Wow, a case between an ISP cartel member and a copyright cartel member. Hard to pick one to root for. Somehow neither side managed to make the other side look like the good guy. You'd think ISPs would do their darnedest to make it so they don't have a way of knowing what content is flowing over their pipes, to help avoid accusations of contributory infringement.
You'd think ISPs would do their darnedest to make it so they don't have a way of knowing what content is flowing over their pipes, to help avoid accusations of contributory infringement.
My understanding is that ISP don't want to be seen as dumb pipes, despite that it would probably absolve them of a lot of responsibility and liability, because dumb pipes are effectively a commodity and thus a race to the bottom price-wise. Of course, in many US markets, they have this locked up and are a monopoly anyway. This is may be (slightly) less of a focus now, but back in the 90s, ISPs saw piled on extra services as differentiators and lock in, which is why you might have had an email address in your ISP's domain, or (in the case of cable internet) Video On Demand services tied in with internet access, or even things like content filtering (to appeal to parents). ISPs have tried a lot of different things to avoid being considered dumb pipes.
Ironically I would probably pay more for a dumber, but more reliable pipe.
My operator changed my modem, and its configuration is "easier than ever!" (says the manual). That is, you have to use your account page on their website to change the wifi settings, and the more advanced part of it you must install an app on your phone. But you can't specify a custom DNS (which I'm sure the default router firmware would let you do).
So yeah, just give me a modem, and I'll do everything myself.
There's EU-wide regulation[1] and local (Dutch) legislation stating that ISPs must allow people to use their own hardware. My ISP (KPN) complies pretty well.
Their website even has clear documentation (in Dutch) with all the configuration details in case you want to set up your own fibre-optics modem, and the details on the VLAN to use for internet and telephony over it[2].
I use the ISP-supplied fibre modem plugged into an OpenWrt router. The router itself does PPPoE via VLAN. I hope you can get your ISP to follow such an example.
While I’m glad I can do it at all, my cable modem requires logging into their webservice, and flip a switch there to turn it from modem + router into just a modem. Still confuses me how that’s not a setting in the modem interface.
Almost every single ISP I've worked with will let you put modem routers into "bridge" mode where it bridges the modem side to Ethernet and there's your mostly dumb modem pushing a public IP to whatever you plug in to it. Cable companies get sneaky sneaky and keep some wifi crap on for their cell phone plans though.
I _love_ those. When I'm trying to configure my modem, nothing gets my nether regions tingly faster than
(1) Visiting the documented configuration site on the back of the modem
(2) It 404s in a new and excitingly different way from every other modem you've owned
(3) After visiting a nearby business with working internet, the helpful interblags populace suggests you need an app, for ... reasons
(3a) At least in the old days, that was just a blatant money grab (harvest your location history and contact list, spam notifications, and so on). Now with halfway decent permissions it's less clear why they're such a-holes about it.
(4) You have the foresight to install the app while internet is available.
(5) Your phone is 2 years old, and the manufacturer views bare-minimum due-diligence updates as a liability and hasn't packaged the latest OS updates. Google simultaneously doesn't trust you to update your phone without somebody-other-than-them taking the liability and also highly recommends all their app developers to only target the very latest Android version. None of the modem-makeitbetter-thingamawhutts work on your phone because 2yo devices are practically fossil fuels in the making at this point.
(6) You surf the interblags for an old version of the modem-makeitbetter-thingamawhutts. After searching long enough, you have a >50% chance of having found the real archive and not a virus, since Google refuses to host old APKs.
(7) By sheer luck, that suffices, and your modem manufacturer doesn't have one of those "your thingamawhutt version is too old" screens in their app.
(8) You go back home, where it's safe and all is well, and when you try to configure your modem you find that the app only works if it can phone home to the mothership. That's a fine thing to want to do, except the only reason you're using the blasted app is because you don't have internet at home and would like to rectify the deficiency.
(9) After a bit of contemplation on the life choices that led you to this moment, you recall that your wife's phone can produce a hotspot that might remedy the immediate clusterfuck.
(10) That doesn't work. Her phone has internet and working hardware, but Google has decided to ask her cell provider if she's allowed to share the hardware she purchased with nearby devices for a single crisis moment. Her cell provider says that she can do anything she would like with the device she paid for and owns, so long as they also get paid repeatedly.
And so on. I'm having fun, reply if you want me to continue the saga, but IME an app is a very poor choice for configuring a modem, and that sort of bullshit has bit me in more ways than I can count. Even when you can get the damned thing to work, like OP mentions, the settings aren't anywhere near as nice as what the firmware natively supports.
