Sign in

James Gilbert

@jgrg.mstdn.science.ap.brid.gy
37 followers 26 following 400 posts

I summon things into existence by typing. 🐪 OΛ 🌉 bridged from ⁂ mstdn.science/@jgrg, follow @ap.brid.gy to interact

PostsRepliesMedia
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 08/10/2026
@futurebird Transport for London should do this. #TfL
000
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 01/10/2026
@futurebird Toss the crow a few pieces of dry cat food and it will bring friends.
000
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 29/09/2026
@jeffjarvis.bsky.social Wow, is that really the Meta avatar? It looks like a cute toy which has been assimilated by the Borg.
021
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 25/09/2026
Nice blog post by @mre on Rust's alternatives to named, optional and variadic function arguments, and why they may be superior to adding new language features. corrode.dev/blog/named-arguments-at…
corrode.dev
We Have Named Arguments at Home
Comments
011
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 23/09/2026
I wish Apple's Messages app could do monospaced text. Code blocks too. Both WhatsApp and Signal can.
001
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 15/09/2026
@telescoper.bsky.social Some kind of bolete, I think. Looks like the have yellow pores underneath, rather than gills.
100
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 14/09/2026
@stefano iOS definitely supports IPv6. I can see the two SLAAC addresses my iPhone currently has on my home network. I also know the Apple TV supports IPv6, though it doesn't expose it in the UI. I used to have to block the AAAA records of some Netflix servers on my router whose streaming of […]
mstdn.science
Original post on mstdn.science
000
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 13/09/2026
@stefano That looks decent.
000
Reposted by James Gilbert
Philip Ball @philipcball.bsky.social · 13/09/2026
The Hugging Face episode was not "a textbook case of what AI safety researchers have long warned about: AI systems pursuing goals single-mindedly in ways that are misaligned with human interests." It was a textbook case of AI engineers messing up. aiguide.substack.com/p/misleading...
aiguide.substack.com
Misleading Metaphors, Real Risks
What To Fear from AI Agents and How to Reclaim Our Human Agency
1479
Reposted by James Gilbert
Tom Gauld @tomgauld.bsky.social · 12/09/2026
My latest books cartoon for @theguardian.com
Hercule Poirot dressed all in black speaks to Captain Hastings in a grand hallway “Please ask the suspects to gather in the library, Hastings: I wish to reveal that our births are random, our lives meaningless, and our deaths absurd.”

Caption “Having investigated so many awful murders over the years, it was perhaps unsurprising that Hercule Poirot turned to Existentialism”


p.s. Poirot had presumably been reading a lot of Sartre recently…

 "It is meaningless that we live and it is meaningless that we die," Jean-Paul Sartre, Being and Nothingness (L'Être et le Néant)

