Re: [off topic] Re: Licensing issues

[Gary Lawrence Murphy <garym@canada.com>]

>>>>> "T" == Terry Dawson <terry@albert.animats.net> writes:

    T> The GPL doesn't talk about cost in a dollar sense. Of course
    T> you can sell those things. The question should be: Should those
    T> derived works be covered by GPL?

It's the "2B or not 2B" :)  meaning of course, paragraph 2b of the GPL (1991)

    T> To be GPL compliant it would have to also contain obvious
    T> notice that you've modified the work.

True, but the work retains its lineage.  If I modify XBill, it has a different
face, but the logic is still just the logic.  In words, I can change the logic
dramatically with the insertion of just a few words, and it still 'works'

    >> I see this as the first problem of the LDP: How do we stay
    >> current?

    T> User demand, how do book publishers stay current?

Then visit comp.os.linux.setup.  The demand is huge.  They are screaming
for new docs.  What modem *do* I use for mgetty?  Don't tell me the XyZel,
because it has long since ceased to exist.

Publishers meet the demand.  The 185th most popular book in all of Amazon
last month was a Linux book.  The author made several thousand dollars per
month basically just doing what the LDP should do: Teach the basics of
installing and running a Linux machine.

    T> Bah, I don't know how many or of what sort of HOWTO you've
    T> maintained, but I can assure you part of the reason I handed
    T> mine off to other people was because the work load was just too
    T> high, not with keeping the document up to date, but with
    T> responding to questions that people had of things they didn't
    T> understand, or that I'd gotten wrong, or with suggestions for
    T> things that should be included.

Exactly.  Now, what if you were being paid to fix it, and someone else
first filtered out the "I didn't understand" comments.  This is what happens
with publishers.  We finish a book and no sooner does it get to the bookstores
when the publisher calls back for a revision.

The royalties on such a book can easily be several thousand dollars per
month.  THe author of the 1989 "Unix Primer" still gets a nice $400 cheque
every single month.  It helps take the sting out of the bother.

Now, if your howto was software, you'd want your copy to perform better,
so if someon sent you a patch to improve the algorithm, you'd use it in a
flash.  Because your howto was a gift, you gave it and then had to hand
it off because it no longer served any purpose for you but bother.

    T> There is a healthy community of LDP consumers out there who do
    T> actively contribute to the accuracy and updating of documents.

And they are to be commended!  I have also hired several to work on
book projects because I want to see some of those royalty dollars
going to reinforce their very generous behaviour.

    T> I don't believe that either. We all write/maintain documents
    T> for a variety of reasons. My prime motivation was to appease a
    T> conscience upset at not being skilled/positioned to contribute

That just renames the same thing.  You gave a gift.  You didn't do it
for yourself and your clients didn't pay for it.  You paid for it and
gave it as a gift.  Out of guilt, perhaps, but it is still a gift.
(thank you)

    T> ... That has intrinsic value that I believe is
    T> quite similar to what most of the software authors achieve.

No.  Some software, but not the really good stuff.  GCC for example
has contributors from every major computer house in the world.  Emacs
is also a who's who of commercial software engineering.  Linux is
constantly improved to scratch real itches.  It's noble to think of
lone hackers in their basements working selflessly for the good of
mankind, but the reality is the vast majority of good free software
is written by and for the people who need it for real, tangible and
bottom-line-dollar reasons.  Even in university research, we used GCC
to save us the cost of SGI C, and hacked at it to support GL (that was
before openGL)

    T> I think such a model would be great if there is an equitable
    T> return on the endorsement that such works receive by being able
    T> to claim any sense of officialness from the LDP.

I believe that comes from having the LDP authors write it.  This is
why anyone involved int the LDP is a hot target for publishers when
they go to a Linux expo.  They have the machinery to publish, but need
the knowledge, we have the knowledge, but need the machinery (which
includes the financing)

    T> People tend not to want to read large works online. I don't see
    T> the LDP or the community gaining much by being to publish a
    T> document online that isn't readily consumable in that form.

Then why are the pre-release books at ORA and MCP so popular?  Is it
because people want to know a book is good before they buy it, or are
they like me and distrust books until repeated use of the online version
tells me it is more economical to just go buy it.  For example, I want
to learn some DTD, and the online docs stink.  The 'answer' given in
the mailing lists is "wait for my book" but i have no reassurance from
the online docs that the printed edition will be any better organized
than the current docs.  For example, having tags defined out of context
in alphabetical order does not solve any of my use-cases.

