A busy week for the courts [LWN.net]

Various courts in the U.S. have handed down a set of decisions in the last week which have strong implications for the free software community. Here is a quick rundown of what the courts have been saying.

bnetd

The developers of bnetd had a straightforward goal: they wanted to be able to engage in networked gameplay, using their legally-purchased Blizzard games, without dealing with Blizzard's Battle.net servers. So they reverse-engineered the protocol used by Blizzard's games to talk to the server and implemented bnetd, which provides the same functionality. bnetd is licensed under the GPL.

Blizzard did not like bnetd. The provision of alternative servers took players of Blizzard's games out of the company's control; it was no longer possible to throw advertisements at players. The Battle.net servers also check the registration key provided by the game client; if the key turns out not to be valid, or if multiple players attempt to use the same key, access to the server will be denied. The bnetd developers never quite got around to implementing the key checks; free software developers have little patience with that sort of thing, and, in any case, Blizzard provides no way for third parties to check the validity of registration keys.

Blizzard's response was to send takedown notices, then file suit with a number of copyright infringement and contract claims. On September 30, a U.S. District Court in Missouri agreed with Blizzard, finding the bnetd developers guilty of breach of contract and violation of the anti-circumvention clauses of the Digital Millennium Copyright Act. The full ruling is available in PDF format.

The contracts in question are the license agreement for the games and the terms of use for Battle.net. Among other things, these contracts forbid reverse engineering of the software and running services that compete with Battle.net. The court found that the EULA and TOU were binding in all respects. Among other things, a license agreement can forbid reverse engineering in all cases and that is just fine with the court.

With regard to the DMCA charges, the court concluded that, by reverse engineering the handshake used to control access to the games' "Battle.net mode," the bnetd developers did circumvent an access control mechanism. In their defense, the developers stated that they fell within the DMCA's exemption for those trying to achieve interoperability. The court disagreed:

The Court find that the defendants' actions constituted more than enabling interoperability. The bnetd emulator developed by the defendants always allows the Blizzard game to access Battle.net mode features even if the user does not have a valid or unique CD Key, because the bnetd emulator does not determine whether the CD Key is valid or currently in use by another player. Unauthorized copies of the Blizzard games were played on bnetd servers. Then, defendants distributed the bnetd program for free. Because the bnetd source code was freely available, others developed additional Battle.net emulators based on the bnetd source code.... Finally, the defendants did not create an independently created computer program. The bnetd program was intended as a functional alternative to the Battle.net service. Once game play starts there are no differences between Battle.net and the bnetd emulator from the standpoint of a user who is actually playing the game.

It is hard to know how to read this reasoning. Interoperability, it seems, is only a defense if the resulting program does not do anything interesting, and if it is not distributed as free software.

The court also found that the developers had violated the DMCA's provisions regarding trafficking in anti-circumvention devices:

The defendants' purpose in developing the bnetd server was to avoid the anti-circumvention restrictions of the game and to avoid the restricted access to Battle.net. Thus, the sole purpose of the bnetd emulator was not to enable interoperability. The bnetd emulator had limited commercial purpose because it was free and available to anyone who wanted to copy and use the program.

This language contradicts the court's statement of the "undisputed facts" in the first part of the ruling:

The users of the Battle.net service have occasionally experienced difficulties with the service. Blizzard has also received complaints about user profanity and users who cheated to win games by modifying Blizzard's software ("client hacks")... To address their frustrations with Battle.net, the defendants joined a group of non-profit volunteer game hobbyists, programmers, and other individuals called the "bnet project."

The above is, remember, an undisputed fact. The court chose, however, to ignore that fact and recast the purpose of bnetd to suit its reasoning. On top of that, the idea that bnetd is a circumvention device because it carries a free license is truly chilling.

The end result is that Blizzard is able to place strong restrictions on the users of its games, preventing them from communicating via any sort of alternative service. Free software developers have been restricted in the sort of code they can develop, and the value of Blizzard's games for its own customers has been reduced. There are certainly problems with the DMCA which allow this sort of thing to happen. This is, however, also a problem with proprietary software; free software users do not have to cope with restrictions of this type. Unfortunately, it may be a long time before we see free games which offer the sort of experience provided by the best of today's proprietary offerings.

