[Fiware-data] Updates/revisions on Query Access GE section (re: Minutes of our last confcall on July 15th)

[Juanjo Hierro]

  Ok.  I will check the proposed wording and take a decision on whether to add the "multimedia" label to this GE.   Nevertheless, I believe it will be worth adding some subsection under "Points still under discussion" where we state that this is a point still under discussion.

  Thanks and best regards,

-- Juanjo

On 19/07/11 10:03, Hutter, Andreas wrote:

Dear all,

I have uploaded our revised version on the Query Access GE;

https://forge.fi-ware.eu/docman/view.php/9/281/FI-WARE+High-Level+Description+integrated+draft+0+1+11-07-06+-++Revised+QA+GE+-+Siemens.doc

(Changes are in the Query Acesss section 4.2.7 , pp. 67 only)

We changed the wording in order to state more clearly, that the QA GE is indeed suited to provide a federative access to "classical/established" data repositories, like SQL DB, Triple stores or XML repositories. As an add-on it provides support for specific search and query functions required in (metadata based) multimedia content search (from a scientific point of view this was one focus area from which also some of the descriptions orginated). However, in several of our applications there is no multimedia data involved.

Therefore we think adding "Multimedia" to the section title does not accurately reflect its capabilities and will give a wrong impression. Also, it is planned, that during the project the integration of other data resources (e.g. NoSQL, …) will be elaborated and realised as far as possible. On the other hand there are certainly restrictions in the current design and implementation which may be indicated by some addition to the section title like e.g. "Basic" (?).

As agreed, the overall objective should be to come up with a generic API for this GE, to provide a unique accesss point for data query. How to best get there and whether or not the current (Basic) QA GE can serve as basis to provide all these functionalities indeed needs to be listed in the "Points still under discussion".

I would also propose that TI/Boris and Siemens have another telco to discuss the best way forward to support the Context Management QueryAccess. We would like to understand better the limitations encountered by Boris.

Best regards,
Andreas



________________________________
Von: fiware-data-bounces at lists.fi-ware.eu<mailto:fiware-data-bounces at lists.fi-ware.eu> [mailto:fiware-data-bounces at lists.fi-ware.eu] Im Auftrag von Juanjo Hierro
Gesendet: Montag, 18. Juli 2011 13:45
An: fiware-data at lists.fi-ware.eu<mailto:fiware-data at lists.fi-ware.eu>
Betreff: [Fiware-data] Minutes of our last confcall on July 15th

Hi all,

  Sorry all, but I thought I had already sent the minutes of our last confcall.   Please find below them.

  Thanks to all for attending the confcall and the great work you are doing.


0. Agenda of the meeting

  Topics that have been addressed during the confcall where:

 1.  Briefing on recent meeting of the FI-PPP Architecture Board
 2.  Review of actions from our last confcall

1. Briefing on recent meeting of the FI-PPP Architecture Board

  Juanjo briefed the team about the last FI-PPP Architecture Board meeting.   This has taken place in Madrid, last July 11-12.

  General impression was that the existing FI-WARE High-level Description was well-received and now UC projects have a better understanding about what we intend to deliver and what a Generic Enabler will be.   They wish to interact more closely with members of the different FI-WARE chapter teams in order to solve doubts about GEs being proposed so that some mechanism for achieving this will be put in place.   Yesterday, WPLs and WPAs of the different chapters meet together and they agreed to manage this through some tracking tool.

  Another general comment that was made by the UC projects is that we should make it explicit when we plan to follow a given standard interface specification for some given GE.   This is actually a comment we should consider in the review of our chapter (making explicit references to OMA specs for example)

  Another point addressed during the FI-PPP AB meeting was that about structure of fields of entries in the FI-PPP backlog the FI-WARE backlog will be part of.   Also how the backlogs are going to be populated and the way UC projects will work to generate input that can be processed by FI-WARE.   Juanjo will send an elaborated email about this topic to the WPLs and WPAs and will forward it to the Data/Context Management team.

2. Review of actions from our last confcall

2.1 Finishing the FI-WARE High-level Description deliverable

  The following actions, agreed during our last confcall, were performed regarding review of the IoT Service Enablement chapter:

    * Guy performed a first peer review and will generate .doc with changes under control by EOB Sunday
    * Boris finally performed a second peer review.
    * Juanjo merged results of both peer-reviews and generated a document that was sent to the IoT team on Wednesday

  We meet together with the WPA of the IoT Service Enablement chapter (NSN) and performed a detailed review of the comments we provided.   Most of them were accepted and have lead to some action points to be done by the editors of the chapter.   Therefore, our peer-review was successfully closed.

