A big day for big data – Like Ideas Conference 2013

Last Friday afternoon I
popped on my conference shoes and headed over to the St Bride Foundation for
the 2nd annual LIKE Ideas conference. Once again SHR sponsored 10 conference
places and our winners have kindly agreed to share their impressions of the day
as guest bloggers for us.

First up is Stuart Lawson @lawsonstu

In this year’s LIKE conference ‘From Big Data to Little Apps’
speakers from a variety of backgrounds talked about how they’ve been using big
data in the workplace. As an information professional I was looking for ways
that their insights could be applied in libraries. For example, Monique Ritchie
from Brunel University discussed big data from the perspective of research data

Dom Pollard from Big Data Insight Group mentioned the importance of
collaboration – we don’t have to restrict ourselves to internal data if sharing
it provides benefits. Libraries now collect a lot of resource usage data,
particularly of electronic resources. Could any of this be shared with other
institutions, perhaps to benchmark or to discover wider trends?

One point that was raised in the discussion following Michael
’s talk on data visualisation is that data must be reliable. Presenting
data in appealing ways helps people to understand it, but there is an ethical
imperative to ensure the data is as accurate as possible or else people may be
misled by it. This is one of the drivers behind the open data movement.

Vast amounts of data are being created every year but we can only
make use of it by asking the right questions. Manny Cohen of RM described a
mobile app as an interface to interrogate a database; apps can give users
access to big data by delivering to them the exact information that they need.

The conference certainly gave me a lot to think about!



One comment to this post

  •  :  Hi. I wasn't able to get to the conference, but your post reminds me of an issue that used to bother me about statistics in the early part of my career when I was a statistics librarian, and I wonder if it came up. Statistics are collected for particular purposes, and may be incomplete or misleading for other purposes. So obviously good metadata is important. In the modern open data movement, government statistics with good metadata should be made available - in a sense they belong to the public. But data exists below the statistical level - eg. computers hold data about processes etc as a by product of their use. Does the same "should" apply at the data level? Hard to imagine they can be made useful without work by the owners of the data, so even if the same obligation applies, is it sustainable? Dion

Leave a Reply to Dion Lindsay Cancel reply

ILX-Recruitment tfpl