[Foss4g2009] Re: Geospatial Integration Showcase - meeting notes
[SEC=UNCLASSIFIED]
shoaib
saburq at gmail.com
Mon Nov 3 08:14:00 EST 2008
IMO: this is how I see the integration showcase
1. Come up with a scenario (involving distributed data sources and
decision makers)
2. Collect the data
3. Host and expose the data as per the scenario (via SOA - OGC)
4. Test the scenario (repeat and refine until perfected)
If we are to be able to make this I expect we should have step 4 done
by April 2009
I don't expect the servers and hosts to stay online into the future -
due to the no of players and cost of on going hosting.
Shoaib Burq
On Thu, Oct 30, 2008 at 10:05 PM, Cameron Shorter
<cameron.shorter at gmail.com> wrote:
> Bruce, Simon, you have valid concerns about minimising OUR scope, and indeed
> I believe this is essential for success.
> The showcase, like open source projects will be successful only if lots of
> people do a little work.
>
> Lots of people contribute to an Open Source project like Openlayers because
> 95% of the functionality exists for my requirements, I only need to add
> another 5%. And after many people contribute 5%, the project becomes
> effective. The key is that an initial investment needs to be made so that
> others can make small additions and get a lot of value.
>
> So base requirement we need to produce for the showcase is:
>
> 1. One server with multiple virtual machines which OTHER people can add
> their products into.
> 2. A marketing pipeline - in the form of conferences (which the OGC & OSGeo
> should be able to provide at no extra cost).
> 3. Suggestions for a scenario, which are used by OTHERS when building their
> presentations.
>
> Base 0 functionality will contain:
> 1. PostGIS database
> 2. Some other database
> 3. Mapserver WMS/WFS
> 4. Geoserver WMS/WFS
> 5. Some other WMS/WFS
>
> I expect that software developer communities will set up these services for
> us.
>
> That is it.
>
> Other vendors can then invest their own time to add extra functionality.
>
>
> mapbutcher wrote:
>>
>> Folks,
>>
>> I have a couple of concerns:
>>
>> 1. Within the context of the conference next year we have a small amount
>> of people at this stage and a growing set of tasks to complete. From
>> comments made by Gavin, Paul, Jeff, Dave etc to this list and from my
>> conversations with others who are regularly involved with running
>> conferences - it will be a significant amount of work - which we shouldn't
>> under-estimate. I don't want to rain on this as I agree with Cameron - a
>> successful integration showcase is a draw card, but given our very tight
>> resources is it going to be realistically achievable?
>> 2. What is it we are actually talking about - I'm keen to avoid
>> discussions about servers/platforms/who owns what etc etc until we are clear
>> on what we are proposing
>>
>> So I would like to propose that we direct discussions in the direction of
>> scope - this will hopefully provide us all with a common understanding of
>> what we are trying to achieve, and it provides us with a clear definition
>> which we can take out to a wider community to canvas interest.
>>
>> The GeoSpatial Integration Showcase might be an event which is
>>
>> :supported by permanent or temporary infrastructure?
>> :supported by an environment which can host multiple operating
>> systems\configurations?
>> :is open to open source projects and proprietary vendors?
>> :can support some kind of benchmarking of system to system performance?
>> :promotes the use of OGC standards as a means to integrate technology?
>> :demonstrates a purpose which is beyond pure demonstration of component
>> integration?
>> :has a common theme which can be conveyed to all participants?
>> :has a common set of data which can be shared by all participants?
>>
>> One approach is to focus on developing a framework which is transferable
>> and continues to be attractive to participants. Where participants can
>> integrate on a level playing field, with a known set of constants
>> (infrastructure, environment, data, scenario)? What are the outputs of this
>> repeatable event? Perhaps the big picture is a project which supports the
>> continuing development of these showcase events, with a website supporting
>> the publication of the annual theme/scenario and data, where the *rules of
>> engagement* are clear for all participants to see, where interested parties
>> can find out information on the purpose of the annual integration showcase
>> and see the results of previous years event? A facility which provides
>> organisations with a set of use cases and demonstrated technology solutions
>> - this is where i think the real value is.
>>
>> On a practical level what might this involve?
>>
>> At FOSS4G 2009 we aim to prove the concept. We do not invest on a
>> permanent basis but rather lay out a simple set of principles and aim to get
>> a number of committed participants (I think even this is ambitious given our
>> time frames, commitments etc). We aim primarily at proving and understanding
>> the process so we can pass it on, the results should change from event to
>> event?
>>
>> My feeling is that these larger questions need to be clear before we go
>> and think about formalising things?
