My 2c
Firstly I agree with Mike that there
are valid use cases, as he describes, where this is a problem, and the
current rules don't allow for a consumable solution (so my vote goes with
it being a valid issue).
If you don't think we should debate
issues that are not accepted, please avert your eyes now ;-)
My preference is to treat the SCA binding
as "always there" rather than "don't care".
If I was allowed to wire from a reference
with a non-SCA binding to a service with an SCA binding, and that meant
that the non-SCA binding was used, then the SCA runtime has to somehow
generate an implicit non-SCA service binding that matches the reference
binding. In the general case I don't think that's possible. For
JMS, for instance, just because I know how a reference is going to send
a JMS message does not mean I can deduce how a service is going to receive
that JMS message. And what about a JCA binding? Also the runtime
that is dealing with the reference may just not support the given service
binding.
So I'd prefer it that in this case the
SCA binding is used, unless there are intents present that require a specific
binding, in which case the wire is not allowed. I believe we should
always be able to derive an implicit SCA binding for the other end of a
wire. I don't believe we can do that for non-SCA bindings.
Regards, Simon
Simon Holdsworth
STSM, SCA Bindings Architect; Master Inventor; OASIS SCA Bindings TC Chair
MP 211, IBM UK Labs, Hursley Park, Winchester SO21 2JN, UK
Tel +44-1962-815059 (Internal 245059) Fax +44-1962-816898
Internet -
Mike Edwards/UK/IBM@IBMGB
28/11/2007 10:20
To
"OASIS Assembly" <>
cc
Subject
Re: [sca-assembly] [NEW ISSUE] Wiring
from a reference with no binding to a service with a binding
Simon,
You're perfectly entitled (as is anyone else) to take the view that a new
issue is not valid for
any reason. That's the thinking behind the process whereby anyone
can raise an issue but
it has to go through a vote before it gets opened.
Anyhow, to the merits of the current issue...
let me try to provide some motivation for the issue, even though I'm not
the raiser of the original
issue, but I do see the thinking behind it.
One of the ideas within SCA is the simplification of the assembly task
- simplifying the connection
of some reference to some service. Binding.sca is one of the tools
used to bring simplification.
It basically says "look, I want to connect THIS to THAT and I don't
particularly care how it is done
- just make the connection and satisfy any intents I've specified"
Dave Booz has always taken the view (which I agree with) that this is the
80% case when doing
Assembly. Only when you're really dealing with the "outside
world" is there a strong need to
get into concrete bindings and the associated specialized endpoint addresses
and other
paraphernalia.
Now there are cases where you can end up with mixtures of bindings. So,
we may have at the
domain level a component A with service S1 that needs to be available externally
over the
Web Services binding. So S1 is configured that way. Along comes
component B with
reference R1 - and the assembler decides to connect that reference to S1
as well.
Currently, for R1, we would have to configure it with a Web Services binding,
with an endpoint
URL pointing to the web service endpoint of A/S1. This is a lot more
complex than simply
stating @target="A/S1" on R1. It is also harder to parse
after the fact.
So, that's the real world case, I think. The argument is being made
that it is a great simplification
for the Assembler NOT to have to go deal with the complexities of the web
service binding.
Stating binding.sca and the SCA target address should be enough. The
SCA runtime can
work out the rest - yes, it would probably use web services under the covers
(at least, if the
components lived on different machines).
Another way of stating this might be to say "binding.sca is ALWAYS
present", except that makes
me uncomfortable for the cases where intents are used to force the use
of particular bindings
where the component code is written using binding-specific APIs. I
prefer to go along with the
idea that it is possible for binding.sca to be "coerced" to the
binding of the other end of the wire.
So, I think the issue is based on:
a) current spec is more complex than it needs to be
b) there is a simpler apporach for the assembler - and the proposal states
it
Yours, Mike.
Strategist - Emerging Technologies, SCA & SDO.
Co Chair OASIS SCA Assembly TC.
IBM Hursley Park, Mail Point 146, Winchester, SO21 2JN, Great Britain.
Phone & FAX: +44-1962-818014 Mobile: +44-7802-467431
Email:
Simon Nash/UK/IBM@IBMGB
28/11/2007 08:35
To
"OASIS Assembly"
<>
cc
Subject
Re: [sca-assembly] [NEW ISSUE] Wiring
from a reference with no binding to a service with a binding
This issue raises a general point about what justification is needed in
order to open an issue.
