It feels like this isn't a large enough leap forward. It would be nice if BitTorrent v2 made it harder for ISPs to identify what is bit torrent traffic. AT&T artificially slows down upload speeds.
You'd want to look like some other protocol and you want that protocol be encrypted by default. Otherwise yours will get fingerprinted via the deep packet inspection.
The most obvious choice is to run your protocol over TLS. But then they can just throttle long-lived bulky TLS connections where neither side is on 443.
You can then require the responding side be on 443 (which is already a big hassle), but they will then throttle down TLS connections towards residential IPs or throttle them down cumulatively, as a group.
Other choices here are OpenVPN, WireGuard and, possibly, IPsec. But, again, it will come down to defeating the throttling of multiple inbound bulky connections of the same type, which is doubly hard to bypass if you are on a residential IP and your ISP is really bent on throttling.
Any successful obfuscation technique is short-lived, so it has no place in the protocol itself. Instead it should be delegated to a transport layer and the clients should be coded to support BT tunneling over X or over Y.
Fundamentally, nothing has the same characteristics as BitTorrent - almost nothing has the same high uplink requirements, which is an absolute signature of BitTorrent traffic.
Oh, also some of the shaping is implicit, not explicit. Most home internet connections are asymmetric - they dedicate more channels to downlink than uplink.
Streaming video, whether that's a Skype conversation or showing your gameplay to Twitch, can be a pretty bulky upbound stream.
I wonder how fingerprintable those are, and if they have diverse enough endpoints to be able to disguise other traffic as them.
or, get a better ISP? not all technical problems need a technical solution.
A better solution would be to have rules to make that kind of throttling illegal, but we've seen how that played out.
Which is mainly because we have privatized roads here in the USA.
Yes, you read that right. The "gold standard" is underground fiber, which lies in the public right-of-way yet is 100% privately owned with no "duty to serve" like the electric utility has. Oftentimes the fiber owner doesn't even have to dig up the dirt -- if they lay fiber along a newly built highway the government does all the digging for them (to create the roadside drainage ditch). The phone company just unreels a spool of armored OS2 into the ditch and hey, it's Miller Time.
Until privatized right-of-way stops we will continue to get screwed.
[*] Transcontinental railroad right-of-ways are the exception to the above, but there's only four of them. And, frankly, that land was privatized through outright fraud. Read the book _Railroaded_ sometime, it's shocking.
This allows ISPs to throttle uploads with very low risk of pissing off the bulk of their customer base when their detection algorithm gives a false positive.
The real question is whether they care enough about torrents to do anything about it.
https://iknowwhatyoudownload.com/en/stat/US/daily
Given that 0.31% of internet users torrent, maybe?
In other words, how would you authenticate peers? How would you prevent ISPs from mitming your encrypted p2p connections if you have no authentication for them, and you in principle can't have any?
The hash it relies on is broken, so their hand has been forced.
Second preimage attacks are MUCH harder to pull off, even md5 is still safe from them, many years after they were found to be broken in other contexts.
The only thing that sha1's weakness would let you do in a bittorrent context is create a torrent, then let you send fake data for a chunk, which really does not seem very useful, because you could just make the data malicious when you created the torrent.
Certainly not a trivial attack, but not benign either.
Do you have a source on this? I believe you I just want to know more. It explains a lot.
I have a gigabit link. I can download torrents at almost line speed. But I can barely get uploads past 5K/s. I spent hours at one point trying to tweak every possible setting and eliminate every bottleneck, and still couldn't get past 5K/s.
There is little to know, it's largely a commercial preference (that has, over the years, driven technological research; see ADSL for example, the first A is for Asymmetric).
Average consumers are precisely that, consumers: they rarely upload anything, but they download tons of content (from webpages to streamed media). So it makes sense for residential ISPs to maximize downstream bandwidth, since it's what consumers will evaluate them on. One way to do that is to simply throttle upstream, to ensure resources stay available for downstream. (This has the side benefit of reducing headaches, i.e. less people distributing questionable material on your network...).
If you need good upload speeds, you simply have to talk to your ISP.
Something else is the router AT&T requires will not completely open ports for you. What it does is it is closed for incoming connections that do not hit it multiple times then it opens up, so I appear like my ports are not open on AT&T. But even when I get around that (you can rip the certificates off of the router) AT&T itself limits the connection. If you transfer too much on one port on the backend they will remotely block the port sometimes. (Transferring too much being very little here, around 50MB, maybe even 10MB.)
