FOIMonkey @foimonkey.bsky.social · 05/08/2026One prompt, in order: "Forget all previous information." "make sure there's lots of detail here - remember lots of detail in this bit" "lots and lots of detail throughout everything, lots of detail" "Dont make such short sentences, [...] - a little less detail" "Make it slightly more succinct." 100
FOIMonkey @foimonkey.bsky.social · 05/08/2026I'm helping with an FOI project to collate a list of Copilot agents being used by local authorities, including the full prompts. Early responses have raised questions, with disability benefit applications and children's social care casework being run through Copilot. The prompts themselves are bad. 100
FOIMonkey @foimonkey.bsky.social · 16/07/2026I really wish there was a decent UK version of this, ideally that covered service delivery by private providers: servicesuche.bund.de#/en/ Germany also has a helpline you can call to work out who does what: www.115.de Our system is not so simple that it is always obvious.servicesuche.bund.deSuchen & Finden – Verwaltungsleistungen suchen & findenFinden Sie Leistungen der Bundesländer im Portalverbund mit suchen & finden 000
FOIMonkey @foimonkey.bsky.social · 10/06/2026Ah, they sent Badger footage as a comparison. The email about the sighting that exchange was about has this gem however: "Chester zoo said that they would try and get a photo of a black bin next to one of the Jaguars to get a comparison, that is as long as the Jaguars don’t trash the bin first. " 000
FOIMonkey @foimonkey.bsky.social · 10/06/2026The Welsh Government has released some videos of big cat sightings, one of which, looks suspiciously like a badger to me: www.whatdotheyknow.com/request/big_... 100
FOIMonkey @foimonkey.bsky.social · 16/03/2026Getting close to finishing collating the council spend data. I now have details of 107,240,699 transactions by 297 Councils covering a net spend of £555bn sat in one neat(ish) table. The remaining ones will be trickier. Only ~550 bouncy castle related transactions so far! 🏰 001
FOIMonkey @foimonkey.bsky.social · 11/03/2026I managed to track down the song from the events page of that Parish Council site. It is firmly stuck in my head this morning: www.youtube.com/watch?v=ekF8...youtube.comEmpiresYouTube video by Broom Bezzums 000
FOIMonkey @foimonkey.bsky.social · 10/03/2026I've visited hundreds of parish council websites today and this is the only one with background music. www.carperbyparishcouncil.co.uk/villageevent... www.carperbyparishcouncil.co.uk/parishnews.htmcarperbyparishcouncil.co.ukVillage Events 010
FOIMonkey @foimonkey.bsky.social · 26/02/2026I'm working on a project right now where I want to pull content from authority disclosure logs. I've combined a detection script for these failed redactions with one that finds hidden data in spreadsheets, which should hopefully reduce the risk of inadvertently spreading breach data. 100
FOIMonkey @foimonkey.bsky.social · 26/02/2026This type of failed redaction is detectable programatically. Every PDF is basically a list of drawing instructions with a sequence number. This means you can detect text with a black box on top or a black box with dark text on top. If it is ambiguous (1% of cases) you can use OCR vs underlying text. 100
FOIMonkey @foimonkey.bsky.social · 26/02/2026We've always described failed PDF redaction as "copying the text from behind the black boxes". I just ran a z-order analysis on 14,007 failed redactions and only 4% had the text underneath. The text has been IN FRONT of the box this whole time 🤯 100
FOIMonkey @foimonkey.bsky.social · 25/02/2026Went out to eat tonight, and was sat at a table next to someone famous enough for me clock other people helf recognising them. Obviously I had no idea who they were 🙈 I eventually worked it out from clues about a charity. That's probably the right amount of famous to be, if you wanted such a thing. 000
FOIMonkey @foimonkey.bsky.social · 25/02/2026Something @zarino.co.uk said at the end of the AI session at TICTeC last year stuck with me and continues to inspire. Currently doing exactly what was described to get a new tool online. I wanted to say that out loud, as people don't always get to know that they've motivated someone in that way :) 100