In general an issue should say that something is wrong and describe what
the problem is. It may also give a proposal for a change to resolve
the
problem. The "something is wrong" could be an error or
inconsistency in
the spec, or it could be a real world use case that the spec does not
address adequately. I don't think an issue should be raised based
on an
opinion that says "I don't like spec rule X and I prefer rule Y instead."
With this issue I don't see any clear description of what is wrong with
the current spec rules. The spec appears to be clear and consistent
as it
stands. All I see is an assertion that the rule in the current spec
imposes an "unreasonable requirement".
Why is this spec rule unreasonable? What real world use case has
a
problem because of this rule? IMO the rule is fine as it stands,
and no
change is needed. However, I'm prepared to be educated on this, and
if
the rule is causing a problem, then we we should look at the problem and
see what spec change needs to be made to resolve the problem. The
solution could be the one proposed, or there could be a different
solution. Without a clear problem statement and use case to drive
the
discussion, the debate will come down to one person's technical opinion
against another's, which is unlikely to be productive.
Simon
Simon C. Nash, IBM Distinguished Engineer
Member of the IBM Academy of Technology
Tel. +44-1962-815156 Fax +44-1962-818999
Mike Edwards/UK/IBM@IBMGB
27/11/2007 15:42
To
"OASIS Assembly" <>
cc
Subject
[sca-assembly] [NEW ISSUE] Wiring from a reference with no binding to a
service with a binding
<This issue is transferred from the SCA Policy TC where it was dubbed
POLICY-34>
RAISER Michael Rowley (original)
TARGET: SCA Assembly Specification
DESCRIPTION:
The algorithm in the policy spec says that it is _not_ possible to wire
from a reference that does not declare a binding (i.e. uses binding.sca)
to a service that declares one or more bindings. However, I think this
should be possible.
It is an unreasonable requirement to say that a service with a binding
can
only be the target of a reference that has that same binding. The default
binding (binding.sca) should be treated as the "I don't care"
binding, and
should work with any binding available within the domain. Or, more
precisely, any binding that can satisfy the required intents.
Section 4.8.1 of the Policy frmework spec states:
The wiring compatibility algorithm below determines the compatibility of
a
wire by a pairwise choice of a binding instance and the corresponding
policySets on each side of the wire.
This should be changed to the following:
If either side of a wire does not specify a binding (or explicitly
specifies binding.sca) the wire is considered to be valid for the purposes
of policy processing. If both sides of the wire use binding.sca then the
policies will be determined by the union of the required intents of both
sides (policy sets aren't used with binding.sca). Otherwise, the bindings
and policies used for the wire will be determined by adding the intents
that are required by the binding.sca end of the wire to the other end of
the wire and then following the section 4.10 algorithm in the Polcy
Framework.
If neither side of the wire uses binding.sca, then the wiring compatibilty
algorithm below is used for determining compatibility. Note that there
may
be many binding instances present at each side of the wire. This algorithm
determines the compatibility of a wire by a pairwise choice of a binding
instance and the corresponding policySets on each side of the wire.
PROPOSAL:
The following should be added to the Wires section of the Assembly
specification:
If either end of a wire does not specify a binding (or explicitly
specifies binding.sca) the wire is regarded as valid. In other words,
binding.sca is regarded as being compatible with
any other type of binding. Where other types of binding are applied
to
each end of a wire, compatibility of the two bindings is determined by
the
specifications for the two bindings
involved, allied to the intent and policies attached at each end. In
general, a wire which has two different binding types at each end (non
binding.sca) is likely not to be valid.
If both ends of the wire use binding.sca then the policies will be
determined by the union of the required intents of both ends (policy sets
aren't used with binding.sca).
Otherwise, where one end of the wire uses binding.sca, the bindings and
policies used for the wire will be determined by adding the intents that
are required by the binding.sca end of the wire to the other end of the
wire and then following the algorithm defined in the Policy Framework
specification section 4.10.
If neither end of the wire uses binding.sca, then the wiring compatibilty
algorithm described in section 4.10 of the Policy Framework specification
is used for determining compatibility. Note that there may be many binding
instances present at each side of the wire. This algorithm determines the
compatibility of a wire by a pairwise choice of a binding instance and
the
corresponding policySets on each side of the wire.
Yours, Mike.
Strategist - Emerging Technologies, SCA & SDO.
Co Chair OASIS SCA Assembly TC.
IBM Hursley Park, Mail Point 146, Winchester, SO21 2JN, Great Britain.
Phone & FAX: +44-1962-818014 Mobile: +44-7802-467431
Email:
Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number
741598.
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6
3AU
Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number
741598.
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6
3AU
---------------------------------------------------------------------
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
Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number
741598.
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6
3AU
Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number
741598.
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6
3AU
