Ecomp Software

Hey guys, HLR here. just a shout out to the empirical/computational Marxists out there, here's a quick survey about what languages, architectural decisions, etc. You'd like to see in economic planning software if people here are gonna be developing it.

docs.google.com/forms/d/e/1FAIpQLSc9OUbD5iCfGdV7GjhA-qk_VZCKm2s8z03HN46K2xrpEJuieA/viewform?usp=sf_link

Other urls found in this thread:

theverge.com/2015/11/11/9719098/fbi-reportedly-paid-1-million-carnegie-mellon-tor
ricardo.ecn.wfu.edu/~cottrell/socialism_book/new_socialism.pdf
hawaiileftreview.wordpress.com/
twitter.com/NSFWRedditGif

Answered and bumped

Bumped again

bumpity bump

have 10 responses so far need a lot more for a representative sample

Nice try CIA.

stfu, what possible use could the cia have for a survey on programming languages. FYI the NSA is already spying on the whole internet, dont like it? get TOR … personally idc tho

TOR has been compromised for a very long time now.

Nice try CIA.

I wish people would stop spreading this urban legend.

So is someone actually developing this?

t. Shill

t. some retard who heard a guy on /g/ call it "botnet" once

Dude you fucking retarded if you think TOR is cia, its open source anyone can look at the code and see what it does

just getting an initial assessment to see what people would like to program it in so we can get the highest amount of people working on it

theverge.com/2015/11/11/9719098/fbi-reportedly-paid-1-million-carnegie-mellon-tor

You don't even need to go that far, just look at all the various non-profits that donated to Tor

You really have to be delusional to think that organisations linked to the Navy would donate to a program that ostensibly evaded government snooping

That vulnerability was patched.

The government isnt that coordinated, the left hand doesn't know what the right hand is doing for example the state department using encryption that the NSA tries to break, etc. One part of the government can fund encryption research while another can try to break it. Plus you aren't getting the point of 'open-source' the whole point is anyone can look at the code and verify what it does, regardless of who contributed to it or developed it. Math is math regardless if the military paid for it or not

IIRC something like 70% of Tor's funding comes from organisations linked in some manner or another to the government

You have to be a certain type of crazy to think that so many organisations wouldn't bat an eye at funding something like Tor

TOR is used by US govt operatives in foreign countries. They need it to be absolutely watertight.

Kind of a retarded thread tbh, but anything memory safe and with solid use in statistical computing. R, Python, and Java come to mind. Looking into probabilistic programing languages such as the recent Anglican, might be a good idea as well.

P2P would be useful if you were thinking about having some sort of a democratic planned economy, or if their was risk of imperialist forces interfering. Otherwise client server would be easier to set up.

Cross platform is almost a necessity for p2p, if client server than I think linux is the obvious choice.

Why would you even want to make economic planning software tho?

market socialism richard wolff cooperatives dual power coordinated through an overall software system with a builtin switch to labor vouchers and cockshott planning techniques. just flip a switch when the rev. happens and its turnkey socialism. Of course it wont be that easy it will have to be adapted to the changing circumstances, but its a start

I don't see what use statistical languages would be for it. It needs to be high performance, probably declarative (highly parallelizable), safe, and have good networking support.

users of proprietary software will be shot after the revolution.

wheredoyouthinkweare.jpg

IMO solid high performance numerical linear algebra / sparse matrix libraries are more important than stats, although stats are helpful.

Pointing out that scaling P2P to millions of users is nigh impossible, even your bittorrent software has like 100 peers at once max, how would you network p2p software scaling to millions of users, or at least hundreds of thousands of socialist 'firms'/coops

even in P2P applications you are not directly connected to every single other node on the network.

That's actually pretty cool now that I think about it, sorry I'm pretty wiped.

Economics isn't exactly a hard science, at best it's statistics.

If economic value is objectively calculable, (or at least scarcity/demand figures), then there is a deterministic algorithm for computing the economy.

druid.io
It works now, it will work in the future. GLPK can be accessed for the purposes of linear optimization as if it were a library via JNI (see the "GLPK for Java" project). In my model that I working on, Druid both serves as a way to connect data input centers and organize data into a vast input-output table which can be treated as a sparse system of linear equations.

