Sounds like a nice example of management desire to tie in charges-per-use is taking priority over unimportant stuff like app performance. Maybe we'll be proven wrong. Bingo. This is someone who just read a book about software-as-a-service and decided "damn, this is what we need to do!" Which is understandable, because SAS is a tempting model. As the provider, you get to milk your customers continuously, instead of just once or every few years. And unlike selling upgrades, you don't have to constantly sell them on the value of version X+0.1 and how superior it is to version X. All you have to make sure is that they're still happy with the service overall.
I'm reminded a little of the proclamations that Sun used to come out with, about how "the network is the computer" or the browser is the new desktop, or whatever. Investors eat that crap up.
no sane court would say that Microsoft and Redhat compete in operating systems sales.
Huh? Sure they do. Just not in the desktop OS market. They are definitely competitors in the server OS market. Microsoft Windows Server vs RHEL; they're probably the two dominant platforms in commercial server deployments.
So, in order for our troops to be somewhere more than 90 days, the Congress has to act. Sure, it may not technically 'declare war,' but it has to do something. Correct. Which they did. It was called the "Authorization for the Use of Military Force in Iraq" and I linked to the Wikipedia article concerning it in the previous comment. It was passed on Oct 11 2002 and signed into law on October 16. The invasion occurred several months later in March of 2003.
The only time I'm aware of in recent memory where you could go after a president for violating the War Powers Act was Clinton during Kosovo, because there was a period during the bombing campaign where he was over the 90-day limit that the Executive branch is allowed with Congressional approval, and Congress hadn't authorized anything yet. By law, all military action should have been halted, but of course it wasn't, and the whole thing basically became a non-issue when Congress went ahead and authorized it. (And there was, by and large, no public or political fallout.) But that situation never existed with Iraq; the AUMF was passed before the invasion occurred.
Similarly, Afghanistan was authorized by Congress in the fairly broad AUMF passed right after 9/11 that pretty much gave the President carte blanche to go after anyone who'd been harboring terrorists, but was pretty specifically aimed at the Taliban.
Both AUMFs have since been de facto renewed by the passage of funding (I don't know whether there have been specific re-authorization bills separate from funding; I assume that this is either not required, or happens simultaneously with funding) since their original passage. Congress could at any time pull the plug on hostilities by simply refusing to authorize more funds, or only authorize funds conditional on a drawdown (which is how they forced a pullout from Somalia in 1998; they just didn't authorize money for anything except protection of U.S. Embassies and other interests), but apparently they're not interested in doing that.
GoLive is one of those programs that gives you just enough rope to hang yourself with, if you don't know what you're doing.
I happen to like it and think it's a fairly decent tool, but I can imagine in the hands of someone who was totally clueless, and only used it in the WYSIWYG mode.... well, you'd get something like what you see above. Garbage. It has a tendency to do the usual WYSIWYG-editor things, like produce weird redundant nested tags, and generally make the code look horrible.
The idea is that it's very easy to switch from "Layout View" (WYSIWYG) to a nice color-coded HTML view, and from there to previewing it in your browser(s) of choice. I don't think the author in this case got the idea.
Except that the President can't start a war. The right to declare war is given to Congress alone.
Which it was. Personally, I would have liked for Congress to call a spade a spade and pass an actual Declaration of War (which Ron Paul proposed, to his credit), but they basically did it under a weasely 'war by any other name.'
I think this is crappy, but the courts have had ample opportunities (including one directly related to Iraq) to say that it's invalid, and they have not. It's the same way that every U.S. military action since World War II has been carried out.
For now, the War Powers Act of 1973, which allows unilateral military action by the President for up to 60 (really 90) days, and indefinite military action with a Congressional use-of-force authorization beyond that, is the law of the land, pretty much without serious dispute.
but remember there's a lot of pissed off people in Iraq who would love to see Bush pay for his crimes against the Iraqi citizens. I'm sure some international arrangement can be done after Bush is out of the office. And he WILL get out. Sooner or later, he will. Riiiight. Man, that must be some good shit you're smoking.
That's so ridiculous as to hardly be worth responding to, but I'll do it anyway: while the guy is startlingly unpopular as a President, I think you'd find a vast amount of opposition to anything stupid like allowing a former President to be prosecuted overseas. It just ain't gonna happen.
The U.S. never joined the Court of Human Rights because the first time you had a U.S. serviceperson being prosecuted by a bunch of Europeans, regardless of the validity of the charges, you'd have people clamoring for the U.S. to bomb the Hague. We'd rather bring them back and execute them ourselves than allow some outside body to give them a lesser punishment, if it means subjugating the U.S. to a foreign authority. It always has been and always will be, for the foreseeable future anyway, a key aspect of both U.S. politics and the national character in general.
U.S. voters are like Red Sox fans: we love to boo our own when they're not doing well, but that doesn't mean we're going to go cheer for the other team.
Or, more bluntly, George Bush -- like every other President that's ever held the office, regardless of his miserableness as a leader -- is only a foreign assassin's bullet away from being a national hero. We're the only ones who get to tar and feather our leaders.
Sure he can. He can be tried and convicted of any breach of the law he might be guilty of, and be sentenced to penalty prescribed by law. The President is not above the law. You're splitting hairs as to what constitutes an 'official act.' While your opinion is valid, I doubt very much that the courts will accept such a minimalist interpretation. I strongly suspect that pretty much everything Bush has done that's considered objectionable or controversial, would be considered official, and be shielded from direct prosecution.
Go read a summary of Nixon v Fitzgerald; that's probably the most obvious case. But there's a whole chain of jurisprudence establishing it. Starting in common law, but made very clear in Mississippi v. Johnson (1867), then again in the Nixon case, and then again in the Clinton one. In Clinton, discussing Fitzgerald, Justice Stevens writes: "As we explained in Fitzgerald, 'the sphere of protected action must be related closely to the immunity's justifying purposes.' Because of the President's broad responsibilities, we recognized in that case an immunity from damages claims arising out of official acts extending to the 'outer perimeter of his authority.'"
So if you wanted to successfully prosecute a President, you would have to show that the act was well beyond the outer perimeter of his authority. I think you'd have a very hard time showing that, because in most cases the Bush administration has been pretty devious in coming up with rationales for their actions, generally via enlarging the scope of the Executive. And as long as it's arguably within the perimeter of Presidential authority at the time, he would have a pretty solid immunity claim.
