level, below the level of building platforms. There really isn't a social science perspective to algos, OS, CPU arch.
2. This is already the case, 50% of classes must be from outside the department with more rules about what is required
3. This is already the case
The author seems misinformed about the nature of CS programs and how departments already work.
My point about ACM pushing garbage content shall remain
Sadly, I think you're right. Not sure if it's an attempt to remain relevant or if the group itself is just waning, but the quality over the past number of years has diminished. It's like they're now just trying to meet minimum word requirements.
Hmm. Not necessarily. If you're speaking strictly from the point of view of sharing knowledge, well, there's nothing really stopping that now. But if we're talking about moving computer science departments into social science departments under the pretext of sharing knowledge, I'm not sure that's necessarily a good thing. It might be, but I'm not sure.
What would the effect be of putting physics in the social science departments? Or mechanical engineering? I think if you can answer that question, you can answer it for computer science, and by extension you could just say you could introduce all of these programs into the social science departments to share knowledge... in which case you could just have a university.
> Despite the title of this essay, I'm not actually advocating for the institutionalized transfer of computing departments into social science faculties—such a move is no doubt highly impractical and implausible—but rather for a change in mentality, a recognition that the field now and in the future will have more affinities with the concerns of the academic social sciences, and fewer with the natural sciences or engineering.
It's true that nothing is stopping the exchange of knowledge now, but the minor, often bad ethics course in CS doesn't give students the impression that social sciences is a field they can and should learn from. That's why students should be sensitized as to which problems they may face with the software they write, which are directly related to and possibly already examined by social science.
In my opinion, this would be similar to the statistics courses I had in my CS studies: Scratch the surface enough to get a feeling for what's realistically correctly solvable and implementable by a programmer with no deep statistical knowledge. That way, no time is wasted with half-correct implementations caused by not knowing that there's a deeper, complex statistical problem underneath.
Would physicists have developed the atomic bomb if they had been required to take a few extra courses in social sciences and if physics departments had an extra multi-disciplinary professorship? I don't believe this would have made any difference.
The suggestions are at the same time radical and pointless. Arguably he is simply trying to colonize other fields, as he accuses CS of doing.
The difference between the atom bomb and CS is that the latter has a ubiquity of access and reproducibility that subject it to completely different ethical and philosophical concerns. The atom bomb doesn't change or influence our understanding of how political discourse operates in modern democracies. Things like social media and GPT-3 do.
I feel like this is coming from a naive socially illiterate nerd stereotype. They are smart people, they might as well know much more than you expect.
The whole point of TFA is that pure intelligence (particularly "computational thinking" type of intelligence) is no substitute for the knowledge and methodology already explored by various humanities and social science fields.
If so, why didn't they make such predictions in 2012 during Facebook's IPO?
This is such a false dichotomy. There exists many, many technical programs that go to field and do that type of "social" research, using the precisely the same methodologies, knowledge of which end up with UX designers, product designers, product managers etc. This TFA readily admits. In other words, anything that social sciences can offer is already being incorporated into those programs. (I am talking about likes of human computer interaction, user experience, information processing science, network science, and tons of other hybrid programs)
Sometimes the things they don't care about are also very deliberate. Google is perhaps the champion of causing harm by not caring about things. This is sometimes ascribed as being caused by their "engineering-centric culture", but I fail to see how engineering excellence can mesh with the real world experience of using their products. A broken clock is still broken even if all its gears are beautiful.
More to the point, I really wonder how exposure to the social sciences may help fix any of this.
The more I participate in the "tech won't fix social problems" kinds of discussions, it dawns on me that it's not the tech that's the problem, but the business models behind that tech's development and deployment. Sure, in small companies, the "tech person" and the "business person" are sometimes the same individual, but I still can't think of a single example where harmful technology was irresponsibly deployed "because it would be elegant/useful/cool" - the problems happen when the motivation is, "because it's easy money".
But that line of thinking reduces the issue to the usual problems of society run by market economy, which is already a well-trodden ground, so you can't find new scapegoats there.
"Google is the champion of circumventing long established norms, such as the prohibition on wiretapping, and evading responsibility for doing so" says a lot more than "causing harm".
Calling Google wiretapping would be like calling your bank a wiretapper because they log every transaction. Those previously listed bad actors with badges agencies doing so would be business as usual.
It is utterly exasperating to see "Big Tech" magically become the magnet for blame of things they don't even remotely do and see people believe any ole bullshit that confirms their preconceptions.
On the other hand, Google drove surveillance vans all over the world, hoovered up people's home network wifi frames, and promoted the engineer who "accidentally" introduced that "bug". If that isn't circumventing a norm against wiretapping, I'm not sure what is.
Outside of that... I would say it's pretty clear that society benefits from research in law, psychology or history, for example.
You can make everything a social science with a bit of imagination.
I'm genuinely curious what you are referring to here?
Computing requires a common language and a common understanding, as does any discipline where you are dealing with the real world. If you extend social theory into computing you WILL end up with people claiming it is discriminatory to privilege any ideas within the field. Sounds like a nightmare to me.
Also, as I've mentioned in another comment, the article is not advocating for computer science and social sciences to 'become the same', but to improve communication and mutual appreciation and knowledge. (The title is a bit unfortunate.)
What would be the cost of from moving CS away from its current home?
Post 1 example of that.
It's more about recognizing that computing now controls a great deal of power and wanting a large slice of that power.
That is not an endorsement of the authors proposal however. I would like to see it go the other way: treat it more like an engineering discipline and as part of that students would be required to attend the engineer ethics course that (almost) all engineering curriculums have.
