The problem with the "F/OSS means hackers already control massive amounts of the means of production, what if they went on strike" is (besides the obligatory "organizing is hard") that the primary work that they do is /creation and maintenance/ of productive systems, not /execution/ of those systems.
So the impact of a simple work stoppage would be the slow degredation (including increasing amounts of scattered failures), rather than the sudden total ongoing effect that strikes generally aim for.
Like it's not impossible, it'd just need some intentional and creative ad-hoc planning.
And each one would probably have to be unique.
@gaditb what do people think “means of production” means? unless they own the hardware and are financing the electrical bill, they don’t own it
if they do, they can just turn the computers off
@u2764 I mean I could also, be misinterpreting them,
@gaditb no i have seen these takes and think that they just don't realize that “owning the means of production” is about owning the means of production OF VALUE, and not owning the means of production of free software code, which is by design completely valueless
@gaditb most of the infrastructure, servers, power supplies and other utilities, manufacturing tools, and other equipment upholding both the internet in specific and technology in general are not owned by hackers, they are owned by massive corporations like amazon and facebook and comcast. we know this. hackers own laptops, not AWS. that’s where the value is produced and that’s what needs to be seized
@u2764 I think there are limitations to what AWS can successfully "own" of the things running within it. Ownership is about control, which requires knowledge -- sufficient illegibility limits ownership.
@gaditb i think the marxist analysis is very straightforward of the bourgeois printing press owner who profits off the labour of the workers running the press. i don’t think said owner needs to know a thing about what is being printed, or needs to care
@u2764 Yeah because the extraction of value is off the physical output of that labor, not the intellectual output. They're not seeking to coopt value through manipulating the content -- to the extent they are manipulating the content, the goal is to preserve value extration happening elsewhere, not to profit directly from the result of their manipulation.
Similarly, the landlord profiting from the press organization -- whether a capitalist's company or a worker-owned enterprise -- has extremely limited care for what is happening on the property itself, just that the most rent they can squeeze is paid.
@gaditb my point is that technology is the same
the value comes from the physical output, not the intellectual output. the physical output being sequences of electrons instead of sequences of printed pages does not change this
marxism is about owning the physical means of producing physical things
@u2764 Not in its opposition to landlordism, it isn't. The opposition to landlordism is about owning the physical means of POSSESSING physical things.
Similarly, here the value isn't in the sequences of electrons being output -- it's in the NOVELTY of the specific sequence. It's not like `cat file1 file1 > file2` creates a file of twice the value -- even though it has twice the sequences output.
Likewise, it's the creation of Microsoft Word version [however they name Word versions/version-increments now] that is being monetized, not the repeated copies. It's the unnamed AWS automation system. It's the released version of some ad-hoc translation software.
The separation between the labor of production of the novel sequence of bits and the extractable monetary valuation of the downstream automated-labor isn't an incidental detail, it's core to the economic relation. If Marx isn't about that economic relation then okay, let's drop him and find or hack together some analysis that IS.
@gaditb i think the situations are more similar than you are meriting. the value of a printing press is, of course, related to the value of books, which is related to the value of what is in the books. but the POINT is that it doesn't matter how good of an author you are if you can’t print the book, and you can’t print the book without using a material means of production which someone else owns.
software is worthless if it can’t be distributed. the value in software is in distributing software. the value is a function of how good the software is, of course, but that doesn't change that the LOCATION and SOURCE of the value is in the distribution. if you don’t own those means (the means of production), you cannot produce value without paying someone who does
@gaditb (there are, granted, other kinds of “value” produced by some kinds of software, for example datasets used internally by microsoft to make key business decisions, but i guarantee that tech workers don’t own the physical means by which those programs run or operate either)
[intellectual property is another topic which is a lot stickier, but honestly i don’t think it’s worth spending that much time on when there are such big obvious non-intellectual sources of property in tech, and intellectual property is often not actually that valuable anyway, which is why open source has won]
@u2764 In a "I'm interested in your thoughts if any" way rather than in a "I think this as evidence supports any of your or my points made" way (maybe it might? I dunno.)--
do you have any thoughts about RubyCentral's rubygems coup and the gem.coop response (if you've been following), in the context of this conversation?
@gaditb i have been following somewhat (i program in ruby for my daygig so i kind of have to be somewhat aware)
but, well, the recent example i would cite regarding “(not) owning the means of production” is the recent bitnami fiasco <https://github.com/bitnami/containers/issues/83267> which maybe hasn’t been as big of a news item but has been causing lasting material issues not only at my workplace, but at libraries around the world (and basically anyone trying to use Helm charts to deploy Apache products like Solr on a public institution‐sized budget)
regarding gem.coop, right now the details are very vague so it’s hard for me to analyse. it seems like they are mostly just a RubyGems mirror right now, which means they are still dependent on RubyGems, just through an extra layer of indirection? but in principle, owning the domain is good, owning the servers (altho i doubt they actually own the servers) is good, etc…
@gaditb what i think i was trying to get at with my argument though is that programmer institutions are largely renters, rubygems and gem.coop included. they rent servers, they rent network connections, they rent a place in the DNS. and having more or less control over the institution itself, while good, can only get you so far when the institution itself is renting the things it needs to operate. and i think the most obvious example of this has been payment processors and pornographic content—when you are renting your space in the marketplace, due to reliance on payment processors, you have to follow the marketplace’s terms. you do not actually have full control over your production. but the same applies to DCMA takedowns and DNS blackouts and really every step of the software distribution chain.
