Bryan

Off to a meeting - I'll come back on this tomorrow and I don't think I'm
confused...just not a scientist.

Regards

James



On 27 September 2010 12:30, Bryan Lawrence <[email protected]>wrote:

> Hi James
>
> I think you confuse content with form.
>
> It is the day job of my science support team to ensure that information
> gets into *whatever* pipeline for metadata etc that exists. I think the
> same statement would apply in all the data centres.
>
> That's why I don't think that you want an admin function, since managing
> data information is what we do, and it's a science activty it's not a
> support/admin activity.
>
> Yes, we need a mechanism to support improvements, but it doesn't happen
> via admin ...
>
> (And, yes, I think the single face is only fed by the data centres, and
> so, no it ought not be possible for data to appear any other way, since
> the data centres own the decision as to whether it is appropriate for
> data to appear that is not held by them, *and* the mechanism.)
>
> Bryan
>
>
>
> > Bryan
> >
> > Likewise
> >
> > James
> >
> > On 27 September 2010 11:07, Bryan Lawrence
> <[email protected]>wrote:
> > > Hi James
> > >
> > > See inline.
> > >
> > > > The main debating point is the 'centralised administrative
> > > > function'. We all know that there is already a NERC Data
> > > > Management Coordinator that has been in place for 7+ years.
> > > > Perhaps what we're recognising is that this role is a key
> > > > *people* component of the IA but given all the other 'pulls' on
> > > > this role, the IA has not yet found the gravity to get more
> > > > specific focus from this role. It was clear back in 2004 that
> > > > the DMC needed support and this was accepted by NERC at that
> > > > time. The NDG MSI spend some funds on improving the management
> > > > information available from the DDS but I'm not sure whether this
> > > > capability is being used to any great effect. There have been
> > > > some issues in sorting out what is a real search on the DDS and
> > > > what is a visit from our friends at Google.  The NDG MSI
> > > > recommended that marketing and communication should be funded
> > > > and improved....
> > > >
> > > > We're suggesting that some sort of central help (not at DMC
> > > > level) is likely to be necessary to do some/all of the above
> > > > (and probably more) from a Single Face of NERC point of view. In
> > > > order to do effective marketing and branding you have to know
> > > > what's going on and what is planned.
> > >
> > > I don't doubt we need governance, and more joined up working, but
> > > I suspect what is being proposed is fighting battles from the last
> > > war. What i see now is a relatively joined up response, and it
> > > certainly needs ongoing governance - what I don't see is how a
> > > central *admin* person for this activity will help. What will they
> > > do day to day?
> > >
> > > It might be we agree on some tasks that need doing in a coordinated
> > > way, but I don't yet understand your vision for this role. Is it
> > > really just about marketing and branding?
> > >
> > > > > But what is the reference number? Grant numbers wont apply to
> > > > > everything, and there is *NO* number which covers what is done
> > > > > in all centres ... but each data centre could and should be
> > > > > reposnsible for project descriptions and a URI scheme.
> > > >
> > > > We know that there is no number that covers everything - some
> > > > will have something others won't. There may be two or more
> > > > numbers (Grant Number, Project Number...) or a number
> > > > accompanied by a type.
> > >
> > > Ok, this is different from what I had understood. I think this  is
> > > ok ... but better would be  a URI to a project description, and
> > > the NERC data centres should maintain those as well (it's patently
> > > obvious we need to do it from existing experience, and it should
> > > join up with EOF).
> >
> > I think we now agree on this but the devil may be in the detail -
> > some RCs definitely want to have searchable project codes in their
> > discovery metadata.
> >
> > > > > > 2. Projects producing data should have the option of making
> > > > > > their metadata (discovery *et al*) and data services
> > > > > > available to the 'Single Face of NERC'.
> > >
> > > Yes, but I think your answers below have me wondering what you
> > > really mean.
> > >
> > > The information is going to have to go *via* the data centres, even
> > > if projects provide their own discovery documents (which is
> > > unlikely).
> >
> > So we are saying that *all* data that appears via the Single Face of
> > NERC can only get their via a NERC Data Centre?
> >
> > > > > >    - Those projects not wanting to make their metadata and
> > > > > >    services
> > > > > >
> > > > > >       available to the 'single face of NERC' should provide a
> > > > > >
> > > > > > documented rationale
> > > > > >
> > > > > >       behind this decision.
> > > > > >       - Those projects that do want to make their metadata
> > > > > >       and
> > > > > >
> > > > > > services available to the 'single face of NERC' should
> > > > > > provide documented estimated
> > > > > >
> > > > > >       timings, volumetrics etc to the administrative function
> > > > > >
> > > > > > supporting the 'single face of NERC'.
> > > > >
> > > > > timings of what? volumes of what?  this has traditoinally
