Please create an account to participate in the Slashdot moderation system

 



Forgot your password?
typodupeerror
×
Slashback

Slashback: Cheats, Entries, Loki 328

Slashback tonight brings you updates to previous stories on computer-class cheating, Smoothwall, AIBO hacking, the Open Source Directory, and the fate of Loki's CVS. Read on below for the details!
Jon Masters was one of the many to write in after recent articles about automated cheat-detection employed in undergraduate CS classes to catch plagiarists. "Hi, cheat detection is hardly new. For example The University Of Nottingham have developed an automated marking/plagarism detection system as part of their CourseMaster software. Personally I don't agree with automated assessment in general, however plagarism detection can be useful."

From the email I've gotten on it, it seems like a whole passel of schools have at least a homegrown solution to CS cheats.

Perhaps the cute dog will end up changing Sony's stance? CodeMonkey555 writes "Here is a story that chronicles Sony's little foray into the DMCA with a hacker who added software for the Aibo robot."

It's nice to see that publications like SciAm are following the results and consequences of the DMCA.

Care to help edit an online software reference? SteveMallett writes "We at Open Source Directory (OSD) have opened the directory to volunteer editors now that we've given app authors and maintainer's a good chance to start and/or maintain their own listings.

Those interested may wish to visit our volunteer page which outlines what we're looking for. Don't worry. We're not that picky. The outline includes guidelines and tips for being a volunteer. Unlike dmoz, which has volunteer editors, we _will_ delete unupdated or neglected editor work in accordance to our Social Contract.

We hope that editors will help fill in the missing apps, take over those listings that they can do a better job of or have become neglected, and find those diamonds in the rough."

See our earlier post about the project if you're not sure what this is about.

Yes, someone has to read all those emails. kcurtis writes "Boston.com's tech site has this AP article about the large response to the Court's request for comment on the MS case's proposed settlement."

Now all they need is a trowel with an emblazoned smiley. enigma48 writes "Looks like the C'T article a little while ago about Smoothwall prompted some changes after all. Juergen Schmidt even gets a little credit. Shadow passwords are now in, but it looks like the ppp secrets file is still open (they describe it as being a "non-vulnerability"). A-patchin' I will go, a-patchin' I will go..."

So you don't have to stop playing your games ... Scott Draeker of Loki has some encouraging words for those who thought the announced (upcoming) closure of Loki would mean the loss of Loki's code and community. Draeker sent word of this a few days ago, but here are more details.

He writes:

"We have prepared tarballs of the public CVS, FAQs, mailing list archives, demos and Loki_Update which will be available for people to host. That's exactly what's going on with icculus.org.

The official repository will be hosted by the SEUL group at MIT. Once that site is set up we'll point the loki domains that direction. They'll also be adding some Loki projects to public CVS which were never completed."

This discussion has been archived. No new comments can be posted.

Slashback: Cheats, Entries, Loki

Comments Filter:
  • by Anonymous Coward on Monday January 28, 2002 @08:04PM (#2917009)
    One of the most interesting discussions of the last week discussing slashdot and editor moderation. [slashdot.org]

    It seems that CmdrTaco and co. are the "benevolent dictators" and don't seem to mind that people are peeved that they are the hidden hand behind the moderation system.

  • by Fnord ( 1756 ) <joe@sadusk.com> on Monday January 28, 2002 @08:15PM (#2917066) Homepage
    No it means the preservation of loki's already open source code (SDL, OpenAL). Loki can't open their game code. They licensed the source from the original Windows versions under proprietary terms so they could port them. Unless Loki managed to convince every company they ported a game for to open the Windows versions as well then its not going to happen.
  • by Danielle Gatton ( 534290 ) <dgatton45@nospaM.hotmail.com> on Monday January 28, 2002 @08:28PM (#2917113)
    Sony didn't (or hasn't yet) pursue legal action. They did hold out the threat of legal action under the DMCA, but they have since reached an agreement with this guy allowing him to put most of his software back on the web. The SciAm article only makes that they still could start legal action, but this seems unlikely given the agreement.

    They do make money from sales of Sony-produced software addons to the Aibo ($150 per upgrade, according to the article), so this guy's software could potentially cost them some revenue. Of course, this might well be offset by the increased sales of hardware AiboPet's software generates. I imagine some Sony bean-counters figured out the balance between the two, and decide it was in their interest to let AiboPet keep at it.
  • by Cato the Elder ( 520133 ) on Monday January 28, 2002 @08:31PM (#2917127) Homepage
    "Encryption was brought about for a variety of reasons, but one of the reasons wasn't to make sure that a product was used in a specific way"

    Yes, but the DMCA was designed so that products can only be used a certain way. A really really stupid, really really bad law, yes, but that's how it is.

