Dear Martin, Yes, maybe I was "too quick". Note I have just answered at the same level than Tobias' issues... quick answers for quick questions ;) However, I like the idea of looking to this more deeply based in examples, as you suggest. Could you provide such examples, please? Please try to provide them before the May 15th deadline. Without those examples, I don't think it makes sense to have the audio today... I think that the time slot of the audio today will be better used for working on that examples. In addition, please find my comments inline. El 09/05/2016 a las 14:10, Martin Bauer escribió: Hi all, We can have the phone call at 16:00, if you want. I think your answers to Tobias points are a bit “too quick” – I think we first need to look at some examples to make sure there are really no problems. Jumping ahead is not helpful for the future development of the NGSI interface. In addition to the JSON-LD related issues identified by Tobias, I see some general problems and I list some examples of problems I see (the list is not complete): - The “specification” provided is not really a formal specification, more a developer’s guide based on examples – this makes consistency checking extremely difficult [FGM1] I think that the document at http://fiware.github.io/specifications/ngsiv2/latest is formal in the sense it provides a full description of the API and all the possible usage options... at least, as formal as a Apiary-based specification can be. I'm not a big fan of Apiary, but let's remember that Apiary is the official API description format for FIWARE APIs, so every API we define in the platform has to use it. Of course, the current version needs some polishing, but basically, all is covered there. If you find some point in which the specification is not complete (apart from the ones you describe in your email), let us know so we can fix it. Finally, don't miss http://fiware.github.io/specifications/ngsiv2/latest with http://fiware.github.io/specifications/ngsiv2/latest/cookbook. The last one is the one which is "more a developer’s guide based on examples" although maybe is a bit outdated by the time being. - If types are in general based on ontology concepts identified by URIs (as foreseen for JSON-LD), using specific strings like “date” does not work [FGM2] I guess you refer to "Special attribute types" section. If you have a better alternative for the "date" keyword, please go ahead. - Information is not complete, e.g. there are error codes listed in the Error responses section, but not all detailed errors are listed there ( e.g. 413 in geospatial properties says “Request entity is too large”, 422 “Unprocessable Entity” is not listed, or in “Retrieve entity attributes” a list of conflicting entities are returned in the error message, but there is no clear specification how. [FGM3] This is the kind of things we need yet to polish. However, I don't see any backward compatibility issue in this part of the spec (I mean, the general "error rules" are pretty clear, so new errors can be added as we go). - There is a section on partial representations, but it is not clear to which operations which aspects apply (“some operations” is clearly not enough) [FGM4] It is explained along with the specific operations descriptions. If the operation accepts the "options" URI parameter and if "keyValues" and/or "values" is included in the acceptable values for "options", then the operation support the format. Eg (from GET /v2/entitites): [cid:part1.6BF3478F.76B2864E at telefonica.com] - “near” only makes sense with one of the modifiers, but it isn’t clearly stated that a modifier is required [FGM5] Not sure of understanding this case... could you cite from the spec in order to clarify, please? - We consider “dateCreation” and “dateModification” critical in cases where there are many sources of changing availability that provide information about the same entity (something that is extremely important to us) – what dateCreation should a broker that does not keep information itself return? The oldest? – this might change with each request depending on the availability of sources … [FGM6] This could be an optional capability, so if the NGSIv2 server is not able to support date-tracking of entities/attributes it can declare so. A mechanism has been proposed for optional capabilities specification, please