I use this at work, but IE needs the ability to add additional zones.
I use Trusted for things like Slashdot, local sites from work, and other sites I need. If it doesn't match this, it falls into the Unknown zone which I have it prompt me for any JS/ActiveX stuff. That includes sites like CNN, NYT, etc... and that gets annoying when you read a lot of consecutive content on these sites, as you're prompted each time. I want the ability to add another zone, maybe "Untrusted" which I can add these sites to, and disable JS and stuff. Thus, sites that I visit often but don't want JS ever, it's an easy fix, while any new sites that I might encounter would get prompted.
But to the problem at hand, disabling JS will only turn off pop-up windows. Interstitials are easily done with standard HTTP commands, and from the description, superstitials are done by standard HTML as well using a trick done by some online comic sites: have calls to all the graphics for the 'next' page at the bottom of the page, but force the size to 1x1. Then, when you click on a URL going out of the site, you go first to the ad page, which then loads all the graphics from cache. Only if you want to catch ANY way of leaving the page (eg 'back') will disabling JS defeat this.
That's the chance you take if you don't take advantage of patents. What is the possiblity that someone else, within those 17 years that would have been granted by the patent, came up with the same general concept, and patented it himself? From my experience in scientific research, it's a good probability that you're building on communicated ideas, that someone else is going to come up with the same idea as you. In addition, you still have the case where the car company does get the device via entirely legal means and patents it themselves, as 'trade secrets' don't have any strong protections (*cough*decss*cough*).
And now, it's 20 years from the date the patent was applied for (as opposed to 17 from the granting date); the application process takes roughly 3 years, but this prevents the 'oops, I forgot to renew the application and have to pay a tiny tiny fine' abuses that some companies did do to artifically extend the patent 3 or 4 more years.
Yes, the court system is at fault here, as that link summary does imply, since megacorp can keep a case going indefinitely. But they you must also argue that the penalties for infringing on a patent willfully are too leninent. $100's of millions is chump change to most megacorps. But imagine if the penalty was in the billions? Or criminal penalties? It reminds me of the rumor when Firestone tires were recalled: the company didn't issue a US recall, feeling that the cost of the recall would be more than the cost of any wrongful death suits that came against it.
But yes, the patent system isn't perfect, and Kearns case is one of those that points out the flaws. But regardless of those flaws, it's still a necessary device.
Software patents work IF the invention is sufficiently unique, not based on prior art, AND has not yet been released to the public before it is approved at to allow competitors and non-competitors alike to understand and use that method. Amazon's 1-click, for example, was quite obviously on their site before they got *awarded* the patent (they had applied for it before they used 1-click, of course), and B&N was sued after Amazon was awarded it, only because B&N borrowed the idea that wasn't obviously patented at the time. Of course, the idea of 1-click isn't sound to begin with. On the other hand, things like compression algorythms which are developed in house and not released to the world until patented are ok.
But, of course, at the speed of computer software, this creates a dilemma - do you hold on to a software idea until you get the patent by which point the idea may be outdated, or do you release it to the world to gain whatever benefits you can without the protection of a patent?
The fact that there are repeat stories within a short time period (a few days) implies either the story moderators aren't reading the site and/or there's too many story moderators and sections of the site. It's completely understandable if a story is repeated after 4 or so months by a different moderator, but on the same day?
What I think needs to be done if they wish to keep all the story moderators is to make it require more than just one person needed to post a story; a moderator sees a story he'd like to post from the submissions pool, does the necessary editing, then goes on to two other editors for 'approval'; the chance that 3 out of 12(?) editors would not recognize a recently repeat story is low, and lower than the chance of 1 out of 12. Sure, this might 'slow' a story down, but/. is not a instant news wire service, only that the content is generally regarding events of the last 48hrs, so this extra moderation step should not be a problem and would improve the declining quality of slashdot.
(and if this goes through, something's wrong with the (post XOR moderate) system)....
But again, what if the guy had no idea he was selling it to a company? Without patents, a company could screw over this guy without the guy knowing how they got their information.
No, you still need to protect the small-time inventor.