Diebold

The Diebold case was the source of another important ruling (PDF). In this case, Diebold attempted to use the DMCA to shut down distribution of leaked internal messages between its employees regarding problems with Diebold's electronic voting systems. The core of the ruling was that Diebold misused the DMCA by attempting to force a takedown of material which was not copyrightable.

The purpose, character, nature of the use, and the effect of the use upon the potential market for or value of the copyrighted work all indicate that at least part of the email archive is not protected by copyright law. The email archive was posted or hyperlinked to for the purpose of informing the public about the problems associated with Diebold's electronic voting machines. It is hard to imagine a subject the discussion of which could be more in the public interest.

The Diebold ruling may not affect free software developers directly, but it should serve to put some limits on the use of DMCA takedown notices.

Kodak

A court in Rochester, NY (Kodak's home town) has found that Sun has infringed upon three of Kodak's patents. Kodak claims that Sun should owe it just over $1 billion for its crime. Intellectual property suits, it seems, are increasingly the strategy of choice for businesses in decline.

The patents (numbers 5,421,012, 5,226,161, and 5,206,951) all read about the same; they would appear to describe any of a number of object request brokers or remote procedure call mechanisms. If they are upheld, Kodak can be expected to begin shaking down technology companies across the U.S.; they would be unlikely to limit themselves to those working with Java.

This looks like a case with a reasonably high likelihood of being reversed on appeal. In the mean time, it serves as yet another reminder of what software patents are doing to the computing industry in the U.S. Until the U.S. patent system is reformed, these lawsuits will be a constant threat. One can only hope that the parts of the world which do not, yet, recognize software patents are paying attention.

SCO

The SCO group had a minor setback in the IBM case when Judge Kimball denied two of the company's motions regarding scheduling. The ruling is up on Groklaw. The judge had little sympathy for SCO's position:

However, there is nothing in the Amended Scheduling Order that precludes IBM from filing motions for summary judgment, and there is nothing in the Scheduling Order that relieves SCO from responding to such motions. Thus, it is puzzling that SCO seeks to "enforce" the Amended Scheduling Order when there is nothing in that Order to justify SCO's request for a significant delay in filing its responses.

The big ruling - on IBM's motion for a summary judgment on its tenth counterclaim (stating that its Linux work does not infringe SCO's copyrights) - is still pending. (What is also pending, incidentally, is the agreement with SCO's lawyers on putting a cap on SCO's legal costs. SCO may have encountered some difficulties in closing that deal.)

(Log in to post comments)

A busy week for the courts

Posted Oct 7, 2004 15:45 UTC (Thu) by mcelrath (guest, #8094) [Link]

It is extremely doubtful that bnetd could implement the CD-check. In order to do that they would have to have the CD-Key algorithm, which is a closely-held secret of Blizzard. If anyone outside Blizzard had the CD-Key algorithm they could generate new, valid CD-Keys and play on Blizzard's servers.

Blizzard has set up their server software so that in order to create "interoperable" software one is forced to also create "circumvention" software. Therefore the user cannot exercise his rights under copyright law, or even the DMCA without violating the anti-circumvention clause of the DMCA.

And *that* is why the DMCA is so dangerous. All control can be handed to the corporation in this manner, leaving the user with no rights under copyright law at all.

-- Bob

A busy week for the courts

Posted Oct 7, 2004 16:34 UTC (Thu) by fergal (subscriber, #602) [Link]

If anyone outside Blizzard had the CD-Key algorithm they could generate new, valid CD-Keys and play on Blizzard's servers.

I would imagine that is not true. From a brief look around the net it seems there is a single CDKey that comes with each licensed copy. This key is needed both for solo play and for network play. This means that the algorithm for _verifying_ the key is embedded in the binary on the CD and could have been reverse engineered. This would not necessarily allow people to generate new keys willy-nilly. That would only be the case if Blizzard were particularly stupid and chose a symmetric encryption algorithm rather than a public key algorithm or a one-way hash function. Even if they were that stupid the fact remains that the bnet guys didn't reverse engineer the key verification part of the software when they could have (just as John Johanssen did with DeCSS).