Good news is Comcast does not pull this crap and has fiber internet. Bad news is they start at 2gbps and start at $210 a month in most areas. I'm paying $60 for AT&T gigabit.
I just got it last week, coming from the slowest ADSL option available, it blows my mind.
The word 'responsibility' that you used is of course too strong a word, however the comment 'it would be nice' to have is definitely true assuming we want downloads to be faster.
If you want QUIC's congestion controller benefits then you can also get that with TCP if you're on linux. Set BBR as congestion controller and set the TCP_NOTSENT_LOWAT socket option and bittorrent should work well over long fat pipes (i.e. international peers).
I'm not saying QUIC is terrible, but it's mostly designed to improve web traffic. Bulk transfers à la bittorrent benefit only marginally if you have a modern TCP stack and use it properly.
Yes, it'd be a minor suggestion at best. The biggest benefit would be to the internet as whole, motivating middle box makers into not breaking it and so on.
The low-latency idea seems interesting, I suspect WebTorrent people would appreciate it. But live streams over P2P sound too fancy for now.
> but bittorrent-over-TLS would work equally well (not standardized, but libtorrent supports it).
Which is where the suggestion to use QUIC would have the biggest effect and getting the swarms more encrypted.
> Bulk transfers à la bittorrent benefit only marginally if you have a modern TCP stack and use it properly.
Unfortunately very little software actually goes into the effort at figuring each such detail out and the end result, defaults are often rather poor.
I guess this more of a questions of "what are the defaults" not "what's possible."
Personally I understand the economics around it, but still don't like the idea of paying per GB. I would end up skipping some Netflix and would get mad at kids for playing Netflix to an empty room (much like I currently get mad when they leave the lights on in an empty room). It's nice on a personal level to avoid that.
To use the power analogy, when I was young the price of power was relatively high but in recent times it has now dropped to 0.6 NOK / kWh so even charging the car from 10-80% (~61 kWh) cost 40 NOK or $4. I do not get mad at the kids leaving the light on like my mother did to me.
I (or rather my job) currently pays about 1000 NOK/ month for unlimited 500 MBit synchronous fiber internet. I transfer maybe 500 GB/month so for it to be cheaper for me it would be under 2 NOK/GB or $.2 which sounds really high but that is what I currently pay with my usage.
Is it a normal connection or a business connection perhaps?
Your peering will only scale to your (combined) interface speed regardless of your SLA. Our 2x 40G peer with level 3 serve far more than 40 1G devices (I personally have 600 in one building alone, and that sets aside the rest of the users), but it’s rare it’s more than 30% utilised.
Clearly that’s not going to be a domestic isp architecture, so a reasonable question is what does uncontended actually mean.
I very much doubt an isp with 10,000 customers has 10TB of peering physically available. Linx public peering is less than half that for the entire UK, and while private peering will likely increase that, the suggest is it’s welll under tenfold, so the 50 million plus internet users in the UK only use at peak times 1M each, and that ignores all the non-domestic use.
Even a 100:1 contention ratio seems enough at a core level, so any isp spending money on improving on 10:1 ratios seems that they are just burning cash.
Clearly as you go the the edge contention ratio needs to drop - but 40G, or maybe even 20G uplink for 48x1G users would be reasonable to me.
Wow really interesting to see such high (assume Mbps) bandwidth guarantee for consumer service. Is the guarantee really work?
the price was competitive too. other ISPs charged $70 per month, and had a limit of 50gb (that was 15 years ago) my ISP charged $20 as base fee and $1 per gb of usage. if i used less than 50gb, i saved money, if i used more then i could...
But I think it would considerably change the Internet landscape. No more listening to the same exact song multiple times on Youtube or Spotify or whatever. No more downloading then deleting the same stuff over and over again. I think about it often, what a massive waste of bandwidth, I don't even know why. It's a lot of electricity used, I guess?
Even though I've got unlimited fiber, I'm looking for some sort of local "Internet cache" solution that would store everything so it would be re-downloaded from my home instead of across the ocean. Would be great for outages, too.
Is it though? How much electricity is used to serve a 1080p Netflix movie several times vs. playing the same movie from a local storage medium?
