Promoting Leftism Online

Do you guys make a habit of going into non-leftist spaces online and trying to voice your opinions? I'm not talking about covert propaganda or anything, just about putting your position out there, starting a conversation and seeing where you get. It's almost become a hobby of mine to go onto Reddit and anger a bunch of libertarian types. It's too fucking easy. Shitting on the alt-right in the YouTube comments is a lot of fun as well.

Does this accomplish anything? I have a sense that it increases my intuitive understanding of political economy, but don't know if what I'm writing is actually reaching anyone. Maybe just having an opposing voice out there can keep people from falling into these right-wing ideological black holes.

pic tangentially related

Attached: namedropping.png (800x518, 136.17K)

Other urls found in this thread:

youtube.com/watch?v=M-_mRC4lKTQ
marxistpedia.org/wiki/Main_Page
theculturedmarxist.tumblr.com/
twitter.com/NSFWRedditImage

no, i think sarcasm is more effective. ex. "because corporation [x] has our best interest at heart", "the market wins yet again". most people are intellectually lazy and aren't willing to listen to new ideas but you can always chip away at whatever bullshit they believe by mocking them

It helps developing your positions. You won't accomplish anything, but it might help you organize your thoughts better.

I think memes are the most effective way. Most people are too lazy and dumb, to grasp new ideas (especially leftist ideas). Without jokes and funny pictures it just doesn't work. It is no coincidence, that the political establishment is using satirical TV shows/late night shows to shill for their agenda.

I think killing is the most effective way. Most people are too lazy and dumb, to grasp new ideas (especially Stalinist ideas). Without commissars and gulags it just doesn't work. It is no coincidence, that the political establishment is using satirical posters to shill for their agenda.

Attached: nJMP_TSSBgQ.jpg (400x614, 66.64K)

stop

Read BABOONS

Nice try, FBI.

show them this

Attached: succdem.png (955x1200, 883.48K)

My email is sage. Contact me.

I think manifesting a presence online is incredibly important. Lately I've been posting leftist memes on non political 4chan boards to disrupt Zig Forums dominance. I don't think it will change anyone's mind but it does chip away at the idea of the new right wing default in those spaces.

Attached: b033622b8fa68aaf5d1e18f73d4b20e7592380b427e4c5780605a8caf2a19030.jpg (1024x576, 173.47K)

Sarcasm and funny pictures sure as hell aren't going to do it. The fascists are much better at those than us. They don't need any actual content to their memes. They can post whatever the hell they like. That's not true for leftists. We're at the disadvantage here.
If we want to gain support, we have to paint ourselves as the collected, responsible, and morally decent side. That means not being afraid to go into the details. Doing that can be a signal as such: These people know what they're talking about. In a "post-truth" society a lot of people will be on the lookout for that. A nuanced and consistent narrative presenting novel solutions. If we can just get that impression out there we'll do well enough.
Jeremy Corbyn's success in the last British elections is proof of this. The conservatives went for the memetic strategy, Corbyn went for real policy proposals. He made big gains over them.

My question is mainly whether there aren't any better places to do this. One medium I've been using is Quora.com. A lot of people use the site and you can just about tell any story you want. There's a lot of snekboys on there and a lot less leftists. Certainly none preaching the Zizekist-Cockshottist gospel.

One thing I've thought about is getting a leftist webm thread going on the /gif/ board or whatever it is. Our webms are our best memes. I'm sure Zig Forumstards will come look out of morbid curiosity and maybe learn to appreciate us some more.
They've also got nazi threads on there that we could disrupt. We just need good videos poking fun at them.

You should post the webm of the Soviets blowing up the Nazi Swastika. That way it fits the theme of the thread yet is still subversive. Just some food for thought

Observe. A non-obvious sage

You have to know your audiance and know which tool to employ. In alot of places online nuanced well thought out narratives will just be ignored due to low attention spans.

Attached: r52lNFJ.jpg (603x636, 39.4K)

That's exactly why I'm here. I used to be a lot further left but now I'm one of these centrist faggots that everyone hates. Though in my defense I'm not really so much center because "both sides are bad" or whatever, I just think that arguing about esoteric forms of government is pointless when AI will completely reshape the political world in the next 10-20 years and none of these ideas will really make any sense anymore in context.

