Skip to main content
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [milo-dev] OPC UA Stack by OPC Foundation
> [Sumit] - Based on RC specs looks like you are about to start this.

In the next couple months probably. I've got a lot of other work I need to do in preparation for the IOP event in September.


> [Sumit] - I am referring to Milo SDK only. I have gone through with Pub-Sub specs and data modelling for sure is quite complex. Doubt which i have if data modelling is present which would be part of stack someday, then technically i dont need Pub-Sub implementation in SDK. Any application (Producer/Consumer) of MQTT or AMQP broker, can easily do the job. Please correct me if i am misinterpreting something.

You technically don't need an SDK implementation, but you'll have a lot of stuff you have to implement on your own, especially if you're going to use security.


> [Sumit] - I am able to produce/consume existing message (for e:x: DataValue) on activeMQ based JMS broker. This was done to perform some POC. It was not as straight forward as i had thought of as i had to use Encoder/Decoder in order to transmit messages via ByteMessage. Using ObjectMessage/TextMessage was out of scope anyway.

Are you asking a question here? This doesn't really seem to be a PoC of PubSub to me... just a PoC using a message broker. If you're not sending the same data structures that PubSub will send it's not really a valid test is it?

On Tue, Jul 25, 2017 at 7:05 AM, Sumit Aggarwal <sumit.kaggarwal@xxxxxxxxx> wrote:
Please see comments inline in blue..

On Tue, Jul 25, 2017 at 6:32 PM, Kevin Herron <kevinherron@xxxxxxxxx> wrote:
> typically how much time it takes you guys to update library in-line with OPC UA foundation stack.

Sometimes we're ahead, sometimes we're behind. Milo's stack isn't based on the foundation's stack, it's based on the specs. 

> A hypothetical question - How long it may take to come out with Pub-Sub support once OPC UA foundation releases final specs.

I honestly can't answer this. I'll begin working on support 1.04 in Milo more-or-less immediately, which will put all the data structures and node types in place to make implementing PubSub possible.
[Sumit] - Based on RC specs looks like you are about to start this.
 
As far as actually implementing it in a client or server goes... that's more complex. It's definitely something to be done at the SDK level, whether you're talking about Milo or the Prosys SDK. I'm not sure when, or even if, PubSub will make its way into the Milo SDK. If you haven't read the spec for it yet I suggest you do so - it's incredibly complex. 
[Sumit] - I am referring to Milo SDK only. I have gone through with Pub-Sub specs and data modelling for sure is quite complex. Doubt which i have if data modelling is present which would be part of stack someday, then technically i dont need Pub-Sub implementation in SDK. Any application (Producer/Consumer) of MQTT or AMQP broker, can easily do the job. Please correct me if i am misinterpreting something. 
 
The immense complexity means it's possible it would require more of my time to support people trying to use the implementation than it would to implement it in the first place. Since I'm not being paid for my time or support that doesn't really sound like a great proposition to me :)
[Sumit] - I am able to produce/consume existing message (for e:x: DataValue) on activeMQ based JMS broker. This was done to perform some POC. It was not as straight forward as i had thought of as i had to use Encoder/Decoder in order to transmit messages via ByteMessage. Using ObjectMessage/TextMessage was out of scope anyway.  


On Mon, Jul 24, 2017 at 9:17 PM, Sumit Aggarwal <sumit.kaggarwal@xxxxxxxxx> wrote:
Hi Team,
One more question - Since eclipse-milo not using OPC UA stack provided by OPC foundation, typically how much time it takes you guys to update library in-line with OPC UA foundation stack.
A hypothetical question - How long it may take to come out with Pub-Sub support once OPC UA foundation releases final specs.

Though i am inclined to use ProSys but eclipse-milo open source would be major reason to go with Eclipse-Milo instead of ProSys. So far time spent on both libraries, features wise both looks similar. Our use case is Pub-Sub instead of Client-Server functionalities.


On Sat, Jul 22, 2017 at 9:24 PM, Kevin Herron <kevinherron@xxxxxxxxx> wrote:
> In the mean while can you answer why eclipse-milo using CompletableFuture so heavily.

Milo was built from the ground up to be async/non-blocking. It's also built on Java 8, which introduced CompletableFuture, a well implemented and nicely composable representation of an async operation/result. So it only seems natural that Milo uses this so heavily.


> Anyways, i am evaluating eclipse-milo and ProSys and so far frankly speaking i am inclined towards ProSys. I would like to reverse this position if we can have a 1-on-1 discussion by any means and have better picture to change my position.

I'll do my best to answer any questions you have here on the mailing list or on gitter.im but I'm not going to try to sell you on using Milo instead of Prosys. 


On Sat, Jul 22, 2017 at 8:42 AM, Sumit Aggarwal <sumit.kaggarwal@xxxxxxxxx> wrote:
Hi Kevin,
I would have love to buy you a beer but we are very far :) and taking a flight for one beer is not very economical business. 

Anyways, i am evaluating eclipse-milo and ProSys and so far frankly speaking i am inclined towards ProSys. I would like to reverse this position if we can have a 1-on-1 discussion by any means and have better picture to change my position.

In the mean while can you answer why eclipse-milo using CompletableFuture so heavily.

 -Sumit

On Tue, Jul 18, 2017 at 7:16 PM, Kevin Herron <kevinherron@xxxxxxxxx> wrote:
The license that the foundation stack is available under is one of the big reasons we can't use it.

If you asked me over a beer I'd also have some other things to say about the foundation stack :)

On Tue, Jul 18, 2017 at 5:44 AM, Ian Skerrett <ian.skerrett@eclipse-foundation.org> wrote:
To clarify, the OPC-UA source code is available to non-members under the GPL v2 license. https://opcfoundation.org/license/source/1.11/index.html

Eclipse Milo is available under the Eclipse Public License (EPL) v1.  

In general, EPL makes it easier to include the code into commercial products.

Ian

On Tue, Jul 18, 2017 at 8:05 AM, Lothar Kimmeringer <job@xxxxxxxxxxxxxx> wrote:
Hi,

Am 18.07.2017 um 13:04 schrieb Sumit Aggarwal:

Any reasons why OPC UA stack by OPC foundation is not being used.

I'm not the one who decided this but the short answer is: The license
the OPC UA stack is available.

Also if above statement is true, any plans to use OPC UA stack in
eclipse-milo instead of home grown stack ?

I hope not since it's the license this project is using why I'm using
Milo instead of OPC UA.


Cheers, Lothar
_______________________________________________
milo-dev mailing list
milo-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/milo-dev



--
Ian Skerrett
VP of Marketing
Eclipse Foundation

Twitter: @ianskerrett

_______________________________________________
milo-dev mailing list
milo-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/milo-dev



_______________________________________________
milo-dev mailing list
milo-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/milo-dev




--
Cheers....
Sumit

_______________________________________________
milo-dev mailing list
milo-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/milo-dev



_______________________________________________
milo-dev mailing list
milo-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/milo-dev




--
Cheers....
Sumit

_______________________________________________
milo-dev mailing list
milo-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/milo-dev



_______________________________________________
milo-dev mailing list
milo-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/milo-dev




--
Cheers....
Sumit
9818621804

_______________________________________________
milo-dev mailing list
milo-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/milo-dev


Back to the top