On the perils of early adoption

["Mcsean, Tony \(ELS\)" <T.Mcsean@elsevier.com>]

Christina Pinkas highlights a problem with some ScienceDirect backfile
articles that we have been aware of for some time.  Long enough, in
fact, to have put in place plans to eliminate it.  Strictly speaking
there are two related issues.

1. Individual pages being scanned poorly.

The Elsevier backfile conversion was a very large project - it 
has already been running 7 years and huge quantities of data have 
been digitised. The backfile collections on ScienceDirect have 
been very well received by customers and users and have brought 
many valuable research articles back into use in current day 
research. In addition, as we've heard on this list in recent 
days, many customers have taken advantage of the backfile 
availability to free up shelf space.  Some missing and 
poorly-scanned pages have slipped through onto ScienceDirect. 
We have added or rescanned these when they have been brought to 
out attention, and will continue to do so wherever we can lay our 
hands on good source material.

2.  Systematic improvements.

During 7 years of scanning, we have learned a lot from the 
experience and have seen important improvements in storage and 
other technologies.  Consequently, some of the decisions that 
seemed reasonable in 2001 are now being revisited. When we first 
started the project resolution levels used were 300dpi, black and 
white. Perfectly good for text, butt less so for images and 
sometimes small-format tables. To implement higher resolution, 
grey scale and colour would not have been feasible, given the 
mainstream storage and line speed then in current use.  In 2004, 
we were able to implement new technology to produce much higher 
quality images and include the full use of colour and grey scale 
without compromising performance or storage requirements.  We are 
currently rescanning where there have been particular problems 
and where colour is of crucial significance for meaning, 
including a complete rescan of some journals, including Brain 
Research, Neuroscience and Icarus.

The rescanning project continues.  You will start to see a marked 
improvement over time in all aspects of image quality and 
continuing elimination of of missing content below the current 
figure of 0.5%. This posting has been unusually detailed because 
we want the process, and the reasons why it has been necessary, 
to be transparent.

We are continuing to invest in improving the usefulness and quality of
the backfiles.  In the meantime, thank you for your feedback and for
your patience.

Tony McSean
Director of Library Relations
Elsevier
London NW1 7BY
+44 7795 960516
+44 20 7424 4242