Example: a guy develops, for all practical purposes, a black box that you can attach to a car engine and increase your gas milage by 200%. Assume the case where he cannot have a patent. Assuming that he's done this for the good of the world, he's going to continue making these boxes; he might only be able to get out 2 a week, and to recover equipment costs of $300 per box, he charges $350 for them. Word's going to get around, and he might end up on a local news story. A car maker employee might see that, notify his company, and they send someone in plainclothes to the inventor, and ask to buy a device. He agrees, the car employee takes the device back to their labs, opens it up and figures out the entire inner workings of it. They then figure that they can make the same device, at a cost of only $100, with a $75 profit on each box, but with a large initial captail to build the mass production plant. Two months later, the car company is heralded as innovative,
while the inventor gets nothing.
With patents in place, the inventor is sufficiently rewarded for having that patent, either by selling the patent or licensing it. If a car company does try to reverse engineer the design, the inventor has legal recourse for that.
The concept of patents are not bad, but they have to be applied right. Physical and chemical inventions are generally done correctly, while most software patents are poorly done. But this is not to say that I haven't seen bad invention patent, nor that there aren't any good software ones, but there does need to be a significant adjustment in how we grant and review them. The BT one, for example, has strong prior art, and should never have been granted. The Amazon.com one-click was one where they were awarded the patent well after the one-click idea was in place in the public, and instead, A.com should have not used one-click until they were granted the patent, such they they can say it was protected.
So now BT's played their hand, and their first target is a middle-ground network? Hello, if I was BT, I'd first target a certain pending mega-merger corp that claims to have the largest subscriber base. (cough*aol*cough) I think that BT knows they are on tenous ground, and their best shot is a company in the similar assets range compared to themselves, since we all know the one with the most money wins every lawsuit.
(and the IDG article is good, as it points out that prior art exists in 1965. Ooops).
Beside the state boundary thing, the judge's ruling questions why anyone who simply distributed the tool is being sued. Which means that the judge understands that the primary use of decss is NOT to pirate movies, and gives the defendents a much stronger case (as only one person could then be allegedly sued) that free speech trumps DMCA.
DeCSS is going to the SC, definitely, because of this.
What/. needs, if not already on the net, is a subsection for "interesting events" that can allows people to find out where things are happening with sufficent notice, and as long as it's a subsection, it wouldn't spam the front page with 'useless' info for those that cannot attend.
Napster involves taking a somewhat weaken fair use clause ('sharing of music is ok as long as it's not for commercial benefit'), and tried to apply it to something that, in the real world, is obviously questionable (allowing for the mass distribution of someone else's content without permission).
Decss involves taking a strong fair use statement, and tried to apply it to something that is normally granted in the real world (the ability to tinker with a purchased item to understand how it works).
While I think Napster was originally a bad idea, it has forced any content distributer to realize the net is here, it's not going away, and they need to adopt or die. But as the above poster puts it, any negative images of hackers will affect any other case involving hackers even if it's a different area of computing altogether.
Really good answers from Garbus here (and not too excessively lawyer-ish since everyone on/. INAL:D) One of the more subtle details that is here and was also in Kaplan's ruling, despite his bent, was that there is something fundamentally flawed about the DMCA and copyrights in general in today's day and age, that while at the current time only affect 'us' (the/. community) may have a bigger impact later. The DMCA needs to be challenged, but where it is to be challenged is in question: Kaplan suggests that Congress needs to look at it again, but Garbus thinks it needs to continue up. I personally believe the latter is true; we need to send a strong case of copyright protection verse fair use to the Supreme Court to get something finalized. The deCSS case is optimal, because the arguement for fair use protection is very strong, and could easily be used to not only repeal the DMCA but to get the CSS standard open and fixed. (Of course, given the current presidental stuff, most future SC rulings for the next few years will probably do minimal tweakage of the law to prevent drastic ramifications). At this point, it's not really about the decss stuff only, but the fact that certain rights we have for physical items no longer seem to extend to virtual/digital items.
Ok, yes, first come, first served, the true fan should keep fandom.tv (heck, he had to pay at least $1000 for it, IIRC the.tv pricing structure). Then we can get into all the abuses of ICANN on domain names, etc etc. We've been there before.
What bothers me more and something that I watch out for is this umbrella'ing of the various sites at one 'official' place for the legal protection specifically mentioned here. Fandom.com is one, but I know several other examples: WB had set up AcmeCity for fan pages with the ability to use graphics and sounds that WB had granted permission. LucasArts set up the Star Wars site for all SW creative efforts. I'm sure there's more. I think from one aspect, this is a great thing, as it shows that said companies realize that fandom DOES exist on the net and it's better to nurture it (with some ability to trim) than to let it run rampent without checks. On the other hand, all they are doing this for is for the publicity and profit, mostly. There's ad banners at most of these pages, and because it's a fandom site, it's really easy to target it. Some sites say the content becomes theirs, even if you leave the site. But again, if there are any other alterior motives, they haven't been made obvious yet.