> > > > > involved some investigation by data centre staff rather than
> > > > > expecting to be told what we need ...
> > > >
> > > > Sticking with timings - we're referring to when components of the
> > > > data (metadata, data services) are likely to become available to
> > > > the 'Single Face of NERC' (not the Data Centres - that's their
> > > > bag).
> > >
> > > Can't see how they get to the "Single Face" except via the data
> > > centres. What do you wman?
> >
> > Is there an instance where data could appear via the Single Face of
> > NERC without going through a Data Centre? My guess is that you're
> > going to say No. Same question as above really.
> >
> > > > Inevitably this will have to involve Data Centre staff as the
> > > > data will become available via the data centre. Whether its the
> > > > Data Centre staff or the Data Providers (probably best to be the
> > > > former) doing the communicating, we feel that there is a need
> > > > for more central coordination.
> > >
> > > Of what? Exactly?
> >
> > Of information coming out of this process. By following the existing
> > process, discovery metadata and related data services could (in
> > theory and practice) just appear in the DDS (which is fine). My
> > question is how would anyone know that this had happened and if it's
> > a particularly important service how can we promote its recent (or
> > predicted - hence timings) appearance. Once it has appeared, what do
> > we want to know about it - number of accesses, downloads, citations?
> > How will such information be communicated up, down and across to
> > demonstrate the value of the IA and the dataset? Do we care as once
> > its there its there? Do we want the community to recommend
> > improvements? How many accesses are there directly through the Data
> > Centre as oppose to through the Single Face of NERC? To me this role
> > would be a resource to help answer some of these questions (and many
> > more) to support better investment decisions in the future.
> >
> > > > > ... but there is no need to invoke a centralised admin
> > > > > function.
> > > >
> > > > A centralised function already exists, albeit not administrative,
> > > > so it's not a massive leap so give that function some support
> > > > (even if it were on a short/medium timescale) to bring this
> > > > together working with the Data Centres (as recommended and
> > > > accepted by NERC in 2004).
> > >
> > > To do what? Exactly? I don't know what this function would be.
> >
> > Perhaps the current DMC should be brought into this debate - I'm
> > pretty sure they aren't on this mailing list. I'm not going to
> > second guess what all the functions could be and certainly not in
> > isolation. I've referred to a few above and a few below -
> > extrapolate that to 6 Data Centres and you might get at least a part
> > time role.
> >
> > > > > > 3. The point made in 2 infers that there is a need for human
> > > > > > intervention as the decisions are likely to be subjective.
> > > > > > Thus an administration role seems to be required that would
> > > > > > oversee the population of the 'single face of NERC', they
> > > > > > would also check that metadata is supplied as and when
> > > > > > expected.
> > > > >
> > > > > Yes ... that would be business as usual if it were done by the
> > > > > data centres, and I can't imagine any centralised solution
> > > > > that would have the relevant expertise to do it.
> > > >
> > > > Ignoring the expertise side as I agree that there is no point in
> > > > having an administrator who is a informatics expert. If you can
> > > > demonstrate how the Single Face of NERC is going to promote its
> > > > capabilities and requirements to the community it intends to
> > > > serve (as recommended in the NDG MSI Final Report) via 6+ Data
> > > > Centres without some sort of central coordination then great.
> > >
> > > Not sure we are agreeing here at all. Maybe we need some
> > > centralised informatics, maybe we don't, but I don't see what the
> > > admin role would do.
> >
> > OK - so what you're saying  (if I understand this correctly) is that
> > your informatics experts are actually performing some administrative
> > task in this process - checking that metadata arrives when it
> > should?; checking that the information in the appropriate fields
> > makes sense?; checking that url references work?; checking that
> > contact details are up to date? etc. Perhaps there is some scope to
> > take some of this administrative burden away from these experts?
> >
> > > Bryan
> > >
> > > ---
> > > Bryan Lawrence
> > > Director of Environmental Archival and Associated Research
> > > (NCAS/British Atmospheric Data Centre and NCEO/NERC NEODC)
> > > STFC, Rutherford Appleton Laboratory
> > > Phone +44 1235 445012; Fax ... 5848;
> > > Web: home.badc.rl.ac.uk/lawrence
> Bryan Lawrence
> Director of Environmental Archival and Associated Research
> (NCAS/British Atmospheric Data Centre and NCEO/NERC NEODC)
> STFC, Rutherford Appleton Laboratory
> Phone +44 1235 445012; Fax ... 5848;
> Web: home.badc.rl.ac.uk/lawrence
>



-- 
James Doughty
Director
Diass Limited
07985 443973
[email protected]
_______________________________________________
NDG-technical mailing list
[email protected]
http://lists.ncas.ac.uk/mailman/listinfo/ndg-technical

Reply via email to