Edward w

@edwardw2.bsky.social

♈︎♉︎♊︎♋︎♌︎♍︎♎︎♏︎♐︎♑︎♒︎♓︎ ☉ⵙ ☽ ☿ ♀︎ ⊕ⴲ ♂︎ ♃ ♄ ♅⛢ ♆ ♇ ☊ ☋ ⊗𐌈 Φⵀ ♡ ⚸ ⚳ ⚷ Cnj ☌ Sextile ⚹ ᚕ Sq □ Tr△ⵠ Opp ☍ ret ℞ 0°1′1″ https://edwardwhite123.blogspot.com https://archive.org/details/@edwwh https://edwardw.bearblog.dev/blog/ https://edwardwhite.mataroa.blog

an interesting argument that the singaporean practise of caning offenders prevents sentence inflation. caning serves to express the community's outrage against the offender- and indeed, express its outrage in a way jail cannot perform, by making the offender feel pain...

Jacob Gifford Head@giffordhead.co.uk · 10h ago

One of the reasons we are struggling to find enough prison spaces is that sentences have increased sharply in length over the last couple of decades. One of the difficulties, I think is that the inflationary pressure on sentences is everywhere within the debate about crime, from all sides.

Within the debates about academic plagiarism, I see a lot of discussion about Turnitin. They seem to suggest its use in the UK is routine & uncontroversial. Have all the arguments against it, particularly about copyright, just been lost? As a student unionist at the time, we were furious about it.

may all sentient beings be at peace om mani padme hum om mani padme hum om mani padme hum om mani padme hum 🐀 唵嘛呢叭咪吽🐂 唵嘛呢叭咪吽🐅 唵嘛呢叭咪吽🐇 唵嘛呢叭咪吽🐉 唵嘛呢叭咪吽🐍 唵嘛呢叭咪吽🐎 唵嘛呢叭咪吽🐑 唵嘛呢叭咪吽 🐒 🐈 唵嘛呢叭咪吽👻唵嘛呢叭咪吽 😈唵嘛呢叭咪吽 🌍 唵嘛呢叭咪吽🌎 唵嘛呢叭咪吽🌏 唵嘛呢叭咪吽 唵嘛呢叭咪吽ཨོཾ་མ་ཎི་པདྨེ་ཧཱུྃ་| ཨོཾ་མ་ཎི་པདྨེ་ཧཱུྃ་| ཨོཾ་མ་ཎི་པདྨེ་ཧཱུྃ་| ཨོཾ་མ་ཎི་པདྨེ་ཧཱུྃ་| ཨོཾ་མ་ཎི་པདྨེ་ཧཱུྃ་|

One supervision in 2.5 years sounds bad but I was once told a tale of a PhD submitted in a foreign language and, when the administrator objected, the student said that his draft chapters had been in that language and the supervisor had said they were fine. His supervisor did not read that language.

Emma Yeomans@yeomans.bsky.social · yesterday

Calls for the Office for Students to investigate Liverpool John Moores University, after it passed Jason Arday's PhD thesis despite plagiarism and the thesis being 38,000 words over word count. He says received just one supervision in 2.5 years. www.thetimes.com/uk/education...

may all sentient beings be at peace om mani padme hum om mani padme hum om mani padme hum om mani padme hum 🐀 唵嘛呢叭咪吽🐂 唵嘛呢叭咪吽🐅 唵嘛呢叭咪吽🐇 唵嘛呢叭咪吽🐉 唵嘛呢叭咪吽🐍 唵嘛呢叭咪吽🐎 唵嘛呢叭咪吽🐑 唵嘛呢叭咪吽 🐒 🐈 唵嘛呢叭咪吽👻唵嘛呢叭咪吽 😈唵嘛呢叭咪吽 🌍 唵嘛呢叭咪吽🌎 唵嘛呢叭咪吽🌏 唵嘛呢叭咪吽 唵嘛呢叭咪吽ཨོཾ་མ་ཎི་པདྨེ་ཧཱུྃ་| ཨོཾ་མ་ཎི་པདྨེ་ཧཱུྃ་| ཨོཾ་མ་ཎི་པདྨེ་ཧཱུྃ་| ཨོཾ་མ་ཎི་པདྨེ་ཧཱུྃ་| ཨོཾ་མ་ཎི་པདྨེ་ཧཱུྃ་|

Project for the rest of the summer is to finish the digital topology book I'm writing (with Dae Woong Lee). I will live-tweet every Figure I make, with no additional context. These are all made with tikz.

may all sentient beings be at peace om mani padme hum om mani padme hum om mani padme hum om mani padme hum 🐀 唵嘛呢叭咪吽🐂 唵嘛呢叭咪吽🐅 唵嘛呢叭咪吽🐇 唵嘛呢叭咪吽🐉 唵嘛呢叭咪吽🐍 唵嘛呢叭咪吽🐎 唵嘛呢叭咪吽🐑 唵嘛呢叭咪吽 🐒 🐈 唵嘛呢叭咪吽👻唵嘛呢叭咪吽 😈唵嘛呢叭咪吽 🌍 唵嘛呢叭咪吽🌎 唵嘛呢叭咪吽🌏 唵嘛呢叭咪吽 唵嘛呢叭咪吽ཨོཾ་མ་ཎི་པདྨེ་ཧཱུྃ་| ཨོཾ་མ་ཎི་པདྨེ་ཧཱུྃ་| ཨོཾ་མ་ཎི་པདྨེ་ཧཱུྃ་| ཨོཾ་མ་ཎི་པདྨེ་ཧཱུྃ་| ཨོཾ་མ་ཎི་པདྨེ་ཧཱུྃ་|

15,000 years ago, in one of the the earliest settled communities in the world, a child collected fine clay from the slow-moving river by her home. She then committed an act familiar to all inveterate play-doh enthusiasts; rolling and and pinching the squishy clay into tiny round balls.

