[Fiware-ngsi] Fwd: VERY IMPORTANT, PLEASE READ: Things to improve for the next educational session

[Juanjo Hierro]

Hi,

  Despite I guess it might be unnecessary, I would like to remark what I said after the infortunate experience during our educational session in Zurich (see mail below).

  Besides the obvious need to avoid any debate, please observe the following questions:

  *   Avoid repeating definition of concepts that have been introduced in the initial session.   Please find enclosed an updated version of the presentation I will make
  *   Avoid unconsistent definition and terminology.   This particularly applies to Boris' presentation, where I remember a slide showing interactions between the Context Broker and the rest of entities in the model which had names of entities and names of operations that do not map with the names we have introduced in the introductory presentation.
  *   Boris has to introduce the ContextML-based Pub/Sub Broker GE API properly.   Some points that should be highlighted:
     *   We all know and have already explained that OMA NGSI didn't specify any binding ... therefore Boris should simply start saying that we will support two bindings, one more RESTful to be described by NEC and another one which is more document-oriented (documents following the ContextML language)   That would make things easier to understand and people would not find any particular inconsistency.
     *   The slide that I was referring previously from Boris' presentation showing the entities and their interaction could then be used but emphasizing that operations are not what is being displayed but ContextML documents exchanged (of course, names of entities should still be updated to armonize with those presented in the introductory presentation).
     *   Using names like setContext, getContext, etc while describing interactions using the ContextML-based Pub/Sub Broker GE API introduces a lot of confussion if we REFER TO THEM AS OPERATIONS.   I would rather consider talking about "document types", establish then what is the mapping between NGSI operations and those document types, and then elaborate on the particular syntax of ContextML documents associated to each document type.

  Hope that this time we will make it better.

  Cheers,

-- Juanjo

-------------
Product Development and Innovation (PDI) - Telefonica Digital
website: www.tid.es<http://www.tid.es>
email: jhierro at tid.es<mailto:jhierro at tid.es>
twitter: twitter.com/JuanjoHierro

FI-WARE (European Future Internet Core Platform) Chief Architect

You can follow FI-WARE at:
  website:  http://www.fi-ware.eu
  facebook: http://www.facebook.com/pages/FI-WARE/251366491587242
  twitter:  http://twitter.com/FIware
  linkedIn: http://www.linkedin.com/groups/FIWARE-4239932



-------- Original Message --------
Subject:        VERY IMPORTANT, PLEASE READ: Things to improve for the next educational session
Date:   Tue, 22 May 2012 15:45:07 +0200
From:   Juanjo Hierro <jhierro at tid.es><mailto:jhierro at tid.es>
To:     fiware-ngsi at lists.fi-ware.eu<mailto:fiware-ngsi at lists.fi-ware.eu> <fiware-ngsi at lists.fi-ware.eu><mailto:fiware-ngsi at lists.fi-ware.eu>


Gentlemen,

  With all due respect.

  I didn't ask FI-WARE partners to avoid any short of heavy debate/discussion in front of UC projects during the Educational Session week.   I thought this was common sense, but apparently I was wrong given the scene we have experienced during the slot about the Pub/Sub Broker GE within the joint IoT/Data session today.   You saw that I tried my best to minimize the debate and, fortunately, it didn't go too strong, so the damage was probably manageable, but please don't repeat anything like this.   I rather believe it can really damage our image in front of the UC projects.    It is not just my impression.   I tell you that I got the opportunity to chat with one of the UC projects' architects while we were going lunch and he told me that he was a bit confused about the whole thing because of the exhibited debate.

  Nevertheless, let's look ahead and simply focus on learn the lesson for the next educational session.   Some points I took note about which should be considered for that next educational session (and, generally, any presentation/workshop we may get involved in):

  *   First of all, AVOID internal discussions going public in this sort of meetings
  *   While we should try to avoid repeating description of concepts that have been described before (because it just increases the probability that we confuse people because delivering different definitions), what we definitively have to avoid is to introduce new concepts and terms that clearly seem to be in collision with the concepts and terms first introduced.    Here, I found particularly harmful the slide that summarized the interaction between a Context Broker and other roles.   There were many roles there that aligned with roles introduced earlier in the session, but were simply named differently and I guess this could have created a lot of confusion.    Saying "we support the OMA NGSI model" and then describe the model you support naming roles differently certainly doesn't help.   I understand that it is easy to reuse existing slides, but we should try to make an effort to homogenize definitions and terminology.
  *   I think that we didn't introduce the ContextML-based Pub/Sub Broker GE API properly.   Some points we should have highlighted:
     *   We all know and have already explained that OMA NGSI didn't specify any binding ... why didn't we simply start saying that we will support two bindings, one more RESTful and another one more document-oriented (documents following the ContextML language) ?   That would make things easier to understand and people would not find any particular inconsistency.
     *   I believe that using names of operations (setContext, getContext, ...) while describing interactions using the ContextML-based Pub/Sub Broker GE API introduces a lot of confussion.   First, because those operations, although similar, are named differently and use different arguments than the operations we introduce when we describe FI-WARE NGSI at abstract level (registerContext, queryContext, notifyContext, ...) ... but, second, because if I understand it well the point is that different type of ContextML documents is what you will ultimately exchange per operation.   I would rather consider talking about "document types", establish then what is the mapping between NGSI operations and those document types, and then elaborate on the particular syntax of ContextML documents associated to each document type.

  Please take this comments constructively.   I'm sure this can be properly addressed in the presentations we will use in the second Educational Week.   I also believe that it would be worth introducing the proposed comments regarding how to introduce the ContextML-based Pub/Sub Broker GE API in the current slides before uploading them to the wiki in the space that will be setup for sharing information for this Educational sessions.

  Best regards,

-- Juanjo

-------------
Product Development and Innovation (PDI) - Telefonica Digital
website: www.tid.es<http://www.tid.es>
email: jhierro at tid.es<mailto:jhierro at tid.es>
twitter: twitter.com/JuanjoHierro

FI-WARE (European Future Internet Core Platform) Chief Architect

You can follow FI-WARE at:
  website:  http://www.fi-ware.eu
  facebook: http://www.facebook.com/pages/FI-WARE/251366491587242
  twitter:  http://twitter.com/FIware
  linkedIn: http://www.linkedin.com/groups/FIWARE-4239932


________________________________

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/fiware-ngsi/attachments/20120605/a1138fd5/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: FI-WARE NGSI Overview 12-06-05.pptx
Type: application/vnd.openxmlformats-officedocument.presentationml.presentation
Size: 707395 bytes
Desc: not available
URL: <https://lists.fiware.org/private/fiware-ngsi/attachments/20120605/a1138fd5/attachment.pptx>