I've been watching how the commercialization of the net has affect net fandom, and to be truthful, I don't think we've lost much. If you try to put long clips (or whole episodes) of a show on the web, you'll be shut down, but short clips as per fair use tend to be ok. You try to put up images or stories that put characters that shouldn't even be in compromising positions into compromising positions and you'll be hit. You try to make money off the visitors to your site using fair use items, and you'll get a nice letter. But these are all obvious cases of reasonable copyright/trademark protections.
What I'd REALLY like to see, instead of an umbrella'ing of sites, is for the various entertainment companies that hold the copyrights to fandom's programs to create a true contract (you'd have to mail it back in and such signed) whereby you and the studio agree that you have a fan site, and that you'll run it within all the usual provisions of fair use and non-infringing copyright as expected, for non-profit use, but on any server that you want, with the agreement that if you do happen to post something that shouldn't have been, the studio would work it out with you as opposed to immediately taking action (in others, some legal protection). In addition, I'd really like to see studios hire 'fandom correspondents', part-IP lawyers, part webmasters, that would be able to work with independent site owners to determine what is fair use and to make a better connection between the fans of the show and the studio.
According to this article, it's not that MS and Sun are fighting over the standards for XML -- that's deeply entrenched in the W3C; neither can extend the XML standard without causing serious problems for any third party developer.
What they are at odds at is how to deliever XML to end users. From the visions that I've read, the concept of.NET or any massively-XML'ed software set is there will be a server that sits there and passes XML back and forth, and apps will communicate with the server to get the XML data that they need. However, the method by which that XML data is requested and set is what MS and Sun appear to fighting over; MS *seems* to want an HTTP-like functionality, Sun's aiming for a SQL-like one. From anyone writing third-party software, it's a significant problem as the CPU-cost of implementing both (or any more than 1) methods is high.
At least, that's the way I'm reading the article...
Obviously, I agree that the NSI action of holding onto domains is bogus, but as others pointed out, NSI theorhetically holds all domains by their updated User Agreement, and therefore they have a right to sell them for such. This part *may* hold up in court.
But this then sounds like the lawfirm has a more important goal in mind: point out the fundamental flaws of NSI, such as retroactive changes in User Agreements, poor customer service, etc etc, such that NSI is penaltized by the court system for their actions. The auctioning of expired names is merely a tip of the iceberg of abused that NSI has done for years.
We're still a ways away from having a good online privacy protection act here in the states, but at least there have been strides to pave the way for this. The first one was to have privacy policies explicitly stated.
But as this CNet article points out, even though you may know if a program reports back to a company, you do not specifically know what is being sent back. I think we (as in the global web community) to start asking these sites to at least explicitly state if they are sending back any identifiable information, or if they are only sending back anonymous data. For example, the Google case, I'd trust google to say that they don't send any identity back and avoid sending back URLs that might have a similar case. But other such people, I would not trust as far if they just say "We do not collect personal information about you." -- that leaves the door open for information about your COMPUTER or the application specifically. Sure, there are cases when I DO want to be identified (online shopping, for example), but even then, it should be explicitly stated even if redudant as to get consistancy across the entire e-commerce web.
IRC is a rather antiquated and outdated protocol for what it needs to do. It's not the concept of multiple servers in leaf/hub connections that's the problem, but the protocol that is used to send messages between servers and the various types of messages that a user can initiate. There's a lack of encryption (in general, some servers have that hacked in), spoof protection, and various other details that while most people think should be added to IRC, no one wants to reprogram all the clients and servers from the ground up to add said features. Previous attempts to try to 'extend' IRC by various tricks, such as MS' infamous Comic Chat, tend to pollute the system more, and if these people set up a standard on their own IRC network, some other networks that are meant for chat only might see weird requests from people trying to leech from EFNet.
If anything, instead of starting with IRC, just develop a new protocol that would add the needed features in a desired P2P setting; don't use the IRC protocol as the model, save for how leaf/hubs work and how a message is propigated across a network.
Ok, definitely they get more specific in the details later. But I don't think that patenting CDDB was their only goal -- the abstract is what is the key -- they want a way to synch SOME info from a remote site while a CD is playing -- this in itself is not a 'bad' patent.