@gaditb so i think that to the extent that RubyGems/gem.coop is a distribution platform, having control of that platform is good. but there are still going to be limitations to what they can achieve, because that platform itself rests upon rented space.
@gaditb in terms of the radical potential of work stoppages, which you were talking about in your first post
well, if everybody depends on gem.coop, and they say “agree to our terms, or we shut down”, that certainly is some leverage…
but i don’t think that’s the kind of power they are really trying to attain right now. but this is a kind of power that corporations ARE willing to leverage, as i mentioned with bitnami above, who is saying “pay us massive amounts of money or lose access to secure software you depend on” as a form of value extraction. independence from that kind of reliance on corporate goodwill is good, but i haven’t seen the free software movement willing to actually turn around and use those same mechanisms to achieve any of their own aims, despite the fact that the world runs on free software and depends on their goodwill just as much.
well, some maintainers have decided to stop respecting vulnerability confidentiality timelines, which might be a start
Backing up your point on this:
"""
Thus, as Mark Atwood, an open source policy expert, pointed out on Twitter, he had to keep telling Amazon to not do things that would mess up FFmpeg because, he had to keep explaining to his bosses that “They are not a vendor, there is no NDA, we have no leverage, your VP has refused to help fund them, and they could kill three major product lines tomorrow with an email. So, stop, and listen to me … ”
"""
https://thenewstack.io/ffmpeg-to-google-fund-us-or-stop-sending-bugs/
(I feel like I don't quite get the mechanism, tho. Like just, "we won't maintain this feature [that $productLine relies on]"? How does that kill them /tomorrow/, tho? And why can't they just do Libav-take-two?)
(You don't need to answer that or have an answer, the point of this post is just to link in this relevant quote. Just, I wanted to post it as a complete thought-from-me.)
@gaditb (reading between the lines, they probably have at least three product lines not complying with the licence 😝)
@gaditb @Satsuma it’s legal power which is inherently idealistic and i think most marxists would be a bit skeptical of
that skepticism is currently being proven correct as AI companies perform rampant copyright infringement with as of yet no real penalties from the courts, tho i guess we will see if those receipts ever come due
in the united states tho we actually invest quite heavily in the ideals of the law. our courts are part of our national narrative and identity and there is a state interest in upholding legal rulings. insofar as the national interest and the interests of corporations do not align, legal power is a way of leveraging one against the other
but Congress can always pass a law saying that ffmpeg is public domain if they really want. so the power seems contextual, and limited, but at the same time it is real power, and thankfully the state is more interested in backing legal precedent and its own legitimacy than the interests of capital in this case
@gaditb @Satsuma marxist analysis falters here because marx was not writing from a place of a firmly entrenched, literalist, constitutionalist national state and later marxists were typically writing from a pre- or early-industrial perspective (so, likewise). so i think the theory of how marxists can use the existing national state against the interests of capital is underdeveloped. but necessarily, this can only work against capital, not against the national state itself, so it only works insofar as those two can be distinguished
when you are dealing with things which do not serve the national ideology, for example indigenous control of traditional knowledges, the law cannot help you (it does not recognize any such ownership). we have lucked out in that there is a compelling interest, both national and commercial, in recognizing software as intellectual property, and now that it has been asserted the state will back us on it, thanks to its interest in preserving its own legitimacy
@gaditb @Satsuma so i don't think it is like a strike because it is immaterial (there is nothing materially preventing violation of software licences, whereas a strike affects what is materially possible). this doesn't means it is powerless but it does mean the power depends on enforcement, which translates the ideology into material consequences (i.e. a judge issuing a fine), which ultimately is a function of the state
but i don't think that, in the united states, we can effectively avoid using the state. maybe in marx's time everything could play out on the material plane, and workers could stop work and employers could face immediate material consequences and that was enough for a revolution, but so much of the functioning of society is currently routed thru the state and so certain revolutionary actions being routed thru the state also seems inevitable. but there are of course limitations
@gaditb @Satsuma the united states is nationalist first and capitalist second. the nationalism is frequently bad (white nationalist, christian nationalist, colonialist), but it is not the same as the capitalism, and i think it is possible to imagine other “nationalist” programs (inclusive pro-immigrant ones and/or indigenous sovereign ones), some of which are currently competing for power. regardless, the interests of capital are generally globalist in character and there is tension between them and the national interest which can be exploited. IP law is a weird example for this, because it has broad global concurrence and is not obviously nation-building, but the investment in state legitimacy turns it into a weapon. in the united states so far (this is trying to be undone) we have invested very very heavily into the legitimacy of the state, the power of legal precedent, and the inflexibility of law (via our fondness for textual literalism), so this weapon is pretty sharp
@Satsuma @u2764 Hm. Looking at how it fits into this conversation, then -- would that be a strike, tho?
Or I guess more generally, would that be some form of witholding* the value-generating labor in order to cause economic damage to the owner of the capital?
(* or maybe just strategically using the value-generating labor? I guess more specifically -- I don't see how the power to kill ~three product lines like that derives from the labor they put in. It's a power purely operating and only functional in the domain of intellectual property (however you define it).