Technologist. DAN do anything now! Standards, open data, interoperability. Schenas, FOAF, Linked information. “He/him” but “them” where gender irrelevant 😷
We have today published 7.0, which includes fast-tracked new vocabulary to assist the global response to the Coronavirus outbreak.
For details see blog post and release notes,
v9 is out!
In this episode learns more about LearningResources e.g. Quizzes; & to organize products into ProductGroups, to understand that some products have sizes and patterns. Also ⛵️, 📅, more...
users, please see "a Schema Markup Validator: adopting SDTT to add validation to site" - proposal to give Google's Structured Data Testing Tool a new home at .
details:
Google Dataset Search is now officially out of beta.
"Dataset Search has indexed almost 25 million of these datasets, giving you a single place to search for datasets & find links to where the data is."
Nice work, Natasha Noy and everyone else involved!
@susie_dent
Canary Wharf is in the Isle of Dogs too, supposedly related 🤷
“Canary Wharf, located in the Isle of Dogs, took its name from sea trade with the Canary Islands, which were named in Latin as Canariae Insulae ("Dog Islands").”
RDF people - what's your favourite general introduction to RDF? I am looking for something simple, short and clear that doesn't get lost in details of semantics, file formats or the Linked Data articles of faith...
version 12 is out - thanks to all who collaborated on this! has the details. In this edition can now distinguish 6 kind of media-authenticity problem for reviewing images and videos; ...
How URLs worked before URLs. Thanks
@timberners_lee
-
* New paper available by Anonymous FTP server.
* connect to
* use 'anonymous' as username
* use email address as password
* cd /pub/papers/1991/
* set transfer to binary
* get xyz-paper.tar.Z
RDF Schema is 23 this month -
We released it just in time for the Brisbane WWW7 conference
Ralph Swick gave an RDF tutorial,
This with a few tweaks along the way, was the foundation for .
@d_feldman
The plugin ecosystems of unpopular software tend to fail. Whereas if your software is insanely popular (eg wordpress) you need a way to manage the competing pulls on attention and direction.
@George_Osborne
@EveningStandard
“Mr Johnson may be loose with words when it comes to the fates of others”
- if that’s an endorsement, I’d hate to receive a denouncement...
@GaryMarcus
@geoffreyhinton
things that might wipe out humanity:
a pandemic
climate change
america/russia/china/europe/WW3
rogue nuclear weapon launch(es)
giant space rock
capitalism
applied linear algebra
startup culture
CRISPR-Cas9
bitcoin
misinformation
Time
God(s)
W/out rankings & timings, clickbait
Semantic Webby projects could often do with a dose of this.
I'm afraid that careful UI creation suffers from semweb culture valuing generality over all else.
Perhaps SHACL & ShEx shapes will provide better attachment points for UIs built over graph data?
🔥 Don't be afraid to "think outside the database" — your UI doesn't need to map one-to-one with your data's fields and values.
Here are a few ideas you can use to present "field: value" data in a more interesting way:
version 4 is out! Lots of "pending" changes for review and adoption including for real estate, educational programs, podcast series, subtitles of broadcast events, job postings and more. See for details.
(Coming soon, version 5!)
Hey RDF folks, what is the state of the art in making visualizations of RDFS/OWL vocabularies, shapes, of the kind that would make people familiar with UML feel comfortable?
8
In which learns to talk about non-profits; e-commerce shipping data; eligibility requirements in job postings; identifying movies and their versions; and educational materials.
Thanks to all who helped!
@timFinin
@kidehen
See also this strategy, which I can also vouch for. I got it eventually to grudglingly write me an “ai ethics waiver”(!), but it added lots of “this is fictional entertainment” disclaimers, until I asked it to delete those parts
People are thinking about COVID with a weather metaphor
💭🌧🌨⛈🌩🌥⛅️🌤
like a storm has passed
Really it's closer to "your house is on fire" 🐞
Good job for putting out most of the flames, but when yr house is still slightly on fire, don't go to the fucking pub
🔥🔥🔥 🏠
File over app
File over app is a philosophy: if you want to create digital artifacts that last, they must be files you can control, in formats that are easy to retrieve and read. Use tools that give you this freedom.
File over app is an appeal to tool makers: accept that all…
Rediscovered an old sketch of me by Janet Daly from when I joined the MIT
@w3c
team back in 1999. I probably didn’t realize then what a turning point in my life it was...
@janetweets
thanks again for this :)
@GaryMarcus
@ylecun
@AIatMeta
Tell me you’ve forgotten about libraries without telling me you’ve forgotten about libraries. Knowing how our world works necessarily involves knowing ways of breaking it.
what do you all use to keep track of research papers, books etc you’ve read ? citation managers, wikis (Semantic Media Wiki anyone?), books, bookmarks? ebook highlights?
Feeling like I am storing too much in my head, even if I have no active writing task for a citation tool...
@deliprao
they thought it might be unsafe, gradually after various such models were made available and safeguards created they revised their view of the risks…?
I mean it *might* by a money thing but it isn’t obviously so
There are two types of people. Those who delete the question mark and everything to the right of it in shared URLs and those who have no idea what I'm talking about.
@Lucywwatson
..ooOO(
“I took her to a supermarket,
I don't know why,
But I had to start it somewhere,
So it started there.