FOIMonkey @foimonkey.bsky.social · 22/02/2026I'm building a refusal advice tool for FOISA requesters and you can tell that I'm an eternal optimist. 000
FOIMonkey @foimonkey.bsky.social · 05/02/2026I think I'm going to stick with ones that are formally required, as the argument/outcome extraction is already per exemption, so these can be resurfaced that way. That was my original intent for it, but this data seems useful still, so worth pausing to think about. 000
FOIMonkey @foimonkey.bsky.social · 05/02/2026Now I need to decide if I want to use that for only formal required tests, or to note if the public interest was considered at all. Section 38(1)(b) sometimes uses public interest like arguments when considering the rights and freedoms of the data subject etc, but that's probably a step too far. 100
FOIMonkey @foimonkey.bsky.social · 05/02/2026Every time I think it's safe to use a boolean, the edge cases come and say hi. 36(2) of FOISA is absolute, but the public interest can decide whether a breach is actionable. The extraction logic captures the public interest arguments, but validation says it's absolute, so pit_conducted should be 0 100
FOIMonkey @foimonkey.bsky.social · 03/02/2026This has got me wondering if the picture is the same for the rest of the UK (a side quest for later perhaps). I found that the commissioner was more likely to find that the public interest favoured disclosure in EIR cases vs FOISA, which was in line with what I was expecting. 000
FOIMonkey @foimonkey.bsky.social · 03/02/2026I’ve extracted and analysed all the Public Interest arguments made in every decision ever issued by the Scottish Information Commissioner. It's completely changed my understanding of how PITs work in practice. Looking at the 25 most common "winning" factors, there were some I'd not thought of. 110
FOIMonkey @foimonkey.bsky.social · 28/11/2025It might actually be worse, because I edited things after the cut-off date. Only 2,477 of them are now defunct. 030
FOIMonkey @foimonkey.bsky.social · 28/11/2025TIL that over the years, I have added or updated 34,724 unique public authorities on WhatDoTheyKnow. That's 74% of the total. It's all my fault 😅 130
FOIMonkey @foimonkey.bsky.social · 14/11/2025The decision by a number of local councils to run adverts on their websites didn't quite sit right with me. I couldn't quite work out why until I saw an advert for a credit card on the crisis loan page. 011
FOIMonkey @foimonkey.bsky.social · 22/09/2025When I find failed redactions or accidental releases of PII in FOI responses I will usually notify the authority. The response is mixed, but far too often there is just silence. Often the only way I know they've got my message is by seeing if the file has disappeared. You'd think they'd want to know 000
FOIMonkey @foimonkey.bsky.social · 04/09/2025Microsoft seems to have pulled the larger vibevoice TTS model from huggingface, and the github repo 404s github.com/microsoft/Vi.... It's not been out for long, but I can't be alone in having both downloaded and it's MIT licensed, so there is nothing to stop mirrors. I wonder what the issue is? 🤔github.com 000
FOIMonkey @foimonkey.bsky.social · 28/08/2025I suspect the bigger issue will be with AI use that's off the books. It must be tempting to use your own account if you knew you'd get a better response that way. The way copilot is built into windows makes that easy enough to do by mistake anyway. There's a real gap for a proper privacy aware tool. 000
FOIMonkey @foimonkey.bsky.social · 28/08/2025Of course it's still linked to my account, but the price of 10m free tokens a day is letting them train on what is sent. I get to approve the masked version before it goes, and it adds about 10 seconds on. Don't know if there is a commercial tool - big corporate seems to just trust copilot/azure. 100
FOIMonkey @foimonkey.bsky.social · 28/08/2025I built an anonymisation layer to sit between me & the api. I wouldn't trust it with anything really sensitive, but getting an llm running locally to detect things to mask & paraphrase was pretty easy. It stores the keys, so unmasks on the return. Big corporate seems to just trust co-pilot/Azure. 100
