[Board] OGC Relationship

Sat Jan 6 03:59:04 PST 2007

I just woke up fretting my comments might have come across as too extreme, and I see indeed Gary saw them as such.  So let me try a different tack-

What I am trying to do is show that a "commitment" to OGC puts us on an ill-defined, slippery slope with respect to the future of our organization.  I am seeing terms like "pseudo-standard" and "offical standard", and I'm not sure these are well-defined at this point.  Consider the recent WMS-T work, the GeoRSS work (older, prior to OGC's involvement), and the GeoTIFF work (years ago now).  All three of these could very likely have been activities of the OSGeo community if begun in 2007 -- would such work have been acceptable in terms of the proposed commitment to OGC?

I do not want OSGeo to be an offical standards body set up like an ISO or an IETF, certainly.  But I would not want to get in the way of the next GeoRSS/WMS-T/GeoTIFF/whathaveyou, and indeed I'd like to see OSGeo play a big part in making it happen.  And if we could later get it "published" with OGC's imprint, so much the better for all parties in the GIS arena.

So, continuing on-

OGC does spec development very differently from the way OSGeo does/would, and so I'm not sure the idea of handing off a draft spec for OGC to approve makes sense [*under current, normal OGC processes*].  Would that entail having OGC's members and WGs be able to modify and vote to approve work that OSGeo's members and WGs already did?  Taking any degree of control out of the hands of the free-to-all-comers-and-lurkers OSGeo membership feels wrong to me.  [I pause to note that Raj's heart is in the right place here, and the MassMarketWG might become the exception to normal OGC processes.  But we don't know how that works yet.  Hence current discussion.]

OGC regularly passes specs upwards to ISO for what seems like thumbs-up/down, almost pro-forma approval.  I'm not sure how that process actually works, but might that be a model worth discussing?

-mpg

________________________________

From: Gary Lang [mailto:gary.lang at autodesk.com]
Sent: Fri 1/5/2007 9:45 PM
To: cholmes at openplans.org; Michael P. Gerlek
Cc: Frank Warmerdam (External); board at lists.osgeo.org
Subject: Re: [Board] OGC Relationship

Exactly.

--------------------------
Sent from my BlackBerry Wireless Device

----- Original Message -----
From: Chris Holmes <cholmes at openplans.org>
To: Michael P. Gerlek <mpg at lizardtech.com>
Cc: Gary Lang; Frank Warmerdam (External); board at lists.osgeo.org <board at lists.osgeo.org>
Sent: Fri Jan 05 09:26:58 2007
Subject: Re: [Board] OGC Relationship

Oh I'm against basically any level of trying to prevent our members from
doing work that might be considered 'standards work'.  I thought that
saying 'we're not in the standards business' would basically mean that
we wouldn't say that anything our members collaborate on is an 'official
standard'.  Just that it's an incubating space for things that may turn
in to real standards, for example when they go to the OGC.  In my
opinion we can do all these activities we speak of, and even let them
become pseudo-standards.  But we as an organization are not in the
business of promoting the standards or trying to get anyone to use them.
  If we want to do that, then we pass along to the OGC.

I was assuming that's about all they wanted out of us?

When I talked to Raj awhile ago the thing he mentioned as by far having
the most value to OGC would be for OSGeo members to actually collaborate
on creating standards.  And I'm pretty sure he's also fine with us
starting them in a 'safer' space like our wiki, and passing them on to
general OGC when we're happier with them.

