Hello everyone,
I think it is important that the computer-assisted planning community develop a piece of planning software which is as easy to use as microsoft excel or one of those simple economy simulators in video games.
I see two main reasons why this should be done.
Argument 1. We need a real world working robust example to point to
Argument 2. We want there to be a real world working robust program that communist/socialist orgs can use
Argument 1 : We need a real world working robust example to point to
I'm sure you've all been in dialectics with other communists, reactionaries, lay people, etc. and brought up a myriad of ways in which economic planning is improved by modern computers. But i'm sure you all have also found yourself getting stuck in theory, in scientific papers, etc. I know I have felt a bit of a let-down being in conversations with people who are on the edge of being convinced and if only i had had a working and easily usable example of planning software to show or have them fiddle with it would have been great.
There are examples of planning software such as NewHarmony by Paul Cockshott but these are not intuitive to the lay person. they are command-line interfaces with outputs and inputs that only make sense to those already familiar with planning software. I consider these great works (NewHarmony is actually what first got me into this area of research) but they are more like proofs of concept.
Being able to tell someone that there is a working, easy to use, and robust piece of planning software they could easily go and use at this very moment themselves or which could be demonstrated is immeasurably helpful.
Argument 2: We want there to be a real world working robust program that communist/socialist orgs can use
This argument is actually two arguments together.
Argument 2.1. if there is such a piece of software, it is incredibly useful for propaganda purposes. each quarter there is economic data published in material outputs. with an easy to use piece of planning software, any of a number of organizations could show how inefficiently and unproductively capitalism is using resources.
something like "Last quarter alone there were 500,000 people seeking work and was a surplus of 1000 tons of lumber. we could have housed this many people" or similar. any communist org could produce an economic plan for a time period to show how easily people's need could be met.
Argument 2.2. there are organizations which could actually make use of such a piece of software right now. tho they are small in number and in scope, there are no doubt organizations for mutual aid or supportive housing or food kitchens which could use planning software. larger nonprofits (tho they are not socialist in almost all their forms) could make use of this planning software. local government services like community health services, public utilities, etc. could also make use of it.
even if the software is not being used explicitly by communist orgs or nation-spanning socialized production entities, it still has a use. for each org that adopts it there is more credibility. it will become easier to argue for it's wider adoption if it is proven on smaller scales.
larger firms like walmart and amazon and koch industries essentially centrally plan their entire operations but their algorithms and software are closed source. if such a piece of planning software existed, it could be used for such orgs.
Some Aspects of the Software which I feel are vital
1. the ability to plan in monetary and non-monetary terms
2. the ability to have real-time communication between a central computer and a distributed network of computers
3. a web-browser interface
4. a GPL or other copyleft license
5. transportation needs to be planned as well even if only in rough terms
7. easily defined and altered user restrictions on inputs/outputs (limiting CO2, not letting production of X good fall short, etc.)
Plan of Action
I'd like to hear how complicated you all think this would be and whether you agree that it would be a worthwhile undertaking.
Would the idea be to simply have a toy model or an actual accounting system tied to this simulation in order for enterprises to actually plan things?
I said to the cibcom people recently that they should check out existing ERP packages: https://en.wikipedia.org/wiki/List_of_ERP_software_packages#Free_and_open-source_ERP_software
There is also ogasdemo.ru which I hear will have a public release soonish. The website's design is a mess unfortunately. And of course it's mostly in Russian but that's a minor gripe (:
Regarding licenses I would suggest AGPL or the nuclear option: the SSPL. Both licenses are deeply hated by the likes of Google.
Even though I'm a maths guy I'd avoid getting bogged down in algorithmic specifics. Those are mostly interesting from a theoretical standpoint. It's enough to know that we can formulate a feasible plan in reasonable time.
@casperadmin I don't think toy models are very useful. The video game world is full of toy models. What would be a lot more useful is real-life experiments.
To expand on the licensing question, another option is the "no license" approach used by some post open source people. The only thing legal departments hate more than the GPL is ambiguous license situations. That might be a double-edged sword however.
@casperadmin i may have mispoke. i don't intend for there to be a simulation but rather a fully fledged piece of planning software that is as easy to use as those toy examples.
i want it to be an actual piece of software that a communist government or even other socialized industries could conceivably use.
@thardin there are a lot of alternatives but sadly unless it's one of the big ones there will be major issues.
in much of the english speaking world and particularly in the US, the actual functioning of a license is largely decided by case law. for smaller licenses, the case law is small and sporadic.
i think the GPL offers the best option in this regard since it's widely used and has case law supporting it.
the no-license option also prevents derivatives from being shared with the same license.