Also, reverse engineering would probably not have been necessary to exclude multiple use of the same key. The bnet guys could have prevented a large part of the illegal usage - where someone buys one copy and then his friends play it on multiple machines connected to their private server. This assumes that the server can persuade the clients to send their keys in such a way that identical keys can be spotted, this should be possible even if the key exchange uses some sort of challenge response mechanism.

Blizzard has set up their server software so that in order to create "interoperable" software one is forced to also create "circumvention" software. Therefore the user cannot exercise his rights under copyright law, or even the DMCA without violating the anti-circumvention clause of the DMCA.

As I said above, I don't think this is actually the case but I'll assume I'm wrong for a moment. If I kick my ball into disused and dangerous mine, it's still my ball, I have a right to it. Does that give me the right to go into the mine without the owners permission? Can I force the owner to allow me in or force him to go down the mine to get my ball?

I think it's possible to have a right which you cannot excercise because circumstances (or someone elses rights) do not allow it.

Of course if Blizzard had a decent key system, they could make the public key public without any worries, thus allowing the bnet people to implement it easily. Perhaps a court could be convinced that that's the correct answer as it causes no loss to Blizzard (in terms of pirated keys) and allows the bnetd guys to exercise all of their rights.

A busy week for the courts

Posted Oct 7, 2004 16:51 UTC (Thu) by allesfresser (subscriber, #216) [Link]

>I think it's possible to have a right which you cannot excercise because circumstances (or someone elses rights) do not allow it.

That's very true. However, the right of fair use is a long-established one and the fact that the DMCA makes it very easy to trounce all over this basic right shows its callous and flawed nature very sharply. The basic understanding of copyrights has escaped so many people that the egregious nature of the DMCA glides right by without protest. The right of the copyright holder (when the Constitution was written and this has been affirmed many times until very recently) is a necessary concession, a legislated injury to the basic right of the public to share information. The idea that copyright holders should be able to have a perpetual monopoly with regard to the work(s) in question would have been (and still is, in many circles) totally abhorrent to any person familiar with the Constitution.

</soapbox> :-)

A busy week for the courts

Posted Oct 7, 2004 17:32 UTC (Thu) by fergal (subscriber, #602) [Link]

I agree with everything you say except for one thing. I don't think this is a fair use issue, it is a right to reverse engineer issue. Fair use is more about being able to install the game on multiple machines - home, work etc. I think this is possible with the game, what's not possible is connecting to battlenet from these machines at the same time, which is quite a reasonable and common restriction in the non-Free (as in FSF Free) world. One might say that reverse engineer is just a sub set of fair use but in that case you cannot argue that it is right that has been around for centuries as reverse engineering is a recent thing.

I think the bnet guys were perfectly entitled to do whatever was necessary to create an interoperating server but I think they were wrong to produce software that enabled piracy when it seems they could have prevented it partially or even fully with some extra effort. Yes it would probably have just been a gesture because sufficiently savvy end users could disable it but if they had made that gesture they might not be in their current predicament - although it seems that judge was so confused it wouldn't have a made the slightest difference, surely someone must have pointed out Redhat as an counter example to "open source has no commerical significance".

A busy week for the courts

Posted Oct 7, 2004 18:09 UTC (Thu) by mcelrath (guest, #8094) [Link]

I think the bnet guys were perfectly entitled to do whatever was necessary to create an interoperating server but I think they were wrong to produce software that enabled piracy

They did not set out to "enable piracy". In fact it is extra work for them to prevent piracy. As such, you are arguing that bnetd should be responsible for enforcing Blizzard's copyright. They did not circumvent Blizzard's copyright, rather, they chose not to enforce it. I think that is reasonable.

This is a dangerous path -- requiring all third parties to be responsible for enforcing other people's copyright. It's very similar to the "common carrier" argument used by internet infrastructure companies. This allows them to avoid enforcing other people's copyright.