Again, most patent lawyers know that the rule of thumb is that if you only want a bathtub full of water, you try to patent the whole ocean, as the excess will be trimmed, leaving you with the bathtub and maybe some more. If you only go for the bathtub, you may only be left with a glassful.
In this case, they are probably aiming for the synched content, and if they can maintain the patent on CDDB like functions, then they're happier. Of course, I doubt that CDDB will stand.
And while there's no obvious fair use, because of the number of years that CDDB was free and non-profit (I know I was using it for at least 5 or 6 years ago, and a patent application doesn't take that long), the info about it is sufficiently public domain that they cannot win that claim alone.
The idea is based of CDDB, but it's NOT the CDDB database directly. The patent describes actions that a browser can take in synch with musical content that is played from a CD player. So, for example, the streaming of music lyrics to a Java applet, the synching of a music video to the music from the CD, or actions in a chat room. Yes, the idea of retrieving the CDDB database entry for a file from this could be covered, but the claims don't appear to do this. (And as others pointed out, they really have no way to defend this aspect in court). The title is very poor, as the patent's more focused on how to time and synch with musical playback, as opposed to fuzzy searching.
User defaults are your friend; style sheets are your enemy.
Ahhhhhh, not quite. Style Sheets are friends too, just don't abuse them. Importantly, do NOT declare any classes as "!important" such that they override any users setting. But as you also say, never ever never ever touch the font size setting, and always use font families as opposed to specific typefaces. But if you are using CSS for maximum readability, the only major things that you are probably playing with is spacing and indentation.
Most boxes have something on them that says "You must agree to the terms of the EULA (contained within) to use this software." Because you bought this, you have no defense to say "I never knew there was no EULA for this software!". (Alternatively, this is why those hacking CueCat software *are* in the clear: the average packaging did not contain any evidence, including a redirected on, on the distribution package -- if you simply plugged it into a keyboard port of a Linux box, you won't ever see a EULA either).
The case in point here is that if you are aware of the EULA and read it appropriately, finding that you can't agree to those terms, what recourse do you have? Can you return it? Can you continue to use it?
If you read Kaplan's decision, he took the position (that was strengthed by MPAA's case) that the defendants were hackers, and therefore his ruling was altered from one which should have been handed out if no bias on the defence was given. Specifically, from a factual point, code is free speech, and there is text allowing for bypassing encryption for interoperability in the DMCA (though there are also ones that say the reverse, so it's up to interpretation). So a unbiased judge *may* have found 2600 to be legally ok to distribute said code. Instead, a bias judge now has any push for a Linux DVD player slowed to a crawl, and potental First Amendment problems with the hyperlink problems. The judge *was* influenced and as many felt, misinterpreted even basic law because of how he was coerced. It can happen anywhere and anytime.
On the Napster thing, among the numerous other defenses that they are approaching, one that I heard them using was that the RIAA groups were overly protecting their copyright to the point where they were behaving as a monopoly, and using that copyright protection to retain their monopoly, thus falling under Sherman act regulations. Of course, the problem here is that RIAA is a group, not one company, so "monopoly" is non-existent. But they are still persuing the concept that agressive copyright protection as to remove fair use rights is a problem.
I've seen you twice now on tour: once when you were headlining for (of all bands) Hootie and the Blowfish (and I felt very sorry for the treatment that you got from those people while you were doing something like "Spy"), and when you were the main feature, leaded off by a heavy-metal group called "Chainsaw Kitties" that pretty much left the audience bored. Do you have a choice of whom you tour with? Or does some manager above you make all those decisions?
Slashdot Mirror
I use Trusted for things like Slashdot, local sites from work, and other sites I need. If it doesn't match this, it falls into the Unknown zone which I have it prompt me for any JS/ActiveX stuff. That includes sites like CNN, NYT, etc... and that gets annoying when you read a lot of consecutive content on these sites, as you're prompted each time. I want the ability to add another zone, maybe "Untrusted" which I can add these sites to, and disable JS and stuff. Thus, sites that I visit often but don't want JS ever, it's an easy fix, while any new sites that I might encounter would get prompted.
But to the problem at hand, disabling JS will only turn off pop-up windows. Interstitials are easily done with standard HTTP commands, and from the description, superstitials are done by standard HTML as well using a trick done by some online comic sites: have calls to all the graphics for the 'next' page at the bottom of the page, but force the size to 1x1. Then, when you click on a URL going out of the site, you go first to the ad page, which then loads all the graphics from cache. Only if you want to catch ANY way of leaving the page (eg 'back') will disabling JS defeat this.