No. I have to deal with people screaming at me with no chance for me to give a counterargument at home already, I dont need more of it in my life.

Does that even register as a sage though?

If you make a striking claim and then have a wall of text backing it up, they'll register plenty of information about you. This is just as much a matter of seeming the nuanced side as actually being the nuanced side.

AI will unavoidably be a mess under capitalism. We'll in effect be put at the service of the machines, instead of the other way around. This can be avoided with socialism. Cyber-socialism is the only way forward.

On the internet they can't stop you from giving a counter-argument, and if they're too dumb to fucking read you castigate them for it. Can't do that at home. If done well it's very satisfying.

I agree that when AI becomes full blown the ideal government will resemble something like communism or socialism but I strongly disagree with your first statement. How would they benefit from that? When machines become better workers than humans, why would humans ever need to work? At best you could argue they'll just wipe out "lesser" humans so they can keep more resources to themselves but I think that the breakneck efficiency of an AI workforce would be able to produce resources much faster than humanity can reproduce and consume them, causing an (at least temporary) elimination of scarcity. Cant guarantee it'll go very well but I'm not convinced that the effort to overthrow capitalism before then will be worth it, especially since capitalism has already nearly brought us to this new age and we'd be setting ourselves back a bit having to rebuild after a revolution.

I've tried debating nazis on those 4chan /gif/ threads, and telling them how the Nazi economy was a huge ponzi scheme or how they left Europe in ruins. Most replies I get are some variation of "t. shlomo" or "Fuck off commie", they simply don't care about facts.

I miss rebel…

Attached: 2b607dddea100bb11f3edba2f152a9fa8e5a2aad93fe1cc704d90c5666f0606b.jpg (800x600, 38.2K)

I mean that in the market-system you have an inherent bottleneck. They'll be put into competition with one another to maximize profits. They'll never reach anything like the "ideal government" you're talking about. Instead they'll get stuck in an endless loop of stock market manipulation. Instead of providing trustworthy information for individuals, they'll try to confuse us into buying dumb overpriced products. Instead of directing the economy more efficiently, they'll try to sabotage the efficiency of their competitors.

There won't be a big "singularity" or a "robotic workforce" taking over all human tasks. That's sci-fi nonsense made up by people who can't think. We'll just see a iterative growth of automation, and use of digital models by large corporations. It'll get us into a big knot we can't get out of. Information and capitalism don't mix.

Nazis are a special case. You just gotta humiliate them with memes. Their fragile ego can't handle a leftist being more of an alpha than them.

I do the opposite

Attached: bushmissionaccomplished.jpg (460x288, 30.5K)

I'm sorry but that's just completely wrong. The singularity is an idea accepted by most of the people in this field and it basically nullifies everything you've said. Once a machine is capable of improving it's own intelligence, which is not at all farfetched, it will obviously begin to improve exponentially until it hits the physical limits. At this point everything we know about the world and especially economy breaks. Presumably it will be able to give us new information about the universe that will change everything about our lives but on top of that, and most importantly, this thing will have essentially infinite power relative to what any human or group of humans could have. Because of this, whatever the goals are of this machine are basically completely unstoppable except with another machine of comparable power (and it will be able to easily stop these from being made assuming they aren't made at the same or nearly the same time). So basically if this machine has some kind of stated goal, no one, no matter how rich, will be able to stop it. If its goal is to create an ideal utopia you can bet its gonna do a damn good job trying. Everything beyond the singularity is truly a mystery but the idea of the singularity is not a wacky idea from people who "cant think", you're dismissing an entire massive field of research that you clearly dont entirely understand without any grounds besides some silly ad hominem. At the very least it is absolutely undeniable that intelligence more powerful than human brains are possible, and pretty much everything I've said still applies even if only those are possible and the singularity (which is loosely defined anyway) somehow is not.

Not Go to leftist planets offline and be silent about your opinions?

Your entire argument falls when the machine is programmed to only think about the profit motive and not the material interests and needs of the worker/robot-worker.

Which field are you talking about? Science-fiction authors?
I always find this line especially stupid. We barely know anything about our world and economy. Economists are overpaid theologians. What's there to break? Will gravity reverse itself? Will bears stop shitting in the woods? This is some prime nonsense.
How the fuck do you even "research" this bullcrap?