@joe I still think the AGPL is the best option, because it keeps Google etc out. What do you think of exploring the existing ERP solutions?
Writing something ourselves is going to be very tricky, because you need to actually be embedded in the relevant social context. You need actual users. I've had in mind to write some kind of collaborative inventory system. Each user publishes a list of tools, buildings, land etc that they want to contribute and the system collects all these lists. That gets you some initial data. It might also enable very rudimentary human-to-human planning to be done.
In my case this list would include lots of electronics tools, some forest and some fallow fields.
I think we should definitely explore them. I was actually entirely unaware of them until you posted them.
Is there a most popular one or a most widely used one? I watched an introductory video for ERPNext.
If we did want to write it ourselves i think that the first thing would be establishing a "philosophy" or "guiding principles" first would be in order. we could look at what features and commonalities there are for existing ERP software and have that inform our work.
Going down the programming route would be time-consuming but (as is likely evident from my post) I think it is more than worthwhile. some of the first ideas i have for guiding principles would be making the entire program modular. making it a collection of smaller, perhaps even single task programs.
if this were the approach to implementing it, we could start at seemingly any level of the planning problem (national, regional, transportational, factory-level, demand response, etc.) without too much worry of how later parts will plug in.
it is extremely complex and average user don't know how lightbulb is working or how even to play openttd
the explaination thus is pointless
https://wiki.openttd.org/en/Community/NewGRF/XUSSR%20Set/Freight%20wagon
making it 'simplified' is even more complex and this work serves no purpose anyway
the people will say 'ahh maybe works' 'maybe not'
this alone will send user into agony (and it is very very very very simple game...)
(they're all incompatible b t w)
https://wiki.openttd.org/en/Community/NewGRF/XUSSR%20Set/Railway%20tracks
no one is born knowing how to play openttd or how lightbulbs work.
ignorance is not unchangeable. but that's a difference discussion.
I've actually played that mod for openttd quite a bit (i combine it with an industry expansion mod and LARP as a soviet planner lol).
There are challenges to planning on a national scale for sure but i think you're analysis falls short in 1 key way: if we make a robust enough piece of planning software (like i was trying to describe but i may not have done it well enough) then it could be used at any level of the economic planning problem. workers in the rail industry would never have to learn the lightbulb industry, the starbucks barista would never have to learn the rail industry. the planning software would be set up so that it handles the general or abstract idea of a planning problem but can be tailored to a specific level or area of the economy.
i've worked with construction guys and contractors in small towns and the knowledge they have even after 1 year doing odd jobs and fixes/repairs will make a newby's head spin. what concrete to use, who sells it in the area, whether the product home-depot sells is the same as direct to consumer or an independent hardware store, what temperature and humidity mean to concrete mix adjustments, how many days to wait between block work and stucco and painting, etc.
altho it definitely seems like workers of an area are only scratching the surface or in mindless single task jobs, that is rarely the case.
workers on the whole either know the ins and outs of their sector or know who does know.
what i see this piece of planning software as being able to do is plug into these existing "networks" of workers at workplaces at whatever level they operate at in the economy.
my fear is a piece of planning software that doesn't work with humans. by letting humans use the software rather than the software using human labor, i think this can be avoided.
I think that in terms of fundamental principles you'd want to develop the accounting system. There are essentially two fundamental parts of such a program that you'll need to get down - the accounting side and then the planning decision making side, which must be built on the former.
Users should be able to create defined categories, like in Quickbooks, for the different inputs and outputs used in production. If we're thinking of this as double entry accounting, for goods that were not acquired on a market you would have to include the inputs required to produce them or traded for them if it's a direct exchange with another enterprise. This also means that when you produce the final outputs for the enterprise, you know what the inputs were required to produce it.
On top of the accounting system you have the planning system, which allows you to estimate what technical coefficients you'll have and iterates on those as it gets real data, and spits out several bundles of different combinations of goods you can produce with your given budget of resources which you can choose your production targets from.
Additionally, maybe you can have a consumption planning software which issues tokens for the goods produced based off of user defined rules. It should be flexible enough to create different tokens/pools of consumption goods.
The accounting side of this can definitely be based off of existing open source software, though most of the UX for those isn't great and will have to be adjusted. The consumption side can also probably be built off of the accounting payroll functions.