I agree. I think its over stated. But I do think there was a more direct customer-disadvantage outcome, albiet increadibly brief. I think a bunch of people like me have now got a better sense our always-on backend is 'brittle' even if very very strong, most of the time. http://www.google.com/appsstatus#hl=en&v=status&ts=1376701087982 suggests it was a disconnection from considerably more than search. I don't believe index analogies jusify some of the scaling/visualization/comparison-to-root-dns things, but I would have been made distinctly uncomfortable in some circumstances by the loss of google backed email, google drive, and their implicit "no local storage required: you're always on" behaviour. An example is when I posted some stuff to the UK from the Post office across from the hotel at IETF, and spend 2 min online searching google mail for the address. Or, given the new "your airline ticket on your phone" model, I might have been trying to checkin at the last 5 minutes onto a flight. Or get into a ball game... Is this "40% of the net offline" ? no. Was it pretty wide reaching? Yes. On Tue, Aug 20, 2013 at 3:29 PM, Matthew Petach <mpetach@netflight.com>wrote:
I'm curious; do people really think that the difference in material indexed between Google, Yahoo/Bing, and others is really that big? I don't mean the heuristics and algorithms used to return the results in a particularly useful order; I mean the sheer raw set of indexed pages. I don't debate that Google found a particularly useful page ranking system; but I question the notion that the loss of Google was akin to the loss of your root directory.
Matt
On Mon, Aug 19, 2013 at 10:12 PM, Jimmy Hess <mysidia@gmail.com> wrote:
On Mon, Aug 19, 2013 at 9:48 PM, Randy Bush <randy@psg.com> wrote:
Without Google, how do you know where anything even *is*?
ask that to 20% of the world's population
Turning off Google is essentially doing a rm -rf http:// www-wide analog to rm -rf / or temporarily loss of the root directory, pending a fsck.
The important stuff is still there, somewhere... it's just becomes a real chore to get to your files without a useful directory provided by the indexing system, until you can get your superblock repaired.
Webcrawler, Gopher sites, and Archie search engine become viable options.
There's also backup on some stacks of tapes somewhere labelled Bing, DMOZ, Yahoo, and a few other misc. unlabelled stacks, various well-known .COM and .EDU domains, which you could probably use to find your materials if you downloaded the old Hosts.txt files; if you look long and hard enough, you can still find the filesystem data you need to relink the directory and get at the files you need; it can just be darn inconvenient sorting out all the spam.
randy
-- -JH