Attached: alice-in-wonderland-caterpillar-smoking.gif (500x390, 487.98K)

Duh. But it's possible to program it for something besides that, is it not? I'm not making any statements about how likely it is that someone will program it to be benevolent.

If you don't know how they research this "bullcrap" then it's even more clear you haven't looked into this at all. There are heaps of organizations doing tons of research on these things. It's not just a bunch of dudes smoking joints and going "woooah robots dude" either, these things can be mathematically modeled. You can feel free to keep on living without looking into any of these things but if you aren't the slightest bit aware of this field of research you aren't really qualified to call it bullcrap as if you're some expert. I'm no expert either, it's not like you couldn't point out holes in things I've said if you knew anything at all about AI safety research.

To get a wage.

There's no reason to think any machine's goal would be toward any sort of place most of us would want to live in, unless humans with ideals similar to ours direct the intelligence toward that goal. Thinking that the goal comes about miraculously at a semi-eschatological singularity is silly. I would even say that the singularity cargo cult is the reified impotence of alternatives within neoliberalism, the hope that technology by itself will solve politics for us.

Who is doing this research? How are they doing it?
That's because Elon Musk likes spending his money on shit like this when he gets high. It's nothing but a bunch of self-important silicon valley twats having a good think about it.
There's nothing magic about "mathematical models." They're completely dependent on the ideological assumptions that go into them. Oftentimes they just function as a way for autists to hide how vacuous and far-fetched their ideas are. To give it some apparent academic rigor. That's 100% what's going on with whatever you're talking about.
We can't even mathematically model our current society, how could we mathematically model what happens when some super-AI takes over? We can't. It's bullshit.

If you want to have an understanding of how capitalist society might react to improving AI technology, start with Marx.

Thats my point, the idea of getting a wage will be outdated by that point because it won't make any sense if machines who dont care about wages will be the ones doing the work. The idea of wages relies on a world without greater-than-human level AI to even make sense at all. But that seems crazy to anyone living in this world because it's all we've ever known. You know there was a time before wages existed right? You think everyone just starved to death then?

I never said it would come about miraculously, the implication was that someone explicitly programmed that as its goal. Obviously they could put in literally any other goal. Or maybe they cant, maybe the changes from self improvement would be so extreme they would eliminate any goal originally programmed in, I dont fuckin know. That was an example not some kind of magical foresight.

Go back to living in your blissful ignorance, this technology will keep moving regardless of your insistence that it won't.

The idea of getting a wage has been outdated for about two centuries now. But between the moment when a social relation is outdated and the moment when it's effectively gone, there's this little thing called a revolution.

I'd agree with you there but who's to say an AI takeover isn't a revolution? It's the ultimate one in my opinion. Much more effective than one wish squishy human-boys

With*

Meaning?

u got me

I occasionally speak up if I see Zig Forumstards spamming things, but that's about it.

precisely

Attached: jeb.jpg (800x606, 78.74K)

Meaning replacement of flawed self-interested humans with robots in government positions. And all other jobs naturally.

I should mention these are all still me advocating AI shit, just switching between phone and PC

I do particularly if a loy of Neo Nazi and other types of Reacrionary users are there so I make sure they don't take over. I managed to at least get a few people interested in leftism. Problem is it also had a bunch of Reactionaries trying to constantly debate me no matter how often they lose.

I try to teach people how to understand IRL fascist methods when they appear.

One thing neo-Nazis like to do around here is put up flyers in public spaces or throw them onto people's lawns weighted in plastic baggies. What invariably happens is that some idiot liberal contacts the media or puts it on Facebook to "warn the community" and it goes viral locally, and then the media contacts the person who found them. This results in the flyers getting in the local news where far more people will hear/read about them than had the media never reported the "shock" story in the first place.

This is an old Klan trick dating back decades. They make it look like there's a Nazi army on the loose when it's usually only a couple of guys or even a single dude who spends his weekends driving around to various towns and tossing flyers. It's media manipulation and this is how they recruit, and also to polarize and set people against each other. The alt-right's methodology is also to "reveal the bigotry of the left" (one way they like to phrase it) where they pull a trick by presenting – say – an anti-immigrant message without overt Nazi symbols, or they'll insert a fasces (say) instead of a swastika, even though it's an actual neo-Nazi group doing it. So the idiotic right-wing Trump people who don't understand this will see the local liberals (who realize it's a neo-Nazi group but overreact) in the process of freaking out and they'll go "see, the left calls everything Nazi these days." Then the liberals call the conservatives Nazis, and it just gets worse.

