Hacker News Reader

Comment by jere | Hacker News Reader

I think the author got it right and d23 hid that in the ellipsis.

>TOR, short for The Onion Router, is an obscure routing network that allows anonymous access to the “darknet” — the vast, murky portion of the Internet that cannot be indexed by standard search engines.

>Estimated to be 5,000 times larger that the “surface” web, it’s in these recesses that you’ll find human-trafficking rings, black market drug markets and terrorist networks.

[edit: To be perfectly clear, I was only highlighting what was actually said. I don't know anything about the dark webs.]

View on HN

It's quite obvious from your very own quote that it's the author of the article who's confused about the three (!) different notions of unindexable contents larger that the indexable one (deep web) vs. anonymous p2p file sharing networks (darknets) vs. Tor hidden services. I'm tempted to call this "a confusion hat-trick". ;-)

http://en.wikipedia.org/wiki/Darknet_(file_sharing)

http://en.wikipedia.org/wiki/Dark_Internet

http://en.wikipedia.org/wiki/Tor_(anonymity_network)

These are three completely different things.

Tor is a tiny wart on the Internet's nose. Even the darkweb of which it is a part probably isn't that large. The statement is just an attempt by the author to add color and drama to his story.

Even so, there's no way in hell the "darknet" is 5000x larger than the "surface" web.

The darknet is a tiny subset of the "deep web," which includes anything that isn't publicly accessible. Examples would include bank accounts, settings pages on Facebook, etc. I'm fully ready to believe that the latter is much, much larger than everything else, but the author clearly has the two conflated.

I guess the way it's written it's technically true - if you count the darknet as part of the deepweb, then it is indeed in the deepweb that you find the nasty bits. But that's extremely disingenuous, and it seems more likely that the author doesn't understand it.

However, the deep web (non indexable content) is several orders of magnitude larger than the surface web. The source I found [0] is from 2001 (sry!) and talks of 400-550 times the surface web. [0] http://quod.lib.umich.edu/cgi/t/text/text-idx?c=jep;view=tex...

Reply on news.ycombinator.com