does application (using SARL) have to be designed with specific methodology?

[tresnaningtiyas purbo]

Hi, I need to build a MAS application, I do interest using SARL to deploy it
Is there any specific methodology for building the application in SARL, or is there any recommendation?

Thanks

[Stéphane Galland]

As most of the general-purpose programming languages, there is not a single methodology for using SARL.

Usually, me and my teams are using a standard agile methodology with several steps that are inspired from our complex-system design methodology ASPECS, when we do not require to use a so complex methodology.

When implementing with SARL, read carefully the document of each function. Indeed, several of them may have a strong impact on the run-time because they are run in parallel to the rest of the application.

Keep in your mind that every "on" block is run in parallel to the other.

From that, you will have to be carefull about the distributed system design of your multiagent system, and its implementation with really parallel features. 

Think event-based implementation.

If you have any question, do not hesitate to come back on this forum.

Stephane.

[tresnaningtiyas purbo]

Thanks for the feedback 👍

[tresnaningtiyas purbo]

Is there any publication that explain ASPECS in detail along with a case study?
I've read the case study about AMP Simulator in ASPECS web site, but I thought the diagram hasn't been explained in detail

[Stéphane Galland]

The reference paper is downloadable at: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.383.2675&rep=rep1&type=pdf

Is it enough for you?

[tresnaningtiyas purbo]

Thanks for the fast answer

I have read the paper and (I am sorry) I still don't understand some part of the diagram
In the POD there is <<concept>> Vehicle and then more feature added in SOD such as Height, Weight, and Speed
But why the relationship between Vehicle and Height, Weight, Speed is association?
Why not using aggregation or composition?

Thank you so much for your help

[galland.stephane]

The model that you have mentionning is a représentation of the ontology of the système. In ASPECS the diagramme is vases on class diagramme with stereotyping related to ontology. Therefore the concept stereotype corresponds to a concept into an ontology.

From a semantical point of view, weight is not a part of a vehicle. So it cannot be associated by a composition. Association is enough to represent the relationship in the example.

Prof. Dr. Stéphane GALLAND

Full Professor of Computer Science and Multiagent Systems 

Co-Founder and Deputy Director of CIAD

Laboratoire Connaissance et Intelligence Artificielle Distribuées

CIAD UMR 7533

Université de Technologie de Belfort-Montbéliard

13 rue Ernest Thierry-Mieg

90010 Belfort cedex, FRANCE

CIAD Site: www.ciad-lab.fr

Web: www.multiagent.fr/People:Galland_Stéphane

Phone: +33 384 583 418 (work office)

Mobile: +33 662 274 442 (private)

-------- Message d'origine --------

De : tresnaningtiyas purbo <tresnanin...@gmail.com>

Date : 15/03/2020 16:47 (GMT+01:00)

À : sarl <sa...@googlegroups.com>

Objet : [SARL] Re: does application (using SARL) have to be designed with specific methodology?

--
You received this message because you are subscribed to the Google Groups "sarl" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sarl+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sarl/785a3121-262b-46f4-a51c-c2d9271bbd67%40googlegroups.com.

[tresnaningtiyas purbo]

Thank you for the explanation

Currently, I am reading the other ASPECS paper titled "The ASPECS Process" 

On Monday, March 16, 2020 at 1:49:00 AM UTC+7, galland.stephane wrote:

The model that you have mentionning is a représentation of the ontology of the système. In ASPECS the diagramme is vases on class diagramme with stereotyping related to ontology. Therefore the concept stereotype corresponds to a concept into an ontology.

From a semantical point of view, weight is not a part of a vehicle. So it cannot be associated by a composition. Association is enough to represent the relationship in the example.

Prof. Dr. Stéphane GALLAND

Full Professor of Computer Science and Multiagent Systems 

Co-Founder and Deputy Director of CIAD

Laboratoire Connaissance et Intelligence Artificielle Distribuées

CIAD UMR 7533

Université de Technologie de Belfort-Montbéliard

13 rue Ernest Thierry-Mieg

90010 Belfort cedex, FRANCE

CIAD Site: www.ciad-lab.fr

Web: www.multiagent.fr/People:Galland_Stéphane

Phone: +33 384 583 418 (work office)

Mobile: +33 662 274 442 (private)

-------- Message d'origine --------

De : tresnaningtiyas purbo <tresnani...@gmail.com>

To unsubscribe from this group and stop receiving emails from it, send an email to sa...@googlegroups.com.