Jayne and Michael continue to chip away at adding reverse calculations to our beloved Egg Timer aka “Parliamentary Time”™, though that work is not quite ready for the limelight. Hopefully next week, they tell us. If they remain true to their word, at that point, citizens of Whitehall lucky enough to have a Minister to call their own will be better equipped should their Minister wish to exercise power X by target date Y. Merely fire up the Egg Timer, tell it the target date and the applicable procedure, and the latest possible laying date will be returned with no further trouble. We’re told this will represent quite the efficiency saving, freeing up both civil servant time and the Whitehall budget for National Trust calendars and abacuses (abaci?).
This week has been mainly spent responding to feedback from testing with users. Which means Iona’s request for some form of notification pertaining to the ‘21 day rule’ has now been met and Arabella’s request for explanation of the assorted procedural calculations and the legislation that sets them out is in progress. Next week we intend to respond to Ben’s feedback, making it more obvious when meeting the Minister’s target date would require some form of time travel, not yet invented.
Increased focus on the old Egg Timer combined with the sudden slew of holiday taking has left us with less time than we’d have liked to concentrate on our groundbreaking Procedure Browsable Space™. If you’ve been following along from home, you’ll know that we’ve entered phase two of development. Phase one saw the assembling of a full and complete ‘browsable information space’, with all the things acquiring a URI and all the lists of all the things being built out. Phase two is taking a long, hard look at the results - sharpening our HTTP chisel and sculpting it into something useable by semi-normal people.
To help with that, we drafted in the marvellous Mr Waddington for a good click around, armed always with the questions:
is this view of interest to the librarians?
is it of interest to other people?
or, is it of interest to no one?
Anything falling into the final category is in the process of being stripped out. Things of interest largely to Jayne and our crack team of librarians, are being left in place with a layer of CSS to hide them from the common or garden punter. Jayne and team have been handed a tiny dash of CSS to be run client side, turning the less interesting views back on again. And hopefully everyone is happy. So far, we’re about fifty percent of the way through making the Waddington inspired changes, and we think it’s already looking much less noisy. Why not take a click? Tell us what you think? You could even run the local CSS and see the world through librarian eyes.
In the course of correcting, it came to our attention that quite a number of our aggregations did not work as intended. Being big fans of listing things, we had 14 different lists of work packages, all ordered by laying date. All of which worked as intended for things that had been laid. Unfortunately, our beautifully managed lists also contain work packages for things that may or may not become treaties, none of which have yet been laid. All 14 queries have now been tweaked to include things both laid before Parliament and things brought to the attention of Parliament. Which is a day we’ll never get back, but does appear to have fixed that problem.
Another quiet week for our procedural cartographers, broken only by Librarian Ayesha’s addition of a new step and accompanying routes to all three remedial order procedures. This one covering Correspondence published by the Joint Committee on Human Rights (JCHR), which is not, in fact, a new step, having lived out the first part of its life solely as part of the CRaG treaty procedure map. Which is why, if you click on the last link but one you’ll find business items actualising that step from both treaty-focussed work packages and remedial order-focussed work packages. Wow, you guys really take your reference data normalisation seriously, you’re probably thinking. Yup.
The long, slow job of painting our assorted applications in corporate colours continues to reap rewards. With much needed help from Designer Mary, we now have access to the assets that go to make up the Parliament design system. Rather than just plonk on links to CSS and fonts, Shedcode James has taken the much more professional approach of hand-rolling a Library design system Ruby gem. The gem includes the design system assets, and also an overspill stylesheet covering styling that is either not yet covered by the design system or, perhaps, will never be.
Last time out, we were delighted to announce that our MNIS Prodder™ was the first application to leave the paint shop. And stunning it looks too. This week, it was joined by our Question Checker™ application, also much improved.
Last time out, we promised we’d be live with by-elections held during Parliament 56 by the end of July. And we came damned close. Librarians Anna and Emily pulled out all the stops, applying their much admired spreadsheet skills to both constituencies and candidates. Unfortunately, it turned out we were short of invalid vote figures for a handful of by-elections. Something we could work our way around with a template change or two, but less than ideal. Researcher Isabelle is currently manning the phones, contacting local authorities and asking politely if they could possibly take a look in their filing cabinets. One gap filled so far, the rest to follow shortly. Do tune in next time.
In less interesting news, we’ve also been hard at work reshaping both code and database to deal with the dreaded Milton Keynes split, this being occasioned by the The Parliamentary Constituencies (England) (Miscellaneous Changes) Order 1990. Usually, interim boundary changes tend to tweak boundary lines only. Unusually, this particular order not only shifted boundary lines, but also turned two constituencies into three. Not a thing we hope to ever witness again.
Our original solution created a whole new set of constituencies for the post-interim boundary change period across the whole of England. But that threw us out of kilter with constituency identifiers in MNIS. So we now have an overarching English boundary set covering 1983 to 1997 and two child boundary sets: one set containing the two expunged constituencies from 1983 to 1992 and one set containing the three new constituencies from 1992 to 1997. And all we have to show for all that effort is a nested boundary set list. Which is not a lot of change for rather a lot of effort. Although we remain confident that we’ve saved ourselves time and trouble in the future. Which is not nothing.
Also now live is our first attempt at a ‘data dictionary’. Rolling like it’s 1995. If you’re a consumer of our election results Datasette offering, it may well be of use. If you’re more of a casual, drive-by website clicker, it should also help explain exactly what it is you’re looking at.
Librarian Emily has also found herself occupied with tidying MNIS records for our Members’ countries of birth. Which had been something of a mess. But is now not. Another bit of MNIS both tidied and documented. Top work, Emily.
As our dear reader will well know, the main output of our crack team of librarians’ efforts is Parliamentary Search. This is, for reasons lost in the mist, only available to users on the parliamentary network. We do, in fairness, make a public version available in the shape of Search Material but that had its functionality restricted in all kinds of unexpected ways and we would not recommend using it.
One side effect of restricting Parliamentary Search to an internal audience was that research briefings “published only on the intranet” could be included. Given we’re on a mission to replace Parliamentary Search with what we still tend to call Jon’s Search, and given that Jon’s Search will be available to all and sundry, the presence of “intranet only” briefings was something of an encumbrance. Happily, colleagues in the Parliamentary Computational Section stepped in and solved the problem. Our Solr instance is now free of internal-only briefings and the pipes into Solr have been tweaked to ensure that no more turn up in the future. Another dusty corner spruced up.