Hot off the hottest of presses, the job ad for our Open Data Technical Lead has just gone live. At the risk of repeating ourselves, if you’re a dab hand with a triple store, fancy helping to reinvent parliamentary information management and would like to turn these frowns upside down, please do click the big, purple ‘Apply’ button.
On the subject of resources of a homo sapiens nature, Librarians Anya, Liz and Steve, together with computational gentleman amateur Michael, have spent the last couple of days interviewing candidates hoping to join our team of crack librarians. Spending five hours a day concentrating on pixel based communications makes one realise quite how much one tunes out in the course of an average meeting. We think they call it ‘active listening’ and, my word, it is tiring. That said, we met some pretty darned fine candidates so can’t really complain.
We are pleased to announce our enabling of enabling Acts is now enabled. This latest missive is being typed on a Saturday afternoon. Earlier this morning, following extensive testing by Librarian Jayne and a green flag from boss bloke Ian at the end of last week, our Jianhan pressed the button that sent data flowing into our live triple store and the welcoming arms of our brand new enabling Act model. Monitoring continues and all appears to be well. It only remains for our colleagues in Software Engineering to add signposting and filtering to the statutory instruments website and we feel sure a pat on the head from JO Jane is assured. Thanks are due to our friends at The National Archives for putting us straight on our modelling efforts. We wave as ever to John, Matthew and Catherine and proffer a pint in the not too distant future.
Last week we reported on our first paddle in the shallow waters of componentised procedure maps. This week, efforts paid off as Librarian Jayne and Michael kicked off the usually labourious task of feeding our logical draft affirmative map to the machines. What used to take the best part of a day was done in a little over two hours because we now have ready-made component maps for EVEL certification, the Joint Committee on Statutory Instruments, the Select Committee on Statutory Instruments and the Secondary Legislation Scrutiny Committee. We can now manage the data once and reuse anywhere, which not only lessens the opportunities for mistakes creeping in, but also makes the job of entering new routes for new procedures a lot less arduous than it used to be. All the data for the logically mapped draft affirmative procedure is now inside the machines and the machines have once again been good enough to draw their own version. Thanks machines. More interviews aside, next week will be dedicated to checking our work. And probably correcting it. At which point, that’s four procedures down, two to go.
Procedure editing got easier still as our Jianhan took time out of his particularly busy schedule to make a couple of tweaks to the procedure editor. Lists of routes now also list the procedure or procedures the routes are in which gives us much better visibility of the changes we’re making.
A wee while back, Jianhan also wrote some code to check that our various step types have the appropriate number of inputs and outputs. That code runs every morning and sends out an email to our team of crack librarians reporting whenever it finds a business step with more than one input, or an AND step with more than two inputs, or a NOT step with more than one output, and so on. It also runs whenever we ask Jianhan to trigger it, a request we’ve been making with increasing frequency. This week, Jianhan added a button to the procedure editor, allowing anyone with access to trigger said email. Which we estimate will save at least 10 Jianhan-hours per week.
Work continued on grabbing petition data and inserting it - via an Atom feed from the data platform - into our Search and Indexing triple store. Anya has tested the feed and confirmed it includes all expected attributes. We now await a convenient time to begin drip-feeding petitions in the general direction of our librarians. Jianhan, Robert and Ian met with Andrew from Unboxed - the folks who maintain our petitions website - and confirmed our petition counts now match theirs. Andrew has also confirmed they plan to add any original government response to their JSON feed wherever the original response has been replaced by something more recent.
Last week we explained our thinking on the removal of clock start steps. Or at least tried to. This week, attentions have been diverted to the question of when clocks end. As tenancious readers of these notes will well know, the question of clocks and how one times procedural eggs is one close to our hearts. We had been resetting clock end steps to the date of revocation whenever an instrument met that end. But a week back, a stack of librarians came to the conclusion that this approach was incorrect. Time being time and legislation being legislation, they decided that the clock end step should never be reset, no matter what fate befalls the instrument in question. This conclusion was communicated to JO Jane who raised no objections. So Librarian Jayne has now updated both scope note and manual, reverted any reset end steps and communicated this change to all interested parties. It only remains for Jayne to raise a ticket with our colleagues in Software Engineering requesting that certain steps are not displayed on the website when certain conditions are met. Thereby preserving accurate data and heading off any confusion on the part of our users.
We have long been under the impression that John was the only regular reader of our outpourings. This week we were delightfully disabused of this notion when Silver introduced us to Wannes. And overnight our readership doubled. Wannes deals with matters of Intergovernmental Affairs at the United Nations. He works with groups and committees and meetings and agendas and papers and decisions and procedures - the flows of information on which decision making bodies depend. This rings a distant bell. He was keen to point out that most of them were still grappling with how to digitise document publishing processes. Which also rings a bell. And that we might be alone in side-stepping documents to model underlying process flows and procedures. So complimentary was Wannes about our procedure modelling work - and our reaction so very British - and indeed Irish (sorry Anya) - that we were still blushing by Thursday morning. It is safe to say our notes have not always met with universal acclamation, so it’s nice to know they’re of use to someone, somewhere, for some reason. Thanks Wannes.