Cue a total blowup on a local subreddit and the like, and people tearing at each other. Meanwhile the Nazi group is chuckling and pinching their sides.

And I know exactly who is putting up these flyers and their names of who they are in many cases, what they're doing and the reaction they're trying to get, so I'll just explain that – and usually I'll get upvoted to the top. In my experience, most people do not like Nazis but they are often ignorant of how they work so they either overreact or end up getting played by their tricks in some other way. I mean I'm not doing much but nobody else is doing it?

Attached: starship.jpg (664x377, 2.46M)

Flyers like this?

Attached: c5c5decf09511bc84082f264bcd90b9d8c30fe6fbc545cd6635d377373999030.jpg (1536x2048, 464.25K)

It's not just media manipulation, it's the liberal media deliberately playing along to promote fascism.

For what it's worth I'm not particularly leftist, rightist or any type of "ist" I just like good ideas that work. I come in here, not to spread my opinions, but to see what people here think and I do this in a number of other venues. Having experienced a wide variety of ideological shitposting boards I can tell you that you'd be surprised how many correlations there are between your various ideologies when you set some wedge issues aside. Basic human motivations are natural and common, the solutions to them are not mysterious nor unique to any one ideology.

If I were to start spouting off opinions that blatantly contradict the motif of this forum it would be disruptive and I wouldn't benefit from it in any productive way. Being opinionated tends to irritate people and results in defensive shitposting so I prefer to "learn the lingo" of each place I visit and introduce my ideas in a more conciliatory fashion.

In this way I find that people are more amenable to what I have to say. I also tend to steer clear of posting 'le ebin memes' since they are fairly empty and let my arguments stand, or fail on their own merit.

...

See, case in point: re defensive shitposting

It questions nothing of merit in order to assert nothing of merit. THis is what you'll get if you wander around the internet shovelling shit opinions into other people's screens. Wasted time and data.

If you're interested in accomplishing anything, you're not going to do it online. Read as many books as you can and find some place in the real world where you can contribute. The internet is just a time killer with some interesting factoids, it overestimates its' relevance and squanders its' potential on frivolities.

What really matters is outside where the metal meets the meat.

The best option is to point to current examples of Socialism Succeeding (Cuba China Vietnam Laos DPRK)

Attached: b4ea6119e4e143a7ad62ae0b18ed5e1d4dab87f2e9e7d1ed6c3942fa431cec36.png (158x143, 29.47K)

How do you get from here:
< AI is created.
to there:
< Robots replace humans.
?

China Laos Vietnam and Cuba are the same economically why do you support Cuba but not China?

I admit I know nothing about Laos. Cuba is (until their capitalist policies take action) more similiar economically to DPRK than China or Vietnam, which is soclialist country. China and Vietnam however recognize private property and wage labour is used to produce stuff for market exchange. No amount of marxists in goverment or absence of liberalism is going to change the fact that it is (state) capitalist regime, NOT socialist regime in any way.

Watch russian Zig Forums youtube.com/watch?v=M-_mRC4lKTQ or read Marx.

It could also be that you're not saying anything substantial. What are some "good ideas that work"? Maybe knowing that will help me imagine what you're accomplishing.
On the other hand, being opinionated also gets their attention, and helps clearly delineate what you're talking about. Your post doesn't contribute anything if they're simply reading their own prejudices into it. Then you're simply part of the spectacle.
This is standard leftist practice. We call it "immanent critique." You have to undo capitalist ideological constructions on their own terms, from the inside. That's how you make progress.

AI is already created dumbass, I'm specifically talking about AI that's smarter than humans. Also when did I guarantee humans would be replaced? I'm talking about my ideology here not a magical prediction of the future. If they're smarter than us it's not like it'll be hard for them to replace us if someone wanted them too though.

Pointing out that these Greek statues are not, in fact, a result of "white culture," will stick with people the next time they see such images. This will make them less prone to take them seriously. Gradually they'll lose their interest in right-wing ideas.

