[dns-operations] What are the data? [was: After Google Mail, Google Docs, Google Wave... Google DNS]
Patrick W. Gilmore
patrick at ianai.net
Mon Dec 7 12:29:17 UTC 2009
On Dec 5, 2009, at 1:32 AM, Paul Vixie wrote:
[...]
> i don't know but i hope not. isp's and enterprises have been anycasting
> and/or centralizing their internal recursive dns servers for decades now.
> anyone who assumes similarity of network locale or geo locale between the
> dns forwarder and the later tcp initiator is being (statistically) silly.
> google has no responsibility to warn anybody that such assumptions are
> silly. CDN's probably do bear some responsibility in that area, but:
"It doesn't matter how beautiful your theory is, it doesn't matter how smart you are. If it doesn't agree with experiment, it's wrong." -- Richard Feynman, 1918 - 1988
Ignoring the case at hand and speaking just in generalities and hypotheticals:
When considering the efficacy of an algorithm, I submit the data should be considered, not comments on mailing lists. Put another way, perhaps we should try it and see what happens?
What do the people who use the algorithm say? What do their users say? What do their measurements say? How well does it work in absolute performance metrics? How well does it work compared to other algorithms for the same type of service? How well does it work compared to the existing method? And I don't mean corner cases, I mean a real test with real users pushing real traffic to see real results.
What are the _data_? While name calling on mailing lists is easy and fun, it is not data.
If the real-world, objective, empirical data show a performance or other gain, then the algorithm is not "silly", and no number of posts from anyone no matter how famous will matter.
Speaking just in generalities and hypotheticals, of course. :)
[Not written by Paul, but Paul elided the attribution.]
>> bet not, and since when have you known any company to point out the cons
>> of using their products...
You have obviously never spoken to anyone at my company.
--
TTFN,
patrick
More information about the dns-operations
mailing list