2022 - Week 41

Leeds days

Week 40 once again saw no notes from us. This time we have a good excuse, your regular correspondents having taken the train for a mini-break in Leeds. Michael turned up at Liverpool Street Station with his usual combination of bright eyes and bushy tail, fully intending to meet with Librarian Anya and computational handmaiden Young Robert for a spot of breakfast. A glance up at the departure board saw him come to the slow realisation that all the trains were off to Norfolk. A thing you’d think he’d know having once lived in Norfolk. One somewhat panicked trip up the Metropolitan Line later, the crack team were reunited at Kings Cross and not long after, wheels down in Leeds. Well, for Anya, Robert and Michael anyway. Librarian Jayne had a considerably shorter distance to travel but was not departing from London, so her journey took much longer. Obviously.

Meeting one was conducted with Tom down at the Adelphi, a boozer with that flavour of filtered light that really lends itself to late afternoon drinking. The next day saw a couple more meetings. This time in offices rather than pubs. Your correspondents carrying slightly sore heads and a renewed belief that - when it comes to yet more Guinness - discretion may well be the better part of valour.

First up was a chat with Russell who’s mired in the mess of process flows for joining a union. Given we’re also in the process flow mapping business - albeit with the rather more pretentious title of ‘procedure’ - we thought it might be good to compare notes on modelling flows in code versus modelling flows in data. Russell’s tool is really rather nifty. Once a new joining process is mapped, a quick press of a button generates forms, links and models. Actual storage being a case of JSON objects in a resolutely non-relational database. From which the union in question can extract the data, transform it to the right shape for their particular CRM and load up the results. A design pattern that makes a lot of sense for a lot of applications, introducing a useful level of pace layering between forms and input storage and processing and output storage. A design pattern we think would work splendidly for our current adventures in the Register of Members’ Financial Interests. More of which later.

On the subject of RMFI, the second meeting of the day saw librarians and computational experts sit down with Paul, Tom and Stuart. We remain hopeful that the Open Innovations lads will be up for building us a quick prototype to evaluate how more structured outputs might be enabled by tweaks to the rules around inputs. The only hurdle now being procurement. So let’s not count our chickens.

On day three, we met with POLIS Charlie who’d expressed an interest in setting his students off on a mission to map how brexit-related statutory instruments might have impacted the UK both economically and socially. We explained that we only covered the bits of the process that touched on Parliamentary procedure and that - since not much ever happens to an SI in Parliament, and SIs are but one stop on the itinerary of policy negotiation, formulation, delivery and impact assessment, looping through parties, government and parliament and out again - his students might find it all a little tedious. It wasn’t all bad news though. Putting down pens, we departed into blacked-out Leeds back streets and once more headed to the pub. This time with both Charlie and Tom. Imagine our surprise when Charlie turned out to be a massive fan of Historic Hansard and, by extension - of young Robert, Charlie even asking for a selfie with the lad. Anyway, when it comes to measuring the economic impact of almost anything, Tom proved to be much more useful to Charlie than we had. Introducing folks in pubs counts as work in our eyes. Especially introducing folks in pubs to be honest.

It’s possible we stayed out drinking rather too long. Nobody expected to see Robert propping up the bar at 2:30 in the morning. Least of all Robert. Still, at least it seemed to clear the cold out of his system. Robert’s cold having moved in with Michael now.

Losing a librarian from the loop

If you tuned in last week, you’ll know Anya has been conducting time and motion studies on her crack team of librarians, looking for information management efficiency savings. One in particular grabbed her attention: the sheer amount of effort - and associated tedium - in resetting scrutiny end dates for both SIs and treaties every time there’s a change to sitting days.

As a first punt at fixing this, Robert’s cold and Michael have added a new form and new URLs to our beloved egg timer, allowing the casual user - or indeed casual machine - to supply a start date, a calculation style and the number of days to count and get back a scrutiny end date. Young Robert chipped in with some much needed pixel polishing. Librarian Jayne and Robert have now checked the work of Robert’s cold and Michael and declared themselves happy.

Elsewhere, Jayne, Ned, Robert, Robert’s cold and Michael have dived back in to our procedure model and are in the process of adding a couple of new classes to cover both clocks and calculation styles. Michael and Robert’s cold have even written a small blob of code to take information from the procedure model, query the egg timer and update scrutiny end dates. Though, so far, Jayne and Robert haven’t checked that. So it might be nonsense. With our Jianhan away - for a month, no less - it’s unlikely these scribblings will transform into working code anytime soon. But at least we have a plan and Jianhan certainly won’t be short of work on his return.

Teaching the machines about legislative reform orders

Librarians Jayne and Ayesha, together with their computational helpmate Michael, spent any and all available downtime adding yet more LRO routes to the machines. The negative half of the procedure is now inside the machines. So that’s a thing. The affirmative half of the procedure is probably 60% done. So that’s another thing. They still have very little to show for their efforts but it feels like there’s only a matter of hours to go. So tune in next week. When there’ll still be very little to show.

In other LRO news, Librarian Jayne has been deep in both legislation.gov.uk and Omnigraffle pixels, adding citations to our LRO map and associated components. Top work Jayne.

Interests financial and otherwise

Young Robert, Robert’s cold and Michael have spent a fair bit of time over the last few weeks with registrars James and Thomas cobbling together mini-models to cover the ten Commons registrable categories. The ten mini-models have now been amalgamated into one ruleset-compliant master model. Another couple of hours were spent checking the combined model and - despite having taken a couple of liberties around identifiers and places - neither registrar emerged looking too queasy.

With Anya on vacation, Michael and Robert’s cold managed to purloin a handful of her crack librarians and set them combing through the written evidence, the oral evidence and the additional material crowdsourced by Steve, for anything that looked roughly like a competency question. Thanks Ayesha. Thanks Anna. Thanks Deanne. It means we now have a spreadsheet of questions all citing sources. Next week’s job will be looking down that spreadsheet and marking all the questions our new model is capable of answering. And all those it isn’t. Wish us luck.