The correct way to go after Bush would be via impeachment, since there, there's no official-duties exception; trying to go after him personally for what are clearly acts as President, rather than as a private citizen, is a non-starter. If he beat his wife or kicked dogs, it'd be different. But starting a war is pretty clearly an official act, since it can't be done by anybody except the President. It's a very, very weak case.
The President can't be tried for treason, at least not in the way that a normal citizen can. He can only be impeached; beyond that he has immunity for official acts committed while in office. Once he's out of office you can't impeach him, and he's still immune as long as he doesn't do anything further after leaving, so there's no reason why a pardon would be necessary.
The difference between, say, Bush and Clinton, is that the things people dislike Bush for are mostly all official acts. They may be abhorrent, but they're official, and thus he's shielded from personal prosecution. Clinton got into hot water (re Paula Jones) because it was alleged that his sexual advances weren't 'official' acts and thus unprotected. Although Clinton argued that everything a President does should be immune, this was rejected by the USSC: acts conducted as the President are untouchable except through Impeachment proceedings; acts conducted as a private individual can still have civil liability (and potentially criminal liability as long as it was prosecuted after the person left office so as not to interfere with their official duties).
I think you'd have a hard time going after Bush personally, outside his role as President, so it's basically a non-issue. As there doesn't seem to be the will to conduct impeachment proceedings, he's free and clear.
When mac software crashes it usually just vanishes, with no user feedback at all. When the OS crashes it blackscreens (like, say, plugging in a firewire drive into Tiger, which they *still* haven't fixed) but I wouldn't say the information it gives is useful at all.. about as useful as a bluescreen.
Huh? When most Mac apps crash it produces that "The Application [ApplicationName] has quit unexpectedly" crashlog dialog box, where it shows you a trace and you can choose to type a friendly little note in and send it away to Apple. this thing.
I don't see it that frequently but I did find a pattern of actions that would repeatedly crash Aperture the other day, and it popped that thing up every time.
Don't know whether it only comes up for Apple applications or what (I don't think so; I remember getting it a few times when Vuescan crashed). Maybe it only comes up as a result of some types of faults, and not all of the fatal ones. But it seems to work fairly well for me.
A lot depends on how you define it; in a way, a PC running its OS and all its applications from immutable media, with a separate storage device for data, has some Harvard-like characteristics, but of course it's using the same RAM for both instructions and data so it's not truly split. A real Harvard arch would have separate memory controllers and separate RAM banks for each. (Which when you get down to it would be cool, since it prevents data buffer overflows from changing executable code, etc. It's a nice security system.)
It's one of those ideas that's just too good to die; it shows up in different forms over and over again, if you look hard at niche markets.
... or, an OS with popularity of BSD, the consistent feel of Linux, the security of a Windows, with the openness and price point of OSX. That's a pretty good description of Vista, actually.
Better than that, why don't we just have one storage area for programs, and a totally separate one for data? You could have your OS, all your applications, basically any and all executable files, stored in one place that was difficult to change, and then all your data, temp files, states, etc. could be stored somewhere else.
Isn't that just the top servers by "article flow", aka "number of posts made by subscribers"? It seems that experienced, heavy posters have their favorite service, which may or may not overlap with the preferred "leeching" news servers. I'm not totally sure. I think that it's some metric which is supposed to measure peering, not necessarily the number of messages originating there. So you could get a high rating either by originating a lot of messages and sharing them with only one other server, or you could do it by sharing a smaller number of messages with a lot of servers. At least, I think that's how it's measured -- I'd be interested if anyone wanted to clarify.
It's entirely possible that UNS.com wins because it's where the traffic originates from, but Giganews contains more leechers; I suppose I could see how that would make UNS look like the more valuable 'peer' (and, in fact, they would be -- you'd want to peer with the place where the content is coming from, obviously). I strongly suspect that if you think about what's required to post content to Usenet, that it's mostly done by experienced users, and a much larger number of leechers consume it.
What I'd really like to see is a map of the peerings, rather than just traffic stats; something similar to the old mid-80s ASCII map you see frequently (you know, this one). There has to be some easy way to dump a bunch of Path headers into a database and have it cough up which server is peered with which other servers on command, and make some nice graphs. This seems to be pretty close, but it's just a tool to do it using a server's logs; nobody seems to be generating and posting recent ones (the example data looks pretty out of date).
I used UNS first before trial-ing and switching to Giganews. Side-by-side comparison showed that while retention was comparable, UNS's completion was awful. PAR files became mandatory on nearly everything I downloaded vs. Giga not needing them on the same exact files. Interestingly, an hour or two after failed requests for articles, it appeared that UNS would go find them from a peer and fill them in. But that's only after a failed/incomplete attempt to download them. Interesting. Well, I can't really say I've had any negative experiences with either; I used UNS' free account for a while a long time ago and thought it was fine, and now I use Giganews just because it's free -- but I basically just use the text groups, no binaries, so it's not like I'm hard to please.
It's not all that distributed where the binaries are concerned. Unless it's your primary business, you don't carry the binaries groups. I wouldn't be surprised if the majority of USENET piracy could be disrupted by taking down less than a dozen USENET providers. You could. At least in the U.S.
On a lark, I went and drew a little back-of-the-envelope map of Usenet based on a random selection of Path headers from some binary postings. It's not scientific or anything (although I'm sure somebody could probably write a little Perl script that would do it to a few thousand messages and produce a nice peering map), but you start to see who the big players are pretty quickly.
I don't even think you'd need a dozen. If you could force maybe four or five major U.S. sites to discontinue the *.binaries.* groups or implement some sort of filtering, you could seriously disrupt U.S. users' access. You certainly wouldn't kill Usenet or even eliminate binaries altogether, but you could probably drastically shorten average retention and completion.
As a text service, Usenet is practically un-killable. Anyone with a typical PC and a broadband connection could probably maintain a usable text-only feed, and peers aren't that hard to find since the bandwidth isn't too high. But when you get into binaries, particularly binaries with long retention times, the network becomes a lot more fragile. It's not a uniform mesh anymore, and that makes it much less fault-tolerant.
I've been trying to imagine what would happen to the traffic flow if you removed the binaries groups from the U.S. Usenet supernodes; at the very least it's an interesting thought exercise. I think a lot of users might try to switch over to European providers, or maybe to smaller U.S. providers that have suck feeds from the big European ones, which would do some wacky stuff to trans-Atlantic traffic. There's probably a good thesis project in modeling that.