This is important to remember: You usually aren't going to change people's opinions by convincing them. Instead try to implant new narratives. The next time they go out into the world, chances are that they'll think back to what you said, even if they aren't aware of it. Then they'll naturally develop similar positions to you.

I hope his fashy daddy dom is splitting his anus asunder as we speak.

Or would could make a classical left wiki so we don't have to do a research project every time we want to respond to someone.

Who would want that and why?

I've been trying to do that but it's hard to get off the ground: marxistpedia.org/wiki/Main_Page

Me, for reasons I've already explained but they can be summed up essentially with: AI could be created to be far superior "leaders" for government than any human ever could because they would theoretically be able to calculate the best solutions to problems much faster and would not be obstructed by self interest, eliminating all the problems that come with all other forms of government. And yes, before you ask, I am well aware that someone could fuck it up and the AI could do things that are non-beneficial. That's not my point. My point is that it is POSSIBLE for it to go very very well, though I am aware it is not necessarily very likely.

You own a smarter-than-humans AI and a robot army to put it on? Who will? Or rather: what class will? What are this class's interests?

Transhumanists are so fucking delusional it's really sad tbh

What do you gain from being a retard and straw manning me? Your question was "who would want that and why". I would want that. I said nothing about owning it. Nor do I know anything about how it would/will happen. For the billionth fucking time I am making no claims about how it WILL happen, only how I want it to happen.

Thanks for your contribution to the discussion, you completely changed my mind, I am no longer a transhumanist.

You should write novels then, not discuss politics.

First of all, we still can't even define or understand human intelligence, so the idea we can create "artificial" versions of something we have a tenuous grasp on suggests a very naive, amateurish understanding of neuroscience and computer science. This is not even consdering the technological renassance that would be needed to produce technology that performs like the human brain but better - or the technology needed to be able to reduce human intelligence to a tractable network.
You just made two valuative judgments about problems and solutions here. Who determines what an actual "problem" is - since some things can be good for one group and bad for another. Furthermore, who determines what's "best" and how? What if the "best" solution to reducing crime in the US is to institute laws that say only people of European and Asian ancestery can live in the US? Certainly, that would drastically reduce crime - but now you've talked yourself into another retarded corner with a dunce cap on because you have to now not only define intelligence, but define what a "best" policy is. Again, this is setting aside the obvious issue with treating a country's "problems" like a math problem on your homework that just needs sufficient computation to figure out the right answer. Is "best" what benefits the most number of people? Is "best" what benefits the economy? Is "best" what minimizes damage, spread out across secors, or what minimizes damage in a particular sector?
The central problem with government is physical coersion, not humans making decisions, and this half-baked science fiction novel you think is reality doesn't do anything to solve it.

Will do.

Correct
This only suggests that we can't create artificial versions of this now, since now, as you said, we only have a tenuous grasp of how intelligence works. This in no way proves it is impossible. What evidence do you have that the brain works with some magical essence that can't be simulated in a computer just like literally all other physical things can be given enough computing power?
The people who create the AI. There is no grand cosmic moral constant. If the people who create the AI want it to instate Sharia law then it will do that. I don't want that.
Then great, we've found a solution, all that matters to me. Though ideally we'd be looking at a global scale and not just one country.
Assuming that's how the AI defines it, then yes. If it were up to me I'd go for something similar to this, a utilitarian approach. And yes I know about the torturing one person to keep dust out of everyone;s eyes bullshit thought experiment, it's invalid because with practically unlimited resources and the ability to put everyone into a simulation there's no reason any retarded situation like that would happen in the first place.
Again, as with all of these things, there is no central grand constant unquestionable answer, its literally just whatever the person/people programming the AI decide to put in and how that effects the end decisions of the AI. "Good" is a human-defined concept, it is not inherent in the universe and so a machine with no built in feeling or sense of "good" will only develop one if it is taught it. Who teaches it that and what they decide to teach it are completely unpredictable. You, like everyone else here, are straw manning the fuck out of me. My argument can basically be boiled down to "a good AI leader would be good" which is an almost entirely meaningless platitude since "good" is subjective, the only real claim of substance being made is that AI would be more effective in instating whatever the arbitrarily defined definition of "good" is than a human or group of humans would. Anything you accuse me of arguing beyond that is pure strawmanning.
Mhmm, definitely no problems with government officials making decisions based on what benefits them and not the people right? Lobbying and other forms of bribery definitely aren't problems right?

holy shit, please stop.

