Dear All,
We are researchers in HP Labs and Duke university. We are currently working on a project related to Middlebox modeling and troubleshooting.
We are currently conducting a survey and gathering feedback from operators.
Can you help us by providing some answers? Please feel free to email us if you have any additional suggestions.
https://www.surveymonkey.com/r/5SFP6G8
Thanks!
-Ying
Why do you keep posting the same thing to the list over and over again?
Do you really think the subscriber count changes that much from week to week?
Hello,
I would offer two suggestions:
First, we heard you the first time. One post was acceptable. Twice was
forgivable. This is the fourth time you've posted this exact same
message to the NANOG list. Please stop spamming.
Second, the people one this mailing list are here for their interest
in Internet backbone networks... where stateful middleboxes can't
possibly keep up and thus are not generally used. Most of us operate
ancillary networks where we use middleboxes of one kind or another but
if you're trying to understand middleboxes on the backbone, responses
littered with that data won't help you. And if you're trying to
understand middleboxes on Internet networks in general, you'll get
badly skewed results by only asking here. We are in no respect a
representative population.
Regards,
Bill Herrin
One issue that stopped me dead in your monkeysurvey was that you asked how
many "Middleboxes" I had without telling me what you consider a middlebox.
Then you go into questions that ask me to delve deep into the whitepapers
of how they work. I work with a team that supports about 100 international
locations on a large MPLS network with Palo Alto, Ipanema, Cisco and
homebrew virtual machines. For me to even try to answer your questions the
way you state would require me to schedule meetings with all network
stakeholders from across the globe. Trust me, we have enough meetings
already. And I'm only on a small network of 30,000 users. I think the
problem isn't what your are trying to learn, it's how you are asking.
There is no motivation for us to answer your survey, there is actually very
good security reasons why we wouldn't. You don't explain what you are
trying to research but asking us to give, gratis, deep inside depth to our
deployments. Most of us would have serious issues with our employers if we
gave out that info.
Hi Joe, William and Daniel,
Thank you for your suggestions and raising the concerns. Sorry for spamming the list. I thought resending will have it be seen by more people who might have missed the earlier ones. I fully understand the security concern as well. I will be more careful posting to the list next time.
Thanks!
-Ying
You don't explain what you are trying to research but asking us to
give, gratis, deep inside depth to our deployments. Most of us would
have serious issues with our employers if we gave out that info.
this is the problem with all the academic surveys on nanog. few, if
any, medium to large networks could answer if they wanted to. so the
surveyors get very biased results and then report them at academic
conferences. this is not really a problem unless you happen to also go
to academic conferences and choke trying to keep your mouth shut.
randy