Hmmm, are you familiar with something called public domain? Maybe you live in a world where everything has a copyright, but I don't. Copyrights expire and for good reason; so the public can take the work and expand it which can enrich the market more so than the original work - think Shakespeare, Dickens, etc. So, is this advertising to pirates? I don't think so. Although, when a pirate sees the above article, that is what they read. Please take your myopic blinders off and remember that the world has been creating content since the beginning of recorded history, not just the twentieth century. Seriously?
Do you have any idea how much public domain content has actually been digitized? I suspect that if you took all the PD music and video that's around in digital form, combined it with all the text, you still wouldn't get close to the capacity of a big Usenet site. (Keep in mind the entire Library of Congress -- which is mostly filled with post-1923 content -- is estimated to be about 20TB; a big newsfeed might take in 3-4TB a day.)
Yes, there is a lot of old stuff around. (In fact, I'm a rather ardent supporter of digitization and preservation of public domain and other old and antiquarian material. And I think the copyright term extension was bullshit.) But you need to put in perspective that while human civilization has been around for a while, a very significant fraction of all the people who have ever lived are alive right now: some 5-10%, depending on which estimate for total population you believe. And widespread literacy is fairly new, so a lot more of those people are in a position to produce content than their forebears.
Even if the content on Usenet was a random distribution of content chosen without regard to date created -- if the people downloading content had just as much interest in copies of the Rig veda as they did in the latest pop music album -- it would still end up containing huge amounts of in-copyright material: because a massive fraction of humanity's total creative output, in absolute numbers, has been produced fairly recently and is trapped in the extremely long copyright terms we've created. (Datapoint: the U.S. alone published 106k books in 1996, and 206k in 2005 -- that's almost doubling in a decade. In the early 20th century it was around 10k new titles per year.*)
And combine that with the fact that most material created prior to the 1970s (to say nothing of material created before 1923) hasn't been digitized yet, yet we're talking about an all-digital medium here, and the public-domain argument becomes even weaker. And what stuff does exist in the public domain tends to be text, rather than video (not that there isn't some movies and video, but there's not that much), so it's pretty small in terms of space.
Look, I really believe in the importance of freedom of information, and hate the pro-copyright lobby that's twisted the laws over and over again to their own advantage, but I hate a downright stupid argument even more. And trying to assert that all the content on Usenet could plausibly be public domain stuff is laughable.
Let's not try to minimize this issue by making arguments that wouldn't last even a few seconds in front of a determined opponent or even a reasonably competent judge. I *wish* that most of the content on Usenet's binaries groups was in the public domain, but it's not and everyone knows it.
Sure, but the fact that they're advertising it as one of the benefits of a service that they charge for might make it difficult to call it "not for profit." Although perhaps you could argue that the infringement per se wasn't committed for gain (the actual uploading of the file by the originating user), the provider is certainly benefiting by it. Even in 1995 I think any decent lawyer would have had a stroke if they were asked if that was a good idea.
Heck, even warez sites and BBSes in 1995 were less obvious than that, at least they seemed to generally realize they were doing something illegal.
It's one thing to tell your customers that you'll respect their privacy and offer totally anonymous access, etc. -- the sort of 'wink wink nudge nudge' that the Ars article alludes to -- but it's another to specifically advertise your service as a source for warez.
Yep. The Usenetserver (or UseNetServer, as they seem to spell it) guys have always struck me as pretty smart. I remember back in the day they switched over from NT to Linux and offered a free account to all Linux users for a while. They seemed honestly interested in understanding what people wanted in their service, and they seem to have done pretty well for themselves as a result. I was looking at Top1000 yesterday and they're actually the top provider in terms of traffic, even bigger than Giganews, which surprised me (because Giganews is the default provider for a lot of broadband ISPs, i.e. Comcast, who don't do their own news servers anymore).
If you look at their site and then compare it to Usenet.com's, it's pretty clear that UNS is going for a different class of customer; they're not trying to bring in the "where's this Usenet-thingy I keep hearing I can get free stuff from?" crowd. In fact, if you didn't know what Usenet was when you went there, you'd have no idea.
I think they're probably safe, but you're right, it's the services who got greedy and sloppy -- catering to users who were just looking for an alternative to P2P networks, mainly -- that are going to get slaughtered if the **AAs decide to go to town on them.
They still need to concern themselves with 512(c)(A)(ii), the part which says "in the absence of such actual knowledge, is not aware of facts or circumstances from which infringing activity is apparent". Although I don't agree with it, I could see someone making an argument that anyone maintaining a multi-terabyte news server would have to have looked at the contents of some of the groups and realized that the bulk of the content of the binaries groups, which itself is the bulk of a modern newsfeed, is copyrighted material. If you found a sufficiently ignorant judge, and could manage to convince them that Usenet in general was basically a 'piracy machine,' then they might invalidate the Safe Harbor provision for anyone running a big newsfeed, even if they're careful and don't violate 512(c)(B) (the 'no profit' rule). That's kind of a nightmare scenario, but I think it bears thinking about if only as a worst-case.
It goes farther than wink wink, nudge nudge if this page is any indication:
http://www.usenet.com/articles/free_download.htm Humm. Yeah, that's a lot more blatant than they were making it out to be in the Ars article. In fact, what the hell, guys? I know it has a copyright date of 2005 on it, but even if that had been written in 1995 it still would have been a little much.
To wit: (in case they take the page down, which I sure would if I were them)
Where Can You Get Free Downloads These Days? Well, we must admit that it is getting harder and harder to find anything free on the Internet these days. File sharing websites are getting shut down, spam is all over the net and free download options are getting thinner by the day. So what is the Internet user who loves to download stuff for free to do in this situation? There is one solution which has existed for a while but not everyone may be informed of just yet. This solution is called Usenet, also known as The Usenet Experience. It is an underground because it is not a website that anyone can randomly access by doing a search in Google, Yahoo, or AltaVista. It is somewhat hidden and restricted because not everyone has access to the free download areas, called newsgroup. So How Do You Get to the Place with Free Downloads? It is easier than you may think. The place which soon may be the only one that offers free downloads is available to everyone through a Usenet service provider company, such as Usenet.com. In order to start downloading all you want, you need to have Internet access (which you probably have already since you're reading this) and a Usenet account. Once you join Usenet.com, you can access the Usenet newsgroups and start downloading all you want without paying an additional cent. Tired of busy file sharing programs such as KaZaa? Then Usenet is the place for you. It's a place that has it all and where you can download it all. Usenet has a much wider selection than any of the other file sharing programs and it is available to you to use 24 x 7, no mater who's online or who isn't. The files are all hosted on the provider company's servers and it is available to all users to view and download. What Exactly Can You Download in Usenet? Anything and everything. Literally. There are movies, mp3s, cartoons, wallpapers, sounds, videos, pictures, warez, games, software and much more. The files (also known as "binaries" in Usenet) are organized by subject in the so called "newsgroups," which makes it really easy for everyone from the inexperienced user to the expert to find what they are looking for.