So, if you want to require other people to enforce my copyright, under what circumstances should that happen? Under what circumstances is it too much of a burden on the third party? There are millions of copyrighted works and it is unreasonable to require anyone except the copyright owner to enforce them all. What about mods and maps? Should bnetd also be responsible for enforcing the copyrights on them in their server?

-- Bob

A busy week for the courts

Posted Oct 7, 2004 18:26 UTC (Thu) by Ross (subscriber, #4065) [Link]

IANAL.

My understanding of fair use does not mean making more than one copy for
multiple machines. That's pretty irrelevant anyway since Blizzard lets you
do that anyway (spawn installs).

Fair use is a right to use a copyrighted work in a way which would
otherwise be copyright infringement, but it does not allow any type of use.

Examples: time shifting a TV show on your VCR, quoting from a newspaper
article to make a point, converting your CD into MP3 format so you can
listen to it while jogging.

There are other doctrines like first sale. You have the right to sell used
books, games, CDs, etc. as if they were new. This would be copyright
infringement (distribution of the copyrighted work without permission)
otherwise.

A busy week for the courts

Posted Oct 7, 2004 18:19 UTC (Thu) by Ross (subscriber, #4065) [Link]

Blizzard tried to make this claim too. The key is encypted now (and has
been since the first patch for StarCraft) with a random number from the
server and the current time on the client. So to a server that doesn't
understand the encryption no two keys look the same, even if they are in fact
the same key.

A busy week for the courts

Posted Nov 2, 2004 9:59 UTC (Tue) by pandaemonium (guest, #25802) [Link]

From a brief look at how things work, one can tell that there must be two seperate CDKey check algorithms.
How I come to this conclusion? It's quite simple.

The first time your CDKey get's checked is when you install the game. So there must be a CDKey check algorithm in the client (or at least the installer software). Now you might happen to have used one of the various (and illegal) CDKey generators that can be found on the web. You can perfectly install the game with such a CDKey and play single player or LAN games.
Now, when you connect to Battle.Net your CDKey is checked a 2nd time. And this time only your valid CDKey from the game's box will pass the check - the one from the CDKey generator will fail (or at least 99.999% of them).

So... What is the conclusion we can make?
There is a CDKey check in the client (or installer) that allows a limited set of CDKeys and there is a check on the Battle.Net servers, that only permits a even more limited subset of those CDKeys.

So the bnetd authors might have been able to reverse the client/installer CDKey check, but what use would it have been of? This check is allready made during install... And without any help from blizzard it's impossible to reverse engineer the check algorithm implemented by the Battle.Net servers themselves...

A busy week for the courts

Posted Oct 7, 2004 17:25 UTC (Thu) by mmarsh (subscriber, #17029) [Link]

The interoperability argument would hold more weight if bnetd provided functionality *other* than a Blizzard server. The better argument is that Blizzard.net wasn't providing the service that people wanted, and many of them bought the game for the express purpose of playing on-line. The server is thus just an enticement for sales of the software, and it shouldn't matter whether people use Blizzard.net or bnetd. As far as I'm aware, nothing in copyright law guarantees an author advertising revenue.

A busy week for the courts

Posted Oct 7, 2004 17:48 UTC (Thu) by Ross (subscriber, #4065) [Link]

Well it does. You can connect to it with an IRC client for example. You
can send "mail" to other accounts. There are numerous features, some large,
many small, which distinguish the two programs. See our court filings for
more details if you wish. There are many advantages for users even where
there is no difference because the user rather than Blizzard can make
administrative decisions (who can have accounts, who is banned, who gets
operator rights, what channels exist, etc.).

A busy week for the courts

Posted Oct 7, 2004 18:14 UTC (Thu) by mmarsh (subscriber, #17029) [Link]

I stand corrected, then.

