Linux-Advocacy Digest #608, Volume #27 Wed, 12 Jul 00 02:13:04 EDT
Contents:
Re: Richard Stallman's Politics (was: Linux is awesome! (T. Max Devlin)
Re: OT, Re: offering escape, Re: Three things not to say to spam victims... (Walter
Dnes)
Re: Linsux as a desktop platform (void)
Re: Linsux as a desktop platform (void)
----------------------------------------------------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: gnu.misc.discuss
Subject: Re: Richard Stallman's Politics (was: Linux is awesome!
Date: Wed, 12 Jul 2000 00:52:47 -0400
Reply-To: [EMAIL PROTECTED]
Quoting Russ Allbery from comp.os.linux.advocacy; 11 Jul 2000 02:19:01
>In gnu.misc.discuss, T Max Devlin <[EMAIL PROTECTED]> writes:
[...]
>> Because, unlike secret source commercial code, it is only copyright law
>> which requires that you agree to the license.
>
>For the time being at least, I believe that's true of commercial code as
>well.
Not at all. Copyright law does not require licensing the end user of
software any more than it requires the licensing of the reader of a
book. You buy the book, you're "allowed", you don't need permission, to
read it. Likewise, when you buy a binary (you give someone money, they
give you the binary, regardless of any other components of that
transaction), you own it and can run it and keep an archive copy and the
copyright law says so and you don't need a license. The only people who
need licenses according to copyright law, so to speak, are distributors
and developers, because it is the responsibility of the author to
enumerate which rights to copy he is selling them when he authorizes
them to copy his intellectual property. If he doesn't, then the
distributor or developer can do anything they want with it, including
*decide who else can distribute or develop it*. Since this ultimate and
control of the intellectual property resides in the copyright holder,
originally the author, they need his "permission" to distribute or
develop. But the user doesn't need any permission to run it, once they
get a hold of a legal copy of the binary.
So the only thing that "requires" software licensing for end users is
the fact that agreeing to that license is the only way they can get
their hands on a legal copy of the binary. But they're not buying that
binary from the author; they're buying it from a distributor, if they're
buying it at all.
> I don't believe there is any special statutory power given to
>shrink wrap licenses at this time, although there's an effort underway to
>do that precisely because it looks like copyright law may not be enough to
>support them. I'm not a lawyer or legal expert, though, and that's very
>much a legal question.
I'm not a lawyer or a legal expert, either, generally. But I have
studied this issue in some detail. To say that "copyright law may not
be enough" is actually pretty close to the reality. It never was
enough, and never would be enough, because when you copyright something,
you generally publish it, and then the reader can do with it whatever he
wants. Except copy it. But all of the meanings of "copy" besides
"distribute or develop" are already allowed by law. Copying it
somewhere in order to use it is OK, as is copying it to an archive
backup, as well as the possibly standard array of fair use exceptions.
>> Commercial software licenses have a very different character, and are
>> not "based on" copyright law, but merely your agreement when you buy it
>> to agree to its stipulations.
>
>This is unclear.
Copyright law neither underlies nor empowers end user software
licensing. These are actually the same as the "trade secret" licenses
which software previously used (before it was protected as
copyrightable, beginning in 1976), except for one minor change. The
first line says "This software is copyrighted." But the restrictions on
only using it on one PC? No network server installs? No home/office
use? No decompile? No other computer (sequentially)? No negative
comparisons or published benchmarks? These are all authorized only by
your acquiescence. These actions are not transgressions of copyright
law (necessarily), and are only empowered because if you don't agree,
you can't get the binary. Even if you didn't agree, you still couldn't
distribute the binary, or develop derivative works. But that's all you
couldn't do, except for the whim of the producer to keep you more
restricted.
>> Users don't need to agree to it other than its the only way they can get
>> the trade secrets which are the source code, because everyone who has
>> the trade secrets is bound by license not to give them or sell them to
>> anyone else.