I missed that detail (it was illustrative and intended to be slightly humorous). A few that come to mind from my last few years:
- The modem in question is in the lower floors of a skyscraper. Nobody has good service. Incidentally, till wifi texting and calling became popular, that made forced SMS 2FA way more annoying than it already was -- run outside to catch a bar of service, try to get back to your apartment before the 2FA code expires.
- The modem in question is in the middle of nowhere (or some approximation thereof, like an hour or two out of Laramie WY). You technically have 50 kilobits per second of download throughput, but that's painful when the modem configuration app is 28 megabytes. It doesn't help that modern sites clock in at megabytes, and most of your software and hardware nowadays assumes anything over 10 seconds to a minute or so per request is a failure and times it out, making browsing some sites _literally_ impossible rather than just slow.
- You recently meddled with your APN settings and tweaked the one field that makes your phone helpfully decide to not display that APN entry, and it coincidentally means you don't have data service. Not having memorized 20 random ports and uris and stringly typed feature flags, you can't fix it without outside help.
- Your cell provider decided to increase your rates without notice when the contract renewed. You don't like being surprised by that sort of thing and had a spending cap on the card to prevent it, but you were instead surprised by a lack of cell service.
- Your cell provider's autopay functionality is broken the one day that year you needed it to work (friend's experience, not mine).
- Your cell provider had a bug double-counting data usage the one month you planned to get near the limit. Without internet you can't connect to their customer support to get that fixed (to their credit, customer support was fast and hassle-free).
- You're pinching pennies to get through college. You only have a phone because your bank won't let you in without SMS 2FA. It's only a smartphone because the cheapest of those are cheaper than the cheapest dumb phones nowadays (and you bought it used anyway). You don't have a data plan because that's an extravagent and unnecessary expense. (Why are you buying internet in this event? No clue, but it's another reason somebody might not have cell data, and I'm sure amongst the 3M people in the US with smartphones without a data plan some of them have self-consistent reasons for getting internet at home).
And so on. Any given day, lots of people have smartphones without data.
> You'd think ISPs would do their darnedest to make it so they don't have a way of knowing what content is flowing over their pipes, to help avoid accusations of contributory infringement.
ISPs don't generally go looking for copyright infringement. In order to keep safe harbor protections ISPs are required by law to register with the US copyright office (which requires them to pay a small annual fee) and provide them with a contact address where they agree to accept DMCA notices. The media industry hires third parties who hunt for IPs sharing files and they report them to the ISPs at their registered address which means they now can't pretend they didn't know about it. Of course those third parties aren't very careful about who they accuse of infringement and innocent people and even innocent network printers (https://archive.nytimes.com/bits.blogs.nytimes.com/2008/06/0...) have been accused as a result, but accusations and unproven allegations are all that matter according to the courts (so far)
> ALSO: The photo of an Internet Pirate is amazing.
And it seems it's not an AI generated image based on the credit. Someone actually sat down for a photographer and chewed on a CD dressed as a pirate. Wonderful!
That argument could, interestingly enough, technically apply to the power company or landlord as well. I mean people are renting a house and buying electricity to pirate right?
On a related note, if you're a renter and utilities are included in your rent, be sure to check the lease if there's a clause that prohibits high electricity consuming activities like running a home media server, a cryptominer, or growing certain plants.
I lived in a building that had the cost of electricity included in the rent for long-term tenants, but newer tenants had to pay for their own electricity use.
One person I knew in the building was using their "free" electricity to mine cryptocurrency.
This was, in my opinion, not far from stealing. Was it legal? Probably? Was it ethical? Not by a long shot.
Possible, but more likely the landlord (or new owner) realized that if you can switch to tenant-billed even with a rent drop, you have a more valuable property overall - if most of the apartments in the area are tenant-billed.
True, rent prices are typically the most important figure for any property. Everything else comes secondary - if you're halfway down the search results because of your prices, almost nobody will even see that utilities are included.
It's the airline ticket problem - you need to be "selling" the same as everyone else, because nobody checks. Included utilities is like included luggage.
"Nobody checks" because it's a seller's market. If the cost of housing dropped tenfold then I guarantee you that landlords would emphasize the value-add.
