Thank you Sreenivas, these are very interesting questions and I will try
and answer some of them based on the presentation they made.

1. That is the answer they are seeking - as to who will latch on to their
data. As of now, they use this data internally - they have a business
analytics team which looks at all the datasets to detect patterns and tweak
their process. However, they are unable to see (and are willing to learn)
how people outside of UIDAI might find this data useful.
As to the filters they apply, I don't know the specific filters they apply,
but one of the principles on which they build their datasets is privacy -
they will not reveal personally identifiable information. So as an outsider
to UIDAI, you might get information at an aggregate level, but not anything
more granular than that.

2. UIDAI is building 3 portals - a Direct Cash Transfers portal, an
Enrollment portal (which is what I have sent the link to) and an
Authentication portal.
-Direct Cash Transfers portal would most probably be under the Ministry of
Finance, and they will decide if it's going to be open or not.
-The Auth portal has not yet been built - they are working on it, and I
*think it will be open. This portal would give data on usage of Aadhaar
numbers by various departments. And by its very nature, it is a
continuously updated portal. For example, if the PDS chooses to use Aadhaar
numbers for delivering rations, then the auth portal will show you how many
people (of the ones who chose to take an Aadhaar number if it's optional or
if the department makes it compulsory then how many people) in a particular
district used the aadhaar number to access these services. Details about
what they bought using the Aadhaar number will not be shared with the UIDAI
and consequently won't be available on the Auth portal.
-That leaves the enrollment portal - the way enrollment works is that a
person enrolls and 10 years down the line, they have to update their
biometrics. So it's not entirely static data. Also, in addition to the
aadhaar numbers, there's a lot of other data that's being generated which
is more dynamic like the device data etc

3. As Nisha was pointing out in a back-channel message, we should invite
UIDAI to the datacamp, and I think they are also very open to explaining
their processes to people. That will help you understand the context in
which each dataset is created. I don't know about the low reliability
levels.

4. And neither do I know about the statistical validity

5. Well, this is a question I can only guess at and I don't want to take
wild guesses :)

6. I think it does - data is available at a sub-district level for all
states and I *think it is both for rural and urban areas. Can confirm

Please keep the questions coming! I will try and answer based on the
presentation they made on 23rd. And if not, I will pass it on.

Best
Laina


On Wed, Jan 30, 2013 at 12:21 AM, Sreenivas KN <[email protected]> wrote:

> Hi Laina.
>
> Greetings and wishes on the meeting you propose to have. interesting to
> see UIDAI go public. some pointers of interest to me:
> 1. the question is who will be latching on to the data that is expected to
> be shared, generated out of public funds (read tax payers money). we cannot
> control the use of data once it goes public. what filters woudl they be
> applying to access and use of data
> 2. what is the use of one time data that UID is generating at the scale it
> is being done?
> 3. what is the rationale behind the data streams generated? it is
> important to understand the context in which data was generated. which ones
> in their perspective have low reliability levels?
> 4. how good is their data? what kind of statistical validity does it
> subscribe to?
> 5. what drives them to put private data (however insignificant it might
> seem) in the public domain?
> 6. does their database cover urban areas also ([?] my current work deals
> with this?)
>
> PS: we met at the ofkn organised meet at nipfp campus near munirka.
>
> let's keep talking on this. there are many inititives that could benefit
> from this interestign database?
>
> it will good to hear their side of the story on these small insignificant
> points.hope this helps.
> warm regards,
> Sreenivas
> On Tue, Jan 29, 2013 at 2:56 PM, Laina Emmanuel 
> <[email protected]>wrote:
>
>> Dear all,
>>
>> As you all know, the UIDAI has launched its enrollment data 
>> portal<https://data.uidai.gov.in/uiddatacatalog/dataCatalogHome.do>.
>> They are keen to get feedback on the kind of datasets that would be useful
>> to researchers and features that can be implemented on their site.
>>
>> To give you some context, broadly speaking, this is the kind of data they
>> collect (on the enrollment side)
>>
>>  -Timings - How much time it took to enroll a person, how much time was
>> spent in each stage of the process (and this can be broken up by registrar.
>> So in which state was the average time to enroll the maximum and so on)
>> - Devices - How many enrollments were done on a particular device, which
>> operator is handling a device and so on.
>> - People involved - Data about the registrar, enrollment agency, and the
>> person who did enrollment, how many mistakes were found against a
>> particular person and so on
>> - Type of errors committed - How many errors were committed by which
>> partner of the UIDAI during the enrollment process
>> - Geography - Data on enrollment by district and sub-district
>> - Demographics of a person - Data on people enrolled. UIDAI has a policy
>> of not disclosing personally identifiable information. However, you can
>> send in suggestions if you think that there's aggregated data which would
>> be useful for researchers
>> - Internal quality checking - There are a number of quality checks that
>> are done by the UIDAI of people, devices and processes. Is there any
>> information there that you think you can use?
>>
>> I would be talking to UIDAI in a day or two. Once they get our
>> suggestions, they will see which of the features can be implemented, and
>> which datasets can be made open.
>>
>> Since a lot of people here are interested in data, your suggestions would
>> be valuable. Please do send it in by end of day today so I can collate it.
>>
>> Best
>> Laina Emmanuel
>>  Communications & Technology Officer,
>> Accountability Initiative <http://www.accountabilityindia.in/>, Centre
>> for Policy Research <http://cprindia.org/>
>>
>> --
>> For more details about this list
>> http://datameet.org/discussions/
>> ---
>> You received this message because you are subscribed to the Google Groups
>> "datameet" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to [email protected].
>> For more options, visit https://groups.google.com/groups/opt_out.
>>
>>
>>
>
>  --
> For more details about this list
> http://datameet.org/discussions/
> ---
> You received this message because you are subscribed to the Google Groups
> "datameet" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> For more options, visit https://groups.google.com/groups/opt_out.
>
>
>



-- 

Communications & Technology Officer,
Accountability Initiative <http://www.accountabilityindia.in/>, Centre for
Policy Research <http://cprindia.org/>

-- 
For more details about this list
http://datameet.org/discussions/
--- 
You received this message because you are subscribed to the Google Groups 
"datameet" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.


<<328.png>>

Reply via email to