> The $1.5 million bill goes to the magnitude of the unauthorized access. Given it's derived from a non-linear pricing system, I don't think that's the cleares

> The $1.5 million bill goes to the magnitude of the unauthorized access. Given it's derived from a non-linear pricing system, I don't think that's the clearest way to express it. And I'm not sure it's accurate -- NY Times says Swartz got 19,856,160 pages. At a flat eight cents per page, that's $1,588,492.80, so it's possible the $1.5MM estimate is ignoring the $3/document maximum and any other pricing rules that exist. If it were about DoS, they would talk about how much bandwidth was taken from "legitimate" users, how many "legitimate" requests were slowed, any downtime or other obvious harm to "legitimate" users, etc. Instead, they quote big dollars, which means some combination of: there was no actual DoS, they care more about lost revenue than serving users, they want to impress the reader/FBI with big money. This "unauthorized access" somehow wasn't worthy of prosecution, not even $1.5 million worth of it. > There's a substantial DoS risk there that warranted investigation. But it sounds like, by the time the FBI got involved, they'd already determined the "DoS" was coming from one user of the free access program and they shut it down. They could have easily firewalled access from outside libraries and rate-limited free requests from inside libraries, to protect the system. Involving the FBI was punitive, not protective. If public read access was threatening the private write side, or the critical users' read access, maybe they're doing something wrong and should spend some of those millions on fixing it. If abuse by people downloading things willy-nilly because they're not paying is a legitimate threat to the system, again there are organizations and people who would be happy to offload that burden, if the courts would give them the data. Carl Malamud used $60,000 of donations to buy 50 years of federal appellate court data to put it online. Instead of helping, they fight by forbidding the use of RECAP[1]. I'll paraphrase: "If we transferred a document to you for free, don't give it to anyone else. Even though the document itself is public, we waived the fee on our cost to deliver it to you, and you are willing to ignore your own cost to deliver it to someone else. Oh, but since our field is built on research, we do allow scholarly work, even though the preceding sounds like it's entirely anti-intellectual, anti-education, and anti-progress." Seriously, it's not even about abusively mass-downloading for free -- ANY re-uploading of something you got for free is forbidden. > The primary purpose of PACER, again, is as a tool for litigants Frankly, I don't know enough about the courts to opine on who should be the "primary user", but if it's not for the public, maybe they should change the name from Public Access to Court Electronic Records, and stop saying things like "Online access makes the public record truly public, which I think is of great value." or "The biggest challenge—and opportunity—lies in the area of preservation of the electronic dockets and opinions for posterity... We will need to... ensure that future generations can access this valuable information."[2] > Absolutely nothing stops whoever wrote the article from hosting the relevant documents at his or her own expense. Sure, but the reader has to trust that the documents provided are unaltered and comprehensive to what the reader is interested in. And per the above, PACER forbids the author from re-hosting if they got it fee-exempt. (In connection, I do see there is $15 worth of exempt access to each account, which would allow some of my hypothetical article readers to follow up on court sources for free -- but, this has happened since Swartz's escapade.) > So PACER fees aren't the government making "huge revenues off of public documents." We have different understandings of "surplus", then. As I understand it, they've got a big pile of money that's doing nothing right now. And I would imagine (or hope) that its use is somewhat restricted, since it was made by the public courts selling public data. If they spend it on something, maybe my position will change, but for now I just see it as mostly-illegitimate and a waste of peoples' money. Your conclusion ignores my arguments that people could make new things with the data, and more cheaply than the courts would. And the courts might not ever make those things, if they're not within the courts' scope (which limited scope I hope you would agree is not the same as being worthwhile in general.) Going back to the article's topic somewhat, people would find, index, and link interesting things nobody would have thought to look for, let alone dig out their credit card for. Maybe as an actual user of the system, you don't find this compelling, but you haven't really specifically said why. Or your reason, that people who don't pay don't really care, is vague and doesn't address the transformative and meta-uses(?) I'm talking about. But my visions are kind of vague, too, I'll grant you. P.S. I noticed the price has risen to $0.10 per page, capped at $3/per document. I suppose not even the courts are immune to the rising cost of digital paper that they have their own monopoly on, and they wouldn't want that surplus to shrink. [1] https://www.pacer.gov/announcements/general/exemptnotice.html https://www.pacer.gov/announcements/general/exemptnotice.htm... [2] http://www.uscourts.gov/news/2013/12/09/25-years-later-pacer-electronic-filing-continue-change-courts http://www.uscourts.gov/news/2013/12/09/25-years-later-pacer...