Yes, those ads are part of the problem, because it hurts their Safe Harbor defense (see my post which quotes the section of the DMCA, further down in the thread). But only insofar as they might show that Usenet.com was benefiting directly from illegal content. And I'm not sure they do that, because the ads aren't that blatant. They basically just suggest that they have a rigorous privacy policy, etc. It's not totally damning.
Where I think they get into trouble is that, in order to claim Safe Harbor, they basically need to be able to claim "hey, somebody put that up onto our system, we didn't know it was infringing, we didn't even know it was there!" And it's a bit tough to do that with Usenet, seeing as how it's about 99% binaries and anyone who's ever opened up the alt.binaries.* hierarchy can tell that it's got a lot of bootlegs and warez in it.
It would be a little comical to see a whole bunch of seasoned network engineers and other greybeards try to claim that they had no idea there was copyrighted material on Usenet. ("Warez? On my Usenet?") But that's sort of the position they have to put themselves in, in order to get a successful 512(c) defense.
They also have to show that in the past they've complied with DMCA takedown orders against content that a copyright holder has pointed out as being infringing, which it seems like they weren't doing. That may also be a problem, although maybe they can argue that they didn't have the capability to delete articles (after all, if they took them out of their store, would they just have come in on a feed from another site that they peer with?). It might be difficult to get a judge to swallow that, though.
I think they're in trouble, but I'm not sure exactly how much trouble just yet.
To be honest, I don't know how Usenet.com can not qualify for DMCA protection, since it's exactly the type of service that the Safe Harbor exception is supposed to protect. The only thing that seems like it could harm Usenet.com is their advertising, which does veer a little into "wink, wink, nudge, nudge" territory. However, damning a company because it says it respects users' privacy, without actually advocating any type of criminal activity, seems like pretty terrible precedent, and I can only hope (although at this point I have little faith) that a judge will see it similarly.
I think the mention in the Ars article about Safe Harbor being related to "transitory network communications" is irrelevant here. Transitory network communications is covered under 512(a) of the OCILLA (which is part of the DMCA); the portion that I would expect Usenet.com to seek protection under is 512(c), "Information Residing on Systems or Networks at Direction of Users".
You can read the relevant section here, but the significant portion, IMO, is:
A service provider shall not be liable for monetary relief, or, except as provided in subsection (j), for injunctive or other equitable relief, for infringement of copyright by reason of the storage at the direction of a user of material that resides on a system or network controlled or operated by or for the service provider, if the service provider--
(A)(i) does not have actual knowledge that the material or an activity using the material on the
system or network is infringing;
(ii) in the absence of such actual knowledge, is not aware of facts or circumstances from which infringing activity is apparent; or
(iii) upon obtaining such knowledge or awareness, acts expeditiously to remove, or disable access to, the material;
(B) does not receive a financial benefit directly attributable to the infringing activity, in a case in which the service provider has the right and ability to control such activity; and
(C) upon notification of claimed infringement as described in paragraph (3), responds expeditiously to remove, or disable access to, the material that is claimed to be infringing or to be the subject of infringing activity.
The major things they're going to have to avoid are that they "had actual knowledge" that the material was infringing (which might be tough -- I mean, anybody who opens up alt.binaries.movies can probably tell pretty quickly that it's full of bootlegs) and that they didn't receive a "financial benefit directly attributable" to the infringing activity. I think that second one is actually a little easier (for Usenet.com) than the former. And, of course, they have to successfully argue/explain that they don't really have the power to remove articles from Usenet, because of the nature of the network -- it would probably help their case if they started at least deleting articles from their spool/store when they receive a complaint.
I suspect that this may lead to a shakedown in the Usenet provider world, if Usenet.com loses. At the very least, the big providers might have to do more in order to maintain a veneer of plausible deniability (deleting some of the more obviously movie and/or warez related groups, perhaps), or move their servers out of the U.S.
Yes, you can do this. (And in fact, I think this is the way to go on a lot of crypto, e.g., PGPfone or OTR Messaging's fingerprint-verification systems that don't require any PKI.)
However, for email, you may and probably do want to talk to a lot of people that you may never meet in person or communicate with any other way. This makes verifying a lot of individual fingerprints cumbersome -- but if you don't have any other method for proving authenticity, you create a massive security hole for MITM attacks.
So you pretty much need some way of verifying that the public key you're being given matches the intended recipient of the message, without going to the recipient and verifying it out-of-band for each new person you want to communicate with. This requires some form of PKI; either a web of trust where lots of individuals verify each others' identity, and you can find trust paths through the web to virtually everyone else (in theory), or you have centralized "trusted authorities" whose reputation is based on verifying others' identities. PGP uses the first method (mostly), S/MIME uses the second (again, mostly). Either one can sort of be used the other way around -- Thawte's personal certificates utilize a web of trust, and you can have psuedo-authorities using PGP by setting the weight of their trust very high, so that anyone they verify is considered OK. But they both function best when they're used according to their designs.
If you only want to talk to one person securely, then sure, you can generate your own certificate, they can do the same, you can exchange them and verify the fingerprints through some hard-to-forge method (like voice phone). But this only works if you can recognize each others' voices. If you're trying to communicate with someone you've never met before, it's vulnerable to spoofing and MITM (you try to call them, but instead of them, you get the attacker posing as them; likewise, they try to call you, and instead get someone posing as you). It's not a scalable solution.
But for instant messages, where you're probably communicating over and over with a relatively small group of people, and even telephony in many instances, it would be fine. But email in particular is probably not a good match for infrastructure-less PK crypto.
True. (And in case anyone else wanted that in terms of linespeeds, that's about 2.3 Gb/s, just barely under an OC-48, but probably too much when you consider overhead. You'd be talking about a dedicated OC-96.)
It's been a long time since I've thought about, or really seen anything written about, the architecture of Usenet. With those kind of requirements I suspect there can't be more than a handful of really complete news servers, and they can't be too generous about peering, because the bandwidth would be out of control. That does hell on what's ideally supposed to be a mesh topology.
