Johannes, Morris and all, Short summary ------------- I fully maintain my position on following technical issue : The OGSA-based 'PGI Use Case Template' is trash. Forget it. I propose a 'PGI Software Engineering Use Case Template' (based on Wikipedia work) at http://forge.gridforum.org/sf/go/doc16024?nav=1 Do you like to waste your time kidding, or are you ready to invest your work into professional Software Engineering ? Detailed argumentation ---------------------- I fully maintain the argumentation provided in my mail below dated 12 July 2010. In addition : I always spend efforts and time to understand the work of other persons, and I usually spend efforts and time to provide constructive comments and proposals in order to improve it, so that we all together win. In particular, I assess that the GROMACS Use Case, though incomplete, contains enough useful information to be improved, and could be used as a basis for the creation of an acceptable template. But the OGSA-based 'PGI Use Case Template' puts focus on concepts which must hidden following the OO principle of encapsulation, and forgets the main concepts of 'Use Cases' : System, Actors, Interactions. Therefore, I most strongly assess that this OGSA-based 'PGI Use Case Template' is very bad, unsuitable, inappropriate, and blocking our work. Using such bad foundation, any structure that we would build would crash, we would fall inside the mud and be drowned. As an example, see my comments inside 'gLite Use Case : Parallel Jobs' at http://forge.gridforum.org/sf/go/doc16027?nav=1 The OGF PGI telephone conference using GoToMeeting on Thursday 15 July 2010 had bad audio quality. In fact, there was NO agreement to use the OGSA-based 'PGI Use Case Template'. Finally : - Writing down this OGSA-based 'PGI Use Case Template' was a simple technical error. - Continuing to pretend, after having read my mail dated 12 July 2010, that this OGSA-based 'PGI Use Case Template' is suitable for Use Case capture, is a professional error. - Requiring that OGF PGI-WG must continue to use this bad OGSA-based 'PGI Use Case Template', just because someone began to use it, is obscurantism. I suggest that each member of PGI-WG studies this critical issue, seeks advise of professionals working daily on Use Case capture, makes his/her own mind, and presents his/her position. Best regards. ----------------------------------------------------- Etienne URBAH LAL, Univ Paris-Sud, IN2P3/CNRS Bat 200 91898 ORSAY France Tel: +33 1 64 46 84 87 Skype: etienne.urbah Mob: +33 6 22 30 53 27 mailto:urbah@lal.in2p3.fr ----------------------------------------------------- On Mon, 12/07/2010 23:03, Johannes Watzl wrote:
Dear Etienne, Mark,
thanks for your comments. I will discuss with Morris and we will come back to you. Maybe we put it on the agenda for the Thursday call.
Best, Johannes
Am 12.07.10 19:36, schrieb Mark Morgan:
Having already put the effort into writing up the Genesis II use cases in the OGSA Use case template format, I'm not thrilled with the prospect of doing so again in a different format. I know that the use case template went out not too long before the use cases were supposed to be due, but if we are going to have this type of discussion about things every time, then we need to take more time deciding and voting on things like templates ahead of time and not rush the process.
-Mark
On Jul 12, 2010, at 11:49 AM, Etienne URBAH wrote:
Johannes, Morris and all,
Short summary ------------- The OGSA-based 'PGI Use Case Template' is trash. Forget it.
I propose a 'PGI Software Engineering Use Case Template' (based on Wikipedia work) at http://forge.gridforum.org/sf/go/doc16024?nav=1
Are you ready to invest your work into professional Software Engineering ?
Detailed argumentation ---------------------- As promised, I have studied the 'PGI Use Case Template' based on the OGSA template.
Regrettably, this OGSA-based template :
- puts the focus on 'Customers', which are very specific roles, and and more useful in 'for-profit business' than in 'scientific data processing',
- puts the focus on 'Scenarios', which is a plural, whereas best practice is to describe in detail only the primary scenario,
- puts the focus on 'Applications', whereas for Service Grids, an Application is just one of the several files given as input to an Activity,
- puts the focus on 'Involved resources', which are internal to the system, and MUST STAY HIDDEN following the very basic OO principle of ENCAPSULATION,
- forgets to request the definition of the SYSTEM and the description of its BOUNDARIES, whereas such definition and description is a core issue for Distributed Data Processing,
- forgets to request the definition of preconditions (apart security, which is mentioned only after the scenario(s) which uses the security context).
So, this template is perhaps suited for dissemination and for submission of 'papers', but NOT AT ALL for professional Software Engineering.
This is perhaps one of the causes of the very moderate success of OGSA.
Therefore, writing comments inside this OGSA-based template in the hope of improving it would be a complete waste of time.
Instead, I have used http://en.wikipedia.org/wiki/Use_case as basis to write my own proposal, which I call 'PGI Software Engineering Use Case Template', and is available at http://forge.gridforum.org/sf/go/doc16024?nav=1
The goal of this template is professional Software Engineering, permitting technical understanding, technical criticism and technical improvement. Therefore, the focus is NOT AT ALL on the scientific applications, but on the precise definition of : - The boundaries of the system for which the 'Use Case' captures one behavior, - The list of actors, stakeholders, participants, roles, - The detailed description of the primary scenario of interactions between the actors and the system.
The Use Cases which I already published as simple text already contain most of the information required by this professional Software Engineering template. I will easily reformat them accordingly.
This 'PGI Software Engineering Use Case template' is to be improved. Do not hesitate to provide comments and suggestions.
Best regards.
----------------------------------------------------- Etienne URBAH LAL, Univ Paris-Sud, IN2P3/CNRS Bat 200 91898 ORSAY France Tel: +33 1 64 46 84 87 Skype: etienne.urbah Mob: +33 6 22 30 53 27 mailto:urbah@lal.in2p3.fr -----------------------------------------------------
On Thu, 08/07/2010 19:45, Etienne URBAH wrote:
OGF PGI telephone conference on 2010-07-08 ------------------------------------------
...
Everyone can add comments to the template for Use Cases http://forge.gridforum.org/sf/go/doc16013?nav=1 - Etienne will add some comments
Following people should upload their Use Cases following the template to GridForge http://forge.gridforum.org/sf/docman/do/listDocuments/projects.pgi-wg/docman...
: - Emmanouil for Globus - Luigi for gLite - Morris for Unicore
Following people should convert their Use Cases according to the provided template : - Etienne for EDGI - Aleksandr for ARC
Next OGF PGI telephone conference on Thursday 2010-07-15 at 15h30 CET