Andy writes in part:
"We need to create a solution that will be implemented in routers and switches (among other network devices). A complicated, bloated solution with way more stuff than we need will not get implemented by many network equipment vendors."
I respond:
Huh? Saying that this "...is a complicated bloated solution with way more stuff than you need" is unfounded and not easily proven. The bulk of the work is already there in supporting XML in the first place. You haven't even considered options like supporting a subset of SOAP, or just responding to mustUnderstand headers, or a lot of other factors. With respect to WSDL, you could easily create custom implementations of it, rather than relying on bulky automatic code generation products. Plus, the advantage of being standards-compliant and being inherently extensible far outweighs any MINOR increase in footprint.
John
John Strassner
Chief Strategy Officer
Intelliden Corporation
90 South Cascade Avenue
Colorado Springs, CO 80903 USA
phone: +1.719.785.0648
FAX: +1.719.785.0644
email: john.strassner@intelliden.com
-----Original Message-----
From: Andy Bierman [mailto:abierman@cisco.com]
Sent: Wednesday, April 16, 2003 3:49 PM
To: John Strassner
Cc: Allen, Keith; xmlconf@ops.ietf.org
Subject: RE: WSDL
At 03:35 PM 4/16/2003 -0600, John Strassner wrote:
>But Andy, you never answered my question. The combination of WSDL, a
>binding (SOAP, BEEP, whatever) and UDDI offer a number of advantages
>over building a brand new protocol, including already having worked
>through some nasty security issues. So I repeat, why build Yet Another
>Protocol?
See my previous email on the different user communities.
The security issues are not a factor. Except for the identification of principals, the security is taken care of by BEEP.
Advantages for who? Windows application developers or
vendors of network devices? We need to create a solution
that will be implemented in routers and switches (among
other network devices). A complicated, bloated solution
with way more stuff than we need will not get implemented
by many network equipment vendors. Code footprint and
complexity are important factors to consider.
>regards,
>John
Andy
>
>John Strassner
>Chief Strategy Officer
>Intelliden Corporation
>90 South Cascade Avenue
>Colorado Springs, CO 80903 USA
>phone: +1.719.785.0648
> FAX: +1.719.785.0644
>email: john.strassner@intelliden.com
>
>
>-----Original Message-----
>From: Andy Bierman [<mailto:abierman@cisco.com>mailto:abierman@cisco.com]
>Sent: Wednesday, April 16, 2003 3:30 PM
>To: Allen, Keith
>Cc: xmlconf@ops.ietf.org
>Subject: Re: WSDL
>
>At 04:06 PM 4/16/2003 -0500, Allen, Keith wrote:
>
>>OK, I'll see the bet and raise the ante. Could you explain succinctly
>>why we need a new protocol at all? Why instead aren't we defining WSDL
>>messages for different functions, and then using SOAP bindings and UDDI
>>to solve the problem?
>>
>>regards,
>>John
>>
>>
>>
>>Umm. Because that would be too easy?
>
>Since it is so easy, how about if you volunteer to create a WSDL
>version of the XMLCONF protocol so we can compare the XML messages that
>would be used with both approaches? Never mind the transport issues for
>now.
>
>It seems to me there are different user communities for the output of
>the netconf WG, which don't share the same goals. One group wants to create lightweight mechanisms that can be used to leverage existing CLI (training by operators, code by vendors). Another group wants to leverage tools which may already exist for application data exchange, but are not particularly lightweight or aligned with
>
>existing CLI implementations.
>
>Both are valid positions, and I am concerned that we will
>not be able to produce a single solution that is completely satisfactory to both camps.
>
>Andy
>
>
>>
>>
>>I must say I asked my question in response to thinking about how we
>>would go about building applications that would use a NETCONF
>>interface. It would be nice to have a WSDL description we could just
>>feed into a web services toolkit. WSDL is extensible, though, and
>>should allow us to write a BEEP binding (much like we would write a
>>SOAP binding), and then use UDDI. I just didn't want to be the first
>>one to write a BEEP binding for WSDL. I actually sent an e-mail to the
>>chair of the BEEP WG to see if anyone else has done this. Then all we
>>would need is a web services toolkit that supports BEEP. Keeping my
>>fingers crossed...