>
>I'm fairly certain that your use of "trade secret" in this context is
>legally incorrect, at least from the formal definition of a trade secret,
>but again that's based on my understanding of how trade secrets work
>(largely derived from the various Church of Scientology struggles over
>this issue) so I could well be incorrect.
Absolutely, yes. My use of trade secret is quite precisely identical to
the formal legal definition. Windows is licensed because it is a trade
secret, not because it is copyrighted work.
>But I don't think the program
>that you receive from MS is properly a trade secret, given that they're
>not keeping it secret.
Yes, they are. By requiring every single person who receives a binary
to agree to a promise not to decompile it.
>The *source code* may be, but note that the source
>code isn't part of the license and isn't an issue at all with proprietary
>software. You're not being given it in the first place.
Note that there is no legal distinction between source code and object
code in terms of copyright. The object code (binary) isn't a trade
secret; everybody knows how that works. But without being able to
decompile, nobody knows how it works. You see?
>> The user of a binary compiled from open source doesn't need to agree to
>> a license,
>
>This is somewhat unclear as well, actually, but that's a longer argument.
>(It's interesting to read the various open source licenses and figure out
>what they say about how the program can be *used*.)
They are all saying the same thing. What is interesting is to see the
various abstractions of the concept "software" they're referring to,
because that changes from clause to clause.
A proper understanding of the existential (not the technical)
relationship between object code and source code is necessary to
understand why GPL is "free" in the way that RMS's politics indicates.
Thank you for bringing us on-topic.
>> because they aren't copying it in any way that isn't allowed by fair use
>> or essential steps exemptions in copyright law.
>
>Wait, hold on, you're going a little too fast. The essential steps
>provision applies exactly the same to proprietary software as it does to
>open source software.
Yea. You see how that works? In other words: end users don't have to
agree to licenses according to copyright law. Proprietary (secret
source, trade secret, closed code) software licenses are different for
end users and for developers. For open source software, they are the
same license. There is no "End User License Agreement" and then a
"Developer License" or "OEM License". There is only the license. And
that is because that license rests purely and squarely on copyright law,
with no aspects of trade secret, even as the trade secret licenses seem
to invoke copyright.
[...]
>The fundamental difference between proprietary and open source software
>isn't at this level. The difference lies in what's required to get a
>legal copy of the program in the first place.
Exactly. You've nailed it.
But with proprietary software, what is a "legal copy" for an end user is
not a "legal copy" for a distributor or developer.
The difference is because the developer must promise to maintain the
secret, while the user must promise to pretend it isn't there. ;-)
>> Could you be more specific?
>
>Kerberos, under the hood, isn't actually that complex as far as large
>software packages go. It's certainly not that hard to reimplement from
>the protocol descriptions. MS's Kerberos implementation is not
>bug-compatible with MIT's, has some very substantial differences at the
>core level with how it interacts with their server software (such as all
>the Active Directory tie-in's), and otherwise does not behave like a
>derivative of MIT's work. It's quite possible that MS programmers looked
>at the MIT code in the process of writing their own, but it's also quite
>possible that they didn't (out of legal paranoia).
Then I would agree with you. Thanks for the response. Again, I never
expected the Kerberos question to be considered an actual case, but a
thought experiment. Even still, I wasn't aware of any reason to suggest
that MS didn't copy the reference. Now that you have provided more than
sufficient reason to question that, I have been answered.
>> Well, there you have a point, I'll admit. But if we're talking
>> copyright law, then it is common to view the work in order to see how
>> derivative it is. And software "developers" that make their profits on
>> secret source don't generally like a lot of people examining their code,
>> which makes it a might hard to consider, let alone determine, just how
>> derivative the code is.
>
>Certainly true. (It's quite possible that there are various commercial
>packages that do contain GPL'd code and have successfully hidden this
>fact.)