>>
>> Simon
>>
>>
>>
>>
>> 2008/10/30 Cameron Shorter <cameron.shorter at gmail.com
>> <mailto:cameron.shorter at gmail.com>>
>>
>> Read from bottom to top: discussion between OGC and OSGeo about a
>> Geospatial Integration Showcase:
>>
>> I'm +1 for server installed with Virtual Machines as suggested by
>> Raj and Bruce.
>>
>> In answer to Raj's question, I believe the OGC should be custodian
>> of the Showcase VM/Server. To attract vendors, the Showcase should
>> not be locked into one vendor, or open source solution.
>>
>> CCed to the foss4g list with permission from Bruce, Raj and myself.
>>
>> Greg, it would be good to move over to an OGC email list soon if
>> you are able to provide that infrastructure.
>>
>>
>> Bruce Bannerman wrote:
>>
>> Cameron,
>>
>> We still need a better definition of what this 'thing' is that
>> we are
>> trying to put together.
>>
>> You have provided a good start with your example, however I
>> find it
>> difficult to believe that all vendors would like their
>> products mastered
>> on an Open Source DVD.
>>
>> Ideally, I'd like to see this run, managed and maintained by
>> the OGC to
>> ensure that the perception of a Level Playing Field is
>> maintained. We
>> can and should help to drive 'it'.
>>
>>
>> For another suggestion as to what 'it' might be:
>>
>> I agree with the need for a scenario driven approach, however this
>> should be portable. We'll want to allow additional scenarios
>> in the
>> future if this 'thing' is to be portable.
>>
>>
>> We could look at sourcing a high performance server and run
>> something
>> like VMWare ESX Server as the host platform. OGC would be
>> responsible
>> for administering the ESX server.
>>
>>
>> We could then offer each vendor the opportunity to configure a
>> virtual
>> server using their Operating System of choice, configured and
>> tuned to
>> ensure optimal performance for their products. The vendor would be
>> responsible for maintaining 'their' virtual server.
>>
>>
>> During the event, ESX Admin consoles would need to be made
>> public to
>> ensure that all participants to see that all virtual servers were
>> allocated equal resources.
>>
>>
>> This would then give us a level playing field with which to
>> compare
>> product performance on the same set of data.
>>
>>
>> This is another take on an 'Integration Showcase'.
>>
>>
>> Bruce
>>
>> --
>> Bruce Bannerman
>> Supervisor, Climate Centre Computing Support
>> National Climate Centre
>> Bureau of Meteorology
>> 700 Collins Street, Melbourne, Vic 3001
>> Australia
>> Tel: +61 3 9669 4093
>>
>>
>>
>>
>> -----Original Message-----
>> From: Raj Singh [mailto:rsingh at opengeospatial.org
>> <mailto:rsingh at opengeospatial.org>] Sent: Thursday, 30
>> October 2008 8:04 AM
>> To: Cameron Shorter
>> Cc: Greg Buehler; Simon Hope; Bruce Bannerman; Jeff McKenna
>> Subject: Re: Geospatial Integration Showcase - meeting notes
>>
>> You talk about creating a server that moves around to
>> different venues. I like this idea as it would do a few
>> things:
>> - deployable on a LAN, avoiding exploitative Internet
>> access costs at conferences
>> - take the speed vagaries of the general Internet out of
>> the equation when examining performance
>> - easier to tweak and administer as it wouldn't be behind
>> a firewall or subject to some rigid sys admin
>>
>> However, who would own it or at least be the caretaker?
>> OGF? OGC? Would it be passed down to FOSS4G conference
>> committees?
>> ---
>> Raj
>>
>>
>> On Oct 29, at 3:57 AM, Cameron Shorter wrote:
>>
>>
>> Raj, Greg, Simon, Bruce, Jeff,
>> I'd like to CC this email to the foss4g email list for
>> transparency and archiving. Could you please confirm
>> to me that it is ok.
>> I have ACTION items listed below.
>>
>> =Geospatial Integration Showcase Teleconference=
>>
>> 29 October 2008
>>
>>
>>
>>
>> http://www.timeanddate.com/worldclock/fixedtime.html?month=10&day=29&y
>>
>> <http://www.timeanddate.com/worldclock/fixedtime.html?month=10&day=29&y>
>>
>> ear=2008&hour=8&min=0&sec=0&p1=240
>>
>> Next meeting in 2 weeks:
>> Tu/We 11 or 12 November 2008. 5pm in New York. 8am in
>> Sydney.