A busy week for the courts

Posted Oct 7, 2004 17:45 UTC (Thu) by Ross (subscriber, #4065) [Link]

If I had reverse engineered their CD key techniques do you not think they
would haul me into court for trade secret, DMCA, and vicarious copyright
claims? How would that be any better? And on top of that, any protections
would be meaningless becuase they could be bypassed with a user adding
"#if 0 ... #endif". It's less than pointless. And normal users are not
affected in any way by the addition of this code. What burdens me to write
copy protection code for someone else's products? Aren't they in the best
position to do that? Would that not be the same as a bank telling random
people walking by on the street that they were in charge of security for
the building? The point is that this is a problem with a perfectly
reasonable technical solution: the game should connect to Blizzard for the
key check no matter which server is being used. This is how games like
Quake work...

There are very few places in the protocol that use cryptography and the CD
key is one of them. You do not speak with any knowledge about what was
possible or attempted with respect to work on CD validation.

The program is not a "crack" and is in no way intended to be a "crack".
If a user connects to bnetd with an illegal copy of a game they have
already succeeded in bypassing Blizzard's copy protections and already have
a working copy running on their computer. They have already broken the
law. Saying bnetd is the cause does not make any sense. The DMCA
supposedly said there was no mandate to implement DRM schemes but the court
has effectively changed that by saying that if you want to reverse
engineer something and interoperate with another product you have to figure
out what types of copy and access controls it has and duplicate them -- and
do so in such a way as to not expose the fundamental insecurities of such
mechanisms.

I'm not even getting into the EULA stuff which is a whole separate mess.

I'm sorry I can't be more specific but I'm still in the middle of
litigation so I probably shouldn't even be posting this much.

A busy week for the courts

Posted Oct 8, 2004 10:31 UTC (Fri) by fergal (subscriber, #602) [Link]

If I had reverse engineered their CD key techniques do you not think they would haul me into court

If you had reverse engineered their key generation techniques then yes (although I doubt that algorithm is embedded in the client binaries). If you had reverse engineered their key verification techniques then what would be their claim? This reverse engineering would not in any way allow any extra use of their CD.

And on top of that, any protections would be meaningless becuase they could be bypassed with a user adding "#if 0 ... #endif". It's less than pointless.

If you had for example decided to relay all the license info to the Blizzard servers for verification and you had requested that people not distribute copies where this has been disabled then I think many responsible people would have complied. Sure, some people would distribute the "loose" version but it would not have been like decss. Decss is necessary for me to use my legitimate purchases on my linux machine. The loose version of bnetd would not be necessary to use your legitimately purchased CD, it is only be necessary if you want to do something illegitimate.

Also, your hands would be clean and I think you would have a more interesting court case. Perhaps it would be easy or even trivial to circumvent but you could at least argue that it is neither you nor your software that is circumventing.

What burdens me to write copy protection code for someone else's products?

Maybe the DMCA does, maybe it doesn't but the DMCA is a terrible law so look at other examples. Many rights come with extra responsibilites. In Ireland, those who have a gun license must also own a certain standard of gun-safe and must keep the guns locked in this safe when not in use. You have a right to drive but you have a responsibility not to speed (even though speeding is often harmless to others).

It's not clear to me that in general you should have a right to replicate with no responsibility to preserve existing copyright protection (where possible). You seem to want a right without a responsibility. Granted the DMCA is a shambles and makes no clear and sensible provisions for this situation. As I mentioned above in another post, maybe the sofware company should be obliged to give details of it's license verification algorithm. In this case it probably would have done no good as the check could be commented out.

If a user connects to bnetd with an illegal copy of a game they have already succeeded in bypassing Blizzard's copy protections and already have a working copy running on their computer. They have already broken the law.

I was not talking about illegal copies, I was talking about legal copies being installed on multiple machines. bnetd allows these copies to get around the spawned copy restrictions. One might argue that the spawned copy restrictions are unreasonable but blizzard would have been perfectly within their rights not to allow spawning at all. Bnetd does cause this, until bnetd arrived there was no way to get around these restrictions.

Blizzard has a particular business model and bnetd disrupts that model. This is fair enough to some extent although I think it's a bit rude and perhaps shortsighted - if you win then blizzard (and everyone else) will have to remove ad revenue from their bugets, leading to either higher prices or poorer games.

