Hi Anuj,
thank you for your thorough review. See below (ME: ).
1.3 Network Types and Characteristics in Focus
----------------------------------------------
(a) "Mobile Adhoc networks (MANET) are self-configuring
_infrastructureless_ networks of mobile devices connected by
wireless technologies."
--> Why is a differentiation being made between MANETs and other
types of wireless networks? Couldn't moblie phones technically
fit into the IoT as well? Is the differentiation only made
based on the _infrastructureless_ part of it? If so, does it
make sense to have categories of infrastructure and
infrastructureless, rather than MANET and non-MANET networks?
ME: I can go with your proposal to use infrastructureless networks and give
MANET as an example.
I think Car-to-car networks is another example but this wouldn't be necessarily
constrained.
I think mobile phones shouldn't be counted as constrained devices.
(b) "Wireline non-constrained networks are mainly used for specific
applications like Building Automation or Infrastructure
Monitoring. However, wireline and wireless networks with
multi-hop or point-to- multipoint connectivity are especially in
the interest of the analysis on the management of constrained
devices in this document."
--> This paragraph is unclear. What is it trying to convey? Is it
saying that only wireline multi-hop and point-to-multipoint
are within the scope of this document? Why not regular
wireline networks that may have constrained devices as part of
a non-constrained network?
ME: By saying they are "especially in the interest of the analysis the text
does not exclude the others.
Non-constrained wireline networks with constrained devices are in scope (e.g.
Building automation).
--> There is talk about bandwidth, but not frame size, which is
likely to have a bigger impact than the overall bandwidth
available in a channel. Maybe this should be mentioned?
ME: In section 3.11. Implementation Requirements we talk on protocol message
size.
Though we don't talk on frame size
1.6. Managing the Constrainedness of a Device or Network
---------------------------------------------------------
(a) "might only be able to support one simple communication protocol,
i.e. the management protocol needs to be possible to downscale
from constrained (C2) to very constrained (C0) devices"
--> Is it important that the same management protocol be
applicable to the entire range of constrained devices? Why not
allow the capabilities of the management protocol to scale up
and down with the type/capabilities of the device at hand?
SOLUTION: A new bullet point stating that different management
protocols work on different class of devices.
ME: It is for sure possible that different management protocols work on
different class of devices. Though it is desirable that the protocol
is prepared for downscaling enabling the use of same type of message and
information exchange as well as easy mapping of protocol commands.
(b) "might only be able to support limited or no user and/or transport
security" AND "might only be able to support very simple
encryption"
--> Simple encryption doesn't make sense, maybe it should be
efficient encryption instead?
ME: I can agree with "efficient for memory and CPU usage". Efficient
only does not explain what we want to say.
(c) "might also need energy-efficient key management algorithms for
security."
--> Not just energy-efficient, but also memory, processing and
etc. efficient. No?
SOLUTION: Change text to only efficient key management.
ME: I think it needs to described what kind of efficiency we mean.
Cheers,
Mehmet
> -----Original Message-----
> From: ext Sehgal, Anuj [mailto:[email protected]]
> Sent: Saturday, January 04, 2014 3:28 AM
> To: [email protected]
> Cc: Ersue, Mehmet (NSN - DE/Munich); Schönwälder, Jürgen
> Subject: Re: Review of draft-opsawg-ersue-coman-*-00
>
> Hi,
>
> As per the call for reviewing the COMAN documents, I have taken a look at the
> latest
> versions and have a few comments/recommendations for both. The reviews, along
> with suggestions on fixing them, are attached to this email in appropriate
> files.
>
> Regards,
> Anuj Sehgal
_______________________________________________
OPSAWG mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/opsawg