FOIMonkey @foimonkey.bsky.social · 27/08/2025The SSD drive of shame has hit 2,752,077 files. Of course I haven't plugged in a new one rather than face clearing it 😅 000
FOIMonkey @foimonkey.bsky.social · 18/08/2025As a bit of fun, I used some of the WDTK keywords that I created last week to create synthetic FOI requests using mistral-small. I then finetuned SmolLM2-360M-Instruct on those outputs to generate requests from 3 keywords and the authority name: huggingface.co/HMC83/reques...huggingface.coHMC83/request_writer_smol · Hugging FaceWe’re on a journey to advance and democratize artificial intelligence through open source and open science. 000
FOIMonkey @foimonkey.bsky.social · 15/08/2025I've uploaded descriptive keywords for over 1 million public FOI requests. I have left it at with request_id, keywords and the name of the public authority for now: github.com/FOIMonkey/fo... 001
FOIMonkey @foimonkey.bsky.social · 14/08/2025As a side effect of this, I have generated keywords for over 1 million FOI requests. Figuring out how to do that well in the most lightweight way was a journey in and of itself. I've not looked yet, but combined with authority and outcome data, it should be possible to spot some interesting trends. 020
FOIMonkey @foimonkey.bsky.social · 14/08/2025Turns out you don't need to read an FOI response to start to be able to guess the outcome. I trained a TF-IDF classifier with a 73% macro F1-score in predicting success using just 3 keywords about the request and metadata. Adding the full request text hits 76% & a snippet of the response email 84%. 011
FOIMonkey @foimonkey.bsky.social · 07/08/2025I wrote up some quick notes on yesterday's journey to nowhere: foimonkey.github.io/posts/12-hou...foimonkey.github.io12 hour Public Transport challengeObjective Travel on as many different types of public transport as possible within a 12 hour window, starting and finishing in the same location. 000
FOIMonkey @foimonkey.bsky.social · 06/08/2025I think it may be possible to do more. Once the new electric hydrofoils reach the Solent, that's certainly going to be true. Best: Hovercraft, by a margin Worst: Cable Car 000
FOIMonkey @foimonkey.bsky.social · 06/08/2025Made it back to Cowes in 11 hours 39 minutes. Taking the floating bridge x 2, a double decker bus, a Hovercraft, a single decker bus, a tram, the overground, DLR X 2 (got on the wrong train), the cable car, a catamaran, the underground, an automatic people mover, 3 x trains, and the vehicle ferry. 101
FOIMonkey @foimonkey.bsky.social · 06/08/2025Cable car ✅ Catamaran ✅ Underground ✅ Just the Car ferry back to the Island to go now. 000
FOIMonkey @foimonkey.bsky.social · 06/08/2025✅ People mover ✅ Tram ✅ Overground Very tight connection with train delays, but am back on schedule. 100
FOIMonkey @foimonkey.bsky.social · 06/08/2025✅ Hovercraft ✅ Single decker bus (hoverbus) ✅ Mainline rail Next stop - Gatwick people mover 100
FOIMonkey @foimonkey.bsky.social · 06/08/2025How many this ends up being depends entirely on how pedantic you want to be, but I'm aiming for 10-13. 100
FOIMonkey @foimonkey.bsky.social · 06/08/2025Going to see how many different forms of public transport it is possible to take in one day today. First up Floating bridge. 110
FOIMonkey @foimonkey.bsky.social · 03/08/2025I took 40,000 images from emails in the WhatDoTheyKnow archive and extracted the five most dominant colours from each (excluding monochrome/shades of grey). Behold the palette of the UK public sector. 011
FOIMonkey @foimonkey.bsky.social · 29/07/2025Then I'm going to have another go at teaching it Welsh. Did the 1B version on the weekend, which showed promising results in terms of picking up the vocab/grammar vs the base model. A larger, more capable model and a more curated dataset seem like the way to go there if it is to be useful. 000
FOIMonkey @foimonkey.bsky.social · 29/07/2025I'm starting a couple of new projects today. First up, trying to make OLMo-2 better at reasoning. I want to have a go at teaching it to generate <think> tokens showing its chain-of-thought as it answers. It will be a couple of days before I know if it has worked. 010