That said, I think I do agree with your right to replicate their server (I didn't say interoperate - what things are now interoperating that were not interoperating before bnetd came along?). But I'm far from convinced that your right to replicate shouldn't also come with a responsibility to protect some of the interests of the original developers (this is a general statement, not my interpretation of a particular law).

Thanks for responding and good luck with your case.

A busy week for the courts

Posted Oct 8, 2004 20:49 UTC (Fri) by Ross (subscriber, #4065) [Link]

Please keep in mind that IANAL.

If I had reverse engineered their CD key techniques do you not think they would haul me into court
If you had reverse engineered their key generation techniques then yes (although I doubt that algorithm is embedded in the client binaries). If you had reverse engineered their key verification techniques then what would be their claim? This reverse engineering would not in any way allow any extra use of their CD.

Really? I don't know for sure how their key validation code works, but I can tell you with certainty that any validation performed by the client would be insufficient because an unlicensed user would already bypassed that very check when installing the unlicensed copy. The format of the key sent to Battle.net is entirely different than the key a user types in, and it is different every time so I don't see how you would adapt the client-side check anyway. The data sent to the server is encrypted or transformed in some way which makes it appear random. I believe they do additional checks on Battle.net which are not present in the client, but how can I possibly know what those checks are, or how the random-looking data is interpreted? We can only reproduce observable behavior. The server will always send the same response to a good key so all we know is what the end result is. Do suggest we should try millions of different key inputs against their servers to try to guess at the algorithm? I can tell you what the end result of that would be (hint, it wouldn't be a working algorithm nor would it have resulted in fewer lawsuits). One data point I have is that it can't be public key encryption because there is just too little data.

Also, to distribute code which verifies or rejects a key would make it very simple for someone to create a brute-force key generator. Do you suggest Blizzard would not consider that a DMCA violation? I really, really, don't recommend that course of action for anyone. It seems at least as likely to displease Blizzard.

And on top of that, any protections would be meaningless becuase they could be bypassed with a user adding "#if 0 ... #endif". It's less than pointless.
If you had for example decided to relay all the license info to the Blizzard servers for verification and you had requested that people not distribute copies where this has been disabled then I think many responsible people would have complied. Sure, some people would distribute the "loose" version but it would not have been like decss. Decss is necessary for me to use my legitimate purchases on my linux machine. The loose version of bnetd would not be necessary to use your legitimately purchased CD, it is only be necessary if you want to do something illegitimate. Also, your hands would be clean and I think you would have a more interesting court case. Perhaps it would be easy or even trivial to circumvent but you could at least argue that it is neither you nor your software that is circumventing.

It's not possible to relay the request because it is based on session- specific information so you can't just be a man-in-the-middle and proxy the communications. I suspect if it were possible that it would not be approved by Blizzard and would have led us to a similar situation today. Blizzard could make this possible but they not want to assist us (see below about Quake-style authentication). And finally, this approach has excatly the same problem as adding a reverse-engineered key check to the server: it can be disabled with little effort so it is meaningless.

DeCSS is not neccessary for you in the same way that bnetd is not neccessary. You could choose to use an official, licensed player for your DVD. You could choose to use an official, licensed server (Battle.net) for your game. To paraphrase the other side's argument: "If you don't want to, tough; you can not buy the {game,DVD}." I see your distinction as false, and I do not see what is illegitimate when privately using properly purchased software in any way you like just like which doesn't involve duplication, distribution, public display, or another exclusive right of the copyright holder. I don't see what is wrong with using properly purchased DVDs any way you like.

I also disagree about the "clean hands". You may not know but that term "dirty hands" has a specific meaning in US law which I don't believe applies here. In any case, I already argue that my software is not circumventing.

What burdens me to write copy protection code for someone else's products?
Maybe the DMCA does, maybe it doesn't but the DMCA is a terrible law so look at other examples. Many rights come with extra responsibilites. In Ireland, those who have a gun license must also own a certain standard of gun-safe and must keep the guns locked in this safe when not in use. You have a right to drive but you have a responsibility not to speed (even though speeding is often harmless to others).

