From: Marcus A Banks <[log in to unmask]>
Date: Thu, 18 Feb 2016 05:00:58 +0000

Agree -- data/text mining requires different understandings. "Normal"
usage as defined by whom, and for what agenda?

Obviously publishers will seek to protect their IP from entities such
as SciHub, but of course the entire debate surrounding open access is
whether that IP is legitimate. Which depends on which side of the
fence you stand on.

The OA debate is now very stale. And the writing is on the wall for
immediate OA in the biosciences -- embargos will become a thing of
history.

I hope that, going forward, the revenue streams for publishers
transition from licensing and APC schemes into licensing tools for
data/text mining on top of an open corpus. -- Marcus


> On Feb 18, 2016, at 12:09 PM, LIBLICENSE <[log in to unmask]> wrote:
>
> From: David Prosser <[log in to unmask]>
> Date: Wed, 17 Feb 2016 11:43:52 +0000
>
> Unfortunately, this method of looking for triggers above ‘normal’
> usage massively inhibits text and data mining.  We are in an era where
> ‘normal’ and ‘excessive’ need to be redefined.
>
> Dave
>
>
>
> On 17 Feb 2016, at 02:18, LIBLICENSE <[log in to unmask]> wrote:
>
> From: "Maher, Stephen" <[log in to unmask]>
> Date: Tue, 16 Feb 2016 16:21:16 +0000
>
> Hi Richard,
>
> In my experience publishers monitor the rate of downloading from an
> institution’s IPs. If the rate exceeds “normal” usage, the publisher
> notifies the institution with information of when the excessive
> downloading occurred and over which IPs. More often than not the
> excessive downloading occurs over the IP address connected to an
> institution’s proxy server. The institution then attempts to identify
> the username(s) associated with the downloads and temporarily suspends
> it.
>
> Hope this helps.
>
> Stephen
> Stephen Maher, MSIS
> NYU Health Sciences Library