  Regarding revision of our chapter, Orange and NSN/NEC provided comments on separate files.   Guy, Boris and myself reviewed the comments and produced a revised version of the two files with replies to comments.   We will perform a detailed revision of comments and our initial reply to those comments in a confcall scheduled with NSN for today, July 15 at 14:30


2.2 Finishing some pieces that were not that complete in the delivered draft:

  We reviewed the different points identified in the previous confcall:


 *   Introduction

          o Several members of the team have reviewed the initial input provided by Federico
          o Juanjo will review the current version and generate the final one to integrate in the deliverable


 *   Section on "Points still under discussion".

We agreed to elaborate on a number of items, generating 3-4 paragraphs per item as we already have done for the Security aspects.
Following is the list of items identified and the status of each point:
Data gathering GE:
Siemens produced the committed contribution and is ready for integration in the deliverable.
Guy will perform a review of it and will also make sure it gets consistent with what is being said in the description of the CEP GE regarding data gathering
BigData Analysis vs Complex Event Processing .
Guy has produced something but have to check whether he has uploaded somewhere at the server.
Support to multiple Query Languages and Query Language Families ...
It has been found that the current description of the Query Access GE is still to focused on multimedia.    Some partners (e.g., Telecom Italia) have some concerns whether to follow to extend the approach described in that GE as to be adopted for all Query Accesses will have serious drawbacks in terms of performance, for example.   Actions agreed:

 *   Rename existing "Query Access" GE as to become "Multimedia Query Access" GE for this release
 *   Create subsection on "Question Marks" section under "Points still under discussion" where we will formulate the need for a Query Access GE that would be used for accessing data/context elements handled by GEs like the Publish/Subscribe, Complex Event Processing or BigData Analysis GEs.   There, we will mention that we have to investigate whether the Multimedia Query Access GE can be adopted as basis, further adapting it, or we should define a separate one.    Juanjo will try to write this section.

 *   GEs dealing with semantic infrastructure:

ATOS confirmed that everything was ok and no piece was missing.

 *   Adding terms to the terms an definition sections

All should provide proposals on terms and definitions they would like to include in this chapter section

 *   Integration of Location-related input from Telecom Italia.
    *   We have agreed, for the time being, to keep the Thales GE as the only Localization GE.
    *   However, we are going to create a section under the "Points still under discussion" based on contents of the contribution made by Telecom Italia, referring to the intend to analyze a more complete Localization Service where the Thales platform may be complemented by additional components

 *   Review additional enablers proposed by Thales not yet integrated in the deliverable
    *   We will add a few paragraphs unders the "Points still under discussion" sections mentioning these enablers and briefly describing them.  A final decision on whether to integrate them or not will be taken for the second release of the FI-WARE High-level Description Deliverable.   Juanjo still assigned to this and hopefully Boris will be able to review the work by Juanjo.

 *   Revision of the description of the Publish/Subscribe Broker GE
    *   This had been already discussed in our previous confcall.  It's just a matter of describing the interfaces, operations and interactions: Juanjo will take care of this

 *   Juanjo to ask the I2ND team to review our chapter, overall in respect to areas where overlapping may exist (S3C GE).   Probably will not be feasible to implement this peer-review for the current official release of the deliverable, but at least would be started for the second release.


________________________________
Este mensaje se dirige exclusivamente a su destinatario. Puede consultar nuestra política de envío y recepción de correo electrónico en el enlace situado más abajo.
This message is intended exclusively for its addressee. We only send and receive email on the basis of the terms set out at.
http://www.tid.es/ES/PAGINAS/disclaimer.aspx

________________________________
Este mensaje se dirige exclusivamente a su destinatario. Puede consultar nuestra política de envío y recepción de correo electrónico en el enlace situado más abajo.
This message is intended exclusively for its addressee. We only send and receive email on the basis of the terms set out at.
http://www.tid.es/ES/PAGINAS/disclaimer.aspx
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.fiware.org/private/old-fiware-data/attachments/20110719/2885d1cb/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: jhierro.vcf
Type: text/x-vcard
Size: 429 bytes
Desc: not available
URL: <https://lists.fiware.org/private/old-fiware-data/attachments/20110719/2885d1cb/attachment.vcf>