Some people were talking about going into industry PL research and I guess I'm realizing most options are at least a little morally insidious, aren't they

People talk about research at AWS forgetting that this is the same Amazon that exploits delivery drivers and warehouse workers and makes literal police surveillance tools with that Ring thing
And with that last one it's not even like an abstract connection like you could be doing PL research at AWS doing software verification that ends up being used to verify software that goes into IOT devices like the Ring

People have mostly forgotten about Facebook but they very much did have the whole thing with selling personal data for political ads and whatnot
But ooh aah they added gradual typing to PHP

I've heard many great things about people working at Jane Street and the research that goes into OCaml but they still are a finance company aren't they? We joke about people in business majors being snakes but the people with the tons and tons of money are Jane Street and the ones at the very top there
Like you're not gonna be an anticapitalist and work full time as a researcher at Jane Street to make their big capital machines run better

Galois itself is cool and fine but they are partially funded by, like, department of defence money? But so are a bunch of academic grants in the US and there's not much you can do about that
At least it's just grant funding and they don't influence what you're doing research on (because you've framed your proposal towards their interests lol) but it's still bad vibes knowing your salary comes from military money

Microsoft either has been smart enough not to have one big scandal I can remember or they're just so old that all of their scandals are from before I was even born
I mean undoubtedly a company as big as Microsoft is gonna be complicit in a bunch of little things
But I'm continually surprised at how much PL stuff comes out of MSR though like they do F# and F-star and Lean? like on the spectrum of making developer tools for programmers on Windows you've got things like C# and Visual Studio on one end and then on the other there's. Lean and its gigantic mathematical community. what's up with that

I don't have a conclusion to this thread lol every industry research lab just is part of a for-profit company that will do its usual insidious for-profit company things however many steps removed from your specific position, or funded by the DoD ig lol

@ionchy There is no ethical production under capitalism

@ionchy i have really just been Not Thinking about what i'm going to do after my phd... i suppose that's not sustainable, though...

@jmgrosen @ionchy I thought I would go into academia, then changed my mind like two weeks ago. I think it's fine 😅

@ahfrom i guess i'm worried, like @ionchy, where i would work--i feel like i'm not aware of anywhere that meets the trifecta of not-too-unethical (by my standards), interesting work, and enough pay to live where i want (my partner's in nyc)...

@jmgrosen @ionchy Yeah, we have a lot of public IT in Denmark, so that's probably easier for me.

@ionchy one thought: academia isn’t funded out of love either. So if you’re worrying about the ministry of defence paying you, perhaps academia is just as bad.

@mangoiv no yeah that's the part about academic grants

@ionchy so yeah. Perhaps the realisation here that’s more important than “people do bad stuff” is that you have to decide on a case by case basis of whether your work causes harm or is actually profitable for society and the environment as a whole.
Doing work that aligns with your values is really the only thing that counts in that case. I think you cannot ask people for more that thinking about whether what they do does indeed align.
I don’t think that the solution is to say “for profit bad” as there are very few things in our system are not for profit. That’s the (sad) truth.
And I guess there’s also that old discussion wrt whether research that can be harmful is bad in general but that’s another point.

@ionchy as an industry compiler dev, yeah, pretty much. Open-source compilers have eaten a lot of the market to actually sell the very expensive output of dozens of highly-specialized engineers (and Microsoft's Embrace-Extend-Extinguish with VS Code and GitHub has eaten a lot of the tools market), so most of the companies that can actually afford to work on compilers are big enough that a multi-million-dollar annual spend on devs pays off in internal efficiencies, and you don't get that big without being at least a bit evil.

IBM is more ethical than you'd think, but a significant chunk of our client base is in the financial sector (with the objections you raise), and most of the compilers org is about selling more mainframes (the marketing material may well be right about them being more energy-efficient than commodity x86, and the executives make good noises about climate responsibility, but I know my bonus for the next few years probably depends a lot on how many new clients we get for the various Watson AI models...).

I realized a couple years into my TT position that I got to pick two of "financial security for my family", "working a non-burnout-inducing number of weekly hours", and "my employment has direct positive social impact". I picked the first two, and keep reminding myself that I'm lucky to only have to give up one.

@bruceiv @ionchy no one will judge you for rejecting ethical concerns in exchange for work life balance and financial security, because unorganized individual choices are ultimately insignificant, contrary to what the capitalists will have us believe. One thing you can do is use your money or energy to help organize the working class

@nguermond @ionchy I wouldn't say "rejecting ethical concerns" -- there's lots of startups probably crashing right now whose big idea is "new language to run computations on the blockchain", and I wouldn't touch something that pointlessly environmentally-destructive with a 10-foot pole. On the flip side, a substantial number of my former students were in the ROTC program -- is it ethical to teach skills that are guaranteed to be used for the US military? Does it change the calculus if many of them came from disadvantaged backgrounds and may not have been able to afford the education otherwise? I never felt entirely settled with those questions.

These are quibbles around the edge though ... at least to my view, teaching people useful skills is a generally socially-positive form of employment, and I miss the work, and the sense of purpose, but not the job. My current job involves making tools for (mostly) banks and insurance companies to maintain database software; something like that job would likely exist in whatever utopian economic system you can dream up, so it's not devoid of social benefit, but I suspect under our current iteration of capitalism the net effect of my labour is likely a marginal transfer of wealth from the working class to the capital class.

Sign in to participate in the conversation

A Mastodon instance for programming language theorists and mathematicians. Or just anyone who wants to hang out.