It would be interesting to put together a modern Usenet map, maybe by scanning a large number of posts and reconstructing it from the Path headers. On second thought, I think that's what Top1000 does already, although they don't produce map-like output, they're more interested in traffic analysis. (Also, I didn't realize that Usenetserver.com was bigger than Giganews. Interesting. I remember when UseNetServer switched to Linux from NT and offered anyone using Linux a free account; I always thought that was cool.)
Just scribbling on the back of an envelope, I think you could seriously disrupt Usenet (in the US, anyway) by taking out a few well-selected major sites. You wouldn't "kill" it, and people who only browse text groups might never notice, but people using the binary groups would. It's pretty obvious if you start drawing a diagram that there are only a few major sites that must have very liberal peering policies that make up the network core, and then a lot of sites with more limited newsfeeds peering with each other out around the edge. (And then there's a whole.edu ghetto that's heavily peered, which is nice to see -- glad some people are keeping Usenet going in the edus -- but I suspect they have rather limited feeds.)
I can't find a link to it online, but I heard a talk recently about a group that was using geological evidence to try and track the sunspot cycle further back than we have human observations. Not sure quite what the method is, or if it's yielded any results.
There are several ways of looking a past climate records. One way is to look at the growth rings on long living (4000+ years!) species of trees. Another way is to look at the deposit layers of ice/snow at the North and South poles, and on sedimentary layers around river deltas. All of these give some idea of what the local climate was like over the years. Cross-reference together from many locations they can give an idea of what the local climate was back then. Deposits of dust/ash/soot at the poles can indicate some serious volcanic eruptions.
Maybe it is one of these? I think Dalambertian (post right below yours) is right. It wasn't rocks, it was ice cores. I'm not sure of the exact physics, and I haven't read any of the relevant papers, but I think it has to do with the production of some very long-lived isotopes due to increased amounts of radiation during high points in the sunspot cycle. This article (which may or may not be totally specious otherwise) suggests that it's Beryllium-10. That's what I heard someone talking about.
The thing I was confusing it with is the geologic evidence of magnetosphere changes (which are recorded in various types of rock as they cool, IIRC).
Yep. Not only that -- the massive storage and bandwidth -- but you need to get a newsfeed. And that's not as easy as it used to be, when you could basically ask the sysop of your local university nicely. I'm not even sure what the commercial news servers would charge for a real UUCP newsfeed, or if they'd sell you one at all (why would they want to create competition for themselves?).
I'm not sure how many high-completion, long-retention news servers are around, but I suspect it's way, way down from what it used to be. It probably wouldn't take too many targeted lawsuits to, if not actually wipe out Usenet (that's impossible), but to at least make it very different from what it's like now. You could definitely make commercial services unprofitable, push it underground, and force people to eliminate binaries or at least shorten their completion/retentions a lot.
Slashdot Mirror
I'm reminded a little of the proclamations that Sun used to come out with, about how "the network is the computer" or the browser is the new desktop, or whatever. Investors eat that crap up.
no sane court would say that Microsoft and Redhat compete in operating systems sales.
Huh? Sure they do. Just not in the desktop OS market. They are definitely competitors in the server OS market. Microsoft Windows Server vs RHEL; they're probably the two dominant platforms in commercial server deployments.
The only time I'm aware of in recent memory where you could go after a president for violating the War Powers Act was Clinton during Kosovo, because there was a period during the bombing campaign where he was over the 90-day limit that the Executive branch is allowed with Congressional approval, and Congress hadn't authorized anything yet. By law, all military action should have been halted, but of course it wasn't, and the whole thing basically became a non-issue when Congress went ahead and authorized it. (And there was, by and large, no public or political fallout.) But that situation never existed with Iraq; the AUMF was passed before the invasion occurred.
Similarly, Afghanistan was authorized by Congress in the fairly broad AUMF passed right after 9/11 that pretty much gave the President carte blanche to go after anyone who'd been harboring terrorists, but was pretty specifically aimed at the Taliban.
Both AUMFs have since been de facto renewed by the passage of funding (I don't know whether there have been specific re-authorization bills separate from funding; I assume that this is either not required, or happens simultaneously with funding) since their original passage. Congress could at any time pull the plug on hostilities by simply refusing to authorize more funds, or only authorize funds conditional on a drawdown (which is how they forced a pullout from Somalia in 1998; they just didn't authorize money for anything except protection of U.S. Embassies and other interests), but apparently they're not interested in doing that.
GoLive is one of those programs that gives you just enough rope to hang yourself with, if you don't know what you're doing.
.... well, you'd get something like what you see above. Garbage. It has a tendency to do the usual WYSIWYG-editor things, like produce weird redundant nested tags, and generally make the code look horrible.
I happen to like it and think it's a fairly decent tool, but I can imagine in the hands of someone who was totally clueless, and only used it in the WYSIWYG mode
The idea is that it's very easy to switch from "Layout View" (WYSIWYG) to a nice color-coded HTML view, and from there to previewing it in your browser(s) of choice. I don't think the author in this case got the idea.
Except that the President can't start a war. The right to declare war is given to Congress alone.
Which it was. Personally, I would have liked for Congress to call a spade a spade and pass an actual Declaration of War (which Ron Paul proposed, to his credit), but they basically did it under a weasely 'war by any other name.'
I think this is crappy, but the courts have had ample opportunities (including one directly related to Iraq) to say that it's invalid, and they have not. It's the same way that every U.S. military action since World War II has been carried out.
For now, the War Powers Act of 1973, which allows unilateral military action by the President for up to 60 (really 90) days, and indefinite military action with a Congressional use-of-force authorization beyond that, is the law of the land, pretty much without serious dispute.
So, sorry, but that's not going to fly.
That's so ridiculous as to hardly be worth responding to, but I'll do it anyway: while the guy is startlingly unpopular as a President, I think you'd find a vast amount of opposition to anything stupid like allowing a former President to be prosecuted overseas. It just ain't gonna happen.
The U.S. never joined the Court of Human Rights because the first time you had a U.S. serviceperson being prosecuted by a bunch of Europeans, regardless of the validity of the charges, you'd have people clamoring for the U.S. to bomb the Hague. We'd rather bring them back and execute them ourselves than allow some outside body to give them a lesser punishment, if it means subjugating the U.S. to a foreign authority. It always has been and always will be, for the foreseeable future anyway, a key aspect of both U.S. politics and the national character in general.
U.S. voters are like Red Sox fans: we love to boo our own when they're not doing well, but that doesn't mean we're going to go cheer for the other team.