have a look to: https://github.com/telefonicaid/fiware-orion/issues/2110. Feedback comments on that issue are welcome. - Some new concepts (compared to OMA NGSI) are introduced that are not defined, e.g. NGSIv2 server (implementation), client, application – we fear that the assumptions are (only) based on the Orion Context Broker architecture (also aspects like “service path” which are GE and not NGSI specific appear in the document). [FGM7] This is another aspect to polish. Maybe a Glossary section at the end of the specification would suffice. - Paging concepts (i.e. limits and offsets) are used in examples, but are nowhere clearly defined. The semantics – what is to be expected as a result – should be specified. We still see issues, especially for a highly distributed system based on the IoT Broker (would probably require caching, but then we would need to specify some request-id to find the right cache? How long would it be valid?) [FGM8] We thought that the concepts were clear enough, as the limit/offset pattern is used in many REST APIs. However, if you think it is not clear we can add a "Pagination" section to the spec. è The document in its current form is not suitable as a basis for discussion in the planned ETSI ISG [FGM9] I'm not involved in ETSI task force, so I'd let the answer to this to some of my colleagues at TID. è I think it makes more sense to first write a formal specification (which we will need anyway) and then take a decision. Otherwise we will run into a lot of problems later and may not be able to ensure backward compatibility. [FGM10] Same comment as in FGM1. APIary specification is (or should be after polishing) a formal specification good enough so implementors of NGSIv2 clients and server could develop their software. Of course, that doesn't preclude the existence of other documents based on it (e.g. the input for ETSI, whatever it be) if the are needed. Thank you very much for the feedback. You are pointing to several places of the NGSIv2 spec that need to be polished, so it is very valuable. However, I don't see any of the issues/concerns be a blocker regarding the plan to froze a a first batch of NGSIv2 for production ready by the end of May. Best regards, Martin ------------------------------------------ Dr. Martin Bauer Senior Researcher NEC Europe Ltd. NEC Laboratories Europe Software & Services Research Division Kurfürsten-Anlage 36 D-69115 Heidelberg Tel: +49/ (0)6221/4342-168 Fax: +49/ (0)6221/4342-155 E-Mail: <mailto:Martin.Bauer at neclab.eu> Martin.Bauer at neclab.eu<mailto:Martin.Bauer at neclab.eu> <http://www.nw.neclab.eu/>http://www.nw.neclab.eu ************************************************************* NEC Europe Limited Registered Office: Athene, Odyssey Business Park, West End Road, London, HA4 6QE, Great Britain Registered in England 2832014 From: Tobias Jacobs Sent: Montag, 9. Mai 2016 12:55 To: Fermín Galán Márquez Cc: marc.capdevielle at orange.com<mailto:marc.capdevielle at orange.com>; fiware-ngsi at lists.fiware.org<mailto:fiware-ngsi at lists.fiware.org>; SERGIO GARCIA GOMEZ; wenbin.li at orange.com<mailto:wenbin.li at orange.com>; JUAN JOSE HIERRO SUREDA; PRIVAT Gilles IMT/OLPS; RAMPARANY Fano IMT/OLPS; t.elsaleh at surrey.ac.uk<mailto:t.elsaleh at surrey.ac.uk>; Martin Bauer; Lindsay Frost Subject: RE: [Fiware-ngsi] Releasing a first batch of NGSIv2 stable functionality as "production ready" by the end of May Dear Fermin, hmm, I am not sure now whether we will have a conference call today. 16.00 (as proposed by Lindsay) would work for me as well. I below give a couple of examples where potential compatibility problems could occur. There are probably more of them. - In the JSON-LD group we had for discussions about how to attach metadata to attribute values, and what we all agreed on in the end was to remove the extra “metadata” layer, but instead make the actual attribute value syntactically nothing more than a specific piece of metadata. This would also have an effect on the “plain” JSON representation. - Another example: representing multiple “value instances” of the same attribute is forseen by JSON-LD and would require a certain JSON-syntax (representing the different values in a list). - Thirdly, I currently do not see how different representation formats (short representation, canonical representation) would be compatible with