    As for Sony's original position, the Scientfic American article leaves out an important factor. Originally, the AiboPet site had backups of Sony's software, obtained through crecking the encryption, available for download. Sony had a legitimate concern that these would be used improperly. Sure, no other hardware platform will run the stuff now, but it does make it easier to make clones or illegal copies of the code. Once Sony turned the lawyers loose, I think they went overboard, basically demanding that the entire site be shut down, originally. After the protests, they moderated their position quite a bit, but the "backups" are still gone.
  • MS comments online? (Score:2, Informative)

    by innocent_white_lamb ( 151825 ) on Monday January 28, 2002 @08:44PM (#2917189)
    The Boston article is ambiguous as to whether the complete comments that the public sent in will be available online.

    I, for one, believe that reading those comments would be a very instructive exercise.

    It seems reasonable to expect that comments solicited online should be published online and not just printed onto thin paper in 3-point type and filed in boxes in the storage room of a courthouse somewhere.
  • by MADCOWbeserk ( 515545 ) on Monday January 28, 2002 @08:56PM (#2917247)
    Reverse engineering is perfectly legal as long as it is done correctly. In order to legally do it, a company must employ two sets of engineers. One to analyse the product and document every input and output of it, the other must have no contact with the product. The second team then takes the data collected by the first team and engineers another product that produces the same data.

    Patents must be honored however.
  • by mfos.org ( 471768 ) on Monday January 28, 2002 @09:21PM (#2917345)
    That is actually a very astute point. I believe that in Texas, it is legal to own your own pickset, and to open your own locks with it, but if using lockpicking skills to commit a crime, that increases the penalties, because you demonstrated specialized knowledge. And you'd think that lockpicking would have a greater potential for breaking the law than would reverse enginerring.
  • by berzerke ( 319205 ) on Monday January 28, 2002 @09:24PM (#2917361) Homepage
    I hope the open source directory takes off and actually takes in editors. I applied to DMOZ for a not too large category and got an instant rejection saying new editors should apply for a smaller category. Ok, I reapplied on for a much smaller category. That was about 2 months ago. (Both categories needed an editor, BADLY). Still haven't heard back. I begin to wonder if DMOZ is even using editors anymore.
  • by Enigma2175 ( 179646 ) on Monday January 28, 2002 @09:25PM (#2917364) Homepage Journal
    Isn't that how PC cloning got started with Clean Room reverse engineering?

    Yes, reverse engineering IBM's BIOS is what gives us the power and low prices we enjoy today on x86 computers. It was totally legal to do this reverse engineering. However now we have the DCMA, which takes away some of the liberties we previously enjoyed. Under the DCMA, in some circumstances it is illegal to even TALK about how to circumvent digital "protection" (i.e. encryption). The US government is a wholly owned subsidiary of USA, Inc.

  • Re:the fall of Sony? (Score:2, Informative)

    by LoseNotLooseGuy ( 554808 ) on Monday January 28, 2002 @10:26PM (#2917552) Homepage Journal

    how much loss of revenue and market share Sony Consumer Electronics Division stands to loose in the next five years

    I am guessing you did not mean that Sony might "let loose or release" loss of revenue and market share. The word you are looking for is lose.

    Congratulations! You are participant #1 in my campaign to rid Slashdot of this error.

  • by periscope ( 20296 ) on Monday January 28, 2002 @10:54PM (#2917646) Homepage
    Hi,

    One of the reasons that I submitted the link to the software used at my University was to point out how routine this kind of thing has become.

    The idea of automated cheat detection is fundamentally a good one, it helps to remove complete weeds from around us. However, at the same time we must remember that software is only as good as what you put in to it. False positives and other negative aspects quickly displace the usefulness of such applications in my mind - especially when you may find that the analysis of "similarity" between submissions is publically available for all to see.

    The problem in my mind is when automated cheat detection develops in to other forms of automated assessment, which in my opinion are wrong. I do not believe that even the best current AI software is able to judge submissions in quite the same way as a human being, we should remember this.

    In any case, I suggest that you guys take a look around at the information each organisation has on its assessment software - it's become quite an interest of mine of late.

    --
    Jon Masters
    http://www.jonmasters.org/
  • by cduffy ( 652 ) <charles+slashdot@dyfis.net> on Monday January 28, 2002 @11:02PM (#2917663)
    I, for one, believe that reading those comments would be a very instructive exercise.

    Your local federal repository library (my local university's library is onesuch) keeps back copies of the Federal Register around. If you're in the US, you probably have one nearby, too.

    I do agree, however, that online copies would be a Good Thing.
  • by Anonymous Coward on Monday January 28, 2002 @11:48PM (#2917838)

    MIT holds an important part in the history of the Free Software Foundation, GNU/Linux distributions and the General Public License and Lessor General Public License. Despite this, Loki games intends to pass redistribution of binaries which violate the LGPL on the MIT and as such get MIT to violate it's own policies [mit.edu].

