Hrefna (DHC)
So one of my problems with RDF for a protocol, which is being highlighted by my current reading (a book by sleepingirl), is the difference between the _map_ and the _territory_.
What I mean by this, let's take a sentence:
"I need to eat lunch tomorrow"
RDF in general and, in specific, ActivityPub allow me to express this concept. But I can communicate almost entirely different things based on _what I mean by each of the words_, _why I am saying it_, and _why my beliefs hold it to be true_
1/
Hrefna (DHC)
Breaking it down we can say:
[I] <need> {to eat} [lunch] [tomorrow]
First question:
Who is "I"?
Is "I" the actor?
Well. The actor isn't eating lunch, are they? The actor is informing you about _someone_ eating lunch, but the actor isn't me.
This seems trivial, but it really, really isn't and the deeper you go down this rabbit hole the more complicated it gets.
I [as in this physical body]…
I [as in this internal psychic vampire]…
I [as in this personality]…
2/
Hrefna (DHC)
This seems pedantic, but these differences _do actually matter_ for having conversations and communicating around these sorts of things, and you and I may mean different things when we say "I."
Which gets worse when we talk about what we might mean by "to eat" and "lunch."
e.g., there's a joke running around on Facebook right now commenting on a "guy" dating a confused lesbian to find out that they are, in fact, themselves a confused lesbian.
The punchline is "well you are what you eat"
3/
Hrefna (DHC)
Even outside of idioms and what can be communicated using the same exact words, there's that everything exists in a context that is _almost always unspoken_ until you interrogate it.
Why are you eating lunch?
Why do you _need_ to eat lunch?
Do you _need_ to eat lunch, what are the consequences for not eating lunch?
_Where_ are you eating lunch?
Are you eating lunch _with_ anyone?
Each layer increases our amount of detail.
But AP asks us to map this ahead of time.
4/
Hrefna (DHC)
Even to the degree we don't need to represent it, to understand others we need a shared contextual understanding of what is being said.
Because if you say "I need to eat" and you mean "I am making a mental note to grab a snack from the cafeteria before my meeting" and I say "I need to eat" and mean "I am craving sex" we are NOT HAVING EQUIVALENT CONVERSATIONS even though our words are the same.
But in AP by and large we'd see them represented the same way, even if they shouldn't be.
5/
Hrefna (DHC)
Someone comes charging up like the coolaid man: WELL ACTUALLY YOU CAN DEFINE YOUR OWN EXTENSIONS AND…
Which is what we did for content warnings, right?
We have a long, rich history of NOT DOING THAT EXACT THING here and then just sort of living with the consequences of not doing that thing.
Let alone cases where we simply don't agree on how they "should" be used (replies coming immediately to mind).
6/
Hrefna (DHC)
Communication is _incredibly_ complex, and here we have something that essentially is designed to map concepts using a variation on _English_ with English thoughtforms and terminology.
This is not how you build a protocol. It also isn't how you build a database. It isn't how you represent thoughts in a way that others will understand at all…
…at least without a previously communicated shared context.
7/
Hrefna (DHC)
So you have a language that is here to map concepts around, that people aren't using correctly, and that has no true method of communicating the _metamodel_ between groups, and no way to establish that two groups mean the same thing, and no way to even engage in meta-communication to explain what is happening.
How is this supposed to work for a social networking protocol? At all?
8/8
GNU social JP is a social network, courtesy of GNU social JP管理人. It runs on GNU social, version 2.0.2-dev, available under the GNU Affero General Public License.
All GNU social JP content and data are available under the Creative Commons Attribution 3.0 license.