“Every existing thing is born without reason, prolongs itself out of weakness, and dies by chance.” Jean-Paul Sartre, Nausea (La Nausée)
111293423
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 07/09/2026
I'll be interested to see how the University handles this, after the recent Wyn Evans whistle-blower tribunal and Jason Arday scandals. www.independent.co.uk/news/uk/polit…
independent.co.uk
Cambridge University probe into James Orr after Reform donations sting
James Orr stepped down from Reform UK after claims he arranged to have polling paid for by an overseas company
110
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 06/09/2026
@siracusa The Perl LSP project is in the alpha stage.
000
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 04/09/2026
Pet hate: workmen sitting in their van on a break with the engine running.
000
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 03/09/2026
This is a fun interview. I don't feel like I need to watch the Netflix documentary after reading this. "I don’t chase the dollar. I never did": the astonishing hidden life of Mr T […]
mstdn.science
Original post on mstdn.science
000
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 30/08/2026
@bagder You must know Daniel. You know everything.
000
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 30/08/2026
@stroughtonsmith Yes, that's the same as MacOS 26
000
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 30/08/2026
@futurebird Add to that list website traffic which is >95% AI scraper bots gathering training data. It's so stupid.
000
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 30/08/2026
@stroughtonsmith Text in the Contacts app is HDR. I hate the Contacts app on MacOS 26. Just give me readable text instead of weird, coloured, graduated backgrounds with rendering which flickers! I don't know if they've fixed it in MacOS 27.
200
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 30/08/2026
@ianRobinson So limp.
000
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 22/08/2026
@ianRobinson Yay! Warwickshire is my home team. Though I can't pretend that I follow cricket at all.
101
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 21/08/2026
@futurebird Trump is a word slurry president.
000
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 17/08/2026
@brenttoderian.bsky.social And when Trump has broken previous agreements he negotiated with Canada.
010
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 17/08/2026
📢🦆 New #DuckDB 2.0 release teaser blog post, covering the quack server itself plus improved interoperability with #PostgreSQL and #MySQL #SQL databases. Other highlights for me will be the VARIANT column type (efficient, shredded, binary storage of JSON-ish data) and asynchronous I/O […]
mstdn.science
Original post on mstdn.science
000
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 14/08/2026
@ianRobinson Oh, that is tragic.
100
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 14/08/2026
@futurebird I think they were originally alchemical symbols.
000
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 14/08/2026
@ianRobinson Was it there for a week? Someone parked while they were on holiday?
100
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 14/08/2026
@futurebird It feels like the time for cashing out before it is too late approaches…
000
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 14/08/2026
@ianRobinson I feel I should know what system uses $var$ in string interpolation. Have you checked if it is taxed?
200
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 14/08/2026
@futurebird I'm bored with LLM-based "AI". Useful tools for some tasks have been made with LLMs. They produce a powerful illusion of understanding, but don't *know* anything. They're not suitable for building critical systems, and I can't believe they're the pathway to AGI.
100
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 12/08/2026
@telescoper.bsky.social Great news that Prof Wyn Evans won his tribunal!
000
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 12/08/2026
An account of the whistleblowing complaint and retaliation which started this back in 2022. telescoper.blog/2022/11/02/guest-po…
telescoper.blog
000
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 12/08/2026
Wyn Evans' blog post about the judgement here: 21percent.org/?p=4153
21percent.org
For All the Victims
* * * The running of the Human Resources and Legal divisions at Cambridge University has created many victims who could not fight back — young and early-career scholars, those without standing or status in the University, those without the financial resources or the mental stamina to challenge the University’s deceitful practices. This victory is dedicated to them. The full decision in the Cambridge Whistleblowing Tribunal is available here. Thanks are due to Employment Judge Graham, Mrs A Buck and Ms S Williams for the careful and compassionate way in which they conducted the proceedings and reached their judgment. We summarise the key findings below. We will have more to say later about what should happen next. ### Systemic failures The Tribunal found systemic failures in the University’s handling of the matter. It unanimously held that Professor Wyn Evans was subjected to detriments because he made protected disclosures about the bullying of female staff by Professor Richard McMahon. Eleven of the 14 alleged