The table is good. The comments/explanations need more work. Did you
want us to review and comment on these now or are you still refining
them?
-Mike-
Rich Thompson wrote:
I like the idea of adding a table
describing
the different types of state. I think another describing the different
coordination mechanisms would also be useful. I have made a start on
two
such appendices (I think an appendix is the right place as it is really
commentary on a particular portion of the spec rather than definitional.
Thoughts? comments?
Rich
Rich,
Yes, we should clarify its use. I just finished a conversation
with some folks in my development team where they completely missed
needing
to use navigational parameters as inputs to internal state. Rather
they equated them more with the current navigational state concept
which
holds state. One of the developers suggested providing a table that
includes all the mechanisms and lists such factors as whether it
represents
managed state or not, whether its for interportlet comm or not, etc.
-Mike-
Rich Thompson wrote:
By the way, sections numbers are always handy ... your page 29 resolves
to section 5.1.17 on page 30 for me.
My comments:
a. The TC decided that "it is always
a Consumer choice to supply them (navigationParameters) on a
particular
invocation". The direct fallout
of this is that the navigationParameterDescription is not allowed to
require
supplying if a particular item. This results in a model where the
Consumer
only supplies new values when they occur and the Portlet. Perhaps what
is missing is a more concrete description (both here and in 6.1.11)
that
navigationParameters are simply an input which the Portlet SHOULD then
store in its navigational state.
b. This sentence is providing guidance about how navigationParameters
are
shared between portlets, but is explicitly noting that the guidance
likely
doesn't apply when two portlets customized from the same POP within the
sharing scope (likely the page).
Rich
Two questions/comments on the navigationParameterDescriptions (page 29
of draft 12):
a. There is a statement that says that
"As such, regardless of the values in the PropertyDescription‘s
capabilities field, navigationParameters are always modifiable and
optional."
Does not this weaken the notion of wsrp:required capability?
b. The intent of the next sentence is not clear to me.
"While Consumer policy governs what navigation parameters are supplied
to each invocation of a Portlet, Consumers SHOULD supply the same value
to Portlets which are related to distinct Producer offered portlet
handles and provide a navigationParameterDescription referencing the
same QName for the parameter’s name."
I think what we want to say is that the consumer should try to supply
the same value to any portlet with a matching name as specified in
navigationParameterDescription. How about changing this text to the
following?
"While Consumer policy governs what navigation parameters are supplied
to each invocation of a Portlet, Consumers SHOULD supply the same value
to any portlet with the same QName for the parameter's name as
specified
in the portlet's navigationParameterDescription."
Subbu
---------------------------------------------------------------------
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
---------------------------------------------------------------------
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
|