OASIS Web Services Interactive Applications TC

 View Only

RE: [wsrp] [wsia] [wsrp-wsia joint interfaces] Search results page,map services, transient entities.

  • 1.  RE: [wsrp] [wsia] [wsrp-wsia joint interfaces] Search results page,map services, transient entities.

    Posted 06-12-2002 14:10
     MHonArc v2.5.2 -->
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    
    

    wsia message

    [Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]


    Subject: RE: [wsrp] [wsia] [wsrp-wsia joint interfaces] Search results page,map services, transient entities.


    I am not questioning whether the design makes sense. I was questioning the phrasing of the problem to demonstrate the
    need for transient entities and I was looking for a convincing business scenario where the need for transient entities
    is clear. As long as there is an alternative, fairly clean solution to the problem, the rationale would not be clear.

    Eilon and I talked about the scenario when the customization call ended early today. We agreed that
    - my solution made sense for the particular phrasing problem
    - and refining the problem where the search results page
    - actually is showing page fragments from each of the map services
    - and the number of results is unbounded (e.g all map services are not part of a single hotel chain).
    makes a good case for transient entities.

    Ravi Konuru

    Rich Thompson/Watson/IBM@IBMUS




            Rich Thompson/Watson/IBM@IBMUS

            06/12/2002 01:22 PM



    To: wsia@lists.oasis-open.org, wsrp@lists.oasis-open.org
    cc:
    Subject: RE: [wsrp] [wsia] [wsrp-wsia joint interfaces] Search results page, map services, transient entities.





    It sounds like you are questioning whether the design proposed for the
    Consumer's search result page makes sense. I would respond that not only
    does it make sense, I have seen the equivalent on the web already (don't
    remember where, but it was the basic scenario Eilon laid out). You are
    basically there when you state that the search page could present a set of
    clickable GIFs ... the question is where do these come from. Two basic
    choices are that the Consumer prefetches them into a local cache and just
    makes a connection to the search result record OR that for each result
    record, the Consumer creates a transient entity at a Producer, sets the
    configuration such that an appropriate GIF is generated and the Producer
    includes whatever logic is needed to deal with clicks on the GIF. I would
    argue that the architecture should not exclude either approach (both have
    their advantages) and that the second approach is a good case where the
    context known only by the Consumer is what dictates whether a persistent or
    transient entity is appropriate.




                         Ravi
                         Konuru/Watson/IBM        To:       wsia@lists.oasis-open.org, wsrp@lists.oasis-open.org
                         @IBMUS                   cc:
                                                  Subject:  RE: [wsrp] [wsia] [wsrp-wsia joint interfaces] Search
                         06/12/2002 10:53          results page, map services, transient entities.
                         AM





    I don't understand why transient entities are created in this scenario. Why
    is the search results page creating the transient entities?
    I can think of a case where the search results page provides a set of
    clickable GIFs that correspond to the hotel services but NOT actually
    create transient entities at the Producer. If the user actually clicks on
    one of them, it is at that time the consumer requests the producer to
    create an entity (transient or persistent) and then embeds the output of
    the entity in its web page.

    Am I missing something?

    thanks,
    Ravi Konuru

    (Embedded image moved to file: pic28419.gif)Carsten Leue/Germany/IBM@IBMDE


    (Embed (Embedded image moved to file:     (Embedded image moved to file:
    ded    pic00419.gif)                      pic30879.gif)
    image                          Carsten
    moved                          Leue/Germa To: Eilon Reshef
    to                             ny/IBM@IBM <eilon.reshef@webcollage.com>
    file:                          DE         cc: "'Alan Kropp'"
    pic116                                    <akropp@epicentric.com>,
    57.gif                                    wsia@lists.oasis-open.org,
    )                              06/12/2002 wsrp@lists.oasis-open.org
                                   06:57 AM   Subject: RE: [wsrp] [wsia]
                                              [wsrp-wsia joint interfaces]
                                              agenda for Tuesday 11 June






    I think that this is a very good scenario where transient entities make
    sense.


    Best regards
    Carsten Leue

    -------
    Dr. Carsten Leue
    Dept.8288, IBM Laboratory B�blingen , Germany
    Tel.: +49-7031-16-4603, Fax: +49-7031-16-4401



    |---------+----------------------------->
    | | Eilon Reshef |
    | | <eilon.reshef@webc|
    | | ollage.com> |
    | | |
    | | 06/11/2002 08:35 |
    | | PM |
    | | Please respond to |
    | | Eilon Reshef |
    | | |
    |---------+----------------------------->
    >
    ---------------------------------------------------------------------------------------------------------------------------------------------|

    | |
    | To: "'Alan Kropp'" <akropp@epicentric.com>, wsia@lists.oasis-open.org,
    wsrp@lists.oasis-open.org |
    | cc: |
    | Subject: RE: [wsrp] [wsia] [wsrp-wsia joint interfaces] agenda for
    Tuesday 11 June |
    | |
    | |
    >
    ---------------------------------------------------------------------------------------------------------------------------------------------|




    Alan,

    I think that one example scenario is a search results page, say for hotels,
    that dynamically displays multiple maps - a map for each hotel found.

