[Platforms] content governance

Wed Apr 4 21:41:23 BST 2007

Thanks Marilynn,

I TOTALLy agree that the list should be the mechanism for moderating code requests.  Please do whatever you can to get those asking for codes independently to join. My one worry is that the list could become a forum in which consensus becomes difficult to reach - but I've been there before and am sure we can come up with a solution should the problem arise.

Chers, Roy.

Cheers, Roy.

>>> Marilynn Sørensen <marilynn at ices.dk> 04/04/07 6:42 PM >>>
Hi Roy,

The remaining codes for POGO will be entered into RECO next week. The content governance issue arises because they are pending final checks due to conflicting information from people who I can see are not on the list and some have no knowledge of existing data from their own country. If I immediately enter a code in RECO everytime someone supplies information, we will have total confusion.

There are many requests awaiting codes and I am still trying to sort out if the requests have gone through the platform list, who has asked for what, whose information is correct and which ones are urgent (POGO).

You have already pointed out that if the harmonizing of lists is to work, it will be essential that all requests for codes and all replies with information must go through the platform list. I agree. It will have to be made clear who needs to accept the proposed code assignment before it is entered officially into RECO. 

Best regards,
Marilynn

-----Original Message-----
From: platforms-bounces at biwebs1.nerc-liv.ac.uk [mailto:platforms-bounces at biwebs1.nerc-liv.ac.uk] On Behalf Of Roy Lowry
Sent: 04 April 2007 16:38
To: platforms at biwebs1.nerc-liv.ac.uk
Subject: Re: [Platforms] G.O.Sars

Hello Marilynn,

To visit the platform subscribers go to:

http://mailman.nerc-liv.ac.uk/mailman/listinfo/platforms 

There is a box at the bottom where you enter your e-mail and password and then you can go to the list.  The list currently has 13 members from

ICES (2)
NIOZ, the Netherlands (2)
JCOMM-OPS (1)
USNODC (2)
DOD, Germany (1)
IMR, Norway (1)
BODC (3)
SISMER, France (1)

My requests for codes is because I'm having problems building the POGO infrastructure.  For example, consider G.O. SARS.  I am trying to link together a repository of CSR forms with the EurOcean research vessel database in which the primary entity is the hull specified by an IMO number.  In the CSR repository I have one code, but in EurOcean I have two.  OK, I could link one hull to G.A. SARS and the other to SARSEN but then the temporal metadata starts to look crazy - in one place it states that G.A. SARS was commissioned in 2003, whilst in another it states that she returned a dozen CSRs dated between 1994 and 2002.

The fundamental problem that we have - and that I am trying to solve - is down to the fact that we are trying to describe a complex entity with a simple code table rather than an accurate hierarchical data model (hull relates one-to-many to name, flag, governance etc.).  The solution I am trying to drive forward is to get around the problem by thinking of the code table as representing a relational join of the tables that make up the data mode, which isn't ideal but will work. 

By all means work to extend the caucus, the hull and multiple names issue is but one of many problems we need to resolve and we're going to need all the help we can get. However, we are under pressure to get a system running for POGO.  The bottom line is that I can hang around for a week or two for these codes, but if they don't come then I'll be forced to break ranks and start allocating codes locally (or more likely under SeaDataNet governance) and our attempt at  ship code reunification will have failed.

Cheers, Roy.

>>> Marilynn Sørensen <marilynn at ices.dk> 4/4/2007 2:29 pm >>>

Hi Roy and platform group,

We at ICES have been discussing the functioning of the platform email group and agree that a broad content governance group must be established to set rules before more codes are assigned. I am contacting other people and am preparing an email which you will all receive next week.

Could you send me a list of the current members (or tell me how to get this information). I have not yet read all the past platform emails, so forgive me if an email exists with this information.  

Best regards,
Marilynn

-----Original Message-----
From: platforms-bounces at biwebs1.nerc-liv.ac.uk [mailto:platforms-bounces at biwebs1.nerc-liv.ac.uk] On Behalf Of Roy Lowry
Sent: 04 April 2007 13:29
To: platforms at biwebs1-2.nerc-liv.ac.uk
Subject: [Platforms] G.O.Sars

Hi Marilynn,

We have what is becoming a very familiar situation for G.O. Sars. The original G.O. Sars was assigned code 58GS.This hull was renamed Sarsen in 2002 and given the code 58GT.  The new G.O. Sars then 'inherited' the code of 58GS.  This is exactly what happened with Tridens/Tridens I, which as far as I can see in Version 1.13 of the RECO codes we haven't resolved. I also can't remember seeing a new code as requested by Michele for BEAUTEMPS-BEAUPRE.

Unless our Norwegian list members object I suggest we follow the same practice for G.O. Sars as we did for Tridens and Pourquoi pas? (but not Meteor) and leave 58GS with the new ship and assign a new code for the old one: whatever we do, we're still going to have to fix our data to some extent.  So, could you please:

1) Assign new codes for the "old" G.O. SARS, BEAUTEMPS-BEAUPRE and TRIDENS

2) Follow Mitch and allocate 64T1 to TRIDENS I

If you need XML for any of these ask Fran.

Cheers, Roy.

_______________________________________________
Platforms mailing list
Platforms at mailman.nerc-liv.ac.uk
http://mailman.nerc-liv.ac.uk/mailman/listinfo/platforms 

****************************************************************************
Denne mail er blevet scannet af http://www.virus112.com
****************************************************************************

_______________________________________________
Platforms mailing list
Platforms at mailman.nerc-liv.ac.uk
http://mailman.nerc-liv.ac.uk/mailman/listinfo/platforms
****************************************************************************
Denne mail er blevet scannet af http://www.virus112.com
****************************************************************************