Re: RFC4941bis: consequences of many addresses for the network

[Alexandre Petrescu <[e-mail]>]

Le 23/01/2020 à 15:10, Fernando Gont a écrit :
> On 23/1/20 10:46, JORDI PALET MARTINEZ wrote:
>> Even not just debugging, but the operational complexity when in some
>> scenarios, certain apps need to "audit" which user did what. Using
>> DHCPv6 is not an option, because Android doesn't support it ... and I
>> will not recommend a customer to use "manual IPv6 configuration" :-(
> 
> To the occasional reader or lurker, this might read as IPv6 folks 
> arguing that, after all, having too many addresses wasn't much of a good 
> idea (?) :-)

If both ends each had multiple IP addresses, and they had a mechanism to 
agree on them all while ensuring one doesnt tell more about self than it 
learned from the other - that would be a good idea.

But here we talk about a poor end user that we technical people try to 
empower with many  addresses vs one Server that must have just one 
address so it can be reachable at a well-known, err, address.

It's strange that in the Privacy discussion no-one worries about the 
fact that Servers must have a unique address such as to be reachable at 
a well-known address.

The side effect of that Server Privacy address is the 'cyber squatting' 
- I hear a Name becomes fashionable, I book the name before the others, 
then I sell it for unreasonable amounts of money.  There are so many 
cases today that way that it is a huge business.  Now some Registrars do 
themselves the same job of cyber-squatters (pre-buy domains, and sell at 
huge prices).  It is a business, but it is not right either - it is 
artificial.

So, I think Privacy and cyber squatting are facets of the same and 
unique problem.

And Spam.

Alex

> 
> Thanks,