Or, more bluntly, George Bush -- like every other President that's ever held the office, regardless of his miserableness as a leader -- is only a foreign assassin's bullet away from being a national hero. We're the only ones who get to tar and feather our leaders.
Go read a summary of Nixon v Fitzgerald; that's probably the most obvious case. But there's a whole chain of jurisprudence establishing it. Starting in common law, but made very clear in Mississippi v. Johnson (1867), then again in the Nixon case, and then again in the Clinton one. In Clinton, discussing Fitzgerald, Justice Stevens writes: "As we explained in Fitzgerald, 'the sphere of protected action must be related closely to the immunity's justifying purposes.' Because of the President's broad responsibilities, we recognized in that case an immunity from damages claims arising out of official acts extending to the 'outer perimeter of his authority.'"
So if you wanted to successfully prosecute a President, you would have to show that the act was well beyond the outer perimeter of his authority. I think you'd have a very hard time showing that, because in most cases the Bush administration has been pretty devious in coming up with rationales for their actions, generally via enlarging the scope of the Executive. And as long as it's arguably within the perimeter of Presidential authority at the time, he would have a pretty solid immunity claim.
The correct way to go after Bush would be via impeachment, since there, there's no official-duties exception; trying to go after him personally for what are clearly acts as President, rather than as a private citizen, is a non-starter. If he beat his wife or kicked dogs, it'd be different. But starting a war is pretty clearly an official act, since it can't be done by anybody except the President. It's a very, very weak case.
The President can't be tried for treason, at least not in the way that a normal citizen can. He can only be impeached; beyond that he has immunity for official acts committed while in office. Once he's out of office you can't impeach him, and he's still immune as long as he doesn't do anything further after leaving, so there's no reason why a pardon would be necessary.
The difference between, say, Bush and Clinton, is that the things people dislike Bush for are mostly all official acts. They may be abhorrent, but they're official, and thus he's shielded from personal prosecution. Clinton got into hot water (re Paula Jones) because it was alleged that his sexual advances weren't 'official' acts and thus unprotected. Although Clinton argued that everything a President does should be immune, this was rejected by the USSC: acts conducted as the President are untouchable except through Impeachment proceedings; acts conducted as a private individual can still have civil liability (and potentially criminal liability as long as it was prosecuted after the person left office so as not to interfere with their official duties).
I think you'd have a hard time going after Bush personally, outside his role as President, so it's basically a non-issue. As there doesn't seem to be the will to conduct impeachment proceedings, he's free and clear.
When mac software crashes it usually just vanishes, with no user feedback at all. When the OS crashes it blackscreens (like, say, plugging in a firewire drive into Tiger, which they *still* haven't fixed) but I wouldn't say the information it gives is useful at all.. about as useful as a bluescreen.
Huh? When most Mac apps crash it produces that "The Application [ApplicationName] has quit unexpectedly" crashlog dialog box, where it shows you a trace and you can choose to type a friendly little note in and send it away to Apple. this thing.
I don't see it that frequently but I did find a pattern of actions that would repeatedly crash Aperture the other day, and it popped that thing up every time.
Don't know whether it only comes up for Apple applications or what (I don't think so; I remember getting it a few times when Vuescan crashed). Maybe it only comes up as a result of some types of faults, and not all of the fatal ones. But it seems to work fairly well for me.
A lot depends on how you define it; in a way, a PC running its OS and all its applications from immutable media, with a separate storage device for data, has some Harvard-like characteristics, but of course it's using the same RAM for both instructions and data so it's not truly split. A real Harvard arch would have separate memory controllers and separate RAM banks for each. (Which when you get down to it would be cool, since it prevents data buffer overflows from changing executable code, etc. It's a nice security system.)
It's one of those ideas that's just too good to die; it shows up in different forms over and over again, if you look hard at niche markets.
... or, an OS with popularity of BSD, the consistent feel of Linux, the security of a Windows, with the openness and price point of OSX. That's a pretty good description of Vista, actually.Better than that, why don't we just have one storage area for programs, and a totally separate one for data? You could have your OS, all your applications, basically any and all executable files, stored in one place that was difficult to change, and then all your data, temp files, states, etc. could be stored somewhere else.
Bet nobody's thought of that before.
It's entirely possible that UNS.com wins because it's where the traffic originates from, but Giganews contains more leechers; I suppose I could see how that would make UNS look like the more valuable 'peer' (and, in fact, they would be -- you'd want to peer with the place where the content is coming from, obviously). I strongly suspect that if you think about what's required to post content to Usenet, that it's mostly done by experienced users, and a much larger number of leechers consume it.
What I'd really like to see is a map of the peerings, rather than just traffic stats; something similar to the old mid-80s ASCII map you see frequently (you know, this one). There has to be some easy way to dump a bunch of Path headers into a database and have it cough up which server is peered with which other servers on command, and make some nice graphs. This seems to be pretty close, but it's just a tool to do it using a server's logs; nobody seems to be generating and posting recent ones (the example data looks pretty out of date). I used UNS first before trial-ing and switching to Giganews. Side-by-side comparison showed that while retention was comparable, UNS's completion was awful. PAR files became mandatory on nearly everything I downloaded vs. Giga not needing them on the same exact files. Interestingly, an hour or two after failed requests for articles, it appeared that UNS would go find them from a peer and fill them in. But that's only after a failed/incomplete attempt to download them. Interesting. Well, I can't really say I've had any negative experiences with either; I used UNS' free account for a while a long time ago and thought it was fine, and now I use Giganews just because it's free -- but I basically just use the text groups, no binaries, so it's not like I'm hard to please.
On a lark, I went and drew a little back-of-the-envelope map of Usenet based on a random selection of Path headers from some binary postings. It's not scientific or anything (although I'm sure somebody could probably write a little Perl script that would do it to a few thousand messages and produce a nice peering map), but you start to see who the big players are pretty quickly.
I don't even think you'd need a dozen. If you could force maybe four or five major U.S. sites to discontinue the *.binaries.* groups or implement some sort of filtering, you could seriously disrupt U.S. users' access. You certainly wouldn't kill Usenet or even eliminate binaries altogether, but you could probably drastically shorten average retention and completion.
As a text service, Usenet is practically un-killable. Anyone with a typical PC and a broadband connection could probably maintain a usable text-only feed, and peers aren't that hard to find since the bandwidth isn't too high. But when you get into binaries, particularly binaries with long retention times, the network becomes a lot more fragile. It's not a uniform mesh anymore, and that makes it much less fault-tolerant.
