4 ms·

The Art of the Propagator (2009) [pdf]

54 points by kylerpalmer 7y ago

voidhorse 7y ago
Nice! I’ve been spending a lot of time with Radul’s thesis, which is also on propagator networks, trying to follow along and implement a reactive framework using its teachings. Fun times
- kylerpalmer 7y ago
  Nice! I just found out about it today. Is your project open-source? Edit: found Radul's thesis: https://dspace.mit.edu/bitstream/handle/1721.1/54635/603543210-MIT.pdf https://dspace.mit.edu/bitstream/handle/1721.1/54635/6035432...
amirouche 7y ago
There is also miniadapton https://github.com/fisherdj/miniAdapton https://github.com/fisherdj/miniAdapton Unlike what is explained OP, the computation is one way.
- kylerpalmer 7y ago
  Cool. Found the paper: https://arxiv.org/pdf/1609.05337.pdf https://arxiv.org/pdf/1609.05337.pdf
agumonkey 7y ago
is Sussman still fond of this ? he was talking about it in meetups IIRC
- kylerpalmer 7y ago
  I've only just heard of it from a talk he gave a few years back [1]. Not sure whether he's continued working with it. [1]: https://www.youtube.com/watch?v=O3tVctB_VSU https://www.youtube.com/watch?v=O3tVctB_VSU
philzook 7y ago
This is also intimately connected with Ed Kmett's Guanxi project https://www.youtube.com/watch?v=D7rlJWc3474 https://www.youtube.com/watch?v=D7rlJWc3474 https://github.com/ekmett/guanxi https://github.com/ekmett/guanxi My understanding is he is using ideas from propagators and other places to make a performant logic programming framework in Haskell.
joshmarlow 7y ago
The section in SICP on circuit simulation has some similar ideas. I like to think that that was an early version of the propagator idea - https://mitpress.mit.edu/sites/default/files/sicp/full-text/book/book-Z-H-22.html#%_sec_3.3.4 https://mitpress.mit.edu/sites/default/files/sicp/full-text/...
- kkylin 7y ago
  The constraint language described in SICP (AFAIK began as a paper by Gerry Sussman and Richard Stallman, continued in the PhD thesis of Guy Steele) is indeed an inspiration / precursor to Alexey's dissertation on propagators.
  - kylerpalmer 7y ago
    Yes, and with Alexey's work (if I've understood it) the propagators are allowed to hold multiple values(beliefs) based on more than one (possibly conflicting) set of constraints.
tempguy9999 7y ago
I won't be able to read the paper, but from the abstract it sounds a bit like a blackboard[0] "a common knowledge base, the "blackboard", is iteratively updated by a diverse group of specialist knowledge sources, starting with a problem specification and ending with a solution. Each knowledge source updates the blackboard with a partial solution when its internal constraints match the blackboard state. In this way, the specialists work together to solve the problem" I can't see the term Blackboard used in the paper. Probably irrelevant but just an observation. [0] https://en.wikipedia.org/wiki/Blackboard_system https://en.wikipedia.org/wiki/Blackboard_system

54 points by kylerpalmer 7y ago

Nice! I’ve been spending a lot of time with Radul’s thesis, which is also on propagator networks, trying to follow along and implement a reactive framework using its teachings. Fun times

kylerpalmer 7y ago
Nice! I just found out about it today. Is your project open-source? Edit: found Radul's thesis: https://dspace.mit.edu/bitstream/handle/1721.1/54635/603543210-MIT.pdf https://dspace.mit.edu/bitstream/handle/1721.1/54635/6035432...

amirouche 7y ago

There is also miniadapton https://github.com/fisherdj/miniAdapton https://github.com/fisherdj/miniAdapton Unlike what is explained OP, the computation is one way.

kylerpalmer 7y ago
Cool. Found the paper: https://arxiv.org/pdf/1609.05337.pdf https://arxiv.org/pdf/1609.05337.pdf

agumonkey 7y ago

is Sussman still fond of this ? he was talking about it in meetups IIRC

kylerpalmer 7y ago
I've only just heard of it from a talk he gave a few years back [1]. Not sure whether he's continued working with it. [1]: https://www.youtube.com/watch?v=O3tVctB_VSU https://www.youtube.com/watch?v=O3tVctB_VSU

philzook 7y ago

This is also intimately connected with Ed Kmett's Guanxi project https://www.youtube.com/watch?v=D7rlJWc3474 https://www.youtube.com/watch?v=D7rlJWc3474 https://github.com/ekmett/guanxi https://github.com/ekmett/guanxi My understanding is he is using ideas from propagators and other places to make a performant logic programming framework in Haskell.

joshmarlow 7y ago

The section in SICP on circuit simulation has some similar ideas. I like to think that that was an early version of the propagator idea - https://mitpress.mit.edu/sites/default/files/sicp/full-text/book/book-Z-H-22.html#%_sec_3.3.4 https://mitpress.mit.edu/sites/default/files/sicp/full-text/...

kkylin 7y ago
The constraint language described in SICP (AFAIK began as a paper by Gerry Sussman and Richard Stallman, continued in the PhD thesis of Guy Steele) is indeed an inspiration / precursor to Alexey's dissertation on propagators.
- kylerpalmer 7y ago
  Yes, and with Alexey's work (if I've understood it) the propagators are allowed to hold multiple values(beliefs) based on more than one (possibly conflicting) set of constraints.

tempguy9999 7y ago

I won't be able to read the paper, but from the abstract it sounds a bit like a blackboard[0] "a common knowledge base, the "blackboard", is iteratively updated by a diverse group of specialist knowledge sources, starting with a problem specification and ending with a solution. Each knowledge source updates the blackboard with a partial solution when its internal constraints match the blackboard state. In this way, the specialists work together to solve the problem" I can't see the term Blackboard used in the paper. Probably irrelevant but just an observation. [0] https://en.wikipedia.org/wiki/Blackboard_system https://en.wikipedia.org/wiki/Blackboard_system