So the ability to write bnetd is an "extra right" which I should have to get permission from Blizzard for, or maybe a government agency? I thought software was speech? How about some examples: Should the WINE or SAMBA projects have to get permission from Microsoft, or even worse, from every single program with copy or access controls which use those interfaces? If Microsoft's website contained key-checking code could they claim that other websites are circumvention tools?

It's not clear to me that in general you should have a right to replicate with no responsibility to preserve existing copyright protection (where possible). You seem to want a right without a responsibility. Granted the DMCA is a shambles and makes no clear and sensible provisions for this situation. As I mentioned above in another post, maybe the sofware company should be obliged to give details of it's license verification algorithm. In this case it probably would have done no good as the check could be commented out.

So any video codec should have to implement all copy protections of other codecs? Or just closely-related ones? Should Linux have to implement the same Trusted Computing guarantees that Windows does? Should free BIOS projects have to implement DRM-encumbered BIOSes? Should Linux be able to disable Intel's unique processor ids? This path leads to insanity.

If a user connects to bnetd with an illegal copy of a game they have already succeeded in bypassing Blizzard's copy protections and already have a working copy running on their computer. They have already broken the law.
I was not talking about illegal copies, I was talking about legal copies being installed on multiple machines. bnetd allows these copies to get around the spawned copy restrictions. One might argue that the spawned copy restrictions are unreasonable but blizzard would have been perfectly within their rights not to allow spawning at all. Bnetd does cause this, until bnetd arrived there was no way to get around these restrictions.

I see. Those aren't technically legal copies, but I see the distinction you are making. The problem with that is that Blizzard already allows this. Their games are perfectly able to check for duplicate serial numbers without any interaction with the server, at least in the same game. If they do not (and I guess that is the case since we are discussing it) I would think Blizzard doesn't see it as an important check. For the case of copies in different games, see my previous comments about a Quake-style authentication server. The simple fact is that Blizzard does not want to implement that because it would separate the functionality from Battle.net thus removing legal excuses for stopping the bnetd project. We can't make them do it and the law doesn't either. So we are stuck with people telling us that we should have implemented key checking, but it is practically impossible to do without Blizzard's assistance. And don't say we didn't ask.

Blizzard has a particular business model and bnetd disrupts that model. This is fair enough to some extent although I think it's a bit rude and perhaps shortsighted - if you win then blizzard (and everyone else) will have to remove ad revenue from their bugets, leading to either higher prices or poorer games.

Blizzard's business model should really have nothing to do with the discussion. But for the sake of argument let me just say that the revenue from ads must be miniscule compared to the bandwidth costs alone.

That said, I think I do agree with your right to replicate their server (I didn't say interoperate - what things are now interoperating that were not interoperating before bnetd came along?). But I'm far from convinced that your right to replicate shouldn't also come with a responsibility to protect some of the interests of the original developers (this is a general statement, not my interpretation of a particular law).

I disagree with your viewpoint but recognize your right to disagree with mine, and also recognize that differing viewpoints can be supported by logical and rational beliefs. But specifically, I do not understand your suggestion that it is an obligation to not produce software like bnetd when it can be used with illegal copies of someone else's software even when there is no easy way to prevent it. Can car manufactures easily prevent people from running over pedestrians? No. Should they stop making cars? No. The things which interoperate are the game and bnetd. This is just like the third-party car part industry. Third-party spark plugs can interoperate with the vehicle. The term interoperation is not restricted to being a third-party mediating or adapting two existing components but also includes a new component working with a single existing one.

A busy week for the courts

Posted Oct 7, 2004 19:16 UTC (Thu) by brouhaha (subscriber, #1698) [Link]

it seems to me that bnetd _is_ a circumvention device

Perhaps, but only incidental to its primary purpose, and only because there was no way that the bnetd developers could build an interoperable server that included Blizzard's access controls. If Blizzard had offered to provide the bnetd developers with software or a network service that could be used in the bnetd server to provide the desired access control, and the bnetd developers had refused to incorporate it, Blizzard would have had a much stronger circumvention claim.