Hi Jacek,

Unfortunately, I will be in the SLAC Director's Review of LSST next
Wednesday all day.  So, I will provide my thinking here as much as possible.

I take exception to building the schema entirely outside of EA, if in fact
these schemas are anything more than throw-away "one offs".  I am assuming
there is enough in these schemas so that is not the case.

One of the goals of consolidating the schemas in EA is to create a
"baseline" schema that is as much as possible NOT dependent on a particular
database, since we really don't know what database we will be using in 2012,
nor do we believe that even that database will be the same in 2022.

I think we should definitely import the schemas (as Jacek did, I believe,
with Sergei's initial schema), generalize them to "basic SQL", and
understand/annotate the areas that will need to be post-processed to a
db-specific implementation.  Since this schema will not immediately be able
to generate db-specific SQL, we can create a separate "realization" via a
static link to a db-specific SQL file in the short term.

Ultimately, the implementation of the "generic" schema will require some
post-processing to map to a particular database; that's what code generators
and post-processing tools are good at.  I believe the EA code generating
module is fairly customizable for this purpose, and we should over time
learn to customize and use it.

Thanks!

Jeff

> From: Jacek Becla <[EMAIL PROTECTED]>
> Organization: Stanford Linear Accelerator Center
> Reply-To: LSST Data Management <[email protected]>
> Date: Wed, 01 Mar 2006 13:04:42 -0800
> To: LSST Data Management <[email protected]>
> Subject: Re: [LSST-data] Re: LSST Database telecon Wed Mar 01
> 
> Jeff
> 
> A lot of what we discussed today at the Database telecon
> (see notes I just sent) is highly relevant to what you just said.
> How about if we come back to these issues at the next
> Database telecon (note special time: next week),
> assuming you can call in, so that we make sure we all agree..
> 
> Jacek
> 
> 
> 
> 
> 
> Jeffrey P Kantor wrote:
>> Hello all,
>> 
>> We have some existing policies that we should be adhering to:
>> 
>> Meeting notes, minutes, etc. should be migrated to lsst-data, not sitting in
>> the plone. This was a condition of using the plone and we need to work on
>> this.
>> 
>> Schemas should be in Enterprise Architect whenever possible, in docushare
>> otherwise.  I understand that UW has some issues with the SQL generated by
>> EA, so if an exception needs to be made, at least there should be a link to
>> the schema within Docushare.  There is a collection in Data
>> Management:Hardware & Software Deliverables.  You can create collections for
>> protoypes in here (as was done for the WCS tool).  It is possible to post a
>> URL as a document link.  So if you want to point to a cvs location that is
>> fine, as long as the location is accessible.  Otherwise, upload the files.
>> (With regard to the EA SQL issues, we need them documented so that we can
>> forward them on to EA support.)
>> 
>> Now that we have docushare, we need to start moving to retire the plone.  We
>> are too fragmented in our information, and while our tools may not be
>> perfect, they are generally serviceable and we all need to use them.
>> 
>> Thanks!
>> 
>> Jeff
>> 
>> 
>>> From: "Russell E. Owen" <[EMAIL PROTECTED]>
>>> Reply-To: LSST Data Management <[email protected]>
>>> Date: Wed, 01 Mar 2006 10:52:54 -0800
>>> To: [email protected]
>>> Subject: [LSST-data] Re: LSST Database telecon Wed Mar 01
>>> 
>>> In article <[EMAIL PROTECTED]>,
>>> Tim Axelrod <[EMAIL PROTECTED]> wrote:
>>> 
>>> 
>>>> Hi Jacek,
>>>> 
>>>> Can you point me to info on these two schemas?   I'm not sure which, if
>>>> either, corresponds to "PrecursorSchema" in the UML model.
>>> 
>>> The UW schema documentation is available officially here:
>>> <http://www.astro.washington.edu/becker/LSST/local/DBDocs/UWDatabase.html
>>> 
>>> However, this needs a password (which has been given out to a lot of
>>> folks, but we also hope to lift that restriction soon). Meanwhile, I
>>> have a temporary unprotected copy here:
>>> <http://www.astro.washington.edu/rowen/DBDocs/UWDatabase.html>
>>> 
>>> Although much of this schema may be relevant to LSST, please keep in
>>> mind that we designed it primarily for our use in prototyping algorithms.
>>> 
>>> Regards,
>>> 
>>> -- Russell
>>> 
>>> _______________________________________________
>>> LSST-data mailing list
>>> [email protected]
>>> http://www.lsstmail.org/mailman/listinfo/lsst-data
>>> 
>> 
>> 
>> _______________________________________________
>> LSST-data mailing list
>> [email protected]
>> http://www.lsstmail.org/mailman/listinfo/lsst-data
> 
> _______________________________________________
> LSST-data mailing list
> [email protected]
> http://www.lsstmail.org/mailman/listinfo/lsst-data
> 

_______________________________________________
LSST-data mailing list
[email protected]
http://www.lsstmail.org/mailman/listinfo/lsst-data

Reply via email to