>>
>>
>>
>> http://www.timeanddate.com/worldclock/fixedtime.html?month=11&day=12&y
>>
>> <http://www.timeanddate.com/worldclock/fixedtime.html?month=11&day=12&y>
>>
>> ear=2008&hour=8&min=0&sec=0&p1=240
>>
>> Present:
>> * Greg Buehler OGC,
>> * Raj Singh OGC,
>> * Cameron Shorter LISAsoft, FOSS4G Australia,
>> * Bruce Bannerman Bureau of Meteorology, FOSS4G Australia,
>> * Simon Hope, ESRI Australia, FOSS4G Australia,
>> * Jeff McKenna, OSGeo Conference/FOSS4G Committees
>>
>> Agenda:
>> * Define Integration Showcase
>> * Identify action items to achieve showcase
>>
>> Wiki:
>> http://wiki.osgeo.org/wiki/Geospatial_Integration_Showcase
>>
>> ==Discussion==
>> Cameron's definition of showcase:
>> # The Geospatial Integration Showcase provides an
>> easily
>> deployable,
>> practical demonstration of standards based
>> interoperability between geospatial applications.
>> # Easily redeployed between conferences # Install
>> software on one server with numerous virtual machines.
>> # Client software on another computer
>> # Open Source LiveDVD will provide core functionality
>>
>> ==Discussion==
>> Over email for the next 2 weeks we need to define the
>> Integration Showcase. Elements that we need to define are:
>>
>> === Scenario ===
>> * What scenario should we use?
>> * Scenario needs to satisfy sponsors
>> * Should be Australian based in order to interest
>> local FOSS4G attendees
>> * Noted that Australia is interested in Climate Change
>> and
>> water (lack
>> of it)
>> * Noted that GEOSS have scenarios (from FOSS4G2008?)
>> that may be applicable.
>> ** ACTION: Greg/Raj to source GEOSS scenarios
>> * CSIRO/BOM have mandate and budget to address Climate
>>
>> change and will
>> likely be interested.
>>
>> === Data ===
>> * Need to source data suppliers
>> * Bruce doesn't see a problem sourcing data from
>> Australian
>> government
>> agencies
>>
>> === Marketing/Sponsorship ===
>> * OGC to use their channels to attract vendors and
>> sponsors
>> * OGC to add name and take a coordinating role so that
>> the
>> showcase is
>> seen as a Standards based showcase, not an Open Source
>> showcase.
>> This is important for attracting proprietary vendors.
>> * OGC likely will be able to provide OGC twiki for
>> collaboration, including twiki, issue tracker, email
>> lists. This is better
>> provided
>> by OGC than OSGeo in order to attract proprietary vendors.
>> * Marketing material should be created, which will
>> likely
>> include some
>> of:
>> ** Recorded webinar presentation
>> ** Professionally created movie including screen
>> captures, anchor voice over, related images, etc
>> * ACTION: Greg to confirm what OGC can provide and
>> costs (if any) associated with OGC provided services.
>>
>> === Targeting Vendors ===
>> * Need to provide attractive value proposition for vendors
>> * Criteria for vendors is that they are Open
>> Standards. Proprietary vendors need to be attracted as
>> well as Open Source. This has been iterated on OSGeo
>> email lists.
>>
>> === Targeting Geospatial Buyers ===
>> * Testbed needs to put capabilities in front of buyers
>>
>> === Collaborators on board ===
>> * Europeans are discussing a "Persistent testbed"
>> which is very similar. We need to engage these
>> participants.
>> * ACTION: Greg/Raj to contact and invite collaboration
>> with
>> Persistent
>> testbed stakeholders.
>>
>> === Infrastructure ===
>> * Still needs to be defined
>> * Potentially a server with multiple virtual machines
>> in a
>> conference,
>> with participants connecting via their clients around
>> the centre
>>
>>
>>
>> --
>> Cameron Shorter
>> Geospatial Systems Architect
>> Tel: +61 (0)2 8570 5050
>> Mob: +61 (0)419 142 254
>>
>> Think Globally, Fix Locally
>> Geospatial Solutions enhanced with Open Standards and
>> Open Source http://www.lisasoft.com
>>
>>
>>
>>
>>
>> -- Cameron Shorter
>> Geospatial Systems Architect
>> Tel: +61 (0)2 8570 5050
>> Mob: +61 (0)419 142 254
>>
>> Think Globally, Fix Locally
>> Geospatial Solutions enhanced with Open Standards and Open Source
>> http://www.lisasoft.com
>>
>>
>
>
> --
> Cameron Shorter
> Geospatial Systems Architect
> Tel: +61 (0)2 8570 5050
> Mob: +61 (0)419 142 254
>
> Think Globally, Fix Locally
> Geospatial Solutions enhanced with Open Standards and Open Source
> http://www.lisasoft.com
>
> _______________________________________________
> Foss4g2009 mailing list
> Foss4g2009 at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/foss4g2009
>
More information about the Foss4g2009
mailing list