Re: [wsrp] Public parameters conformance statement?

[Rich Thompson <richt2@us.ibm.com>]

The sentence following the one leading
to this discussion reads "In the
absence of a Consumer supplied value for a public parameter the Portlet
listed in its portletDescription,
the Portlet SHOULD use a default value".
The intent of this conformance statement is to cause Portlets to have some
fallback for when a public parameter is not supplied by the Consumer. It
also allows the Consumer to distinguish between the full set of values
(including null) and using a default value.

Rich

Subbu Allamaraju <subbu@bea.com>

05/31/05 02:48 PM

To	wsrp <wsrp@lists.oasis-open.org>
cc
Subject	Re: [wsrp] Public parameters conformance statement?

Rich Thompson wrote:
> 
> Would you find it better if the sentence read "In general, Portlets
do 
> not store these items in any portion of their state as the Consumer
> supplies the items on each invocation"?

That sounds better, but I'm still not clear on the intent. It sounds 
speculative, and does not seem to add value.

On a related note, since the Consumer may or not may supply all/some 
public parameters with each request, shouldn't we add a 
statement/guideline that portlets be prepared to work reasonably well 
even when those parameters are not supplied?

Subbu


> I don't think conformance language is appropriate as we can think
of a 
> case where storing the last public parameters values does make sense
> (use as default value for any non-supplied item). On the other hand,
> unless we think Portlets should generally manage defaulting value
in 
> that manner, I wouldn't call it out in the spec either.
> 
> Rich
> 
> 
> *Subbu Allamaraju <subbu@bea.com>*
> 
> 05/31/05 10:10 AM
> 
>                  
> To
>                  wsrp
<wsrp@lists.oasis-open.org>
> cc
>                  
> Subject
>                  Re:
[wsrp] Public parameters conformance statement?
> 
> 
>                  
> 
> 
> 
> 
> 
> Just to clarify, I find the sentence "Portlets do not store these
items
> ..." a bit unclear. Are we saying that "portlets don't generally
store
> these parameters ...", or are we saying that "we don't recommend
that
> portlets store ...".?
> 
> I suggest that we tigheten the language either with a conformance
statement
> 
> "Portlets MAY NOT ..."
> 
> or a with a guideline
> 
> "We recommend that Portlet developers do not store ... for such
and such
> reasons."
> 
> Regards,
> 
> Subbu
> 
> Rich Thompson wrote:
>  >
>  > Subbu raised a question about changing "Portlets do
not store these
>  > items in any portion of their state as the Consumer supplies
the items
>  > on each invocation. In the absence of a Consumer supplied
value for a
>  > public parameter the Portlet listed in its portletDescription,
the
>  > Portlet SHOULD use a default value" to say Portlets
either SHOULD NOT or
>  > MAY NOT store these items ...
>  >
>  > I considered this when adding this text, but decided against
it as the
>  > one reasonable use case I could imagine for a Portlet storing
the
>  > current public parameters was to use them as the default
values should
>  > the parameter not be supplied on subsequent invocations.
Do people think
>  > we should leave this as-is or make it a conformance statement?
>  >
>  > Rich
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe from this mail list, you must leave the OASIS TC that
> generates this mail.  You may a link to this group and all your
TCs in OASIS
> at:
> https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php
> 
> 


---------------------------------------------------------------------
To unsubscribe from this mail list, you must leave the OASIS TC that
generates this mail.  You may a link to this group and all your TCs
in OASIS
at:
https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php