General NAO Challenge Discussion

132 views
Skip to first unread message

Accountability Hack 2014

unread,
Oct 27, 2014, 12:24:17 PM10/27/14
to acch...@googlegroups.com
To enter the NAO Challenge, your project must focus in one of these three areas:

  • Best prediction tool 
  • Best API created from NAO datasets 
  • Best tool to increase transparency in processes

Nick M Halliday

unread,
Oct 28, 2014, 6:30:49 AM10/28/14
to acch...@googlegroups.com
Here is a list of some of the more detailed ideas that have been generated so far that could feed into our three challenges:

External suggestions


Challenge

To combine publications.parliament.uk/pa/cm/cmregmem… with allaregreen.us to show who funds our MPs


Challenge

Is there a "ready reckoner of reckonings" available? Could ONS should make one available  e.g. If we think that the wages and salaries growth rate for 2013/14 was 3% and then find out that it was really 3.1% (in 2014/15) but also predicted it would be 2.9% and 2.8% (in 2012/13 and 2011/12 respectively) then clearly it would be useful to get one's hands on a table/grid that showed the years being predicted (column headers) the years of the prediction (earlier year at the top, later years below), with the relevant %s in the intersecting cells.  

Why?  Because if you happen to be auditing forecasting models and you need to know about the variations in key growth rates modellers use (RPI, dividend rates, blah blah blah), that snapshot would be VERY useful.


Challenge

I want to be able to search, be e.g. borough (or a given boundary) OR by sector (education/seconday/Academies) OR by any other field, for all public bodies, who their auditor was, what the fee was, when the accounts were signed, opinion type (there are 4 categories: unqualified, emphasis of matter, except for, qualified PLUS a regularity opinion on top of that...).


Challenge

To obtain an accurate snapshot of UK #localgov borrowing & lending & investments…

(developers should be v wary of data that falls under that > £500 - those kinds of outputs are NOT COMPLETE and relate to CASH payouts in a given period not the expenditure relating to that period (these are not the same thing).


Challenge

Can someone do a really simple app for all Cttes plus APPGs and APGs?  Not sure how to make it "really cool" as they say because no one wants to see a regurgitation of what is already on the parliament website...


The Private Finance Initiative Challenge

A tool to enable collaborative research into Private Finance Initiative projects.


You are a local activist who is interested in PFI. You want to see what PFIs are active in your local area, jump to a repository of existing documents and research and collaborate with others to establish whether the projects are value for money.


  1. Create a slick, interactive map of PFI projects across the UK. Based on the Treasury database here. Somewhat following the pilot we’ve made here, showing all the Department of Health PFIs (~10% of the total). Probably move to Google Fusion Tables but open to suggestions.

  2. Assign a web page/infrastructure to each of the PFI projects to use to host relevant documents and research through Document Cloud. For example, FOI disclosures of contracts, digitised paper copies of historic business cases, public minutes and company accounts etc. A Wiki model seems to make sense.

  3. Create a workflow for processing this information to allow people to share research. More of a stretch goal - but with a few hundred pages of board minutes hosted online, how can you allow people to extract the information so that that people aren’t analysing the same information again and again.


This creates something intuitive and usable from a public dataset, saves a large amount of duplicated effort on the part of local researchers working in isolation (and local government answering the same FOIs all the time), generates new datasets across projects and once established is applicable to any other public interest research projects.


It’s as easy or hard as you want to make it. Anything from adding a bit of flair to our mapped data, creating a basic web infrastructure to host research all the way to designing a research workflow to break up tasks into manageable chunks. See here for a basic plan.


Richard Palmer

unread,
Oct 31, 2014, 7:44:48 AM10/31/14
to acch...@googlegroups.com
Not massively useful but I wrote a while ago a perl script (not dead yet) to convert the PFI data into a SQLite DB (or database of your choice), intended for use with a PFI app or somesuch. I've tidied up a bit and uploaded to github - https://github.com/atiro/PFIDB - haven't verified it all imports correctly though yet.

Jo Carter

unread,
Nov 4, 2014, 9:39:05 AM11/4/14
to acch...@googlegroups.com
MARCH MADNESS!

We aren't going to be there over the weekend, but we're starting to look at how government spending increases during the last few weeks of the financial year.  Here is a link to the project on Pipeline and here are some notes from a session on this at GovCamp Cymru 

Research from America from 2004 - 2009 shows that spending in the last week of the financial year equated to 9% of the annual budget, which would have been 1.9% had it been evenly spread throughout the year.  The research also found that the quality of spend was reduced as a result of speed purchasing.

Whilst government data is now sometimes available, actually making sense of it all is difficult for a non-techy like me.  Could something be developed to help make the process of looking at Government spending much easier?  

In the long run, with data which is easier to digest and easily accessible by the public, might this bad practice reduce? 

Ludovica Rogers

unread,
Nov 8, 2014, 7:15:08 AM11/8/14
to acch...@googlegroups.com

Ludovica Rogers

unread,
Nov 8, 2014, 7:29:23 AM11/8/14
to acch...@googlegroups.com
Reply all
Reply to author
Forward
0 new messages