    Assuming that maps are remote services, and assuming that the number of
    results is dynamic, the Consumer needs to create multiple copies of the map
    service. If we allow the Producer to determine the persistence state of
    those maps, that would mean that someone will have to take care of the
    lifetime of those maps. The Consumer can't, because the page may be gone
    without the Consumer never knowing about it (the user closes the browser
    window). The Producer can't, because it can't tell whether the Consumer has
    stored a reference to those maps as part of a design-time description of a
    user page.

    Is that along the lines of what you were looking for?

    Eilon
    -----Original Message-----
    From: Alan Kropp [mailto:akropp@epicentric.com]
    Sent: Tuesday, June 11, 2002 2:18 PM
    To: 'wsia@lists.oasis-open.org'; 'wsrp@lists.oasis-open.org'
    Subject: RE: [wsrp] [wsia] [wsrp-wsia joint interfaces] agenda for
    Tuesday 11 June



    I find that the question of a Consumer creating a persistent vs. a
    transient
    entity to be a little troubling.


    Isn't the Consumer primarily interested in getting a handle to an
    _entity_
    (be that transient or persistent), and then using that handle to
    direct
    subsequent requests to?


    Could we not consider persistence and transience to be entirely
    Producer
    concerns? I understand from the sequence in the spec (pp. 6-8) that
    the
    only supporting scenario for the Consumer calling
    "createTransientEntity" is
    out of an effort to maximize the efficiency of the conversation with
    the
    Producer. This is a worthy goal, of course, but the Producer should
    advertise its interaction behavior in the meta-data, right?
    Presumably the
    Producer uses the meta-data to hint to the Consumer the way(s) the
    Consumer
    SHOULD structure invocation requests efficiently.


    I'd love to see some more convincing scenarios to support pushing the

    persistent/transient aspect of the conversation out to the Consumer.
    Otherwise, I'd love even more to limit our specification to a single
    "createEntity" call.


    Alan





    -----Original Message-----
    From: Carsten Leue [mailto:CLEUE@de.ibm.com]
    Sent: Tuesday, June 11, 2002 2:19 AM
    To: Gil Tayar
    Cc: thomas klein6; 'wsia@lists.oasis-open.org';
    'wsrp@lists.oasis-open.org'
    Subject: RE: [wsrp] [wsia] [wsrp-wsia joint interfaces] agenda for
    Tuesday 11 June






    Gil -


    1-3: the consumer MUST explicitly create entities (transient or
    persistent). It can decide to create sessions explicitly if it wants
    to
    share them amongs entities, otherwise it can let the producer create
    sessions on the fly.
    4: the decision to have multiple entities per session is up to the
    consumer
    who make this choice based on the producer's metadata. e.g think of
    portlets that form a shop together. These would want to share a
    session
    with each other. But the consumer could decide to have two shops of
    the
    same type on the same page. Then there would be two sessions and only
    the
    consumer could know what portlet shares what session.
    5: The consumer should create a session before issuing a getFragment
    call.
    The entity is created whenever necessary.
    6: A Session is a bucket that holds some data (physically on the
    provider)
    and times-out after a while. A transient entity is a remote portlet
    that
    does not store any data in a DB.


    Hope that helped.


    Best regards
    Carsten Leue


    -------
    Dr. Carsten Leue
    Dept.8288, IBM Laboratory B�blingen , Germany
    Tel.: +49-7031-16-4603, Fax: +49-7031-16-4401






    |---------+---------------------------->
    | | Gil Tayar |
    | | <Gil.Tayar@webcol|
    | | lage.com> |
    | | |
    | | 06/11/2002 09:28 |
    | | AM |
    | | Please respond to|
    | | Gil Tayar |
    | | |
    |---------+---------------------------->

    >
    ---------------------------------------------------------------------------

    ------------------------------------------------------------------|
    |
    |
    | To: Carsten Leue/Germany/IBM@IBMDE
    |
    | cc: "'wsia@lists.oasis-open.org'"
    <wsia@lists.oasis-open.org>, "'wsrp@lists.oasis-open.org'"
    <wsrp@lists.oasis-open.org>, Thomas|
    | Klein6/Germany/IBM@IBMDE
    |
    | Subject: RE: [wsrp] [wsia] [wsrp-wsia joint interfaces]
    agenda
    for Tuesday 11 June
    |
    |
    |
    |
    |

    >
    ---------------------------------------------------------------------------

    ------------------------------------------------------------------|






    Carsten, could you clarify some things for me?


