Reflections on computability of discourse

Message ID
DKIM signature
Download raw message
May 2015,

I'm attending to a Blockchain talk of sorts at this obscure meetup type
of an event full of bearded people going on and on about how efficient
they are at what they do.

The presenter himself was demystifying Merkle trees or something.

You know how the usual complete lack of insight and brilliancy from the
audience during the Q&A phase of so many talks is commonly associated
with how truly shallow are the people who attend these?

(Well, yeah, if you're brilliant it's not like you're going around
listening to them Enterprise duck tales all day.)

On the contrary, I believe the speakers are to blame.

Wouldn't it be wise to address that most talks are truly boring and
quite shit, in the unabridged sense of the word? BAD jokes or lazy
jokes that are simply meant to be BAD but in fact are 100% horrible
nonetheless don't make it any better.

Multiple nuances aside, here I'm sitting with thirty-something people
and trying to "get" consensus.

But strangely enough, it was all about money.

Making fools out of banks.

Six years and countless promises later, the world is yet to benefit in
any capacity from the transactional quality that maketh Blockchain for
what it is.

A data structure.

Perfect— to reason about all things text.

The foundation of mathematics effectively suggests that for any text to
be meaningful, it has to a foremost extent express proof of its own

Now, there's quite a bit of Gödel talk going around of whether that's
possible and under what circumstances; yet for all intents and faithful
purposes of rule-following, language can be expected to be sensible, and
discourse itself —computable.

Or at least, shall I say— that was my reasoning back when I was doing
things like writing blackpaper pamphlets about discursive blockchains.
Distributed peer review, proof-of-stake fact-checking, if not outright
gambling on truthfulness.

You name it.

I maintain still that certain discourses are very computable; that said
the computation can be done, should be done, but to a massive
disappointment— not being done.

And very specific kind of computation at that!

If you were to model arbitrary discourses long enough, they would look
an awful lot like screenplays, in which actors propagate the plot one
proposition at a time.

	It is clear that in the description of the most general form of
	proposition, only what is essential to it may be described;
	otherwise it would not be the most general form.

	That there is a general propositional form is proved by the fact
	that there cannot be a proposition whose form could not have been
	foreseen (i.e. constructed).

	The general form of a proposition is "This and that is the case."

[^] Tractatus Logico-Philosophicus, 4.5

Then again, if you were to model arbitrary discourses for real, like
approach them analytically, you would see that soon enough there's no
sense in talking types no more!

When the "non-essential" type information significantly outweighs what
can be expected from language-in-use by means of convention, it speaks
not of the system complexity, but rather of encoding inadequacies
brought forth by notion of type.

Hence what's commonly referred to as meta-language is effectively an
attempt at a better encoder to what's by that logic referred to as
the language opposing it.

Reply to thread Export thread (mbox)