[Sois-tcons] Re: SOIS Intra-Network Service

[Jane K. Marquart]

Steve,

I've attached an editted document with my comments, etc.  I didn't know how 
you wanted it so I accepted all changes in your original.  There is too 
much so this shouldn't be too oppressive.
General comments:
-  We need a managment section (4.0?) that I believe Rick is working on?  I 
guess it could also be a separate document, but this seems like an 
incomplete story without it.
-  Where are we going to define the channel set-up, binding, etc?  Is that 
part of the managment?  Or is that part of the user protocol and outside 
our scope?
-  I agree we need "failure metadata" info.  I think that was one of my 
document comments too.
-  I await Dia's answers to your questions, because I have the same.

Do we want to set up a telecon next week to discuss?   Maybe Thursday 10/20?

Jane
At 10:08 AM 9/27/2005 +0100, Steve Parkes wrote:

>Dear All,
>
>
>
>The Intra-Network Service document was still in something of a mess at the 
>end of the Atlanta meeting&
>
>I have thoroughly edited the book cutting out all the stuff that was not 
>at all relevant, writing missing sections and making everything 
>consistent. This has taken a while because much of the material was from 
>our original transport/network layer document. Anyway we now have what I 
>think is a reasonable draft. Note that I have tracked changes just so you 
>can see the changes from the Atlanta version. I suggest that before making 
>modifications or comments you accept all changes, otherwise we will end up 
>in a Microsoft mire.
>
>
>
>Dai,
>
>I understand that you have some funding from ESTEC to work on the TCONS 
>red books. Would you please read through the Intra-Network Service red 
>book and let me know if the service interfaces are ok and if you have any 
>other comments? Are the primitives sufficient? What about registering the 
>receiving protocols (End Systems) so that they can be passed the data? 
>What about identifying the send (e.g. with a transaction ID) so that the 
>sender knows which error is for which send? Is this just assumed in the 
>primitives or do we need to specify it explicitly? Do we need to include 
>some sequence diagrams?
>
>
>
>Greg, Jane,
>
>Would you please read through and let me have comments? Would you then 
>please start work on the API book which should provide the API for the 
>four Intra-Network services? This will raise many questions that we have 
>not addressed, like do we register a protocol to receive data from the 
>intra-network service. My initial thought is that we should make the API 
>as close as possible to the previous one that we had for the TCONS 
>transport and network layer. Try to work it through and find our what the 
>questions are. We can then have a teleconference to discuss the particular 
>issues.
>
>
>
>Rick,
>
>Would you also please let me have your comments? I think that the document 
>is now as you intended. One thing that I would appreciate you working on 
>is the Failure Metadata . In your original work on the Intra-Network 
>service you had gone into substantial detail on this, but it was not put 
>into the red book. I think that it would be good to add it in.
>
>
>
>Max,
>
>Your comments would be very welcome too.
>
>
>
>I need everyone s input on this by the middle of October (17th) so that I 
>can put all the comments together and have the red book ready to send out 
>to SOIS etc by the start of November. The API book can be finished after 
>this, but the main thing is to have made a good start on it so we know 
>that it will fit in with the Intra-Network Service book and will not need 
>changes to the Service book.
>
>
>
>Regards
>
>
>
>Steve
>
>
>
>------------------------------------------
>
>Steve Parkes
>
>Space Technology Centre
>
>Applied Computing, University of Dundee
>
>Dundee, DD1 4HN, Scotland, UK
>
>Tel: +44 1382 345194
>
>Fax: +44 1382 348838
>
>Mobile: +44 784 138 3779
>
>
>
>--
>This message has been scanned for viruses and
>dangerous content by <http://www.mailscanner.info/>MailScanner, and is
>believed to be clean.
>MailScanner thanks <http://www.transtec.co.uk/>transtec Computers for 
>their support.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TCONS-Intranet_ Service_rgs_9_26
	Parkes_jkm.doc
Type: application/msword
Size: 457728 bytes
Desc: not available
Url : http://mailman.ccsds.org/pipermail/sois-tcons/attachments/20051013/2c77dbf3/TCONS-Intranet_Service_rgs_9_26Parkes_jkm-0001.doc
-------------- next part --------------
Jane