[synctext]

The article mentions "hundreds of millions of new users". That would be great news for privacy on The Internet.

But it would surely slow the Tor network to a halt. Users trying out privacy-enhancing technology would be disappointed guaranteed with 30+ seconds page load times.

Tor has roughly 1000 exit nodes, all traffic flows through them. It needs to have much more capacity to handle that kind of load. Who is going to pay for 100x or 1000x server capacity?

[tedks]

That is precisely the topic of the article. For a more in-depth view, you can read this tor-talk mailing list thread: http://archives.seul.org/or/dev/Sep-2014/msg00091.html

(This is the original source for the Dot article.)

A summary: Raw bandwidth isn't the limiting factor in scaling Tor; before you just plug in traffic to the network, you'd need to optimize Tor's internal protocols to make expansion of the network even possible. However, those hurdles are fairly small, and once they're handled, it's just a matter of funding to pump up more nodes, which is much more simple.

Further, not all Firefox users (if the article's speculation is correct) will be using private browsing at once. So the actual increase is probably 5-10% (at most? anyone know actual statistics about private browsing?) of that number.

[cryowaffle]

While I agree with you I think there are two factors here (1) only a small fraction of the hundreds of millions would actually try out a Tor-private-browsing-mode and (2) tor won't grow until it is further strained (yes I know it is strained now). This would be great and would push innovation to solve the traffic problems.

[cyphunk]

If the added use case bring too much strain on the network and cannot be solved through interest and need than Tor could simply split it off or itself into an alternate network.

I wish someone would integrate crypto currencies into the layers of the Tor onion. I could imagine paying a few cents for each unlayering. Basically you just include a private key for a wallet holding 0.0000n cents in each layer with perhaps 0.000n for the exit. But I guess this will take some time for intuition to adapt to it.

I would assume that the organisation supposedly integrating Tor would have to fund tor exit and bridge nodes in mass. Or just force users using this alternate private mode to also share bandwidth in a p2p/skype like manner.

[mike-cardwell]

See the part of the article starting: "There are several key challenges that Tor’s developers face when it comes to growing network capacity."