Chris

Michael P. Gerlek wrote:
> No, our charter doesn't say we're in the standards business -- but
> nonetheless, I see a lot of standards-like discussions going on via
> OSGeo channels.
>
> I think it would be a mistake to provide a commitment to OGC that we'd
> not do standards work, for two reasons.
>
> First, if we tell our members that they can't do standards within OSGeo
> channels, then they'll just route around us in classic internet fashion
> and find some other venue.  And so if we really want to serve our
> community's needs and interests, why would we force them to go somewhere
> else?
>
> Second, who's to say what "developing and supporting our own standards"
> means?  Discussions on a mailing list about possible extensions to WMS?
> Collaboratively writing up a WCS change proposal on our wiki to submit
> to OGC?  Implementing a prototype of something in MapServer?  How could
> you possibly enforce this?  Who's gonna be the one to say "Sorry, guys,
> but you can't use the osgeo IRC channel to talk about testing WMS-T
> because WMS is an OGC spec" or "Folks, could you please stop talking
> about possible alternatives to ebRIM on the geodata mailing list,
> because we don't want to look like we're contravening the OGC spec"?
>
> I'm all in favor of working in with OGC, but not to the point of making
> foundation-level commitments that go against our members' obvious needs
> and wants.
>
> -mpg
>
> 
>
>> -----Original Message-----
>> From: board-bounces at lists.osgeo.org
>> [mailto:board-bounces at lists.osgeo.org] On Behalf Of Gary Lang
>> Sent: Friday, January 05, 2007 11:01 AM
>> To: Frank Warmerdam (External); board at lists.osgeo.org
>> Subject: RE: [Board] OGC Relationship
>>
>> "I'm not sure what we would offer OGC in return.  I suspect ultimately
>> what would be most valuable to them is some sort of commitment to not
>> become a a "standards development" organization.  This avoids
>> duplication, confusion in the marketplace, and what they
>> might consider
>> competition."
>>
>> >From talking with David, I think is exactly what we would usefully
>> offer, and for two reasons:
>>
>> a) we're not in the standards business. It's not in our top
>> priorities,
>> last I looked
>> b) exactly what you said - "avoids duplication, confusion in the
>> marketplace"
>>
>> If a) is true, then providing comfort and clarity around b) seems a
>> no-brainer.
>>
>> Gary
>>
>> -----Original Message-----
>> From: board-bounces at lists.osgeo.org
>> [mailto:board-bounces at lists.osgeo.org] On Behalf Of Frank Warmerdam
>> (External)
>> Sent: Friday, January 05, 2007 11:03 AM
>> To: board at lists.osgeo.org
>> Subject: [Board] OGC Relationship
>>
>> Folks,
>>
>> I'm taking the liberty of moving (at least part) of this discussion to
>> the public board list.
>>
>> I am favorable on the idea of a formal liason relationship with OGC
>> though I don't consider it particularly critical to us or them since
>> there is already extensive cross membership and cross pollination.
>>
>> If we are to have a formal liason relationship, one benefit I
>> would like
>> to see is the ability for us give some developers access to
>> working OGC
>> documents, and for those developers to be involved in OGC
>> testbeds, and
>> working groups as OSGeo representatives.  I believe the OGC portal
>> allows members to setup accounts for individuals to access the portal.
>> We could manage a list of developers-with-access via this mechanism,
>> with the understanding that we would never have more than some fixed
>> number of developers (or users really) so authorized.  I think even
>> doing this for 5-10 would be plenty since most OSGeo project
>> folks with
>> an interest in OGC work already have access through corporate
>> memberships.
>>
>> I'm not sure what we would offer OGC in return.  I suspect ultimately
>> what would be most valuable to them is some sort of commitment to not
>> become a a "standards development" organization.  This avoids
>> duplication, confusion in the marketplace, and what they
>> might consider
>> competition.
>>
>> I'd be agreeable with this, but it must be understood that
>> self-organizing working groups within and between OSGeo projects are
>> likely to develop specifications such as GeoRSS, or the web tile
>> specification whether we encourage it or not, and I don't
>> want to be in
>> the position of discouraging
>> that.  So we must be careful that such activities are not precluded.
>> At
>> most I think the board could offer to not develop and support our own
>> standards development process - understanding that we won't supress it
>> either.
>>
>> The other angle might be some sort of more active involvement of OSGeo
>> projects in OGC testbeds and other IE efforts.  However, it
>> is hard for
>> us to force project involvement.  It might be appropriate for the
>> foundation to provide some modest supporting funding for project
>> involvement in OGC testbeds and interoperability experiments.  For
>> instance, providing travel funding.
>>
>> Note that there are at least a few people who would like to see OSGeo
>> become a sort of light weight agile standards development
>> organization.
>> I'm not keen on that, but it might be prudent to give these folks a
>> chance to make their case.
>>
>> Best regards,
>> --
>> ---------------------------------------+----------------------
>> ----------
>> ---------------------------------------+------
>> I set the clouds in motion - turn up   | Frank Warmerdam,
>> warmerdam at pobox.com
>> light and sound - activate the windows | http://pobox.com/~warmerdam
>> and watch the world go round - Rush    | President OSGeo,
>> http://osgeo.org <http://osgeo.org/> 
>>
>> _______________________________________________
>> Board mailing list
>> Board at lists.osgeo.org
>> http://lists.osgeo.org/mailman/listinfo/board
>>
>> _______________________________________________
>> Board mailing list
>> Board at lists.osgeo.org
>> http://lists.osgeo.org/mailman/listinfo/board
>>
> _______________________________________________
> Board mailing list
> Board at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/board
>
> !DSPAM:1003,459f1490125821362196140!
>

--
Chris Holmes
The Open Planning Project
http://topp.openplans.org <http://topp.openplans.org/> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/board/attachments/20070106/2211d533/attachment.htm>