With no sense of irony, this blog is written on a platform that allows some Nazis, algorithmically promotes publishers, allows comments, and is thus only financially viable because of Section 230.
If you actually want to understand something about the decision, I highly recommend Eric Goldman's blog post:
I'd never heard of Ghost.org, but looking at their website, they appear to be much more than just a free newsletter platform. It's not what I'm looking for, but it does look pretty cool for somebody that needs a professional publishing stack. They show their competitors as Medium, WordPress, Substack, and Patreon, and if I were considering any of those I'd give Ghost a closer look.
It's such a vital telecommunication service that the government should give them subsidies to build out networks and provide subsidies to people with low incomes to purchase their services.
But when it comes to regulation, they want to argue that they are no different than Gmail and that neither the states nor the FCC have any authority to impose basic common carrier rules.
That's not even mentioning the oddness that for the sake of wiretapping, ISPs are considered common carriers subject to CALEA and, like phone services, have to use equipment that complies with federally created specs that allow real-time wiretapping.
But unfortunately the federal judiciary is on a campaign to undermine Congress and federal agencies, so we're at the point where the judiciary is saying that the Federal Communications Commission has no jurisdiction over the most important communication system ever invented despite Congress clearly giving it an authority to do so and providing a framework.
Q.com has recently been going around our town laying fiber. This is CenturyLink.
Back in 1995 I was working at QWest (now Century Link), and there was a neighborhood in West Omaha that was a trial for fiber to the home. QWest got a tariff increase approved, to pay for deploying fiber to the home for rural customers, by the year 2K. They never seemed, in any real way, to take steps toward that.
In fact, in the later '90s they were actively NOT deploying fiber to DSLAMs in neighborhoods, to give good DSL, because doing so would allow CLECs to have access to put equipment in neighborhoods. So we all had to suffer with whatever DSL we could get on cable runs that went back to one of two switch buildings serving the town.
Now, 30 years later, they're finally laying fiber. This is after the city is just finishing up laying their own fiber lines to every home and business in town, and I'm imagining that is partly responsible for the telco upgrades.
The hard thing with this kind of search engine is that if it gets it wrong, the user is the one who takes on the liability.
I would pay a license fee and so would a lot of smaller sites if you had insurance and indemnification if they got sued.
However much I dislike Unsplash, and that's quite a bit, I do think there's a interesting benefit to their paid offer where if you use an unsplash image and it turns out that actually copyrighted (e.g. someone who is not the photographer who uploads the image and then the photographer sues you for using the image), Getty will defend you / pay your legal fees.
Right now I'm very particular about using things labeled out of copyright or uncopyrighted. Even things that are in the Flickr Commons or on NASA's site can be suspect.
For instance I just ran across a company logo uploaded by the San Diego aviation museum that's listed as out of copyright in Flickr Commons, but I strongly suspect it's still copyrighted. On NASA's site, I've seen images attributed to SpaceX and don't trust them at all not to sue.
Given that the penalty for being wrong on copyright can be insanely high even if you have the best intentions, I would really like a source that I could trust and would pay to be indemnified.
I ran into this recently. I was looking for a image of clown fish on Flickr with their license based search. I found a great one. I downloaded it and then started making the attribution where on I clicked through to the link of the photographer, or rather the their flickr account. It was immediately super clear it was a fake account with a few 1000 stolen images since there was just too much work and too much diversity in the images. I reverse searched for a few and found some of them on flickr and posted much further in the past and they were "all rights reserved"
Flickr is one of the better websites for this. I've seen a few websites with CC work where uploaders barely had any profile information. A simple "I take all the photos I upload and I know what CC means" would suffice. Ironically, the one that felt the "safest" was someone who linked to their Instagram where they posted photos. Instagram doesn't provide CC info, so this person went through the effort of creating an account in a different website just to reupload their photos that are CC.
I was making some photo editing tutorials and I ran into a similar problem. I needed a photo of a person. Turns out photos of people have their own additional set of legal problems (right to one's own image). This made me very worried because surely all these photographers who are marking their photos as CC have the consent of the models, right? Otherwise nobody would be able to use the CC photos. What would be the point of a photo being CC if it was illegal to copy it anyway? I think some website had a license that said I couldn't portray the person in the photo in bad light, or something of sort, which was particularly problematic for me since I was looking for a photo of someone sufficiently ugly to make a tutorial about skin retouching.
There are people who do deliberately publish images of themselves under a permissive licence. Sometimes seen with photos for specific topics on Wikipedia (published on Wikimedia Commons). I can only guess at their motivations, but there seem to be a good deal of exhibitionists and copyleft idealists in that group.
If a photo is used on Wikipedia with the licence you need, there is a solid chance the vetting process was done thoroughly enough. For photos of a person this would the first place to look for me.
