2021 - Week 35

We must start by apologising for the lack of notes last week. Both of your regular correspondents were feeling a little under the weather. Or the worse for wear. Take your pick. Nevertheless, we have been hard at work, our assorted Trello boards a blur of activity. Which means we now have two weeks of frantic paddling to cover. But please don’t worry, the work has been particularly dull in nature. So we’ll try to keep notes short and sweet. Though we have said that before.

The return and revenge of Librarian Phil

This week we were delighted to welcome Librarian Phil back to the warm embrace of the Indexing and Data Management Section of the House of Commons Library. Or ‘our team of crack librarians’ as you may know them. Phil has already served on the frontline of IDMS but took time out to manage the system used to catalogue our many and varied books and journals. He’s now back in the world of attempting to teach parliamentary procedure to machines. Where the real action is.

Arithmetising the procedure maps

In week 34, Librarian Jayne and computational dilettante Michael put the finishing touches to the arithmetising of our assorted statutory instrument procedures. Both maps and data for all of our procedures are now considered complete and have been handily collected onto a single page for your delectation. The page comes complete with our hand drawn efforts, the dot files that the machines have kindly spat out and SVGs produced from those dot files. So, if you’re in the market for tinkering with dot files, this is your one stop shop. Tony.

Logical and arithmetic data safely tucked up inside the machines, week 35 saw Jayne and Michael checking hand drawn maps against machine drawn SVGs. Draft negatives, made negatives, draft affirmatives, made affirmatives and the treaty procedure as set out by the Constitutional Reform and Governance Act 2010 have all been painstakingly checked. With the emphasis firmly on pain. If you’ve ever fancied attaining a hypnagogic, trance-like state but never been one for repetitive beats, we can heartily recommend procedure data map checking.

Our Jianhan also leant a helping hand, clearing out the fragments of our mis-mapped draft composite procedure and ensuring that the orchestration of step descriptions is switched on. Which means we have descriptions for every business step in every procedure safely inside our triple store. If not yet on the website(s).

Before heading off on a much needed vacation, Jianhan also ticked off the work required at our end to push published drafts under schedule 8 paragraph 14 of the European Union (Withdrawal) Act 2018 into our live triple store. At some point in the not too distant past, we collared Matthew and he helped us with a bit of a rewrite on our legislation model. What had been a Statutory Enabled Thing is now just an Enabled Thing and quite a bit of the new modelling hangs off that. We had been waiting for colleagues in Software Engineering to update a query in a way that would allow us to transition to Enabled Thing without breaking the website. That work is now done. So Jianhan has added published drafts to the physical ontology, turned on the feed of letters laid from Solr to our procedure editor database and plumbed in the pipes for published draft data from said database into our live triple store. Top work Jianhan. Enjoy your vacation.


Away from the drudgery of checking and re-checking procedure data, there have been a couple of glimmers of genuine interest. We have continued on our quest to understand the various types of Members - we count six types but do let us know if we’ve missed any - and whether those Members might move between seats of the same type and between seats of different types. This week our sketch of Member movements benefitted from the keen eyes of House of Lords Anna and Kirsty. And gained so many new lines it begins to resemble a butterfly. Lovely stuff. Thanks Anna. Thanks Kirsty.

On the nature of places

Way back in week 32, we were forced to call on Carl for some remedial geography lessons. We wanted to understand how best to model geographic areas which are bigger than constituencies. And found our knowledge of such matters sadly lacking. This week Librarians Anya and Phil and computational ‘experts’ young Robert and Michael met with Carl. A whiteboard was commandeered for the first time in several months and we think we’re beginning to understand geographies. And how ridiculously hard they are. We now have a rough outline of a data model that we hope to decant into our more relational efforts once folks are happy. Mark has already pointed out that, whilst constituencies forming part of a combined authority are currently wholly contained by that Combined Authority, this may not be the case in the future. So that’s probably more join tables we need to add. No matter. If there’s one thing we love, it’s a join table.