And yet Stallman is attacked for seeming to be vindictively aggressive
against linking. (I don't know if you ever levied such claims; I'm
merely remarking.) Would not a commercial product which only
successfully worked when linked to a GPL library not be considered a
valid reason for questioning if it is derivative, and insisting on
looking at the source code to verify this?
>Correct. However, it's *not* clear whether simply looking at the other
>work makes your work a derivative work. It's rather hazy legal territory,
>as far as I understand, and open to interpretation.
It is clear that looking at the other work is not sufficient to make
yours a derivative work, when considering software as IP. We read from
books, and write our own. It is only the nascent level of software
which we are dealing with in this day and age which makes it appear as
if there can be some question of whether something is derivative.
Essentially, using the same turn of phrase in literary work would be
considered infringement in software, in the current state of the art.
But for that very reason, of course, we must consider that using the
same turn of phrase *is* sufficient reason for considering plagiarism.
It isn't so much that whether a work is derivative is hazy, but what
derivative means, which requires more precedent to consider clear.
>> You're right that the GPL can't draw the line. You're wrong to think
>> that this means that copyright law can draw that line very easily or
>> simply.
>
>I'm not arguing that. I'm just pointing out that MS is a company that
>pays a great deal of attention to legal issues (and takes advantage of
>them) and was aware going in that their Kerberos implementation was going
>to undergo quite a bit of scrutiny in that department. I don't like MS's
>management either, but they're not stupid, not about things like this.
Yes, but you also thought that the Windows empire was based on
copyright, didn't you? :-)
Again, MS was aware of scrutiny all over, and still committed obviously
incriminating actions. But I do understand your point; they must make a
habit of CYA, they just make mistakes sometimes.
>> "Derivative work" is a concept that only makes sense when everybody can
>> view the work itself, such as literature, artistic images, and source
>> code. So if your source code isn't "published" (open), there's no
>> reason to assume that a work that could be derivative, isn't derivative,
>> as I've said.
>
>Well, yes, there is. Innocent until proven guilty, which is a rather
>fundamental assumption underlying our entire legal system (even the civil
>court system; the plaintiff still bears the fundamental burden of proof in
>most situations).
Well, you're over-generalizing (or, rather, mis-applying through
over-generalization) the concept of innocent until proven guilty. You'd
have more luck with the 5th Amendment, to be honest. Innocent until
proven guilty doesn't mean we're required to believe unreasonable
contentions. Only that we allow for reasonable doubt.
>> If MS was known for not being sloppy, Windows would work well, and they
>> wouldn't have held on to those incriminating emails.
>
>Purging e-mail is remarkably difficult, and MS *isn't* sloppy.
Doing anything on Windows is pretty tough. The sloppy isn't in failing
to purge the emails; the sloppy was in developing business strategies
based on "cut off their air supply" or "the intent is for the user to
get confused".
>If MS were
>sloppy, they wouldn't be where they are today.
Convicted? How about where they'll be in five years; check back with me
then.
>They're very large, and
>they're driven by marketing and sales goals rather than strong
>engineering, but that is not the same thing.
Yes, it is. Stupid criminals are common, and even smart criminals are
stupid enough to commit a crime and get caught. Yes, there might be a
master criminal out there who has never been suspected, and yes MS had
enormous "success" in their nefarious schemes. But they weren't hard to
spot; just hard to prove. Open source makes them easy to spot, so a
hint of GPL would have scared MS away from Kerberos in its entirety;
they'd never even consider implementing the protocol.
>Just because MS's software
>isn't up to the reliability standards that many of us would like to see
>doesn't mean that the company isn't good at what it does.
No, but it is a reasonable supposition, to be honest. Because what it
does, regardless of all of the floating abstractions which are rampant
in today's thinking, is produce software. And reliability standards are
practically the most important thing in software. The most perfect
software product in the universe is worthless unless it interoperates
with other software products reliably.
>MS is *very*
>good at what it does; it's just that producing high-quality,
>high-reliability software isn't what the company was constructed to do.
