[dns-operations] Oh, What a Fragile Web We Weave: Third-party Service Dependencies In

[Shane Kerr]

Witold,

On 26-06-18 18:45, Witold Krecicki wrote:
> On 26.06.2018 18:12, Jim Reid wrote:
>>
>>
>>> On 26 Jun 2018, at 16:52, Andrew Sullivan <ajs at anvilwalrusden.com> wrote:
>>>
>>> An important observation in here is
>>> that, in the concentrating-services Internet economy, people have
>>> dependencies on others that they don't even know about.  In the
>>> physical world, people have often specified redundant transit only to
>>> find themselves hosed when all their redundancy traverses a common
>>> peering point somewhere upstream.  This result is a similar one, only
>>> in a much more ubiquitous (and, sometimes, fragile) way.
>>
>> We are in violent agreement Andrew.
>>
>> All I was saying is this is hardly news. Or shouldn't be. And of course it's increasingly difficult to fully evaluate or audit the level of redundancy is as expected. That shouldn't surprise anyone either.
 >>
> There's a difference between gut feeling and a proper research.

Yes indeed.

Also, we should respect science because of the questions it asks, not 
because of the answers that it gets to those questions.

Checking a widely-held belief seems like valuable research, regardless 
of whether the belief is confirmed or not. Especially if the research is 
more nuanced and contains things a taxonomy and analysis of that 
breakdown, like in this paper.

Cheers,

--
Shane