JSON-LD. Maybe they are, but we have not yet taken this question into account. Best Regards Tobias From: Fermín Galán Márquez [mailto:fermin.galanmarquez at telefonica.com] Sent: Mittwoch, 4. Mai 2016 22:29 To: Tobias Jacobs Cc: marc.capdevielle at orange.com<mailto:marc.capdevielle at orange.com>; fiware-ngsi at lists.fiware.org<mailto:fiware-ngsi at lists.fiware.org>; SERGIO GARCIA GOMEZ; wenbin.li at orange.com<mailto:wenbin.li at orange.com>; JUAN JOSE HIERRO SUREDA; PRIVAT Gilles IMT/OLPS; RAMPARANY Fano IMT/OLPS; t.elsaleh at surrey.ac.uk<mailto:t.elsaleh at surrey.ac.uk>; Martin Bauer; Lindsay Frost Subject: Re: [Fiware-ngsi] Releasing a first batch of NGSIv2 stable functionality as "production ready" by the end of May Dear Tobias, As far as I understand, the JSON-LD NGSI encoding proposal was designed in a way fully aligned with "plain" NGSIv2 from the very beginning. At least based in the examples at https://docs.google.com/document/d/1lWH_a-JkwfSR9uE03KQbKe09uajpJqsPXk7nHFJPYQs/edit#heading=h.vx9h5b418g4u, basically JSON-LD payload = "plain" NGSIv2 payload in keyValues render mode + @context and other JSON-LD stuff. However, I understand you are proposing this call because you feel there is some "risk" of non-alignment. Is that correct? Could you share with us before we meeting the points in which you think the alignment is broken, please? Regarding the conference itself, I'd prefer May 9th, 15:30 CET but I think that I could make if most of the people prefers 14:30 CET. In parallel, we would be more than happy to know your feedback after you re-evaluate the current status of the (JSON-LD independent) NGSIv2 proposal. Thanks! Best regards, ------ Fermín El 04/05/2016 a las 17:06, Tobias Jacobs escribió: Dear all, In the confcall that has just taken place we have not been able to discuss this issue below, but we propose to have a dedicated conference call on the relation between JSON-LD and the plans to make the current Orion NGSI-v2 a “stable version”. The proposed time slot is Monday, May 9th at 14:30 CET Could you please confirm your availability? Sergio or someone else, could you provide the conference bridge? Thank you! Best Regards Tobias From: Tobias Jacobs Sent: Dienstag, 3. Mai 2016 15:44 To: 'Fermín Galán Márquez'; <mailto:marc.capdevielle at orange.com> marc.capdevielle at orange.com<mailto:marc.capdevielle at orange.com> Cc: fiware-ngsi at lists.fiware.org<mailto:fiware-ngsi at lists.fiware.org>; Tobias Jacobs; SERGIO GARCIA GOMEZ; wenbin.li at orange.com<mailto:wenbin.li at orange.com>; FERMIN GALAN MARQUEZ; JUAN JOSE HIERRO SUREDA; PRIVAT Gilles IMT/OLPS; RAMPARANY Fano IMT/OLPS; t.elsaleh at surrey.ac.uk<mailto:t.elsaleh at surrey.ac.uk>; CAPDEVIELLE Marc IMT/OLPS; Martin Bauer; Lindsay Frost Subject: RE: [Fiware-ngsi] Releasing a first batch of NGSIv2 stable functionality as "production ready" by the end of May Dear Fermin and all, >From NEC side we have been rather concentrating on the JSON-LD discussions since the beginning of this year, so we have to re-evaluate the current status of the (JSON-LD independent) NGSI-2 proposal before we can make a statement. In the past we have been insisting on not making implicit assumptions on the system architecture when specifying the features of the query language (e.g. not assuming that the broker is a centralized system which has all data in a local database). We will analyze the current version and prepare a more detailed statement. Besides, there is yet another issue, which has to do with the current JSON-LD serialization: We would like to ask all partners how they see the interrelationship between the current JSON-LD discussion and the NGSI-2 interface as defined by Orion on GitHub. On the JSON-LD side we are having very constructive discussions, and I am optimistic that we are about to agree on the last open points soon. For me the main question here is: Are we targeting to align the ‘plain’ JSON message format with what we have decided for JSON-LD? Concretely, aligning would mean that the plain JSON body is what you get when you remove the @context from the JSON-LD body. In our view this alignment would be very useful, so that the JSON-LD bodies are fully backwards compatible with the plain JSON-bodies (which was anyway one of the primary design targets of JSON-LD), and the JSON-LD serialization does not become a decoupled “NGSI 3.0”. By the way, there will be a conference call on JSON-LD next Wednesday (tomorrow) at 15:00, where we could discuss this question. Best regards Tobias From: fiware-ngsi-bounces at lists.fiware.org<mailto:fiware-ngsi-bounces at lists.fiware.org> [mailto:fiware-ngsi-bounces at lists.fiware.org] On Behalf Of Fermín Galán Márquez Sent: Dienstag, 3. Mai 2016 07:56 To: marc.capdevielle at orange.com<mailto:marc.capdevielle at orange.com> Cc: fiware-ngsi at lists.fiware.org<mailto:fiware-ngsi at lists.fiware.org> Subject: Re: [Fiware-ngsi] Releasing a first batch of NGSIv2 stable functionality as "production ready" by the end of May Dear Marc, Thank you very much for the feedback! As you suggest, let's continue discussion on the specific issues a github. That's the way to go :) Best regards, ------ Fermín El 02/05/2016 a las 17:57, marc.capdevielle at orange.com<mailto:marc.capdevielle at orange.com> escribió: Hi Fermin, The freeze and backward compatibility promise of the NGSIv2 is welcome and clearly a sign that the specification has matured. I have two remarks I would like to discuss before the freeze. 1) Roles definitions The current work on NGSIv2 seems to be missing some roles definition (or profiles ?) to clearly identify how different components in the FIWARE architecture (or others) can handle NGSIv2 (i.e. making some parts of NGSIv2 optional to some roles). A broker like Orion can implement the full specification but it seems logical to have some components only implementing a subset of the specification. By having multiple roles clearly identifying what subset of the specification it support, we could then define specialized components as a simple composition of these roles. A component like the IoT Agent node library<https://github.com/telefonicaid/iotagent-node-lib>, handling protocol adaptations to/from NGSI, only supports a subset of NGSI : - as a client it sends data received from devices to a broker (Context Producer) - as a server it registers entities to receive commands and accept some limited querying for "lazy" attributes (Context Provider) Much like it, Fiware-Cepheus<https://github.com/Orange-OpenSource/fiware-cepheus>, handling gateway-level data aggregation, also only supports a subset of NGSI v1. I have added some comments to the ongoing discussion here : <https://github.com/telefonicaid/fiware-orion/issues/1459> https://github.com/telefonicaid/fiware-orion/issues/1459 2) Optional features Moreover, some orthogonal features of the NGSIv2 could also be made optional in some contexts to lighten up the API "surface". The proposition was made here : https://github.com/telefonicaid/fiware-orion/issues/2110 Let's discuss all this on the corresponding Github Issues. Regards, Marc Capdevielle Le 26 avr. 2016 à 18:13, Fermín Galán Márquez <<mailto:fermin.galanmarquez at telefonica.com>fermin.galanmarquez at telefonica.com<mailto:fermin.galanmarquez at telefonica.com>> a écrit : Hi, We started to work on NGSIv2 (both at specification level and reference implementation in Orion Context Broker) last summer. Since that time, the NGSIv2 API has been in "beta" status, thus available for early bird testing (which is good) but not suitable for production usage (which is not so good). But now, more than half a year later, we think that big part of the specification has achieved a very mature status so it should progress from "beta" to "production ready" status. In particular, this "batch" will be focused on the context management functionality (AKA "NGSI10" in the old OMA-NGSI terms) as described right now in the specification at http://telefonicaid.github.io/fiware-orion/api/v2 (e.g. create/update entities, get entities attributes, filters, notifications in NGSIv2 format, etc.). The releasing date for this first batch is expected by the end of May. Of course, this doesn't mean that the NGSIv2 work (both specification and implementation) will stop after May. There are still exciting areas (context management availability, JSON-LD, etc.) under discussion. In addition, some advanced features in context management (e.g. filters on structured attribute values) will not be ready by the end of May. Thus, at some point we will "branch" the spec document, to "freeze" a snapshot corresponding to the stable batch and, at the same time, continue to work on the new topics (and, in the