Omg I lived in a house with electricity included, got the contract pre-covid, as a flatshare, we were all respectful of a fair unwritten usage quota, you know, then covid passed, and since in Netherlands rent prices went to the stars, the landlord tried to evict us with bad excuses, failing (you can't evict in NL tenants of more than 1 year 8 months without good cause), then he tried to raise the rent more than 2% per year (you can't, that's why he wanted to evict us, so you can rent the same place at any arbitrary price), then he tried to raise electricity bills by removing them from being included in the rent (you can't), anyway, we got tired, I think we turned on 5 TVs per day every day all day, PC on all day every day, there was a power price crisis in NL for a while during Ukraine war, I think we bankrupted the guy, it was amazing to see him to try to squeeze any cent from us failing without recourse
Tenancy is a contractual relationship. It doesn't include ethical sidequests like making sure your landlord doesn't pay large electrical bills which he has agreed to pay as part of the contract. Electricity is metered, making it the landlord's responsibility to put an "up to X KW a month" clause in the contract. That electricity isn't free, the scare quotes don't change that. It's cost-included.
Your opprobrium appears to hinge on the idea that this tenant would only mine crypto if the power is provided on a single-bill, all-you-can-eat basis, which I don't see as a safe assumption here. The tenant got a good deal. That happens sometimes. If landlord ever notices, he'll put another clause in the contracts going forward. Live and learn.
I wonder how running several computers would compare to other high-current draws that we would consider normal use, like using an electric oven a lot (3kw, plausibly 10 computers).
The oven won't be on 24/7 and even if you're baking the elements cycle on and off (that's why you hear the clicking relays). You'll have to integrate it over a month and compare that to the constant consumption of a mining rig.
It’s a residential lease, but the power use is for a non-residential purpose. So arguably yes.
I wonder what a court would say. It’s not clear to me either way. As a practical matter it seems like leases with utilities included shouldn’t need to enumerate all the non-residential things you aren’t allowed to do with those utilities.
Violating the contract would be a clear legal question, ethics is more fuzzy and down to personal interpretation but usually isn't dependent on the letters in the contract.
There is a valuable lesson there for whatever fool wrote the contract. Don't promise someone uncapped use of something like that for a fixed price. Especially not electricity.
The more realistic lesson is don't apply programmer logic to legal contracts where the courts will apply reasonableness tests, side with the landlord, and stick you with a many thousand dollar bill.
It's like how pricing mistakes on websites are in the consumer's favor when they're small but in the sellers favor when they're large because it's unreasonable to assume someone would be selling a $450 item for $4.50.
If an idiot landlord gets into court because a tenant took their contract literally, they're a fool who has hopefully learned a valuable albeit very expensive
lesson about writing contracts.
How does that work for shared spaces and excluded areas? For example I live in the bottom half of a house, and my and the other tenants leases very clearly specify what parts of the house can by used by whom and what decks and access gates are shared etc.
I would still expect the xeon server to go pretty low once you make sure power saving is enabled. High double digit watts on average. Maybe they need to cut down to two CPUs but I can't imagine that hurting the performance.
It's significantly more than a NUC but even a single mining GPU would use more and a real mining rig is an order of magnitude higher.
And in terms of grow lights, it gets you a couple square feet.
Nah. The ISP is the actual medium whose network was used to copy the alleged infringing content. So that's not analogous to all the utilities in use by the infringer.
I think the key differentiator is that Cox [somehow] knew that the customer was using their network to copy the alleged infringing content. The electric company doesn't know this, and (to use another example) the hard drive manufacturer doesn't know that their customer is storing the content on the drives the make.
If I were Cox, I suppose I would be incentivized to operate kind of like a VPN and just not store logs and not do any kind of traffic inspection or analysis. That way they might be able to argue that the plaintiff couldn't prove contributory infringement.
How they "know": The copyright holders join a torrent swarm and log all IPs that give them pieces of their copyrighted content. They then send infringement complaints to the IPs' ISPs alleging that someone with that IP was infringing on their copyright at the given date and time.
There are a number of reasons an ISP might not know who had a given IP at a given date and time. DHCP and CGNAT are two big ones. And the rare individuals who choose to fight do sometimes use that as a defense.
However, those technologies have a purpose other than anonymizing activity. A residential ISP who took active measures to obscure their customers activity would have a harder time defending against a complaint of negligence.
DHCP can be set up to make things harder, but most people's IPs don't change very often. ISPs don't want to make it impossible to tell who had an IP address at a given date/time because they need that information to fight other forms of internet abuse (spammers, phishers, hackers, malware, child porn, etc) as well as other legal requests they see from police and certain three letter agencies who file the proper paperwork with the ISPs legal department. This is generally a good thing and makes the internet a safer/better place.
Any ISPs who made it impossible for themselves to tell which customer had an IP address is going to act as a safe haven for bad actors and they'll get blacklisted from many other networks pretty quickly. Only the largest ISPs can ignore abuse issues on their network without fear of being blacklisted and their IP space null routed.
