[wdmmg-discuss] Still no joy with HMRC

Wed Apr 14 12:27:10 UTC 2010

Hi,

I just got off of the phone with Vilosh Brito, the CEO of Information Edge,
the company that created Terasolve and Camelot/Prove, which is the platform
that COINS sits on.

He was very helpful and agreed to forward on an email from me to an official
at the Treasury, to see if it would be ok for him and Information Edge to
assist with the design of an extract of the data which meets all the caveats
that have been specified in the previously denied FOI requests.

Apparently, the relevant people at the Treasury are bored of spending all
their time with lawyers! So it makes a lot of sense, if permission is given,
for the designers of the system to build a simple query (a rough quote of
1/2 day's work) that excludes all the sensitive data.

This might include:
- public corporations
- military spending
- secret service spending
- future forecast data
- "what if" modelling data
- data that is known to be incorrect

 Information Edge had issues with the schema of the database being released,
because that represents their intellectual property. That seems fair enough
really, seeing as they have designed a system that lets you run a country!!!
What this means is that some of the dimensions could well be stripped away.
But as long as the fact data, ie. the transactions, are kept, then it will
still be useful.

Another issue was that the data would almost definitely not reconcile with
the national accounts without that IP being present. I think this explains
the Treasury's belief that people would complain as a result of the data
being released. I think clear notices of this fact when the data is
presented would mitigate that risk.

The data could potentially be extracted as an XMLA file, but Information
Edge might have a better suggestion.

So, I've got to describe to Vilosh, in detail how he can help us, so that he
can get permission to do so. Does anyone have anything they want me to
include?

Cheers,
Donny.

On 14 April 2010 12:32, Lisa Evans <lisa.evans at okfn.org> wrote:

> You may remember the request I made for the type of databases used to store
> tax data, and that it was rejected on account of requiring greater than 3.5
> days work to find this information.
>
> So I narrowed down the requests to just one tax type (income tax) and asked
> for the schema and supplier separately:
>
> http://www.whatdotheyknow.com/request/schema_of_income_tax_database
>
> http://www.whatdotheyknow.com/request/supplier_of_income_tax_database
>
> I did the same with national insurance contributions:
>
> http://www.whatdotheyknow.com/request/schema_of_national_insurance_con
>
> http://www.whatdotheyknow.com/request/supplier_of_national_insurance_c
>
> All were rejected.
>
> It is not entirely clear to me the basis on which they are rejecting these
> requests. It seems like a mixture of merging the requests into one, saying
> the time required will take more than 3.5 days and suggesting I be more
> specific -- which I would love to be but I've not yet found, if it does
> exist, any public information that will allow me to be more specific.
>
> Any suggestions most welcome or discussion on the okfn chat tonight.
>
> _______________________________________________
> wdmmg-discuss mailing list
> wdmmg-discuss at lists.okfn.org
> http://lists.okfn.org/mailman/listinfo/wdmmg-discuss
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.okfn.org/pipermail/openspending/attachments/20100414/8c94d3d4/attachment.html>