    T> I believe the community takes licensing issues seriously as a
    T> general rule.  I don't believe the community is necessarily
    T> well versed in the technicality of licensing though, and that
    T> it will err on the side of caution == freedom where there is
    T> any doubt. Certainly I will.

We are solving that :) On of my authors is teaming with a legal
expert to write a book on open source licences.

    T> Hmm.. I think hardware vendors who distribute binary only
    T> kernel modules will find growing resistance and pressure to
    T> release source. Do such binary modules get included with
    T> Linux's kernel source or endorsed by him?

You see? Even you appeal to the 'official kernel' :)  I will wager that
the first winmodem module, even if binary, will get wide distribution on
Linux CDs by all distros.  The itch is too great not to scratch it.

    T> The reality of the situation will probably be a diversity of
    T> works sitting somewhere within that spectrum. All the while,
    T> the publishers are able to claim implicit endorsement by the
    T> LDP while the Linux community gets a work that it can read
    T> online, or perhaps print for itself, both of which are
    T> impractical because of the size/nature of the work.

Would you really even consider printing an entire book?  I don't even
print entire HowTo's ... I just print the parts I need.  If I gave you
my Kernel book, would you need the SMP chapter?  What if all you needed
was the SMP chapter?  What if all you needed was one table in the SMP
chapter?

    T> I think if publishing companies feel strongly about that, that
    T> they should do it, without expecting/requiring endorsement from
    T> the LDP. I'm sure the LDP would be happy to maintain an index
    T> of material published online in this fashion without making any
    T> claims that those works were works of the LDP.

Then let's see it because ORA, NewRiders, Coriolis and Sams are already
doing this.

    T> Several independent publishers of LDP material exist in
    T> countries around the world. I do believe they would be missed.

To repeat, I am not advocating a blanket policy on all docs, just on
these that are 'donated' by the publishers.  The micro-publishers of
the SAG &c can still do that.  Also, the licence says you cannot publish
"without permission" and there are many precidents for such permission.

    T> ... unless you are proposing that
    T> the LDP additionally have the right to select/refuse
    T> incorporation of particular documents or that it have some
    T> mandate to determine what documents a publisher will produce.

Oh, wouldn't that be something!  Why wouldn't it work? It works for
Linux CD's

    T> I appreciate the value of having access to those skills. Still,
    T> if you're a volunatary author sometimes the last thing you want
    T> is somebody telling you that this sentence is unclear and
    T> perhaps you ought to mention fardnargle there.

So don't work on these projects.  It is that simple.  Stick to producing
little handouts.  No one is saying all docs must fall under this plan.

    T> The publishers will want control.

Oh and the LDP wouldn't? ;)  That is what a relationship is all about!

    T> Yes they are, and you yourself have already pointed out why.
    T> Motive.

No motive.  They are racking in cash now, all of them.  They don't need
the LDP's endorsement, really.  Even MIS publishes Linux books.  All they
need do is gain endorsements from any of a thousand other Linux orgs.


BTW: Exactly how many publishers have proposed this arrangement?  Remember
that I am not speaking for a publisher, but just talking about one that I
know.  Are they really beating down your door?  So far, I have been on this
list for a month and have not seen a single mention of any such deals.

    T> The LDP barely exists. I truly don't know why some group of
    T> like-minded people haven't gone off an established their own
    T> Linux documentation project, picked a catchy name, developed
    T> some sort of constitution and actively sought to encourage
    T> existing author to defect or at the very least contribute to
    T> both.

Hmmm ... "opendoc.org" appears to be taken (the webpage is a placeholder)

I know of several such projects outside of the Linux world. The LDP has
the advantage that all the publishers have it stuck in their heads you
are it: Just about every book on the shelves lists you as a primary
reference for "more information", in many cases, because the author
really didn't understand that part of their book ;)

    T> if (stalled) diverge(); else unify();

    T> That's how it works isn't it?

provided no one has set a SIGALRM :)


-- 
Gary Lawrence Murphy <garym@canada.com>  TeleDynamics Communications Inc
Business Telecom Services : Internet Consulting : http://www.teledyn.com
Linux Writers Workshop Archive: http://www.egroups.com/group/linux-hack/
"You don't play what you know; you play what you hear." -- (Miles Davis)


--  
To UNSUBSCRIBE, email to ldp-discuss-request@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org