You're by no means incorrect but the strategy you're speaking of here to determine where warez is going is .. about 25+ years behind and even then network engineers used it as kind of a "look, they're not netengs, they're not going to understand how we encap/decap packets and uniquely identify their chain of custody, so just say something simple so they get the idea."
It also fell flat on its head when it hit the "is an IP a person?" argument.
Every file crossing the line is fingerprinted and various ISPs give various copyright-track software vendors or licensors direct access to track those pre-determined fingerprints and report problem to ISP for customer to mitigate or help prepare for subpoena. There's a lot of other ways this stuff is tracked I won't get into and the environments even more direct now with all of the ML tools. Webhosts and social media apps do the same thing so they can have smaller abuse departments digging for CSAM content. Data exfiltration is a huge space.
As soon as a new lady gaga video is uploaded to Vevo, the fingerprint was already determined, and the copyright bots have alread been scanning the internet for that file for days/weeks. Watch a new music video release on youtube, you'll see attempts to copy it get hammered immediately.
DPI and MITM-TLS which a lot of people blindlyaccept by accepting various free ISP features basically takes most of the end users privacy away in the end anyway. They're all reselling your data, ISPs and most VPNs are scum.
These are scenarious that someone with experience at a major CDN would be very familiar with. We do the same thing tracking every unique object on the entire platform.
I suspect willful blindness to infringement wouldn't help. Willful blindness is such a problem in law that most laws are phrased using the words "known or should have known". Someone who should have known is said to have constructive knowledge.
There's also business and legal reasons why ISPs need to log and inspect traffic. "No logging" means "no abuse desk" - i.e. if someone's using your ISP to send out spam or hack people, and you aren't logging or inspecting traffic, then you can't investigate that spam or that hacking. At a minimum it means every other online service is going to distrust your users - i.e send them through reCAPTCHA hell; up to an outright ban. Think like what it's like to use clearweb services through a Tor exit node, and now imagine that for a bunch of nontechnical users. They're going to switch to the ISP that doesn't result in them getting group punishment. But actual criminal actors are going to flock to the service, because they need someone in the middle to launder their packets for them, which makes them have a worse reputation, and so on.
VPN providers, especially non-logging ones, keep their abuse handling practices secret. They all claim to have "ways" to catch abuse on their network - I suspect it just comes down to a combination of flagging accounts/logging events following certain types of behavior along with data collected from abuse complaints.
Seems possible enough to keep logs of possible abuse for an account without keeping IP logs around. I'd also guess they have a lower standard for evidence needed to suspend/close an account for abuse than an ISP would. It's just not worth the trouble to keep a suspicious account open when it risks getting their IPs blacklisted.
VPN companies already do sketchy shit to get around Netflix's georestrictions[0]. I've heard rumors like they route traffic through residential IP blocks somehow. If they're able to do this, then IP reputation doesn't matter as much. They're already burning IP addresses to connect to Netflix with anyway...
[0] BTW, how is this not a massive industry-wide DMCA 1201 lawsuit? Advertising a service as being capable of getting around georestrictions sounds like it should be as illegal as advertising a DVD decryptor.
IP reputation matters, but users only care when it stops them from doing what they want. Residential connections often have certain ports blocked and some uses restricted by the ISP which makes resi connections less ideal for some types of abuse, but great for things like using streaming services. An IP that doesn't work for some services may still work great for others. It seems like a lot of people using VPNs know to try a few different servers/connections until they get the kind of access they want.
> how is this not a massive industry-wide DMCA 1201 lawsuit?
Is it illegal to bypass geoblocking in the US? I'd think that in the US using a VPN to get around geoblocking could count as 'circumventing a technological measure that controls access to a work' or whatever and could be considered illegal under the DMCA, but perhaps there's also a difference between using a VPN to access content offered outside of your own country you were never authorized to view vs using it to access the same content you'd normally have access to if you weren't just on vacation? I don't think I've heard about any lawsuits against VPNs for that.
Either way, I'd agree that advertising a service as a means to bypass geoblocking is probably a bad idea. As long as they don't outright tell people they should use the service to do that though I think an argument could be made that a VPN just provides internet access, and the VPN's ToS/AUP requires users to not to do anything illegal with the connection, and so the user is responsible for what they do with it.
how would it be a lawsuit? its farcical that there even are georestrictions in the first place. Netflix doesnt care about them either. The only reason they exist is to please rights holders with weird demands. It's not either netflix or the customer's responsibility to care about upholding them (other than netflix's own contractual requirements)
In other news; netflix themselves allow the usage of VPNs[0] provided you dont use the free (ad-supported) plan. Presumably to prevent the sort of thing I used to do with spotify years ago (be from a region where no advertisers paid them, so the space where adverts would usually be was just blank)
They didn't just know about it, they actually said they didn't want to perm disconnect accounts they were told have been repeatedly infringing because they didn't want to lose the money those accounts would bring in. They screwed themselves by talking about it over email (including saying outright "Fuck the DMCA").