And now, it's 20 years from the date the patent was applied for (as opposed to 17 from the granting date); the application process takes roughly 3 years, but this prevents the 'oops, I forgot to renew the application and have to pay a tiny tiny fine' abuses that some companies did do to artifically extend the patent 3 or 4 more years.
But yes, the patent system isn't perfect, and Kearns case is one of those that points out the flaws. But regardless of those flaws, it's still a necessary device.
But, of course, at the speed of computer software, this creates a dilemma - do you hold on to a software idea until you get the patent by which point the idea may be outdated, or do you release it to the world to gain whatever benefits you can without the protection of a patent?
What I think needs to be done if they wish to keep all the story moderators is to make it require more than just one person needed to post a story; a moderator sees a story he'd like to post from the submissions pool, does the necessary editing, then goes on to two other editors for 'approval'; the chance that 3 out of 12(?) editors would not recognize a recently repeat story is low, and lower than the chance of 1 out of 12. Sure, this might 'slow' a story down, but /. is not a instant news wire service, only that the content is generally regarding events of the last 48hrs, so this extra moderation step should not be a problem and would improve the declining quality of slashdot.
(and if this goes through, something's wrong with the (post XOR moderate) system)....
But again, what if the guy had no idea he was selling it to a company? Without patents, a company could screw over this guy without the guy knowing how they got their information.
Go to the "Power" button, drop menu down to "Power Plants", and plop down a coal plant for $6,000 in some cleared space.
Or better yet, raise some terrain, and add a water space for $40, and then a hydrodam for $400, and viola, more power!.
Example: a guy develops, for all practical purposes, a black box that you can attach to a car engine and increase your gas milage by 200%. Assume the case where he cannot have a patent. Assuming that he's done this for the good of the world, he's going to continue making these boxes; he might only be able to get out 2 a week, and to recover equipment costs of $300 per box, he charges $350 for them. Word's going to get around, and he might end up on a local news story. A car maker employee might see that, notify his company, and they send someone in plainclothes to the inventor, and ask to buy a device. He agrees, the car employee takes the device back to their labs, opens it up and figures out the entire inner workings of it. They then figure that they can make the same device, at a cost of only $100, with a $75 profit on each box, but with a large initial captail to build the mass production plant. Two months later, the car company is heralded as innovative, while the inventor gets nothing.
With patents in place, the inventor is sufficiently rewarded for having that patent, either by selling the patent or licensing it. If a car company does try to reverse engineer the design, the inventor has legal recourse for that.
The concept of patents are not bad, but they have to be applied right. Physical and chemical inventions are generally done correctly, while most software patents are poorly done. But this is not to say that I haven't seen bad invention patent, nor that there aren't any good software ones, but there does need to be a significant adjustment in how we grant and review them. The BT one, for example, has strong prior art, and should never have been granted. The Amazon.com one-click was one where they were awarded the patent well after the one-click idea was in place in the public, and instead, A.com should have not used one-click until they were granted the patent, such they they can say it was protected.
(and the IDG article is good, as it points out that prior art exists in 1965. Ooops).
Don't have one? Throw them out.
Then ask for their karma. Then apply the simple formula: R = (1,000,000 - UserID) + Karma*20000.
Whoever has the largest R value is your perfect candidate!
(that was a joke, in case you couldn't tell).
DeCSS is going to the SC, definitely, because of this.
What /. needs, if not already on the net, is a subsection for "interesting events" that can allows people to find out where things are happening with sufficent notice, and as long as it's a subsection, it wouldn't spam the front page with 'useless' info for those that cannot attend.
Napster involves taking a somewhat weaken fair use clause ('sharing of music is ok as long as it's not for commercial benefit'), and tried to apply it to something that, in the real world, is obviously questionable (allowing for the mass distribution of someone else's content without permission).
Decss involves taking a strong fair use statement, and tried to apply it to something that is normally granted in the real world (the ability to tinker with a purchased item to understand how it works).
While I think Napster was originally a bad idea, it has forced any content distributer to realize the net is here, it's not going away, and they need to adopt or die. But as the above poster puts it, any negative images of hackers will affect any other case involving hackers even if it's a different area of computing altogether.