future, produce new batches). However, what is important to remark is that any future evolution or addition to NGSIv2 has to be done in a way in which it is compatible with the existing stable batch, i.e. preserving backward compatibility. In summary, the plan is as follows: * To polish and branch the spec at http://telefonicaid.github.io/fiware-orion/api/v2, in two documents: stable and wip. * To align reference implementation (Orion Context Broker) with that "stable" spec (most of that work is done already). * To "freeze" spec and reference implementation by the end of May. * In parallel with the above points, discussion on ongoing topics (e.g. JSON-LD) will continue and its "output" will be available in coming stable batches (post-May) of NGSIv2. Any feedback is welcome. The deadline to receive such feedback is May 15th. Thanks! Best regards, ------ Fermín ________________________________ Este mensaje y sus adjuntos se dirigen exclusivamente a su destinatario, puede contener información privilegiada o confidencial y es para uso exclusivo de la persona o entidad de destino. Si no es usted. el destinatario indicado, queda notificado de que la lectura, utilización, divulgación y/o copia sin autorización puede estar prohibida en virtud de la legislación vigente. Si ha recibido este mensaje por error, le rogamos que nos lo comunique inmediatamente por esta misma vía y proceda a su destrucción. The information contained in this transmission is privileged and confidential information intended only for the use of the individual or entity named above. If the reader of this message is not the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this transmission in error, do not read it. Please immediately reply to the sender that you have received this communication in error and then delete it. Esta mensagem e seus anexos se dirigem exclusivamente ao seu destinatário, pode conter informação privilegiada ou confidencial e é para uso exclusivo da pessoa ou entidade de destino. Se não é vossa senhoria o destinatário indicado, fica notificado de que a leitura, utilização, divulgação e/ou cópia sem autorização pode estar proibida em virtude da legislação vigente. Se recebeu esta mensagem por erro, rogamos-lhe que nos o comunique imediatamente por esta mesma via e proceda a sua destruição Since January 1st, old domains won't be supported and messages sent to any domain different to @lists.fiware.org<http://lists.fiware.org> will be lost. Please, send your messages using the new domain (<mailto:Fiware-ngsi at lists.fiware.org>Fiware-ngsi at lists.fiware.org<mailto:Fiware-ngsi at lists.fiware.org>) instead of the old one. _______________________________________________ Fiware-ngsi mailing list Fiware-ngsi at lists.fiware.org<mailto:Fiware-ngsi at lists.fiware.org> https://lists.fiware.org/listinfo/fiware-ngsi _________________________________________________________________________________________________________________________ Ce message et ses pieces jointes peuvent contenir des informations confidentielles ou privilegiees et ne doivent donc pas etre diffuses, exploites ou copies sans autorisation. Si vous avez recu ce message par erreur, veuillez le signaler a l'expediteur et le detruire ainsi que les pieces jointes. Les messages electroniques etant susceptibles d'alteration, Orange decline toute responsabilite si ce message a ete altere, deforme ou falsifie. Merci. This message and its attachments may contain confidential or privileged information that may be protected by law; they should not be distributed, used or copied without authorisation. If you have received this email in error, please notify the sender and delete this message and its attachments. As emails may be altered, Orange is not liable for messages that have been modified, changed or falsified. Thank you. ________________________________ Este mensaje y sus adjuntos se dirigen exclusivamente a su destinatario, puede contener información privilegiada o confidencial y es para uso exclusivo de la persona o entidad de destino. Si no es usted. el destinatario indicado, queda notificado de que la lectura, utilización, divulgación y/o copia sin autorización puede estar prohibida en virtud de la legislación vigente. Si ha recibido este mensaje por error, le rogamos que nos lo comunique inmediatamente por esta misma vía y proceda a su destrucción. The