[LINK] ICT Predictions 2009 - little OT (gov't reports example)

Anthony Hornby anthony.w.hornby at gmail.com
Tue Jan 13 10:32:58 AEDT 2009

Hi Tom & Linkers,
some comments interfiled below.

On Tue, Jan 13, 2009 at 8:21 AM, Tom Worthington
<Tom.Worthington at tomw.net.au> wrote:

> There does not appear to be any legal requirement for government agencies to
> send paper copies of reports to libraries. Changing to electronic copies
> would save about two tonne of paper per year*.

I think they are  required to ... not that this means we always get them ...

Under the LDS Australian Government departments and agencies are
required to provide one copy of each publication that they produce to
each participating deposit library. Deposit libraries comprise the
National Library of Australia, State Libraries and publicly funded
universities identified under the Higher Education Funding Act 1988
[External Site].

The LDS supplements legal deposit [External Site], a statutory
requirement under the Copyright Act 1968 [External Site] and
equivalent State legislation, administered by the National Library of
Australia and State reference libraries respectively. Please note that
by complying with the LDS, publishing agencies also meet their legal
deposit requirements.

> If the electronic version of a report is the "Best copy" (the technical term
> for the most complete version of the work), then that would be acceptable to
> supply. Instead of actually mailing, or e-mailing, electronic copies, it
> should be sufficient to put them on the web and tell the libraries where
> they are.

The format would be acceptable to me but isn't the major problem. As
to "putting them on the web" we'd like to harvest the information from
some well ordered source this would save everyone time.

> AGIMO could change the procedures to encourage electronic distribution, with
> the paper option retained for the few items where this is not suitable. If
> the agencies put some metadata in the reports to identify them, then
> specialized crawlers can find them (such as the NLA and AGIMO ones). The
> obvious identifier to use is an ISBN in the appropriate Dublin Core element.
> The agencies would then not need to send any report forms or any printed, or
> electronic, copies anywhere: just publish to their own web sites and let the
> crawlers find them.

Problem is government agencies merge and change names constantly so
basic web links would likely break fairly quickly. Far better to have
them in repositories somewhere for preservation / persistent links.
State libraries / NLA / Government run would all be good options.
Universities also all have repositories now and if the central
repository was setup for simple harvesting of items everyone else
could replicate whatever they wanted to preserve locally very simply.
As to ISBN's why not ditch that and give departments handle
(handle.net) prefixes and allow them to use whatever suffix they like.
Failing that let them buy into the commercial DOI scheme which seems
to just be a commercial implementation of handle.net anyway
(http://www.doi.copyright.com.au/). Handle is a far more flexible
discovery mechanism than ISBN for many reasons and used in the bulk of
Australian university repositories and worldwide.

> Agencies could help the process by improving the quality of their reports,
> making them easy and efficient to read online, instead of just facsimiles of
> printed documents. This would reduce the cost of production and
> distribution.

Seems perfectly reasonable to provide layout and presentation
appropriate for the medium.....

>> All government reports should also be released under a creative commons
>> licence to allow us to copy and preserve them with ease. ...
> The Queensland Government seems most willing to consider this, perhaps
> because the Australian CC experts are at QUT in Queensland. Jessica Coates
> talked about it at the Open 2020 Summit last year:
> <http://tomw.net.au/moodle/course/view.php?id=9&topic=11>.

Nice advantage of CC is it is simple to end users to understand and
also simple to allow metadata harvesters to bring back the rights also
as these are nicely standardised.

> will be about 2,231 t.

Most of it waste *sigh*.

Regards Anthony

More information about the Link mailing list