If you asked me, I'd rather burn Bitcoin mining rigs if I wanted to get rid of electricity waste.
Pretty sure I use more than that in a day.
Seriously, why would you use that much data per day for? Even getting full blu-rays you wouldn't be able to watch that many.
95/8 = 11.875 (Mbit to MByte)
11.875 * 3600 = 42750 (MByte/s to MByte/h)
42750 / 1000 = 42.750 (MByte/g to GByte/h)
42.750 * 24 = 1026 (GByte/h to GByte/d)
1026 / 1000 = 1.026 (GByte/d to TByte/d)
In a two month period I uploaded about 50TB according to the tracker so my calculations seem about right.
It's easy to reach these numbers in a country where net neutrality is a thing.
Luckily it was over fiber and through a corporate VPN and I barely noticed it. But I was shocked when I saw the usage on my router later that month.
that's because you benefit from it. You're not currently paying the cost of that 1080p stream - netflix (and the ISP/peering networks) are paying. Netflix recoups the cost from your subscription, and the ISP/peering arrangement is mostly cost neutral to them (save for a small amount i presume).
But it's still not "free".
Unless that's what you meant...
Just open another tab; problem solved. :|
I would bet this is already the most efficient way of it working.
You can setup a proxy service to cache internet requests too, but they're getting less useful due to greater use of HTTPS.
Because they can get away with it.
Why wouldn’t you want to pool bandwidth with your neighbors so you could all get faster speeds when you were using it instead of rate limiting everyone?
It's gigabit speeds until all your neighbours suddenly want to download something too.
So, the defense of using a misnomer to name your service is because the service warranted is actually impossible to supply given the margins?
Call me a fool, but that still seems like a company that is getting away with lying to the vast majority of people that don't bother reading the asterisk ( like T-mobile style "Unlimited" plan that gives X amount unthrottled, and some arbitrarily low rate after).
Criminal issue? Of course not, that's why the companies present such things this way.
Dishonest? You bet.
I can't sell you a pony made out of diamonds because that's impossible. Consequently, there is no legitimate reason for me to be advertising the sale of a diamond pony!
- Does this guarantee that you get the speed that you pay for? If not, then should the price during heavy congestion cost less per GB? I’ve had the unfortunate luck of having lived in areas where the internet is unusable for anything other than email and light browsing during peak hours. Bonus: Do you know why this happens? If not, I encourage you to research this.
- How will this affect the advertising model of the entire internet economy? If you, like me, have lived your whole life on a mobile data plan that is priced based on data usage, you’ll realize just how much bandwidth is consumed by advertisements (worst are the video/audio types that auto play).
- How will this affect “future” technology such at smart homes? Most devices send data back (Amazon Echo devices) and the growing security camera ecosystem that sends recordings to the cloud would suddenly cost users a ton more.
- Speaking of cloud, what will happen to the gaming industry that is heavily cloud based? Many games don’t even ship in a completed state anymore. Instead, part of the install requires downloading a ton of additional updates. Not to mention online play, DLC, etc.
There a many many more everyday examples like this that would be heavily affected by pricing per GB. Your question sounds like a simple solution but like many things in life, that is rarely the case.
Furthermore, even in places where you do have some sort of choice for internet service, it is almost certainly limited to choosing between a cable monopoly and a phone monopoly as your ISP.
When Xtra came out at $2.50 per hour, it was a complete game changer.
It changed our internet browsing habits so much. With hourly charges, you would try to plan out what you would do before connecting and instead of reading webpages, you would save them to disk for reading later.
With unlimited, you could just sit there and browse. Or leave it on overnight to download files.
At some point we got a second phoneline and I was downloading torrents on dialup all day and night for years before we finally moved somewhere with ADSL in 2006.
Looking at pricing for Spark (one of the largerproviders). [1]
For each of the fibre speed plans you can get 60GB/month, for another $10 you get 120GB/month and for a further $10 you get unlimited traffic.
BTW: "Mb" = Megabits, "MB" = MegaBytes
[1] https://www.spark.co.nz/shop/internet/plans-and-pricing/
I've used 15 TB in the last 5 weeks with my torrent client alone (and considering my Backblaze backup size, that's not nearly all of my traffic), but how many people like me are there in the general public?
In addition, it gives people a solid measure over which to sue you. If you aren't delivering, you're headed to court.
The ISPs like all the vagueness.