What bothers me more and something that I watch out for is this umbrella'ing of the various sites at one 'official' place for the legal protection specifically mentioned here. Fandom.com is one, but I know several other examples: WB had set up AcmeCity for fan pages with the ability to use graphics and sounds that WB had granted permission. LucasArts set up the Star Wars site for all SW creative efforts. I'm sure there's more. I think from one aspect, this is a great thing, as it shows that said companies realize that fandom DOES exist on the net and it's better to nurture it (with some ability to trim) than to let it run rampent without checks. On the other hand, all they are doing this for is for the publicity and profit, mostly. There's ad banners at most of these pages, and because it's a fandom site, it's really easy to target it. Some sites say the content becomes theirs, even if you leave the site. But again, if there are any other alterior motives, they haven't been made obvious yet.
I've been watching how the commercialization of the net has affect net fandom, and to be truthful, I don't think we've lost much. If you try to put long clips (or whole episodes) of a show on the web, you'll be shut down, but short clips as per fair use tend to be ok. You try to put up images or stories that put characters that shouldn't even be in compromising positions into compromising positions and you'll be hit. You try to make money off the visitors to your site using fair use items, and you'll get a nice letter. But these are all obvious cases of reasonable copyright/trademark protections.
What I'd REALLY like to see, instead of an umbrella'ing of sites, is for the various entertainment companies that hold the copyrights to fandom's programs to create a true contract (you'd have to mail it back in and such signed) whereby you and the studio agree that you have a fan site, and that you'll run it within all the usual provisions of fair use and non-infringing copyright as expected, for non-profit use, but on any server that you want, with the agreement that if you do happen to post something that shouldn't have been, the studio would work it out with you as opposed to immediately taking action (in others, some legal protection). In addition, I'd really like to see studios hire 'fandom correspondents', part-IP lawyers, part webmasters, that would be able to work with independent site owners to determine what is fair use and to make a better connection between the fans of the show and the studio.
At least, that's the way I'm reading the article...
But this then sounds like the lawfirm has a more important goal in mind: point out the fundamental flaws of NSI, such as retroactive changes in User Agreements, poor customer service, etc etc, such that NSI is penaltized by the court system for their actions. The auctioning of expired names is merely a tip of the iceberg of abused that NSI has done for years.
But as this CNet article points out, even though you may know if a program reports back to a company, you do not specifically know what is being sent back. I think we (as in the global web community) to start asking these sites to at least explicitly state if they are sending back any identifiable information, or if they are only sending back anonymous data. For example, the Google case, I'd trust google to say that they don't send any identity back and avoid sending back URLs that might have a similar case. But other such people, I would not trust as far if they just say "We do not collect personal information about you." -- that leaves the door open for information about your COMPUTER or the application specifically. Sure, there are cases when I DO want to be identified (online shopping, for example), but even then, it should be explicitly stated even if redudant as to get consistancy across the entire e-commerce web.
If anything, instead of starting with IRC, just develop a new protocol that would add the needed features in a desired P2P setting; don't use the IRC protocol as the model, save for how leaf/hubs work and how a message is propigated across a network.
Again, most patent lawyers know that the rule of thumb is that if you only want a bathtub full of water, you try to patent the whole ocean, as the excess will be trimmed, leaving you with the bathtub and maybe some more. If you only go for the bathtub, you may only be left with a glassful. In this case, they are probably aiming for the synched content, and if they can maintain the patent on CDDB like functions, then they're happier. Of course, I doubt that CDDB will stand.
And while there's no obvious fair use, because of the number of years that CDDB was free and non-profit (I know I was using it for at least 5 or 6 years ago, and a patent application doesn't take that long), the info about it is sufficiently public domain that they cannot win that claim alone.
Ahhhhhh, not quite. Style Sheets are friends too, just don't abuse them. Importantly, do NOT declare any classes as "!important" such that they override any users setting. But as you also say, never ever never ever touch the font size setting, and always use font families as opposed to specific typefaces. But if you are using CSS for maximum readability, the only major things that you are probably playing with is spacing and indentation.
The case in point here is that if you are aware of the EULA and read it appropriately, finding that you can't agree to those terms, what recourse do you have? Can you return it? Can you continue to use it?
On the Napster thing, among the numerous other defenses that they are approaching, one that I heard them using was that the RIAA groups were overly protecting their copyright to the point where they were behaving as a monopoly, and using that copyright protection to retain their monopoly, thus falling under Sherman act regulations. Of course, the problem here is that RIAA is a group, not one company, so "monopoly" is non-existent. But they are still persuing the concept that agressive copyright protection as to remove fair use rights is a problem.