I agree.

If the people who define what the problems are do a decent job of it then yeah. I highly doubt it will suggest genocide unless some dumbass decides to tell it to "do what's best for the white race and no one else". You expect me to believe a human run government is any less likely to suggest ethnostates and genocide? How's that been working out for you so far? Doesn't look like too well considering history.

Good job debunking all the points I made instead of just picking one statement out of context and ignoring the actual point made after it. My mind is completely changed, I'm an anarcho-primitivist now guys, technology is evil and building things smarter than us could never possibly be helpful to us in any way. Any more smart ideas?

You don't have arguments or points, you have sci-fi fantasies that, if they ever actually manifest, will take place so far in the future the context under which they emerge will be so radically different from the world you know now you might as well not bother with predictions, plans, or half-baked debates.

Fair enough, debating isn't gonna change what happens, I suppose we must just wait and see.

...

My main point of contention is whether "an AI smarter than humans" will be invented at some discrete point in time and subsequently revolutionize anything. I call bullshit on that. Instead we're going to see a gradual improvement of AI taking over more and more human tasks. Whether we live under capitalism or socialism is going to define the character of this evolution.
There's a good case to be made that the architecture of the human brain exhibits properties that are too difficult to capture digitally. Of course, we really don't know at this stage.

Okay I'll give you that one. When I started talking about the only substantive claims being made I was primarily talking about the topic on hand at the time which was not as broad. But you're right that I had made much larger claims before.

I actually agree with all of this as well. I think the only distinction I would make is that, although the change will be gradual, it seems like a reasonable assumption based on what we know that the pace of change will pick up a bit as time goes on, looking a bit like an exponential curve. This is for a lot of reasons, towards the end it might be because the AI itself is doing research on how to make AI more effective, but earlier on it might be because AI taking over jobs will leave a lot more people open to pursuing AI research instead of whatever they would have been doing otherwise. But of course there's a fuckload of other details to consider and it will depend heavily on how we handle jobs being handed over to machines.
I also agree that it might be impossible to create something as powerful as a human brain artificially, but I feel pretty confident that its possible. Do I have any solid evidence for this? Nah, it's just my opinion. But there are a lot of researchers who agree and who could probably explain a lot better why it might be possible, I'm just a stupid CS undergrad.

Attached: liberal.jpg (600x337, 45.93K)

The closest thing I get to that is simply promoting social democracy
Italian easier for normalfags to swallow than socialism, since ☭TANKIE☭s give socialism a bad name

I also like making fun of 56% white neo nazis

I have a leftist tumblr and every day I hate being on tumblr more and more.

share, was thinking of making one and need people to follow

theculturedmarxist.tumblr.com/

This.
Placing yourself in a position where you have to actually articulate your views comprehensively is one of the best things you can do to stimulate critical thinking. Just realize that you can't change anyone's mind online.

at this point i think promoting leftism must go further than just online. it's easy to dunk on people on the internet and usually consequence free. i think more socialists need to do things outside the internet including charity and community service, these are things that would not only boost your popularity and make you look good but there's also literally no way to have anyone smear you or your group if you're doing something like feed/clothe the homeless

Attached: stalin.jpg (620x413, 34.61K)

This is true. How do you suggest we organize physically though? Join some existing leftist org? Usually this board comes to the conclusion that they're all shit.

I've learned through experience to never trust this board with organizing, only discussing of theory and praxis should exist here

This is still true. Even using our most advanced technology we can't duplicate every detail of a bird's wing.

followed nerd
I don't even like tumblr that much

You could change some people's preconceptions as to what communism may be, but that doesn't mean they're going to start waving red flags, attending black bloc rallies or quoting Zizek. While humour can certainly appeal to certain groups, going out and debating people on purpose isn't a garuntee in making them turn leftist.

What it CAN achieve however, is you learning to articulate your points/ strengthening any other future arguments.

ah yes I see you like ropeplay
anyways

nah, people think DPRK is like hell on earth, despite any information to the contrary

It's not.