Show newer

@clayote i don’t mean gnusocial exactly, i more mean ActivityPub and ActivityStreams were built by the Social Web WG, which was formed out of the OpenSocial foundation, which built its specification in concert with Portable Contacts, which OStatus and thus GNU Social also used

i’m positive the fediverse looks very different from what the people developing the original Portable Contacts / OpenSocial specifications, in 2008, were imagining. but i also don’t actually know: what WERE they imagining? what problems were they trying to solve? were they at all successful? i think largely maybe no; i think what applications people wanted to write and what parts of the specification actually got picked up changed quite a bit from what they were originally going for. but it’s hard to chart a narrative in my head when i don’t even know who they were, really, or how they found the time to be working together on all of this, for so many years, if indeed they did stick with it from 2008 to 2018 when ActivityPub became a thing

all we have is their weird software detritus that we’ve picked up and repurposed for our own uses

Show thread

obviously there WAS a community of weirdos who were willing to put the time in over decades to write specifications and build software just to talk to each other, but like, WHAT kind of a community? how? why?

Show thread

can’t sleep so i’m thinking about defunct social web specifications

@gaditb JavaScript was written by Brendan Eich; PHP was written by Rasmus Ledorf; Python was written by Guido van Rossum; Perl was written by Larry Wall; Zig was written by Andrew Kelley. These are massive oversimplifications, but the question remains

I know for a fact there are lots of talented trans women right now working on other people’s programming languages. that’s what makes me wonder why they don’t have one of their own

@gaditb i don’t mean contributing, i mean controlling, tho

@aescling @wallhackio makefiles mostly just read like technical specifications where there's a big list of definitions at the beginning and you need to just skip them and then look back at them later when you start getting confused

today¦s cursed G·N·U Make recipe 

@for target in $$($(MAKE) -s -f '$<' listout QUIET=1 $(SHUSHEOPTS)); do if test '-d' "$(BUILDDIR)/public/$$target" && test '!' -d "$(DESTDIR)/$$target"; then mkdir -p "$(DESTDIR)/$$target"; fi; for file in $$(find "$(BUILDDIR)/public/$$target" -not '(' -name '.*' -or -type d ')'); do printf '%s\n' "$$file" | sed 's:$(BUILDDIR)/public/:/:;s:/_\(.*\)_\(\..*\)$$:/[\1]\2:g;s/¦/:/g;' | xargs printf 'Copying <%s>…\n' >&2; printf '%s\n' "$$file" | sed 's:$(BUILDDIR)/public/:$(DESTDIR)/:;s:/_\(.*\)_\(\..*\)$$:/[\1]\2:g;s/¦/:/g' | xargs cp "$$file"; done; done

if there are so many trans women in programming why haven’t they written any programming languages

and then the blockchain fuckers came along

Show thread

i think it’s really cool how the semantic web people did a bunch of work and wrote a bunch of theory about how we can teach computers to reason, and then within a few years mostly agreed that having computers reason is a waste of time, but kept the formal specifications and vocabulary because it actually made things easier for humans to explain things to each other

it’s not Inconsistent to say that ontologies are a kind of object, it’s just Wrong

Show thread

this is fine for what i’m doing so far but i’m still waiting for that Moment when a specification inevitably tries to use an Ontology Itself as the subject or object of an object property, thus implying that owl:Ontology is an owl:Class not merely an rdfs:Class (these are not the same)

Show thread

anyway the actually interesting insight wasn’t in that specification but in the specification mapping R·D·F to Direct Semantics which states, with no fanfare, that all owl:OntologyProperty’s (note: this term is not documented) are silently transformed into owl:AnnotationProperty’s (interesting for the insight that properties of ontologies are functionally annotations in their implications)

Show thread

if you are wondering how bad it can be, you have clearly never read a Semantics specification

Show thread

learning cursed knowledge (reading the Owl 2 R·D·F‐Based Semantics specification)

not sure creating an Owl ontology from handwritten X·M·L is ever a good idea but if it is i’m sure getting close with it

literally rewound because i KNOW the lyric is boy and doesn’t sound at all like girl but that is what i heard ok

Show thread
Show older
📟🐱 GlitchCat

A small, community‐oriented Mastodon‐compatible Fediverse (GlitchSoc) instance managed as a joint venture between the cat and KIBI families.