RE: Shrewd University OA Policy Advice from the Antipodes

[<Toby.GREEN@oecd.org>]

I recognise only too well the work being done by Professor Sale 
and can empathise with what he says about working with faculty. 
Why? Because at OECD Publishing we have been facing exactly these 
challenges since the 1960s when OECD first started publishing its 
authors' outputs for a wider audience. OECD has had a 
self-publishing mandate since the 1960s, which I guess equates 
very much to the idea an institution self-publishing via 
institutional repository, as being tried in Tasmania and in other 
places now. OECD still uses an internal publishing entity to 
handle the publication of its reports, databases and more. Since 
last year much of what we publish freely, for example our working 
papers, has also come under our responsibility. We've just 
completed a project that has improved how we manage and publish 
our working papers. In view of this and of our long experience of 
publishing for an internal author community, I'd like to echo 
some of what Prof Sale says, challenge one point, and add some 
new ones:

1. Authors need support - lots of it. We've learned that 
hand-holding never stops. This is because authors are generally 
unskilled at preparing or uploading documents and adding the 
correct metadata. They are also largely uninterested in learning 
too - imagining that the magic of the 'web' and 'google' will 
overcome all problems (that's if they even think of an audience 
beyond their own circle). This is wrong. We've recently took over 
the publishing of all OECD's working papers because the authors 
were so bad at doing it themselves. The result is a marked 
increase in downloads and in quality of output.

2. Another benefit of centralising the loading with an expert 
loader is lower costs. Why? because we've got a junior staffer 
doing all the loading and OECD pays her less than our expert 
authors. This may not matter to management if you've only got few 
authors, but becomes significant if you've got lots of authors. 
(Incidently, our staffer loads the working papers into an 
internal database which then uploads the metadata into more than 
one website (currently two but more in the pipeline) 
automatically - so we get more bangs for our loading buck and 
wider dissemination than we'd achieve via a single 
website/repository.)

3. OECD Publishing does all the transformation of manuscript into 
publishable PDF. This ensures a standard level of consistency and 
quality-assurance across the Organisation's outputs. This is 
especially important if the papers have a high level of graphics. 
If left to their own devices we've found that authors often 
forget to add simple stuff (like the institution's name) or use 
internal language/acronyms (eg "Joe Blow, ECO") in their 
affiliations. They have also been known to leave links to 
documents on the intranet in external documents. Centralising 
management also means we can build tools to add value for 
readers: in early 2007 all working papers will have reference 
linking via CrossRef and downloadable citations (compatible with 
EndNotes et al).

4. We have quarterly meetings with each department - don't 
underestimate the need to communicate regularly. You'd be amazed 
at the 'churn' rate among authors (ie new authors coming and old 
ones leaving) and the consequent need to educate and re-educate - 
even if you're doing most of the publishing work yourself.

5. Do worry about metadata quality - without it dissemination is 
severely compromised. In our experience authors are not good at 
managing metadata.

6. Even when you've got the mandate, don't think it'll mean you 
have to let up in convincing authors to stick with the mandate's 
requirement. Managing authors is like herding cats.

7. If our experience is anything to go by, your IR will become a 
significant outreach tool for your institution (assuming it is 
successful!). This means your management will begin to worry 
about the image it projects as well as the cost of running it. 
They will ask questions about cost/benefit and will begin to ask 
for reports on the number of downloads and running costs. This, 
in turn, means pressure to think about the quality of what is 
being loaded (and how to reject low-quality content), whether the 
papers carry the correct institutional 'branding' and so on. It 
may also mean that you will have to think about how you can boost 
download numbers.

8. The last point suggests that running an institutional 
repository will evolve to become little different from running an 
online publishing platform. So, hiring someone with a publishing 
background to manage both the platform and the marketing thereof 
might be worth thinking about.

Toby Green

-----Original Message-----
From: owner-liblicense-l@lists.yale.edu
[mailto:owner-liblicense-l@lists.yale.edu] On Behalf Of Stevan Harnad
Sent: 02 May, 2006 5:40 AM
To: JISC-REPOSITORIES@JISCMAIL.AC.UK
Subject: Shrewd University OA Policy Advice from the Antipodes

Professor Arthur Sale of University of Tasmania has rapidly
become the planet's premiere strategist of successful University
OA Self-Archiving Policy. Apologies for cross-posting -- but
ignore at your own peril! -- SH

[SNIP]