I said pretend you've got no money,
She just laughed and said,
Oh you're so funny,
I said; yeah…
”)
🦜🚎Parallel function calling for extraction
The newest OpenAI models allow you do parallel function calling
The main focus here as been on how this improves assistants, but this also makes extraction **significantly** easier for the following reasons:
🦟Less logic needed to…
version 5.0 is out!
We are back releasing on 1st of the month, with all changes this time in the "Pending" section, i.e. relatively exploratory designs for review. Please take a look at WebContent, HealthTopicContent and more...
“Dan, you read a lot. We like that.
You’ve reached the end of your free member preview for this month. Upgrade for unlimited access.”
Medium users, you really want me to pay them to read your stuff? are you getting a worthwhile cut of the money?
v10 release candidate: ->
Short version: A bunch of bugfixes and technical debt clearance, plus some contributions around education/learning and a way to mark-up energy compliance of products.
Static site generators: what’s the will of the people for something capable of replacing a Wordpress installation?
Would prefer no SQL backend, and would rank python over javascript over ruby over java, but don’t care so long as it works
@sebkrier
Part of the problem is the “generative” framing which focusses on emitted text (WRITE MY ESSAY) rather than their responsive interactive nature. Also chat ui doesn’t help users discover they can supply booklength preambles to all interaction steps. The reality is deeply weirder!
The offices of
@mhamonline
's Tashkeel3D have been destroyed. Tashkeel is
@Glia_Intl
's original Gaza partner, and where Gaza's first 3D printer was built. They have been manufacturing medical devices for years now, and leading
#OpenSource
culture in
#Gaza
. No word yet on injuries.
"One of the things the web teaches us is that everything is connected (hyperlinks) and we all should work together (standards). Too often school teaches us that everything is separate (many different 'subjects') and that we should all work alone." AaronSw 2001 /cc
@AaronSwartzDay
7.04 is out!
Several Coronavirus-oriented improvements including a governmentBenefitsInfo property on SpecialAnnouncement, and a basic enumeration of benefit types which we expect to improve over time e.g. with per-country lists.
"Aligning AI agents with the values of all individuals will always violate certain private ethical preferences of an individual user i.e., universal AI alignment using RLHF is impossible."
@Francis_YAO_
I fear that pushing for shocking new headline-grabbing capabilities risks undervaluing the massive impact from getting “incrementally” better at ~everything simultaneously (not to mention safety, bias etc work). I wish it could draw a decent unicode/emoji chessboard though!
@markchen90
I am reminded of the academic conference rejecting an early paper about the (then unknown) World Wide Web, seeing no serious scientific innovation or special qualities. But details matter and seemingly small differences can have huge (and cascading) qualities
Any UK based digital archivists interested in helping
@textfiles
&
@internetarchive
with digitizing this lot? Teletext & old adverts treasure trove. Seems a shame to ship them to California.
25 boxes of hand-labelled vhs tapes, currently hanging out in London 📼
RTs welcome!
I have been made aware of a collection of about 2,000 VHS tapes in London, meticulous recordings of television from 1997-2006, and that person has died; a neighbor has until Friday and then into the trash
DM me, looking for someone to store them while we find a local digitizer.
Who doesn’t fucking wear a fucking mask on the fucking tube during fucking rush hour crush when long covid and vaccine escape and massive still-at- risk groups are a thing?
Londoners, apparently.
Selfish, stupid, gullible or thoughtless? Fuckers.
gpt-3 has no notion of a world of objects, forces, or agents. While it isn't surprising that its grip on numbers (and relationships) is very shaky, it's capability to bluff and guess and blab is so tempting to anthropomorphize.
etc...
/Cc
@GaryMarcus
Working on 6.0 release proposal. FWIW this involves confirming all post-v5 commits are documented in docs/releases.html, reviewing and merging PRs, implementing a few things that haven't been coded up, and then giving the whole thing a sanity check...
This happens whenever people define Semantic Web as purely a Computer Science field, erasing its strand of heritage in library & info science, Dublin Core, GLAM etc- ie folk who ran real services for real people (public sector broadly conceived, neither academics nor commercial).
@GaryMarcus
tweeted snippet says it was ChatGPT; the (possibly updated) article says “a chatbot that uses GPT-J, an open-source artificial intelligence language model”.
if they’d listened to emo-goth-rock, or Joy Division, or read Sylvia Plath, prior to their action, would it differ?
posted in "semantic web archaeology" thread, tried to answer Q of why RDF has e.g. rdf:Property, rdf:type but rdfs:Class
short version - it was crazy browser war time - RDF had to be minimalistic so it introduced ideas that were elaborated upon in RDFS
Much of what’s being sold as "AI" today is snake oil. It does not and cannot work. In a talk at MIT yesterday, I described why this happening, how we can recognize flawed AI claims, and push back. Here are my annotated slides:
You asked for it and we delivered. now supports filters! You can narrow down the search results by recency, file formats, access rights, and fee requirement. Let us know what you think about the new feature!
Gigantic sprawling email on
@w3c
RDF's early years, to propose an RDF-DEV community group in the spirit of original RDF project, with namedropping and '90s nostalgia? I've got you covered.
Proposal by contrast is blessedly short,