Re: SRU [was: Re: WMO Information System to make metadata available via GEOSS]

[TOYODA Eizi]

Hi Bruce,

 

I'm afraid I don't recall the conversation, but I'll try analysing the situation.

 

As a matter of regulation, the Manual on WIS states

"4.9.1 This specification requires that each
metadata catalogue host shall support the Search and Retrieve via URL (SRU) specification of the
ISO 23950 Information Search and Retrieval Protocol."

The word "shall" means mandatory.  The manual also mandates some indices of SRU.  You might know SRU has no technical relation with ISO 23950 at all (please check specs if you doubt), but the manual text is a compromise between those who needed clear meaning and those who wanted some feeling of authority or neutrality (I don't know).  Anyway SRU is a nice open standard, and I love to have open interface for sense of transparency.

 

Regarding value, however, there is no answer yet to the question "why it is necessary, or even useful, for each WIS catalogue to have SRU interface?"  Standardization of the search interface is meaningful if we need a parallel (distributed) search to different catalogues, which are not present in current WIS since people really pursued global syndication of metadata catalogues.  In simpler words, "Why we don't just let people search at a GISC (instead of DCPCs) as it has entire set of WIS metadata?"  Or now we can say "Why we don't let people search at the GEO Portal as the WIS metadata is a part of GEOSS metadata?"  In short, the manual is still in the evolving process.

 

Please note I'm not trying to abolish anything (Dave, don't get mad please).  The two approaches (distributed search by SRU and syndication by OAI-PMH) are both invested in the library science community.  It might also be useful if a couple of GISCs provides mutually backed-up service of SRU to other parties if machine-readable search gets more popular.  Nothing wrong in research.

 

But for many people right now, more pressing agenda exist somewhere else, for example launch of WIS centres or successful migration from existing catalogues like Volume C1....  I think "not pursuing" is something like that.

 

Hope my thoughts are useful.

 

Eizi

----- Original Message -----

From: Bruce Bannerman

To: wmo-i...@googlegroups.com

Sent: Monday, July 30, 2012 7:13 AM

Subject: SRU [was: Re: WMO Information System to make metadata available via GEOSS] [SEC=UNCLASSIFIED]

Jeremy,

I recall a discussion at our Exeter IPET-MDI meeting that we were not pursuing SRU.

Is this no longer the case? Perhaps I misunderstood the conversation.

Bruce



On 28/07/12 1:22 AM, "David Thomas" <dth...@wmo.int> wrote:

Hi Jacques
I have added your SRU server to the SRU training page today. It should be available soon. See  http://www.wmo.int/pages/prog/www/WIS/wissruform.html 

And look for Toulouse
thanks
Dave

On 25 July 2012 18:28, Jacques.Roumilhac <jacques....@meteo.fr> wrote:

Hi Dave,

Toulouse and Exeter obviously support SRU.
Your url is the good one but you need to add <portal.sru> to plug to the sru server
For Toulouse: http://wispi.meteo.fr/openwis-user-portal/srv/en/main.home/portal.sru <http://wispi.meteo.fr/openwis-user-portal/srv/en/main.home/portal.sru>
An example of sru question on the GISC Toulouse OpenWIS SRU server (question about La Reunion Island): http://wispi.meteo.fr/openwis-user-portal/srv/en/main.home/portal.sru?operation=searchRetrieve&version=1.1&query=%22reunion%22&startRecord=1&maximumRecords=10&& <http://wispi.meteo.fr/openwis-user-portal/srv/en/main.home/portal.sru?operation=searchRetrieve&version=1.1&query=%22reunion%22&startRecord=1&maximumRecords=10&&>

Best regards
Jacques

David Thomas a écrit :

Hi Eizi,
Good point and fair question.
GEOSS is harvesting from Offenbach. Timo is working with GEO to establish a backup GISC.
You will note that the news release highlights that this is available both ways (eg find GEOSS from WIS too) as required by congress quoted in the article. Interestingly, I have had trouble finding the GEOSS connection from the GISCs. So far I can see the following do from the SRU search

DWD - http://gisc.dwd.de/SRU2JDBC/

but the following don't (or at least not obvious to me)
CMA http://wisportal.cma.gov.cn/wis/jsp/srusearch/sru.jsp <http://wisportal.cma.gov.cn/wis/jsp/srusearch/sru.jsp>
JMA http://www.wis-jma.go.jp/meta/sru.jsp?version=1.1 <http://www.wis-jma.go.jp/meta/sru.jsp?version=1.1>
Toulouse http://wispi.meteo.fr/openwis-user-portal/srv/en/main.home <http://wispi.meteo.fr/openwis-user-portal/srv/en/main.home>
Exeter http://wis.metoffice.gov.uk/openwis-user-portal/srv/en/main.home <http://wis.metoffice.gov.uk/openwis-user-portal/srv/en/main.home>

I am sure this requirement is probably met by all GISCs but it is not easy to find in any. I am not even sure how Toulouse and Exeter support SRU. I suspect it is hidden under the user interface, but I have not fully explored both sites at this time. It would be good if each GISC explained how to search GEOSS (and other remote SRU servers) from their site so we can put it on the main GISC page www.wmo.int/giscs <http://www.wmo.int/giscs>  <http://www.wmo.int/giscs>. Normaly I use the SRU training page when showing this facility, then use DWD as the example GISC.
regards
Dave


On 12 July 2012 18:28, Eizi <toyod...@gmail.com <mailto:toyod...@gmail.com>> wrote:

    Hi Dave,

    WMO issues news release that WIS Discovery metadata is available
    at GEOSS.  Congratulations, first of all.  That should be a great
    step.

    http://www.wmo.int/pages/mediacentre/news/index_en.html <http://www.wmo.int/pages/mediacentre/news/index_en.html>
    https://twitter.com/WMOnews/statuses/223425025963868161 <https://twitter.com/WMOnews/statuses/223425025963868161>

    Which gisc works as gateway?  I'm more than simply curious.  It
    should be good to know that for stable operations I think.

    Best,
    Eizi

    ------------------------------------------------------------------------

    The information contained in this electronic message and any
    attachments are intended for specific individuals or entities, and
    may be confidential, proprietary or privileged. If you are not the
    intended recipient, please notify the sender immediately, delete
    this message and do not disclose,  distribute or copy it to any
    third party or otherwise use this message. The content of this
    message does not necessarily reflect the official position of the
    World Meteorological Organization (WMO) unless specifically
    stated. Electronic messages are not secure or error free and may
    contain viruses or may be delayed, and the sender is not liable
    for any of these occurrences.
    ------------------------------------------------------------------------

    SAVE PAPER - Please do not print this e-mail unless absolutely
    necessary