information contained in this transmission is privileged and confidential information intended only for the use of the individual or entity named above. If the reader of this message is not the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this transmission in error, do not read it. Please immediately reply to the sender that you have received this communication in error and then delete it. Esta mensagem e seus anexos se dirigem exclusivamente ao seu destinatário, pode conter informação privilegiada ou confidencial e é para uso exclusivo da pessoa ou entidade de destino. Se não é vossa senhoria o destinatário indicado, fica notificado de que a leitura, utilização, divulgação e/ou cópia sem autorização pode estar proibida em virtude da legislação vigente. Se recebeu esta mensagem por erro, rogamos-lhe que nos o comunique imediatamente por esta mesma via e proceda a sua destruição ________________________________ Este mensaje y sus adjuntos se dirigen exclusivamente a su destinatario, puede contener información privilegiada o confidencial y es para uso exclusivo de la persona o entidad de destino. Si no es usted. el destinatario indicado, queda notificado de que la lectura, utilización, divulgación y/o copia sin autorización puede estar prohibida en virtud de la legislación vigente. Si ha recibido este mensaje por error, le rogamos que nos lo comunique inmediatamente por esta misma vía y proceda a su destrucción. The information contained in this transmission is privileged and confidential information intended only for the use of the individual or entity named above. If the reader of this message is not the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this transmission in error, do not read it. Please immediately reply to the sender that you have received this communication in error and then delete it. Esta mensagem e seus anexos se dirigem exclusivamente ao seu destinatário, pode conter informação privilegiada ou confidencial e é para uso exclusivo da pessoa ou entidade de destino. Se não é vossa senhoria o destinatário indicado, fica notificado de que a leitura, utilização, divulgação e/ou cópia sem autorização pode estar proibida em virtude da legislação vigente. Se recebeu esta mensagem por erro, rogamos-lhe que nos o comunique imediatamente por esta mesma via e proceda a sua destruição Best regards, ------ Fermín ________________________________ Este mensaje y sus adjuntos se dirigen exclusivamente a su destinatario, puede contener información privilegiada o confidencial y es para uso exclusivo de la persona o entidad de destino. Si no es usted. el destinatario indicado, queda notificado de que la lectura, utilización, divulgación y/o copia sin autorización puede estar prohibida en virtud de la legislación vigente. Si ha recibido este mensaje por error, le rogamos que nos lo comunique inmediatamente por esta misma vía y proceda a su destrucción. The information contained in this transmission is privileged and confidential information intended only for the use of the individual or entity named above. If the reader of this message is not the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this transmission in error, do not read it. Please immediately reply to the sender that you have received this communication in error and then delete it. Esta mensagem e seus anexos se dirigem exclusivamente ao seu destinatário, pode conter informação privilegiada ou confidencial e é para uso exclusivo da pessoa ou entidade de destino. Se não é vossa senhoria o destinatário indicado, fica notificado de que a leitura, utilização, divulgação e/ou cópia sem autorização pode estar proibida em virtude da legislação vigente. Se recebeu esta mensagem por erro, rogamos-lhe que nos o comunique imediatamente por esta mesma via e proceda a sua destruição -------------- next part -------------- An HTML attachment was scrubbed... URL: <https://lists.fiware.org/private/fiware-ngsi/attachments/20160509/3026e90a/attachment.html> -------------- next part -------------- A non-text attachment was scrubbed... Name: bcifejdbfcfeelhd.png Type: image/png Size: 4338 bytes Desc: not available URL: <https://lists.fiware.org/private/fiware-ngsi/attachments/20160509/3026e90a/attachment.png>
You can get more information about our cookies and privacy policies clicking on the following links: Privacy policy Cookies policy