    1. If sessions and entities are orthogonal, what SHOULD the Consumer
    do?
    2. SHOULD it create a transient entity _and_ a session?
    3. MUST it do so?
    4. Who gets to decide whether there are multiple entities per session
    or
    multiple sessions per entities?
    5. When SHOULD a Consumer create a session and when a transient
    entity?
    6. What _is_ a session, logically? What _is_ a transient entity?


    Gil


    -----Original Message-----
    From: Carsten Leue [mailto:CLEUE@de.ibm.com]
    Sent: Tuesday, June 11, 2002 08:45
    To: Alan Kropp
    Cc: 'wsia@lists.oasis-open.org'; 'wsrp@lists.oasis-open.org'; thomas
    klein6
    Subject: Re: [wsrp] [wsia] [wsrp-wsia joint interfaces] agenda for
    Tuesday 11 June






    Alan - some first thoughts


    1. the general concept is that entities (transient or persistent) and

    sessions are orthogonal. There might be mulitple entities per session
    and
    (at least for persistent entities) multiple sessions per entity. An
    example
    for the first case would be that entities share session data within
    the
    same user request. An example for the second case are portlets on a
    common
    page (group page) that are access simultaneously by multiple users.
    A transient entity differs from a persistent entity in that its data
    does
    not persist and its lifecycle is coupled with the session life cycle.


    2. Eilon pointed out in a comment to the interface proposal that
    batch
    processing is already part of SOAP. I was not aware of that and will
    look
    it up until the call.





    Best regards
    Carsten Leue


    -------
    Dr. Carsten Leue
    Dept.8288, IBM Laboratory B�blingen , Germany
    Tel.: +49-7031-16-4603, Fax: +49-7031-16-4401






    |---------+---------------------------->
    | | Alan Kropp |
    | | <akropp@epicentri|
    | | c.com> |
    | | |
    | | 06/11/2002 01:54 |
    | | AM |
    | | Please respond to|
    | | Alan Kropp |
    | | |
    |---------+---------------------------->


    >
    ---------------------------------------------------------------------------

    ------------------------------------------------------------------|
    |
    |
    | To: "'wsrp@lists.oasis-open.org'"
    <wsrp@lists.oasis-open.org>, "'wsia@lists.oasis-open.org'"
    <wsia@lists.oasis-open.org> |
    | cc:
    |
    | Subject: [wsrp] [wsia] [wsrp-wsia joint interfaces] agenda
    for
    Tuesday 11 June
    |
    |
    |
    |
    |


    >
    ---------------------------------------------------------------------------

    ------------------------------------------------------------------|






    I think we'll have plenty of discussion around the merged document
    Rich and
    Carsten put together.


    The two main issues I see that have arisen on the email lists are:


    1. What is the difference between transient entities and sessions,
    and is
    there enough of a distinction to warrant including both in the
    specification?


    2. There are efficiency concerns around the use of arrays in the
    method
    signatures, basically to enable batched requests for network
    efficiency.





    Call-in numbers:
    USA Toll Free Number: 877-718-0936
    USA Toll Number: +1-712-923-6878
    PARTICIPANT PASSCODE: 563151





    Alan





    ----------------------------------------------------------------
    To subscribe or unsubscribe from this elist use the subscription
    manager: <
    http://lists.oasis-open.org/ob/adm.pl>








    ----------------------------------------------------------------
    To subscribe or unsubscribe from this elist use the subscription
    manager: <
    http://lists.oasis-open.org/ob/adm.pl>








    ----------------------------------------------------------------
    To subscribe or unsubscribe from this elist use the subscription
    manager: <
    http://lists.oasis-open.org/ob/adm.pl>


    ----------------------------------------------------------------
    To subscribe or unsubscribe from this elist use the subscription
    manager: <
    http://lists.oasis-open.org/ob/adm.pl>








    ----------------------------------------------------------------
    To subscribe or unsubscribe from this elist use the subscription
    manager: <
    http://lists.oasis-open.org/ob/adm.pl>









    #### pic28419.gif has been removed from this note on June 12 2002 by Ravi Konuru
    #### pic11657.gif has been removed from this note on June 12 2002 by Ravi Konuru
    #### pic00419.gif has been removed from this note on June 12 2002 by Ravi Konuru
    #### pic30879.gif has been removed from this note on June 12 2002 by Ravi Konuru

    GIF image



    [Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]


    Powered by eList eXpress LLC