ontologies

2021 - Week 33

It had promised to be another quiet week in distributed Westminster. As young Robert returned from his Battersea based vacation, Librarian Anya took off on hers. Leaving Librarian Jayne at the helm for what we’d assumed would be another week of slogging through the arithmetising of our picture-perfect procedure maps. Then recall happened and the feces hit the fan.

By Thursday, our team of crack librarians had almost 1000 written questions - tabled through recess - to subject index. But stops were pulled out and, by Friday, all questions came complete with subjects. Top work Librarians Ayesha, Jayne, Jayshree, Bosede, Emily, Emma, Sarah, Claire, Ned and Jason.

Arthithmetising the procedure maps

Our regular reader will know that we’ve spent the last couple of weeks mired in the mess of adding arithmetic steps to our statutory instrument procedure maps. Taking what we learnt from mapping the CRaG treaty procedure and our parsing code’s newfound ability to add, increment and equalise, we’ve ticked off both flavours of negative procedure and come very close to ticking off the draft affirmative. Very close. Within inches. Twice we ticked that House of Lords box and twice we unticked it as we realised we’d had some longstanding misunderstandings of procedure.

On the transmogrification of motions

The first realisation was of our misunderstanding amendments to motions. Unlike in the House of Commons, Lords’ Members may table motions - both fatal and non-fatal - against an affirmative instrument. Which always felt a little redundant, given both a debate and a question on the approval motion are guaranteed. We came to the rather late realisation that, by the time the approval is debated, the fatal and non-fatals act as amendments to the main motion. And we had thought, given they are amendments at that stage, it wasn’t possible to table one until the approval motion itself had been tabled. Amendments to something that doesn’t yet exist feeling somewhat odd. Luckily we have both JO Jane and TO Matt to put us right. It turns out - and get this - the nature of a motion transmogrifies over time. A Member can table a motion prior to the tabling of the approval motion and it is merely a motion. When the approval motion is tabled, the other motions become amendments to that motion. Should the approval motion be withdrawn, the amendments go back to being just motions. And when a new approval motion is tabled, they’re back to being amendments. All of which blew Michael’s mind. Though, as Anya pointed out, it should not have come as a major surprise. The nature of a thing being an amending thing being expressed as a relationship to the thing it amends and not some intrinsic property. So that put Michael right.

All of this is to say that Jayne and Michael have now made the tabling of both fatal and non-fatal motions dependent on the laying of the instrument rather than on the extantness of the approval motion. A job that took longer to explain than to do.

On our misreading of standing orders

The second realisation came when Librarian Jayne sat down and actually read House of Lords public Standing Order 73.1 We had been under the impression that an approval motion for an affirmative instrument could not be tabled until either the Joint Committee on Statutory Instruments had reported, or standing order 73 had been dispensed with. Dear reader, this is not the case. It is perfectly possible for the Government to table such a motion. What is not possible is that motion being moved before the scrutiny reserve conditions have been met. So that’s a job for next week. And then we should be done. Should be.

Parsing the procedure maps

With Jayne, Robert and Michael finally reunited, time was found to cast third, fourth, fifth and sixth eyes over Michael’s procedure parsing code. A good hour was spent squinting at lines of Ruby, making our usual adjustments to comments and generally sense checking the code for parsing our new step types. So intent were Robert and Michael on making the comments pitch perfect, that it took Librarian Jayne to point out that an equals step taking two different values should probably output a FALSE. Which it now does. Good spot Jayne.

Preparing the ground for rewriting our parsing code in a slightly more professional fashion, our Jianhan has finished moving laying dates from layings to the more generic business items. Where all the other dates are kept. Which means when we do next run the parser for an instrument that’s been laid, it shouldn’t tell us that a thing that should happen is a laying.

Jianhan has also saved both himself and assorted librarians some time by adding a new link to our procedure editor which, when clicked upon, returns a delightful dot file for the purposes of procedure map checking. Michael has taken the the dot files and added them to our shiny new page of procedure maps, together with our hand-drawn efforts and the SVGs rendered from the dot files. So, if you’re in the dot file market, fill your boots.

Modelling the gubmint

In slightly less procedural news, we continue to be bamboozled by how we might model assorted government organisations and the people that run them. This week Robert and Michael were joined by both Alice and Philip from the Institute for Government. We attempted to explain our problems and they attempted to explain theirs, Philip being kind enough to show the ERD for the IfG ministerial database. Mention was made of why we don’t all just use Wikidata for this. The Government being notoriously poor at mapping itself.

In truth, we are only marginally interested in the structure of government - should it indeed have a ‘structure’ - and more interested in the bits of government and other organisations that poke into Parliament. Be that Members being incumbent as ministers and secretaries of state, opposition spokespeople shadowing departments and things that used to be departments, bodies with a duty to be questioned by Parliament, bodies that are at arms-length from those bodies but to whom ministers turn to help answer questions, bodies with a duty to lay papers into Parliament and so on. All of which is a somewhat baggy and saggy set. The Church of England not fitting into any wider category that we can spot. It does feel like we may end up with a class of things with a title like ‘bodies with a duty toward Parliament’. Which doesn’t narrow the field too much.

And that was our week. How was yours?

  1. We would usually link directly to the order, but sadly the Lords’ standing orders lack fragment identifiers. So you’ll need to scroll down.