Hi Jordi,

> 
>    Hi Jan,
> 
>    Great doc, and very readable. Thanks to all the authors for their input. 
>    I went over the doc with the eyes of a newbie to this stuff and found some 
> areas for clarification.
> 
>    In section 3, we mention "end-user" for the first time, without explaining 
> what an end-user is in this context.. In our courses we have to make that 
> explanation sooner than later, because different audiences read different 
> things in the word “end-user”.
> 
> ⇒ Can’t find “end-user” in the current version. I think we used end-customer 
> and the goal was always “residential/household end-users”
> 
>    The abbreviation of CPE is nowhere explained or fully written.
> 
> ⇒ Right, some other documents use CE. Is the same (Customer Edge or Customer 
> Premises Equipment). We can point here to the definition at RFC7084, because 
> this document is being updated, so if we need to clarify that (don't think so 
> is the case, right now, but just in case), we could take the opportunity to 
> do so …

I talked to Jan, don’t think this is needed, if we explain the intended 
audience better. “operators” was a bit too vague. By explaining that this 
document is for an entity/organisation that provides internet connectivity to 
residential and business customers, the problem is solved. 

> 
>    Can we provide a link to an explanation of the Neighbor Discovery 
> exhaustion attack?
> 
> ⇒ RFC6583 maybe
> 
>    3.1.3
>    where we mention “This method may be seen as easier to implement, but it 
> also brings some drawbacks such as difficulties with troubleshooting”, can we 
> make it a bit more explicit by stating that link-local addresses don’t appear 
> in a trace route, for example?
> 
> ⇒ Can write some text, of course …    

Yes, Jan took care of that, I believe.

> 
>    3.2.1
>    “It should be remembered that some mechanisms use a default /48 prefix 
> size “ 
>    Do we have examples?
> ⇒ Yeah, users of tunnel brokers (some delegate /48), 6to4, ULA, all them use 
> by default /48.    

ULA is not recommended you wrote, 6to4 is on the way to be deprecated and 
tunnel brokers are not the intended audience of the doc, right? I would just 
leave this out….

> 
>    3.2.3.
>    I would make 
> 
> 
>    “There is a clear exception to this rule when assigning addresses in a 
> cellular network. I n this case a /64 will need to be provided for each PDP 
> context for cellular phones, whereas for LTE modems/routers it will still be 
> necessary to choose a /48 or /56, in accordance with the aforementioned 
> considerations.“
>    the paragraph after “assigning a /64 or smaller prefix is highly 
> discouraged”
> 
> ⇒ Ok.    
> 
>    4.
>    “Static assignment means that a prefix is assigned to a customer 
> (typically an AAA) “
>    What is an AAA? Try to explain abbreviations.
> ⇒ I’m starting to think that the target of the document is not clear. AAA is 
> something obvious for any network operator. The document doesn’t target 
> end-users … or I got something wrong? AAA -> Authentication, Authorization 
> and Accounting    

The target audience was indeed not explained well at the beginning ;)


> 
>    4.1
>    “The easiest way method” remove way or method.
> ⇒ Ok.   
> 
>    4.2 
>    “If the CPE knows that the delegated prefix has changed it should send out 
> RA packets”
>    Write Router Advertisements. 
> ⇒ Same as above, but we can write the complete text each time we use each 
> abbreviation.

That is good practice in writing documentation, the first time write the word 
out full, later use the abbreviation. 

> 
>    Did the authors consider images to explain certain concepts? For example 
> bit boundary? RIPE NCC can help with that, if you wish.
> 
> ⇒ I’m fine and happy with that support, but again, who is the target for the 
> document? Need that target all this?

Like I said, don’t know. Wonder how the rest feels…

Cheers,
Nathalie


Reply via email to