This guy gets its, although he should read the Bread Book on why currency must be abolished immediately.

Vulgar market "economics" maybe. This is precise - you put in numbers, you predictably get out numbers according to what your objective function determines is the optimal allocation given in-kind supply and demand.

so what you are saying is their exists a hypothetical algorithm which isn't probabilistic in nature which can model the sporadic demand of extremely large groups of people and applies equally well to a finite but extremely large set of commodities? I can't possibly imagine how that could be to be quite frank.

ricardo.ecn.wfu.edu/~cottrell/socialism_book/new_socialism.pdf

There exists an algorithm to optimize the allocation of resources and choice of productive processes given a certain set of available inputs and demand figures for ideal outputs. The mathematics of this are outlined in both "Towards A New Socialism"(posted by ), where Cockshott and Cottrell present a system based upon labor time calculation, and my personal favorite, Cockshott's essay "Calculation In Natura: From Neurath To Kantorovich", where he advocates the more rational model of accounting for things without a single common unit (aka in-kind planning or planning in natura). Modeling the sporadic demand of extremely large groups of people by definition requires a probabilistic algorithm. We aren't shooting for market simulation like Lange, though. We're going the whole way with in-kind calculation.

I should also add that we're not predicting the demands of people in the long term. Rather, the accepted model decided in the cybernetics thread is that we measure the rate of depletion of surplus stocks of goods as people take "according to their need". This rate of depletion, in turn, informs the program of what demand for goods is in material, in-kind terms. Update everything every couple of seconds like any of the programs of the companies ( Alibaba, Airbnb, Cisco, eBay, Netflix, Paypal, and Yahoo) which use Druid. Problem solved.

…if the demand isn't exactly known (i.e. not everyone supplied was surveyed for demand) or future demand needs to be predicted some time in advance.

just a reminder to people posting on the thread: answer the survey in OP :) thx

Fuck off, FBI. Not using Google to answer this.

If that were true, then the market would not function. How do things work in, say, a supermarket's grocery aisle? Management takes inventory at the beginning of the day and at the end of the day. What is taken is then resupplied from a surplus stock, and it's extrapolated from this rate of depletion how many carrots, bananas, apples, endives, etc. must be ordered to keep up with demand.

Remove the mystifying cover of money, and we see that we already have a basis for in-kind calculation in the mechanisms of today's mass-consumption economy! We're just standardizing the best practices of modern production and reorienting them to serve humanity's needs in a sustainable, rational manner.

it's just asking what feckin programming language to use, not what your social security number is. if you're paranoid use tor.

I assumed this calculation program would be hooked up to some kind of storehouse frontend, where individuals can request items remotely?

Firms continue to sell things to people (and for goods for which supply outstrips demand, they're simply given away). It doesn't matter if stuff's requested remotely or if a person goes into the firm itself. Amazon operates by the same restocking rules as Safeway today - I don't see why their cyber-syndicalist counterparts would be any different. The only difference is that numbers are punched into a computer which gives back new ones for production, instead of being relayed up through corporate bureaucrats. After all, we're proposing a generalized economic system which has potential to grow out of the decline of the current one if we create the right material circumstances and institutions, not a utopian dream without basis in material reality where people just accept every little thing you planned out for them.

I don't want my actual name attached to this if it can be done. If I log in with Google and answer that thing, it's going to have that little piece of data saying "if so-and-so answered this question, then put so-and-so on watchlist".

it doesn't ask for your name. if you're worried about ip addresses and aren't already browsing this site with a proxy then you're fucked.

I do have a proxy :^)
Google collects metadata anyways.

so? your isp collects metadata and so does every other site. You think Holla Forums doesnt collect ur data?

obvious.jpg

not to some people.

Cultural Marxism is a capitalist conspiracy

>.>

Literally who?

I can't access google docs. Can somebody copypaste what it says?

an obscure blogger hawaiileftreview.wordpress.com/


its a poll so copying it wont really help. it basically asks what programming language, software architecture decisions you would like to use in making economic planning software.