    The Loki demos and updates contain executiables which are statcially linked to glibc, libSDL and OpenAL. Each of these libraries are covered by the LGPL. Unlike the GPL, this license does allow for both dynamic and static linking with close-source binaries. But unlike a BSD or X style license, there are still other requirements which must be followed. For example, a statically linked work which displays copyright banners must also display the copyright information for the statically linked libraries. Each update contains a statically linked executiable which displays either an about or title screen with copyright information. But for whatever reason Loki has decided to exempt itself from the reasonable request of displaying copyright notice. None of the statically link binaries will ever display the copyright notices for glibc, libSDL or OpenAL.

    Then there is the primary reason for the LGPL, to ensure modification of the library is possible by allowing the modified library to be relinked with the programs that use it. Loki has choosen to only partically follow this. They do provide an execitable which is dynamically linked to glibc. But honoring this part of the LGPL for libSDL and OpenAL is something Loki again choose not to follow. There is no way to relink modifications to libSDL to HereticII including the updated one from the Loki ftp site. There is also no way to relink modification to OpenAL including the latest update.

    So is all of this theoretical problems? Not really. Violating the LGPL has practical problems. For example, all joystick handling in HereticII is passed through libSDL. The Logictech WingMan Extreme Digital 3D has five axises of which HereticII only recognizes two (X and Y-axis). It would be desirable to be able to use the other axises such as the third axis which registers twisting the joystick clockwise or counter-clockwise to control strafe left and strafe right. The HereticII layer which uses the libSDL layer will support 15 joystick buttons where the Wingman Extreme normally has 7 and the libSDL layer is capable of recognizing all the joystick axises. So, if the additional axises are each translated as two additional buttons (one button which is on when the axis is negative and another button which is on when the axis is postive) then strafing using the twist axis would be possible.

    As a proof of concept, I have written kernel code to present the axises as additional buttons and HereticII does then allow strafing left and right using the WingMan twist. But this code will never be released and will never be accepted into the kernel. It suffers from too many probelms. Such as it only effects USB joysticks, to generically support these "virtual" joystick buttons would also require changing the serial and game port joystick code to also "create" them. And in addition to having to modify three different locations in the kernel, the creations of virtual joystick buttons in kernel space ends up being messy. Finally, this type of modification bloats the kernel with code that really should be handled in user-space.

    According the LGPL, Loki must allow that this type of modification be permitted in user-space by allowing a "virtual joystick button" version of libSDL to be relinked. I even have such a version of libSDL. But Loki has decided to lock the user into one specific implimentation of the libSDL thus locking the user from making joystick code modifications in user-space code. A modification lock-out that the LGPL says can't legally be redistributed but Loki and MIT appear to be willing to do so anyways.

    Maybe it is Scott Draeker/Loki's inablity to read/follow licenses and contracts that contributed to them going out of business?
  • I think you've misunderstood what "fair use" is, it is meant for the owner of the media to be able to copy the original data incase of damage. Nowhere is disseminating the procedures on how to copy the data is part of fair use. It *only* means that if you make copies, never share them with anyone else, you can't be busted if you are caught with them (also have some library exclusions, etc.). It never says anywhere under fair use that the maker ever has to make it easy or even feasibly possible to make a copy, just that you can't be prosecuted for having a personal copy of the data.

    So in fact the DMCA does NOT trample on fair use, the guy could muck with Aibo as much as he wanted, and if you figured out how to make copies of DVD's for yourself you could copy to your hearts content. What the DMCA prevents is the ability to share the knowledge of *how* to do things, which is completely different than fair use.

    Don't get me wrong, I don't agree with the DMCA, it paints with too wide of a brush; but it pains me greatly when people don't understand the items that they are arguing for/against.

    You might want to checkout http://fairuse.stanford.edu/ it is an excellent site
  • by MEGASTeP ( 5506 ) on Tuesday January 29, 2002 @03:36AM (#2918331) Homepage Journal
    For the exact purpose you mention (compliance with the terms of the LGPL license), Loki started to release both static patches (on which technical support was available), and dynamic patches - dynamically linked binaries against all the LGPL libraries like SDL, OpenAL, glibc, etc...
    Of course those were unsupported because it's just goddamn impossible to try to support the zillions of possible combinations.

    Call it a trade-off if you wish, but Loki was doing their best to comply with the GNU licenses, while still trying to make everybody's life easier on the support side of things...
  • by Anonymous Coward on Tuesday January 29, 2002 @05:53AM (#2918524)
    The funny thing about "compliance" is that you are only in compliance when you meet all the requirements. LGPL does not give an exception to LGPL Section 6 requirements to statically link binaries just because a dynamically linked one exists. Prominent notice was still required for the statically linked work and not given. During execution the statically linked work display copyright notices without including the copyright notice for the Library among them. The availablity of dynamic use of glibc does not provide compliance for the statically linked binaries. The statically linked works is not in compliance with LGPL Section 6 and is still be redistributed without permission provided by the license and is still a copyright violation.

Kleeneness is next to Godelness.

Working...