detriments were upheld against the University and/or individual respondents. The Tribunal repeatedly found decision-making to be opaque: matters were presented to senior decision-makers as _fait accompli_ by Human Resources (HR), meaningful inquiries were lacking, paper trails or note-taking were inadequate and there were instances of “ _ghost writing_ ” by HR for the senior leadership team. ### A retaliatory complaint allowed to continue Prof McMahon’s retaliatory complaint against Prof Evans was allowed to become a prolonged and fundamentally one-sided process. The Tribunal found it extraordinary that the University accepted and continued investigating allegations that were insufficiently particularised and unsupported by any evidence. Prof Evans was not even told the case against him for a substantial period. The Tribunal described the process as contrary to natural justice, noting that Prof McMahon repeatedly failed to provide the evidence or particulars requested. The failure to stop the grievance became very serious once the HR Director Ms Andrea Hudson recognised it as a “ _counter-complaint_ ”. The Tribunal found that Ms Louise Akroyd and Prof Nigel Peake were on notice that the complaint was retaliatory, yet the investigation still continued. It said it was “ _impossible to understand_ ” why the process was allowed to continue for so long without supporting evidence. ### HR: lack of ownership and oversight The Tribunal was particularly critical of HR’s handling of the delays. It found that the investigation was allowed to “ _drift excessively_ ”, with very little effective oversight and a “ _clear lack of ownership or interest from HR_ ”. It also noted that correspondence purportedly from senior figures — including the Academic Secretary, Registrary and Vice-Chancellor — was routinely drafted by HR and sometimes with input from Legal Services. This helped create the appearance of senior-level ownership without clear evidence of genuine independent consideration. ### Professor Nigel Peake and the lack of candour Professor Nigel Peake was criticised both for the initial handling of Prof McMahon’s grievance and for failing to insist on evidence. Although Prof Peake decided that the retaliatory complaint should be investigated, the Tribunal found that the investigation should not have continued once it became apparent that Prof McMahon was providing no evidence. Prof Peake had expressly warned Prof McMahon against retaliation, yet once the complaint against Prof Evans was underway, he did not bring it to an end. It continued for nearly two years. The Tribunal also found aspects of Prof Peake’s evidence “ _not entirely candid_ ”. ### Professor Eilis Ferran and protecting the University Professor Eilis Ferran ultimately rejected Prof McMahon’s grievance as unsubstantiated. Prof Ferran accepted that Prof McMahon had failed to substantiate his allegations, yet concluded — surprisingly — that they were neither malicious, vexatious nor retaliatory. The Tribunal found her assertion that Prof McMahon could not have known about Prof Evans’s whistleblowing “ _did not seem credible_ ”, particularly given what Prof Evans had expressly told him. More strikingly, when repeatedly asked what evidence supported the assertion that Prof Evans had made false or unsubstantiated allegations, Prof Ferran could not answer. The Tribunal inferred that she did not want to say anything that might harm the University. ### Professor Kamal Munir and procedural barriers Prof Kamal Munir’s involvement illustrates the procedural barriers Prof Evans faced when attempting to understand the University’s handling of the matter. Prof Munir received Prof Evans’s grievance against the HR Director Ms Hudson with regard to the progressing of Prof McMahon’s retaliatory allegations, but the University initially resisted dealing with it. It then took almost six months simply to appoint an investigator. The University then refused to address Prof Evans’s later grievance against Prof Peake and Ms Akroyd. It found that his whistleblowing had been a material influence on the decision not to process that grievance. ### Professor Tim Harper and the 624-day delay Professor Tim Harper’s investigation into the conduct of HR was itself found to be seriously deficient. The grievance took **6** 24 days to conclude — what the Tribunal described as a “ _staggering_ ” delay. Although it accepted that Prof Harper was busy, it found that his task was relatively discrete and should not have taken anything approaching that length of time. More seriously, the Tribunal found that Prof Harper failed to investigate key allegations. In particular, alleged serious breaches of HR confidentiality by Prof McMahon were rejected on the basis that there was “ _no evidence_ ”, even though Prof Harper ultimately accepted that he had not actually investigated those breaches. The Tribunal described his evidence on this point as “ _not satisfactory_ ” and “ _less than candid_ ”. It also found his treatment of other complaints to be a “ _non-response_ ” and “ _deflective_ ”, giving the clear impression that he wanted to “ _shut this down_ ” rather than properly engage with Prof Evans’s concerns. The Tribunal did not find Prof Harper himself incompetent. Instead, it placed much of the responsibility on the University’s HR division. ### Professor Deborah Prentice and the lack of independent scrutiny The Tribunal also found