That is common, but dangerous, thinking. MS was chartered by the
government to make software (bet you didn't know that.) It was allowed
to sell shares in order to raise capital to do so. It doesn't have a
purpose other than that, and the reason it is allowed to make a profit
is so that it can continue to make software and pay a dividend to its
investors, not to act as a free-floating greed spirit sucking money from
whatever unsuspecting consumers it can in order to profiteer even
further and call that "success".
--
T. Max Devlin
Manager of Research & Educational Services
Managed Services
[A corporation which does not wish to be identified]
[EMAIL PROTECTED]
-[Opinions expressed are my own; everyone else, including
my employer, has to pay for them, subject to
applicable licensing agreement]-
====== Posted via Newsfeeds.Com, Uncensored Usenet News ======
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
======= Over 80,000 Newsgroups = 16 Different Servers! ======
------------------------------
From: [EMAIL PROTECTED] (Walter Dnes)
Crossposted-To: news.admin.net-abuse.email
Subject: Re: OT, Re: offering escape, Re: Three things not to say to spam victims...
Date: 12 Jul 2000 04:53:52 GMT
Reply-To: [EMAIL PROTECTED]
On 7 Jul 2000 20:21:01 GMT, Cameron L. Spitzer, <[EMAIL PROTECTED]> wrote:
> In article <[EMAIL PROTECTED]>, Ron Ritzman wrote:
> >On 7 Jul 2000 06:19:40 GMT, [EMAIL PROTECTED] (Cameron L.
> >Spitzer) wrote:
> >
> >>Posed that way, it's not "Nyah, nyah, my OS is better than yours"
> >>any more, it's "welcome to my community, enter when you're ready,
> >>the door is open for you." It's pure empowerment, not a put down.
> >
> >I was amazed as to the hoops that one had to jump through just to
> >do simple things such as getting the sound card working. Windows
> >plug n pray had spoiled me.
>
> It's awful, and Red Hat and Corel are not helping by claiming
> otherwise.
> Linux is still half-baked for the non-technical desktop.
> Sound card installation support, X Window System configuration,
> and security are all still in a state which would not pass
> any respectable software quality assurance organization.
> The documentation is disorganized and contradictory.
> I've been very careful, in evangelizing, not to mislead.
The basic problem is that the average newbie user is *NOT* capable of
installing *ANY* current OS (be it Win98, NT, or linux) from scratch.
Windows succeeded because Bill Gates managed to maneuver his OS into
being installed on just about every Intel box shipped in the past 10
years. If newbies were given a boot floppy, an install CD, and a
computer with a blank harddrive, you'd be hearing all sorts of horror
stories about Windows installs.
--
Walter Dnes <[EMAIL PROTECTED]>
SpamDunk Project procmail spamfilter at http://www.waltdnes.org
I'm not repeating myself; I'm an X-Window user... I'm an ex Windows-user
------------------------------
From: [EMAIL PROTECTED] (void)
Crossposted-To: comp.sys.mac.advocacy,comp.os.ms-windows.advocacy,comp.unix.advocacy
Subject: Re: Linsux as a desktop platform
Date: 12 Jul 2000 05:05:24 GMT
On Tue, 11 Jul 2000 23:35:01 -0400, T. Max Devlin <[EMAIL PROTECTED]> wrote:
>
>Any app grabs control, and can screw it up royal you are correct, *when
>the foreground application yields*. All good foreground applications,
>of course, yield on a routine basis. But that only highlights the lack
>of necessity for pre-emptive multitasking, when it is assumed that all
>of the programs running on one computer are under the cognizance and
>control of one operator. In multi-user systems, obviously, this
>wouldn't work at all. But on a multi-tasking single user operating
>system, it does, in fact, make sense to put the user, rather than the
>software, in charge of what's important.
Single-user or not, nobody wants their computer locked up because one
application has a serious bug. Operating systems should be resilient
against programmer error, because bugs happen and they happen a lot.
