[oostethys_dev] Re: NDBC SOS server vs. your Perl SOS server?

[Jeremy Cothran]

Thanks for the updates and sharing, I haven't upgraded my perl output formatting scripts for either DIF or Oostethys style SOS since the beginning of 2009.  Glad to upgrade the scripts to support either or both where supporting the same relational database lookup/resultset approach where the output scripts are at a stable version point.

Jeremy

On Fri, May 21, 2010 at 1:31 PM, Eric Bridger <ebri...@gmri.org> wrote:

Emilio,

That would be excellent! Like I said, OOSTethsy is very much a grassroots effort so any additions are always welcome.
I'm hoping to be able to post some robust PERL SOS client code there as well as a result of the RCV project.
There is another ASP version of the OOSTethys Toolkit which I hope to post next week, developed by Felimon Gayanilo <FGay...@med.miami.edu> for GCOOS.

Eric

P.S. I added the oostethys_dev list to the CC. I know Luis will appreciate that.

On May 21, 2010, at 1:00 PM, Emilio Mayorga wrote:

> Hi Eric,
>
> Thanks for your extensive comments, and no problem with expanding the cc'd list.
>
> I should've clarified (to save you time in your reply) that I remember
> perfectly well that your update to the Perl SOS server last Nov.
> included the option to have DIF output; that's the major driver for my
> intent to upgrade to that version. But that it outputs to both DIF and
> SWE is another reason why I like that server.
>
> It sounds like you're committed and attached to this SOS server, then,
> and that NERACOOS and other groups are using it and have no plans to
> switch. That's good enough for me, so I'll stick with it.
>
> As I mentioned, my adaptation of the SOS server works on the Xenia
> schema (or my version of it...), and IMHO I've made some useful
> enhancements to its integration with Xenia. I plan to do a few more
> integration enhancements when I update to your latest code base over
> the next few weeks. I'd be happy to post my code to
> http://code.google.com/p/oostethys/  If prodded, I can also describe
> the main areas -- and reasons -- where my code diverged from the
> trunk.
>
> Thanks again. Cheers,
>
> -Emilio
>
>
> On Fri, May 21, 2010 at 9:21 AM, Eric Bridger <ebri...@gmri.org> wrote:
>> Hi Emilio,
>> (I added some interested parties to the CC I hope that's O.K.  This is a subject dear to my heart)
>>
>> Rick is correct.  At this point, the RCV is focused primarily on, what I call the DIF SOS version.  Mainly because both NDBC and COOPS are using it.  But my plan for the long term, in the RCV is to in fact support both SOS versions, both the DIF GML responses and the OOSTethys SWE responses.  I believe this is what Jeff DLB wants.  And that is what the OpenIOOS SOS map had always done.  At this point the main differences between the two are in the GetObservation responses.
>>
>> (I am keeping a list of issues I'm finding with the DIF response)
>>
>> Also, just to point out, the funding for the initial phase of the RCV is limited but the hope is that more funding will be available going forward to add enhancements.
>>
>> The RCV, of course is from an SOS Client perspective.  The OOSTethys project (which is also suffering from lack of funding but has always been more of a grass roots endeavor) is mainly about SOS Servers.  I don't this that it is going to go away. Luis and I have an NSF proposal submitted for OOSTethys improvements and hopefully that will get approval, but we'll see.
>>
>> So I guess the short answer re: the RCV, is that Regions which implement the DIF version will be easier to integrate in the short term, but that I will make every effort to also include  SWE type SOS's. SWE has many advantages.  Broader use internationally, more OGC compliant, etc.
>>
>> The OOSTethys PERL code at http://code.google.com/p/oostethys/  (see the downloads TAB)
>> (while not yet integrated with the Xenia schema, that would definitely be a major improvement) has continued to improve even without funding.
>>
>> 1) The most recent version of the PERL code (1.0.1) in fact includes support for both DIF and SWE outputs.  This relies on the ability of the SWE version to support requests like:
>> OBSERVEDPROPERTY=wind_speed,wind_direction
>> There is a configuration setting up top:
>> $use_DIF = 1;  to get the DIF output.  There are also settings for using DIF names or CF names, etc.
>> NERACOOS is using this PERL code:
>> SWE:
>> http://www.neracoos.org/cgi-bin/sos/V1.0/oostethys_sos.cgi?REQUEST=GetCapabilities&service=SOS
>> and
>> DIF:
>> http://www.neracoos.org/cgi-bin/sos/V1.0/oostethys_sos_dif.cgi?REQUEST=GetCapabilities&service=SOS
>>
>> 2)  Last winter I spent quite a bit of time ensuring that the PERL OOSTethys code could pass the OGC's CITE testing: http://cite.opengeospatial.org/
>> There is a not yet posted version which will pass this testing (for the SWE output) but I have not yet posted it to Google Code.
>> I'm reluctant because I found that it was very nit-picky and no forgiving.
>> E.g a lot of the tests where checks for required parameters, e.g. service=SOS in every request and then whether you had the proper Exception when the tester sends service=SOSX
>> Kind of silly.  OOSTethys had taken the approach of being completely forgiving for any parameters which were non essential
>>
>> Anyway probably way too much information.
>>
>> My personal opinion, and what I'm going to use for NERACOOS is the OOSTethys PERL code with the DIF setting.
>> But I will probably include SWE GetObservation support in the RCV.
>>
>> Eric
>>
>>
>> On May 21, 2010, at 2:12 AM, Emilio Mayorga wrote:
>>
>>> Hi Eric,
>>>
>>> Rick Blair tells me that as part of your IOOS RCV project, you're
>>> working extensively with the NDBC SOS server (PHP-based?), and for
>>> that and other reasons that SOS server is gaining a lot of momentum
>>> among the RA's. I've been delaying the update of my implementation of
>>> your Perl SOS server (at NANOOS-APL, on top of my adaptation of
>>> Jeremy's Xenia on PostgreSQL;
>>> http://code.google.com/p/xenia/wiki/NANOOSXenia) to your Nov. '09
>>> release, to which I have to add my own modifications. But I need to do
>>> this very soon and make this SOS service operational once and for all.
>>>
>>> So, my question to you is, what are your plans/thoughts for your Perl
>>> SOS server? Are you planning to abandon it in favor of the NDBC SOS
>>> server? I'm cc'ing Jeremy b/c when I adopted that code last Summer,
>>> SECOORA was using it, too; feel free to chime in, Jeremy. Before
>>> investing more time on that code base, I wanted to get a feel for its
>>> future and whether it's best to just ditch it now and follow the herd
>>> with the NDBC code. Personally, I like yours and had no problems with
>>> it.
>>>
>>> Thanks! Good luck with your IOOS RCV work; I look forward to seeing it
>>> in action. Cheers, -Emilio
>>>
>>> -Emilio Mayorga  |   NANOOS
>>> Applied Physics Laboratory
>>> University of Washington
>>> Seattle, WA 98105-6698  USA
>>> Tel: 1-206-543-6431
>>> may...@apl.washington.edu
>>> http://staff.washington.edu/emiliom/
>>

--
You received this message because you are subscribed to the Google Groups "oostethys_dev" group.
To post to this group, send email to oostet...@googlegroups.com.
To unsubscribe from this group, send email to oostethys_de...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/oostethys_dev?hl=en.

[Eric Bridger]

An interesting thread I wanted to share with the OOSTethys Community:
======================================================================