No ISP is happy about the fact that they're forced to spend significant amounts of time and money working as "copyright police" for the media industry for free, and they aren't happy that (so far) courts have insisted that they need to permanently disconnect people from the internet over nothing but unproved accusations from questionable sources, but it was a bad idea to talk openly about that in email because all of that was later used against them in court.
> The electric company doesn't know this, and (to use another example) the hard drive manufacturer doesn't know that their customer is storing the content on the drives the make.
This is a sufficient argument in many situations. To shore it up for cases where the electric company gets informed (by a court, for example, with or without the court actually ordering the electric company to do something about it) that subscriber Jackie Doe has been sued and penalized for copyright infringement, there has to be a complementary argument about whether placing mandatory enforcement on multiple providers in the service stack (website -> internet service provider -> electricity) is a reasonable legal framework. How many turtles on the way down should be forced to restrict known copyright infringers? Before internet access became nearly as necessary as it is today, former Senator Orrin Hatch proposed a law authorizing destruction of copyright infringers' computers by copyright holders. (In the service stack I mentioned previously, "computer" would be a link between internet service provider and electricity.) Meanwhile, my takeaway of the featured article is that bans of internet service are not a fringe idea, at least among big copyright publishers.
This thread reminded me of an EFF article "Beyond Platforms: Private Censorship, Parler, and the Stack" [2] about the content moderation tech stack (social media site -> site hosting provider -> internet service provider). The article focused on voluntary moderation actions by site hosting providers and ISPs but is a useful prototype for examining tech stacks which involve third-party liability laws, including in copyright contexts.
This is completely true, and yet many many countries place a fee on any writable media (drives, cards, disks) and funnel that bribe directly to the RIAA and MPAA by law.
https://en.m.wikipedia.org/wiki/Private_copying_levy
Given so many people with strong feelings about copyright law, has there been any widely publicized occurence of jury nullification on a case like this?
Has there been any widely publicized occurrence of jury nullification on any case? I feel like it's something that certain people on the internet bring up a lot, but in reality pretty much never happens.
Here you go. This happened in the UK. Protesters admitted they had done the damage, the judge said there was "no defense in law" for the protesters actions but the jury acquitted anyway. It was definitely widely publicised - the case was across all the media in the UK when it happened.
Generally speaking you can't just admit to jury nullification without potentially undoing the nullifcation.
All you can really do is say "not guilty". If you want to find jury nullification you have to find cases where the evidence is overwhelmingly in favor of conviction but the jury didn't convict anyway.
And that can be hard to say, since outside of the court you often don't know exactly what the jury saw and the variation in juries can be pretty astounding.
ianal but I believe so, it's the jury's duty under the law to decide based on the law and the facts presented at trial, anything else is a violation of that duty. Jury nullification works because there's no way to prove the why for a juror's decision. Arguably jury nullification is a feature of the right to trial by a jury of peers.
Piracy isn't stealing anyway. It's just copyright infringement.
DRM and other consumer hostile practices are just more reasons to add to the growing pile of reasons why copyright infringement might not be unethical no matter what the legality is.
In the UK there's a law covering this specific case.
Theft is deigned "taking with the intent to permanently deprive", so joyriders were defending their actions by returning the vehicle after just this.
That's not 100k per violation on the download side it's 100k per work. Each work could be downloaded multiple times. The infringement damage is also the cost to make whole the copyright holder not necessarily the value to the person who did the pirating.
Consider you have a painting that's worth $5m and I steal it.[1] If I go to sell it, I'm not going to be able to go to a reputable dealer so I can only sell it for less. Say $1m. Now say I get busted and I have to pay you restitution. I won't only have to pay the value to me ($1m) I have to pay the value that you lost ($5m), because that's the principle of equity - that you should be made whole as far as possible the damage you have suffered. Some value ($4m in this example) has been destroyed but the court doesn't actually care about that - that's my problem because I have to find that from somewhere to pay you back.
[1] I know there are meaningful differences between piracy and theft of a physical item I'm just using the physical example because it makes clear how damages work in the legal sense.
Disappointing that the ISP is liable for anything. They shouldn't be required to disconnect customers based on unproven allegations made by a random company, without any due process involved.
Internet service should just be a human right. If you still really want to punish someone for copyright infringement, fine them. Taking away their internet access is cruel and unusual punishment given how much the internet is needed to function in society these days.
ALSO: The photo of an Internet Pirate is amazing.