But even admittedly that's not seeing the forest for the trees. The questionable systems that are built that are the result of many people working willingly together across a variety of disciplines (to the author's suggestion even). I'm not so sure making computing more multi-disciplinary is the right answer because in fact and in practice it's already that way.
Especially the "Web development" example in section 2 seems to be quite off. In fact the counterargument for section 2 seems to be very convincing!
CS/IT is such a broad, rapidly progressing field that you cannot be prepared enough on technical subjects.
As someone who has no formal CS education I'm impacted immensely by this. I don't only need to keep up with current technology (mostly in my free time) but also often had to catch up on topics that my peers learned in University.
I'm not complaining though. I love it. But to say that there is _easily_ enough room for additional, non-technical/mathematical subjects seems to be quite unrealistic! Especially if we look at the hiring practices of our industry. Apparently having a degree is not nearly enough even from a technical standpoint.
As part of my major, I was required to take: two calculus courses, two physics (or chemistry–but almost nobody did) courses, linear algebra, and two discrete math courses. I can confidently say that aside from some very basic physics from the first of the two classes, I have used absolutely none of the knowledge in any of those courses in my life, and if we are just talking about my career (mostly working on Etsy and Trello), then I haven't even used the physics. The rest, I have effectively forgotten completely.
Some of my CS electives also proved not to be that useful for much; for example, my courses on computer graphics, compilers, and operating systems. I retain some basic knowledge from those courses which sometimes come in handy for quickly understanding tangentially-related concepts, but definitely nothing to spend entire semesters on.
Of course, I'm not arguing that nobody who studies computer science should learn the things I've listed, but it seems silly to me to make them required. I took both my calculus and physics courses during my first year of college, and by my third year, I had probably forgotten almost 100% of it. In terms of my career trajectory, and even in terms of the value of knowledge/education for the sake of being knowledgable/educated, it was a complete waste of my time and money.
Often times, I wish I was able to learn a wider variety of things in college.
I think the bigger issue here is that computer science is so tightly coupled with software engineering. I hope that as time goes on, we get better at separating the disciplines in the same way that we've separated other sciences from engineering (e.g. chemistry vs chemical engineering, physics vs mechanical/civil engineering, etc.).
Computing may need more exposure to social sciences, but it absolutely does not belong within the social sciences. (More accurately, product management or program management needs more exposure to the social sciences - those who decide what programs are going to do and how they're going to interact with people and society. Computing in general - algorithms, programming techniques, and so on - doesn't even need that.)
Allying it with the bit of universities that have abandoned any idea of the value of ideas being linked to actual evidence though is a terrible idea.
Make it part of the engineering department, those guys get the meeting place between a technical discipline and something that has to actually exist in the real world.
At MIT, every undergraduate is required to eight courses in arts, humanities, or social sciences. That works out to taking one arts, humanities, or social science class per semester. Specifically, you have to take one from art, one from humanities, and one from social science. You also have to pick a particular field and take a series of courses in that field.
Caltech is similar, but not quite as structure. Every undergraduate must take 36 units of humanities (Caltech has a different unit scale than most others--one term of most classes earns you 9 units, and there are 3 terms per academic year) , 36 units of social science, and 36 units of humanities or social sciences.
For someone who took the minimum number of units possible to get a CS degree at Caltech, 22% of their coursework would be in humanities or social sciences. 20% for someone who took the average CS course load.
I'd just assumed that it was like that at most other schools too.
This obsession with teaching (tech) workers ethics is a bourgeois pastoral fantasy.
Someone needs to read some Foucault.
Shouldn't engineers who design social media algorithms have some basic knowledge of social dynamics and psychology?
Firstly, let’s not reduce decision making to something only software engineers do because typically product managers have huge, even the most, responsibility on product decisions.
Secondly, what makes you think they (incl. engineers) don’t have that basic knowledge? Most software people I know are autodidacts at heart and are interested in social domains just as much. These people are also fed the warm data from their social software systems and can robustly hypothesise on them (they have to) because features based on those hypotheses is applied back to real life next quarter.
If they are following dark patterns, it is not for lack of knowledge on social dynamics or psychology, but because they would follow normativities of growth and engagement rather than social good.
Looking to social sciences, there seems to be a huge variability on how reputable and reliable their knowledge making machinery nowadays works. Hoax paper crises demonstrated that there are at least some weak spots that are very gameable.
I don’t mean this as an insult to social sciences but we could also say if engineers attended church or participated in any religious or spiritual practice to deepen their moral understandings it could be a force of good, but we can’t prescribe that exposure because it would be too ideological. I fear the same might be the motivation here.
Seriously, there is the same pattern of vague claims to benefits, one-sided elective requirements while poisoning the well to call anyone who disagrees with them close-minded and evil. Sadly/fortunately interdisciplinary never explicitly teaches this sort of bullshitting. Sadly because it could be useful in these fights, fortunately because we are all better off with a minimium of this crap wasting our lives.
The claim that computing is a social science because it has big effect on society is simply bollocks (This was explicitly claimed in the article, heck it had a subtitle "Why Computing Is a Social Science").
The effects of computing to society definitely is social science. But social sciences have exactly zero to offer on how to, as an example, define an upper bound for certain algorithm.
Just as physics has massive impact on the world the actual act of doing physics had nothing to do with social sciences. How on earth does social sciences help me to solve some particularily nasty partial differential equation? How the results are used are yet again part of social sciences as that talks about what happens to society, but the actual physics is completely out of it.
And as I said at the start, the track record of Social Sciences is not too good. I'm quite doubtful they can even say anything valid on the actual matters that fall under it.
I really don't get your point.
Seems to work relatively well.
Classic hardcore CS, where people make distributed databases, new algorithms, and robot control systems, worry about proof of correctness, and study automata theory, is part of mathematics.
Machine learning is a branch of statistics.