| Re: [iot-wg] Agenda for next IoT WG Meeting |
|
Hi Jiaxin, some remarks from my end to your points: ·
First of all they all make sense to me. They are IMO not specific to Industrie 4.0 (which is focusing on manufacturing), but
are also relevant for other IoT scenarios as well. ·
Regarding Big Data: IMO it would not make sense for Eclipse to start Big Data projects if this would be in competition to
the very good Open Source tools which are available in that space already - just thinking of the “usual suspects” from Apache as Benjamin also mentioned below. @Benjamin: Sorry, but I was not able to attend the meeting last Wednesday. Best regards, Michael From: iot-wg-bounces@xxxxxxxxxxx [mailto:iot-wg-bounces@xxxxxxxxxxx]
On Behalf Of Benjamin Cabé Hi Jiaxin, It was great to have you on the call yesterday. I will be commenting inline. I hope others interested in the testbed will feel free to jump in the discussion as well :-) 2017-06-08 3:41 GMT+02:00 Yinjiaxin (Jason) <yinjiaxin@xxxxxxxxxx>:
I agree that we will need to come up with more than just data collection / device control although there is certainly already value in showing that we can solve the connectivity aspect of industrial IoT, like we discussed in the call yesterday.
These are all very valid suggestions and I certainly hope the testbed can address at least some of them. This would certainly require a more specific use case than the one depicted in the presentation so far, that is to say agree on a particular
manufacturing process that we aim at optimizing vs. simply address the "let's connect all the machine" side of things.
Correct. However, I would like to make it clear that the Eclipse IoT Open Testbeds should be an opportunity for us to demonstrate how we can interface with
other existing open source projects (and for everything big data, I certainly hope to build bridges with the great stuff coming from Apache, as an example). Also, it is an assumed goal of the testbeds to help
highlight the value add of the participating member companies, and their commercial offering around (Eclipse) IoT. So if Company A has a bleeding edge commercial solution for data analytics and would like to showcase how it can get data off of e.g Kapua
and process it, it would be great. Same if Company B would connect their proprietary PLCs to the Kapua backend, etc. You're making a very good point regarding Eclipse Kura Wires, and like Pascal pointed out in a comment on the presentation, it would certainly be nice to use the testbed as a way to showcase the edge/fog computing capabilities of Eclipse
Kura (and possibly other projects like Eclipse ioFog?).
I think it would be great to get Eclipse Ditto to play a role indeed (although the code hasn't been contributed to Eclipse yet, fwiw). Eclipse Vorto also comes to mind.
You just contributed, FWIW ;-) I see many ways for companies that do not directly contribute to or lead any Eclipse project to get involved, one being helping out with defining the use case and overall scope of the testbed, like you just
did! For a company like Huawei, I guess it would be interesting to see, for example, if it would make sense to showcase NB-IoT for low-power communication of wireless industrial equipment? So maybe that's another possible area of contribution? Thinking out
loud here but another angle would be running PPMP on top of LiteOS and show how easy it would be to then connect devices running LiteOS to the testbed backend? Certainly happy to have a quick chat with you and brainstorm further :-) Thanks again for the valuable feedback, and looking forward to hearing back your thoughts as well as others'! Benjamin -
-- Benjamin Cabé – IoT Program Manager & Evangelist
|