Can you find a way to make cooperative multitasking as robust as
preemptive multitasking?
Anyway, the type of multitasking and the behavior of the GUI are not as
tightly coupled as you think.
>Isn't it nice the way the design for what was intended to be an open
>application architecture platform encourages, no, demands, cooperation
>amongst all application programmers?
No, it's lousy. Even when written with the best intent in the world,
programs screw up and fail to yield the processor. This scheme also
results in less efficient use of processor time, which in the end means
that things take longer.
These things degrade the user experience, drastically in the first case,
subtly in the second.
>I think the PC could benefit from
>this lesson; it would have made more sense for MS to follow this model
>than the one they did. For a desktop system, which isn't even being
>used as a workstation (though it might still be more appropriate, TBH)
>level box, it just doesn't make sense to do it the same way as a
>multi-user/host/server system.
I disagree, and so do Apple, Microsoft, Be, and anyone else who might be
producing new desktop operating systems.
>> Typing something in a news readers uses what? 2% of the
>>CPU? Yet if you're decompressing something in the background, it will
>>get dramatically slower.
>
>Better something in the background gets slower than my typing into my
>newsreader, you betcha, damn right. Whatever *I* am interacting with
>should have absolute first shot at every cycle it needs.
But on unix, the background task will only slow very slightly, while
interactive apps are still nice and responsive. You should read up on
the algorithm used to do this, it's quite clever.
>> Huge amounts of processing power are simply
>>wasted. In the real world, PMT just works better on the desktop, and I
>>say this as one of biggest Mac fans you'll ever meet.
>
>In the real world, huge amounts of processing power are simply wasted on
>all desktop systems. Or used, if you're of a more functional mind set.
>I'd rather that power get 'wasted' in my direction than because it is
>theoretically better to do it different.
No. There is no benefit that comes with this waste. There are
scheduling algorithms that can take advantage of unused cycles without
slowing down interactive processes noticeably.
>It wouldn't surprise me if it was a Linux which allowed adjustment to
>just how pre-emptive the multi-tasking is.
That's because you overestimate the role of multitasking in determining
how the GUI functions.
>Because it doesn't make any sense, when the primary
>purpose of a computer is to provide a user interface, that that user
>interface, and whatever interaction the user is executing, should always
>have first dibs. Modal dialogs aren't any worse than BSOD or cascading
>segmentation faults.
I can't tell what you're saying here.
--
Ben
220 go.ahead.make.my.day ESMTP Postfix
------------------------------
From: [EMAIL PROTECTED] (void)
Crossposted-To: comp.sys.mac.advocacy,comp.os.ms-windows.advocacy,comp.unix.advocacy
Subject: Re: Linsux as a desktop platform
Date: 12 Jul 2000 05:12:33 GMT
On Wed, 12 Jul 2000 04:44:14 GMT, ZnU <[EMAIL PROTECTED]> wrote:
>
>The amount of processor time an app should yield depends on what else
>the system is doing, which is something the app can't know. Thus, CPU
>time should be allocated by the system.
Well put.
>The needs of a desktop system
>are different than those of a multiuser system or a server, of course,
>but PMT still wins out, it just needs to be tuned differently (e.g.
>things like user input need to be giving high priorities).
A modern scheduler is sufficently self-tuning to work well on servers or
desktops without modification. When my FreeBSD machine runs its nightly
2 am maintenance scripts, I can only tell if I'm physically present to
hear the disk drive's noises.
>Virtually every PMT system allows priorities to be assigned to tasks.
>You just need to give user interaction tasks very high priorities, so
>the system remains responsive under load.
It tends to do that anyway, because interactive tasks spend a lot of
time waiting for user input, which causes them to accumulate priority.
--
Ben
220 go.ahead.make.my.day ESMTP Postfix
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.advocacy) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Advocacy Digest
******************************