An ISP selling Internet access to regular folks sells access to the entire Internet. To do that the ISP connect its network to the rest of the internet via a transit connection.
All your traffic goes through that unless you peer with other networks. When you peer with them you send less through your paid transit connection. So both parties benefit when they interconnect so long as they send a decent amount of traffic back and forth.
There's no such thing as a bandwith hog network. Netflix sends traffic to your network because your users asked for it and they pay you to deliver that traffic.
The notion that traffic ratios have anything to do with whether it makes sense to peer and whether someone should pay as long been debunked in the internet context. Those ideas are just remnants from the phone network which operates on a very different economic model, the one that make phone calls cost dollars per minute.
Here's a very clear presentation from 2005 from a NANOG meeting explaining exactly why you're wrong.
It's simply the principle that the network that you pay to get online doesn't get to interfere with what you do online. That encompasses lots of behaviors including interconnection.
You have not done a good job explaining/proving how they are wrong. Most of your response is only addressing a single paragraph that mentioned Netflix.
> The notion that traffic ratios have anything to do with whether it makes sense to peer and whether someone should pay as long been debunked in the internet context.
Do note how the comment does not mention peering ratios. An ISP being a "hog" does not need to be determined by the peering ratio.
This is a very good article, but it does not directly address the above. It is very specific to arguments about peering ratio. If you have no opinions on peering ratios, you have to read between the lines to get opinions on the original comment.
In Argument #2 counter argument #1: "This is a valid observation ... This is not however an argument for using Peering traffic ratios to restrict Peering."
> And you're also wrong but what neutrality is.
Just saying they are wrong is not helpful. You provide no evidence that "Net Neutrality" has not shifted in meaning since the 90s.
Net neutrality was first used and defined as a term in 2002 by Tim Wu in a paper called A Proposal for Net Neutrality.
The first FCC work about it was a speech by then FCC chairman Michael Powell in February 2004 at the The silicon flat irons where he outlined the four freedoms, which included the right to use whatever application you want, access whatever content you want, and use whatever device you want.
In other words net neutrality as a concept that people talked about did not exist in the '90s.
> Cloudflare's lawyers should have told Crowd strike to kick rocks.
Not all ISPs use provisions in the DMCA that let them put the burden back on the claimant. A few do.
In general, if ISPs or CDNs have a free plan, they can't, as bad actors leverage these free plans in bulk.
But ISPs or CDNs that charge actual money to known customers will generally not take down until all legal avenues to keep their client online are exhausted or someone upstream from them blinks which threatens the rest of their customers.
It's not a question of getting what you pay for so much as being sure that everyone using the same provider is paying, and having a discussion with the provider before it happens instead of during. You also need all links to play it this way, or you have to host in a different jurisdiction, which may not be possible for some data/content.
There are ISPs, CDNs, DNS registrars, data center facilities, backbone providers, who don't take down before asking questions, so if you need to be in the USA, find those.
// I have been both a provider refusing to take a client down for nonsense, and a client of those upstream who refused to take us down when our clients were under threat. And yes, when this would happen we spent money rather than cave if the mega corp insisted to go to court, yes the mega corps lost (typically instantly), and yes we donated to EFF.
McSherry told Ars that major service providers like Cloudflare can become "chokepoints," where lawful speech gets taken down because, regardless of intent, "they can't be precise in what they do" once their platform gets too big and reports get too numerous. She agreed with Senk that a few large companies controlling vast amounts of content online can be problematic.
"It is true that there are a relatively small number of companies that have outsized influence over what is available online," McSherry said. "And that can be really difficult or create a real problem because sometimes they don't have any choice but to sort of overcensor."
"In this case, "there was just big companies using big processes and not being careful," McSherry said. "And that is not acceptable."
> Cloudflare could have told these clowns to go kick rocks without incurring any liability
If Cloudflair didn't remove the content and the content was infringing they could lose their safe harbor protections [1].
In this case the website is obviously parody. This highlights the problems with DMCA. Fraudulent DMCA requests incur cost but are almost never penalized.
Not if the DMCA takedown notice wasn't valid in the first place.
(By valid I mean it correctly follows the requirements in the DMCA, one of them being that it must be for copyright. It does not apply to other kinds of IP, nor does it apply to other violations of the DMCA such as the anti-circumvention provision cough youtube-dl cough)
This is correct and exactly why Cloudflare should have thrown this in the green compost bin the moment it showed up, and written back to say if you send us this kind of trash again, we will sue you.
If you actually want to understand something about the decision, I highly recommend Eric Goldman's blog post:
https://blog.ericgoldman.org/archives/2024/08/bonkers-opinio...
reply