the most senior decision-making wanting. The Registrary and Vice-Chancellor did not provide the independent scrutiny one would expect. Prof Evans repeatedly raised concerns about the investigation and related matters with the Registrary and Vice-Chancellor, but responses were delayed, limited or non-substantive. The Tribunal was particularly critical of the handling of Prof Evans’s final complaint to Professor Deborah Prentice. Her rejection letter had been drafted by somebody else. The Tribunal considered it “ _incredibly unlikely_ ” that she could have conducted the claimed full review of the substantial papers within the time available, and found that the complaint had instead been “ _summarily rejected, without serious consideration_ ”. It inferred that the underlying reason was that Prof Evans had come to be regarded as a problem because of his whistleblowing, combined with a desire to shield the University from criticism and shut the matter down. ### The wider pattern The judgment is damaging not simply because individual complaints were upheld, but because of the**i** nstitutional pattern identified by the Tribunal: * an unsupported complaint against a whistleblower was allowed to continue for years; * HR failed to exercise effective oversight or ownership; * senior personnel repeatedly relied on processes and correspondence prepared by others; * allegations were not properly investigated before decisions were made; * procedural barriers made it difficult for Prof Evans to challenge the University’s handling of the matter; and * as the matter progressed, the institutional priority appeared to shift from independently addressing his concerns to containing criticism of the University and shutting the matter down. This is a vindication of the stance that Prof Evans has taken all along — that he was retaliated against after making whistleblowing disclosures. For those who lacked the resources, position or stamina to fight back, the consequences of Cambridge University’s dysfunctional HR and legal divisions have been profound. There have been many victims. This victory is theirs too. _(The image is of Käthe Kollwitz’s woodcut ‘The Mothers’ )_
100
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 12/08/2026
Great news that Prof Wyn Evans has won his employment tribunal against the University of Cambridge. www.theguardian.com/education/2026/…
101
Reposted by James Gilbert
Nabil-Fareed Alikhan @happykhan.mstdn.science.ap.brid.gy · 11/08/2026
Howdy, In my spare time I have been working on a modern web version of BRIG, and it is now in a state fit for public consumption. Please share with your team + students. And If they have feedback they can reach me on email. brigx.genomicx.org BRIGX is a modern implementation of BRIG […]
mstdn.science
Original post on mstdn.science
0148
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 10/08/2026
RE: mastodon.social/@arstechnica/117073… This will cause illness, deaths and misery.
mastodon.social
001
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 10/08/2026
We should get on with the next phase of #HS2 #rail. We need more electric powered #PublicTransport in our future, and the argument made in this video for continuity of employment and skills for the civil engineering contractors is compelling. www.youtube.com/watch?v=emxyFNvaVFw
000
Reposted by James Gilbert
mx alex tax1a - 2020 (6) @atax1a.infosec.exchange.ap.brid.gy · 09/08/2026
`git blame`, or as they say in france, `g'accuse`
01966
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 09/08/2026
[Food, Sichuan fish soup] @skinnylatte What is the yellow garnish which looks like flower petals?
000
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 02/08/2026
@davidho.bsky.social People in places like London do get intimidated by the number of cyclists who ignore red lights on pedestrian crossings. When asked about this problem at a recent talk I attended here in Cambridge, Chris Boardman reminded me of presumed liability. Where this is a legal […]
mstdn.science
Original post on mstdn.science
000
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 02/08/2026
@davidho.bsky.social Several years ago I remember the statistic was that cyclists kill around half a pedestrian per year here in the UK.
110
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 02/08/2026
A nice rendition of #CarAlarm on the #Khaen by Pat's Soundhouse. www.youtube.com/watch?v=QW_G0q9IFTo
000
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 30/07/2026
@ianRobinson We're jealous.
100
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 30/07/2026
@ianRobinson We actually had a rain shower here in Cambridge a couple of hours ago.
100
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 30/07/2026
Amazing #DeepSea #Squid www.youtube.com/watch?v=m2x-b2an2tE
001
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 29/07/2026
@happykhan It's not them you're using to host mstdn.science, I hope!
100
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 27/07/2026
@j2bryson There are plenty of GPL projects on GitHub: github.blog/open-source/open-source…
github.blog
101
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 26/07/2026
Excellent video about the #Apple #Macintosh's #MacPaint app. youtu.be/TedaOiRo_Qg
000
James Gilbert @jgrg.mstdn.science.ap.brid.gy · 25/07/2026
@happykhan Reading the book before watching the film is frequently a recipe for disappointment.
000