I've been trying to imagine what would happen to the traffic flow if you removed the binaries groups from the U.S. Usenet supernodes; at the very least it's an interesting thought exercise. I think a lot of users might try to switch over to European providers, or maybe to smaller U.S. providers that have suck feeds from the big European ones, which would do some wacky stuff to trans-Atlantic traffic. There's probably a good thesis project in modeling that.
Do you have any idea how much public domain content has actually been digitized? I suspect that if you took all the PD music and video that's around in digital form, combined it with all the text, you still wouldn't get close to the capacity of a big Usenet site. (Keep in mind the entire Library of Congress -- which is mostly filled with post-1923 content -- is estimated to be about 20TB; a big newsfeed might take in 3-4TB a day.)
Yes, there is a lot of old stuff around. (In fact, I'm a rather ardent supporter of digitization and preservation of public domain and other old and antiquarian material. And I think the copyright term extension was bullshit.) But you need to put in perspective that while human civilization has been around for a while, a very significant fraction of all the people who have ever lived are alive right now: some 5-10%, depending on which estimate for total population you believe. And widespread literacy is fairly new, so a lot more of those people are in a position to produce content than their forebears.
Even if the content on Usenet was a random distribution of content chosen without regard to date created -- if the people downloading content had just as much interest in copies of the Rig veda as they did in the latest pop music album -- it would still end up containing huge amounts of in-copyright material: because a massive fraction of humanity's total creative output, in absolute numbers, has been produced fairly recently and is trapped in the extremely long copyright terms we've created. (Datapoint: the U.S. alone published 106k books in 1996, and 206k in 2005 -- that's almost doubling in a decade. In the early 20th century it was around 10k new titles per year.*)
And combine that with the fact that most material created prior to the 1970s (to say nothing of material created before 1923) hasn't been digitized yet, yet we're talking about an all-digital medium here, and the public-domain argument becomes even weaker. And what stuff does exist in the public domain tends to be text, rather than video (not that there isn't some movies and video, but there's not that much), so it's pretty small in terms of space.
Look, I really believe in the importance of freedom of information, and hate the pro-copyright lobby that's twisted the laws over and over again to their own advantage, but I hate a downright stupid argument even more. And trying to assert that all the content on Usenet could plausibly be public domain stuff is laughable.
Let's not try to minimize this issue by making arguments that wouldn't last even a few seconds in front of a determined opponent or even a reasonably competent judge. I *wish* that most of the content on Usenet's binaries groups was in the public domain, but it's not and everyone knows it.
Sure, but the fact that they're advertising it as one of the benefits of a service that they charge for might make it difficult to call it "not for profit." Although perhaps you could argue that the infringement per se wasn't committed for gain (the actual uploading of the file by the originating user), the provider is certainly benefiting by it. Even in 1995 I think any decent lawyer would have had a stroke if they were asked if that was a good idea.
Heck, even warez sites and BBSes in 1995 were less obvious than that, at least they seemed to generally realize they were doing something illegal.
It's one thing to tell your customers that you'll respect their privacy and offer totally anonymous access, etc. -- the sort of 'wink wink nudge nudge' that the Ars article alludes to -- but it's another to specifically advertise your service as a source for warez.
Yep. The Usenetserver (or UseNetServer, as they seem to spell it) guys have always struck me as pretty smart. I remember back in the day they switched over from NT to Linux and offered a free account to all Linux users for a while. They seemed honestly interested in understanding what people wanted in their service, and they seem to have done pretty well for themselves as a result. I was looking at Top1000 yesterday and they're actually the top provider in terms of traffic, even bigger than Giganews, which surprised me (because Giganews is the default provider for a lot of broadband ISPs, i.e. Comcast, who don't do their own news servers anymore).
If you look at their site and then compare it to Usenet.com's, it's pretty clear that UNS is going for a different class of customer; they're not trying to bring in the "where's this Usenet-thingy I keep hearing I can get free stuff from?" crowd. In fact, if you didn't know what Usenet was when you went there, you'd have no idea.
I think they're probably safe, but you're right, it's the services who got greedy and sloppy -- catering to users who were just looking for an alternative to P2P networks, mainly -- that are going to get slaughtered if the **AAs decide to go to town on them.
They still need to concern themselves with 512(c)(A)(ii), the part which says "in the absence of such actual knowledge, is not aware of facts or circumstances from which infringing activity is apparent". Although I don't agree with it, I could see someone making an argument that anyone maintaining a multi-terabyte news server would have to have looked at the contents of some of the groups and realized that the bulk of the content of the binaries groups, which itself is the bulk of a modern newsfeed, is copyrighted material. If you found a sufficiently ignorant judge, and could manage to convince them that Usenet in general was basically a 'piracy machine,' then they might invalidate the Safe Harbor provision for anyone running a big newsfeed, even if they're careful and don't violate 512(c)(B) (the 'no profit' rule). That's kind of a nightmare scenario, but I think it bears thinking about if only as a worst-case.
http://www.usenet.com/articles/free_download.htm Humm. Yeah, that's a lot more blatant than they were making it out to be in the Ars article. In fact, what the hell, guys? I know it has a copyright date of 2005 on it, but even if that had been written in 1995 it still would have been a little much.
To wit: (in case they take the page down, which I sure would if I were them)The hell with it: They're pretty fucked.
Yes, those ads are part of the problem, because it hurts their Safe Harbor defense (see my post which quotes the section of the DMCA, further down in the thread). But only insofar as they might show that Usenet.com was benefiting directly from illegal content. And I'm not sure they do that, because the ads aren't that blatant. They basically just suggest that they have a rigorous privacy policy, etc. It's not totally damning.
Where I think they get into trouble is that, in order to claim Safe Harbor, they basically need to be able to claim "hey, somebody put that up onto our system, we didn't know it was infringing, we didn't even know it was there!" And it's a bit tough to do that with Usenet, seeing as how it's about 99% binaries and anyone who's ever opened up the alt.binaries.* hierarchy can tell that it's got a lot of bootlegs and warez in it.
It would be a little comical to see a whole bunch of seasoned network engineers and other greybeards try to claim that they had no idea there was copyrighted material on Usenet. ("Warez? On my Usenet?") But that's sort of the position they have to put themselves in, in order to get a successful 512(c) defense.
They also have to show that in the past they've complied with DMCA takedown orders against content that a copyright holder has pointed out as being infringing, which it seems like they weren't doing. That may also be a problem, although maybe they can argue that they didn't have the capability to delete articles (after all, if they took them out of their store, would they just have come in on a feed from another site that they peer with?). It might be difficult to get a judge to swallow that, though.