BildBildBild

Exactly. This whole thing seems to be superfluous discovery due to incentives created by intellectual property law. Had Anthropic continued to train their models on pirated content, they could do so without depleting the knowledge commons. But for legal reasons, they destroy books instead. 🧵

As Anthropic trained successive LLMs, it became convinced that using books was themost cost-effective means to achieve a world-class LLM. During this time, however,Anthropic became “not so gung ho about” training on pirated books “for legal reasons” (Opp.Exh. 19). It kept them anyway (e.g., Opp. Exh. 17 at 93–94; CC Opp. Exh. 35 at -0273474).To find a new way to get books, in February 2024, Anthropic hired the former head ofpartnerships for Google’s book-scanning project, Tom Turvey. He was tasked with obtaining "all the books in the world" while still avoiding as much "legal/practice/business slog" as
possible (Opp. Exhs. 21, 27). So, in spring 2024, Turvey sent an email or two to majorpublishers to inquire into licensing books for training AI. Had Turvey kept up thoseconversations, he might have reached agreements to license copies for AI training frompublishers — just as another major technology company soon did with one major publisher(e.g., Opp. Expert Malackowski ¶¶ 50, 64). But Turvey let those conversations wither. Instead, Turvey and his team emailed major book distributors and retailers about bulk-purchasing their print copies for the AI firm’s “research library” (Opp. Exh. 22 at 145; Opp.Exh. 31 at -035589). Anthropic spent many millions of dollars to purchase millions of printbooks, often in used condition. Then, its service providers stripped the books from their bindings, cut their pages to size, and scanned the books into digital form — discarding the paper originals. Each print book resulted in a PDF copy containing images of the scannedpages with machine-readable text (including front and back cover scans for softcover books).Anthropic created its own catalog of bibliographic metadata for the books it was acquiring. It acquired copies of millions of books, including of all works at issue for all Authors.To summarize the analysis that now follows, the use of the books at issue to train Claude and its precursors was exceedingly transformative and was a fair use under Section 107 of the Copyright Act. And, the digitization of the books purchased in print form by Anthropic was also a fair use but not for the same reason as applies to the training copies. Instead, it was a fair use because all Anthropic did was replace the print copies it had purchased for its central library with more convenient space-saving and searchable digital copies for its central library — without adding new copies, creating new works, or redistributing existing copies. However, Anthropic had no entitlement to use pirated copies for its central library. Creating a permanent, general-purpose library was not itself a fair use excusing Anthropic’s piracy.
maia arson crimew 🏴@crimew.gay · last wk.

the solution to anthropic ripping up millions of books isnt more copyright by the way, theyre doing this Because of copyright

may all sentient beings be at peace om mani padme hum om mani padme hum om mani padme hum om mani padme hum 🐀 唵嘛呢叭咪吽🐂 唵嘛呢叭咪吽🐅 唵嘛呢叭咪吽🐇 唵嘛呢叭咪吽🐉 唵嘛呢叭咪吽🐍 唵嘛呢叭咪吽🐎 唵嘛呢叭咪吽🐑 唵嘛呢叭咪吽 🐒 🐈 唵嘛呢叭咪吽👻唵嘛呢叭咪吽 😈唵嘛呢叭咪吽 🌍 唵嘛呢叭咪吽🌎 唵嘛呢叭咪吽🌏 唵嘛呢叭咪吽 唵嘛呢叭咪吽ཨོཾ་མ་ཎི་པདྨེ་ཧཱུྃ་| ཨོཾ་མ་ཎི་པདྨེ་ཧཱུྃ་| ཨོཾ་མ་ཎི་པདྨེ་ཧཱུྃ་| ཨོཾ་མ་ཎི་པདྨེ་ཧཱུྃ་| ཨོཾ་མ་ཎི་པདྨེ་ཧཱུྃ་|

My favorite research success story, the moral of which is: always be nosey. I am writing on a subject on which everyone cites a prize-winning MA thesis that seems to be *un-gettable.* I contacted the archivist at the university where the thesis was written, and they got nothing. 1/6

this is actually something I try to keep in mind all the time when doing Chinese history; it's very normal to quote a source from, say, 1520 when referring to something in 1550, but that was a generation and a half apart!

Paris copy of the 1516 polyglot Genoa Psalter, in which a sixteenth-century reader has added the Lord’s Prayer and Sūrah al-Fātiḥah in Arabic and Latin, both copied from Guillaume Postel’s Grammatica Arabica (ca. 1538). BnF - Arsenal FOL-T-188

BildBildBildBild