>>
>>
>>
>>
>>
>>Keith Allen
>>
>>SBC Technology Resources
>>
>>9505 Arboretum Blvd.
>>
>>Austin, TX 78759
>>
>>(512) 372-5741
>>
>><<mailto:kallen@tri.sbc.com>mailto:kallen@tri.sbc.com>kallen@tri.sbc.c
>>om
>>
>>
>>
>>
>>
>>-----Original Message-----
>>From: John Strassner [<mailto:John.Strassner@intelliden.com>mailto:John.Strassner@intelliden.com]
>>Sent: Wednesday, April 16, 2003 3:54 PM
>>To: 'Andy Bierman'; Allen, Keith
>>Cc: xmlconf@ops.ietf.org
>>Subject: RE: charter proposal - rev C
>>
>>
>>
>>OK, I'll see the bet and raise the ante. Could you explain succinctly
>>why we need a new protocol at all? Why instead aren't we defining WSDL
>>messages for different functions, and then using SOAP bindings and UDDI
>>to solve the problem?
>>
>>regards,
>>John
>>
>>John Strassner
>>Chief Strategy Officer
>>Intelliden Corporation
>>90 South Cascade Avenue
>>Colorado Springs, CO 80903 USA
>>phone: +1.719.785.0648
>> FAX: +1.719.785.0644
>>email: john.strassner@intelliden.com
>>
>>
>>
>>
>>-----Original Message-----
>>From: Andy Bierman [<<mailto:abierman@cisco.com>mailto:abierman@cisco.com>mailto:abierman@cisco.com]
>>Sent: Wednesday, April 16, 2003 2:46 PM
>>To: Allen, Keith
>>Cc: xmlconf@ops.ietf.org
>>Subject: RE: charter proposal - rev C
>>
>>
>>
>>At 03:09 PM 4/16/2003 -0500, Allen, Keith wrote:
>>>All,
>>>
>>>It seems to me, based on my admittedly limited knowledge in the area,
>>>that a main output of the working group should be a WSDL description of
>>>the network configuration service. Is this something that needs to be
>>>part of the charter or is this something the WG would decide once it is
>>>underway?
>>
>>This could be discussed by the WG. So far, I have only heard interest
>>in WSDL from one person. If there is enough WG interest in this extra
>>deliverable, then it could be done by the WG at some point. It could
>>also be done separately, outside the WG, as an Informational RFC.
>>
>>
>>
>>>Keith Allen
>>>SBC Technology Resources
>>>9505 Arboretum Blvd.
>>>Austin, TX 78759
>>>(512) 372-5741
>>>kallen@tri.sbc.com
>>>
>>
>>Andy
>>
>>
>>
>>>--
>>>to unsubscribe send a message to xmlconf-request@ops.ietf.org with the
>>>word 'unsubscribe' in a single line as the message text body.
>>>archive: <<<http://ops.ietf.org/lists/xmlconf/>http://ops.ietf.org/lists/xmlconf/>http://ops.ietf.org/lists/xmlconf/>
>>
>>
>>
>>--
>>to unsubscribe send a message to xmlconf-request@ops.ietf.org with the word 'unsubscribe' in a single line as the message text body.
>
>>
>>archive:
>><<<http://ops.ietf.org/lists/xmlconf/>http://ops.ietf.org/lists/xmlconf/>http://ops.ietf.org/lists/xmlconf/
>>>
>
>--
>to unsubscribe send a message to xmlconf-request@ops.ietf.org with the word 'unsubscribe' in a single line as the message text body.
>
>archive:
><<http://ops.ietf.org/lists/xmlconf/>http://ops.ietf.org/lists/xmlconf/
>>