I think they're in trouble, but I'm not sure exactly how much trouble just yet.
To be honest, I don't know how Usenet.com can not qualify for DMCA protection, since it's exactly the type of service that the Safe Harbor exception is supposed to protect. The only thing that seems like it could harm Usenet.com is their advertising, which does veer a little into "wink, wink, nudge, nudge" territory. However, damning a company because it says it respects users' privacy, without actually advocating any type of criminal activity, seems like pretty terrible precedent, and I can only hope (although at this point I have little faith) that a judge will see it similarly.
I think the mention in the Ars article about Safe Harbor being related to "transitory network communications" is irrelevant here. Transitory network communications is covered under 512(a) of the OCILLA (which is part of the DMCA); the portion that I would expect Usenet.com to seek protection under is 512(c), "Information Residing on Systems or Networks at Direction of Users".
You can read the relevant section here, but the significant portion, IMO, is:
The major things they're going to have to avoid are that they "had actual knowledge" that the material was infringing (which might be tough -- I mean, anybody who opens up alt.binaries.movies can probably tell pretty quickly that it's full of bootlegs) and that they didn't receive a "financial benefit directly attributable" to the infringing activity. I think that second one is actually a little easier (for Usenet.com) than the former. And, of course, they have to successfully argue/explain that they don't really have the power to remove articles from Usenet, because of the nature of the network -- it would probably help their case if they started at least deleting articles from their spool/store when they receive a complaint.
I suspect that this may lead to a shakedown in the Usenet provider world, if Usenet.com loses. At the very least, the big providers might have to do more in order to maintain a veneer of plausible deniability (deleting some of the more obviously movie and/or warez related groups, perhaps), or move their servers out of the U.S.
Yes, you can do this. (And in fact, I think this is the way to go on a lot of crypto, e.g., PGPfone or OTR Messaging's fingerprint-verification systems that don't require any PKI.)
However, for email, you may and probably do want to talk to a lot of people that you may never meet in person or communicate with any other way. This makes verifying a lot of individual fingerprints cumbersome -- but if you don't have any other method for proving authenticity, you create a massive security hole for MITM attacks.
So you pretty much need some way of verifying that the public key you're being given matches the intended recipient of the message, without going to the recipient and verifying it out-of-band for each new person you want to communicate with. This requires some form of PKI; either a web of trust where lots of individuals verify each others' identity, and you can find trust paths through the web to virtually everyone else (in theory), or you have centralized "trusted authorities" whose reputation is based on verifying others' identities. PGP uses the first method (mostly), S/MIME uses the second (again, mostly). Either one can sort of be used the other way around -- Thawte's personal certificates utilize a web of trust, and you can have psuedo-authorities using PGP by setting the weight of their trust very high, so that anyone they verify is considered OK. But they both function best when they're used according to their designs.
If you only want to talk to one person securely, then sure, you can generate your own certificate, they can do the same, you can exchange them and verify the fingerprints through some hard-to-forge method (like voice phone). But this only works if you can recognize each others' voices. If you're trying to communicate with someone you've never met before, it's vulnerable to spoofing and MITM (you try to call them, but instead of them, you get the attacker posing as them; likewise, they try to call you, and instead get someone posing as you). It's not a scalable solution.
But for instant messages, where you're probably communicating over and over with a relatively small group of people, and even telephony in many instances, it would be fine. But email in particular is probably not a good match for infrastructure-less PK crypto.
True. (And in case anyone else wanted that in terms of linespeeds, that's about 2.3 Gb/s, just barely under an OC-48, but probably too much when you consider overhead. You'd be talking about a dedicated OC-96.)
.edu ghetto that's heavily peered, which is nice to see -- glad some people are keeping Usenet going in the edus -- but I suspect they have rather limited feeds.)
It's been a long time since I've thought about, or really seen anything written about, the architecture of Usenet. With those kind of requirements I suspect there can't be more than a handful of really complete news servers, and they can't be too generous about peering, because the bandwidth would be out of control. That does hell on what's ideally supposed to be a mesh topology.
It would be interesting to put together a modern Usenet map, maybe by scanning a large number of posts and reconstructing it from the Path headers. On second thought, I think that's what Top1000 does already, although they don't produce map-like output, they're more interested in traffic analysis. (Also, I didn't realize that Usenetserver.com was bigger than Giganews. Interesting. I remember when UseNetServer switched to Linux from NT and offered anyone using Linux a free account; I always thought that was cool.)
Just scribbling on the back of an envelope, I think you could seriously disrupt Usenet (in the US, anyway) by taking out a few well-selected major sites. You wouldn't "kill" it, and people who only browse text groups might never notice, but people using the binary groups would. It's pretty obvious if you start drawing a diagram that there are only a few major sites that must have very liberal peering policies that make up the network core, and then a lot of sites with more limited newsfeeds peering with each other out around the edge. (And then there's a whole
There are several ways of looking a past climate records. One way is to look at the growth rings on long living (4000+ years!) species of trees. Another way is to look at the deposit layers of ice/snow at the North and South poles, and on sedimentary layers around river deltas. All of these give some idea of what the local climate was like over the years. Cross-reference together from many locations they can give an idea of what the local climate was back then. Deposits of dust/ash/soot at the poles can indicate some serious volcanic eruptions.
Maybe it is one of these? I think Dalambertian (post right below yours) is right. It wasn't rocks, it was ice cores. I'm not sure of the exact physics, and I haven't read any of the relevant papers, but I think it has to do with the production of some very long-lived isotopes due to increased amounts of radiation during high points in the sunspot cycle. This article (which may or may not be totally specious otherwise) suggests that it's Beryllium-10. That's what I heard someone talking about.
The thing I was confusing it with is the geologic evidence of magnetosphere changes (which are recorded in various types of rock as they cool, IIRC).
Yep. Not only that -- the massive storage and bandwidth -- but you need to get a newsfeed. And that's not as easy as it used to be, when you could basically ask the sysop of your local university nicely. I'm not even sure what the commercial news servers would charge for a real UUCP newsfeed, or if they'd sell you one at all (why would they want to create competition for themselves?).
I'm not sure how many high-completion, long-retention news servers are around, but I suspect it's way, way down from what it used to be. It probably wouldn't take too many targeted lawsuits to, if not actually wipe out Usenet (that's impossible), but to at least make it very different from what it's like now. You could definitely make commercial services unprofitable, push it underground, and force people to eliminate binaries or at least shorten their completion/retentions a lot.