[
Lists Home |
Date Index |
Thread Index
]
All fine and good as long as:
1. You indemnify that code. Essentially,
you will absorb the cost of any claims against it.
2. You release your rights to make claims to
any IP you contribute, and therefore, your
right to trade on it. How bounded that is is
a topic for another thread.
The first item is a condition that your buyer
can make on you. The second is a condition
that you make to be a contributing member
of the open source community. It may be the
case that the community will obligate you
to indemnify it and by aggregation, those
who profit by your code will cite your
indemnification but that may not be sufficient
given as said earlier, you may not be able
to cover the total costs of liquidation
of a given customer. My guess is that the
companies profiting by the code will have
to indemnify. That would be your group
all the way up to IBM and Red Hat.
I know the legal cases are bizarre. That isn't
anything new, but there is no misunderstanding of
needs. There is a conflict in the business
models of certain producers and consumers.
The producers of open source are so far
unwilling to manage the risks their model
exposes their customers to and some customers
are unwilling to obligate themselves to those
risks.
That is really what this is about, not the
means by which you make your code better
from version to version. There are other
parts of the contract, such as performance
clauses, liquidation clauses, etc. that
cover issues of performance and quality.
Attempts in the press or elsewhere to
shift the topic to FUD about Microsoft
or to plead that only open source can be
better, or to wrap open source in a
patina of moral majesty will not change
the facts of standard business contracts.
They only irritate the judge.
Yes, some customers will forgo indemnification
and yes, they will get what they pay for. It
may be higher quality than the last version
of that product, but with unknown risks of
claims.
The open source community must provide
indemnification. Acquisition of IP is
a separate topic, related in that it has
value and therefore can be traded on, but
not essential to managing risk.
The articles I've read on the SCO/IBM dispute
contradict your interpretation, but I don't
know the details so can't comment. As I
said, it is not the issue but an incident that reveals a
weakness in the open source business model.
Note well, it is the business model and not the
product quality that is at issue.
len
From: Rick Marshall [mailto:rjm@zenucom.com]
i think there's a basic misunderstanding here between the "needs" of
bigger business from a corporate perspective and the "needs" of the
programmers, contractors, consultants trying to make things happen
across a myriad of businesses, large, small, and non-existent (or
personal).
but before i comment further, there is something very bizarre about the
latest in the sco round, and it is a warning to all of us.
if i understand their latest demand correctly they assert that ibm
developed some ip, contributed it to unix, and that it is therefore now
owned by sco, even though they didn't pay for it. ibm no longer is
considered free to do with their ip as they like - ie contribute it to
linux. this may go further than ip and to the very heart of contract law
and the nature of a contract - does there have to be a valuable exchange
for there to be an enforceable contract?
open source, and standards, make it easier for large numbers of
professionals, like myself, to build larger, better, more innovative,
and more reliable systems because we share expertise. some of it's
operating systems, some of it's ideas, some of it's programs. this is
really about the workers working better, not the corporations getting
richer. but it is working very well, unofficially, for the corporation.
and often it's the simple tools that are really good - xsltproc and my
own xml design for multi page forms has transformed the nightmare task
of getting good postscript. thanks to the w3c, xml, and gnu.
len, you asked earlier about what i do to protect my ip. the answer
really is nothing. granted that in australia my rights are protected by
default as the author under australian copyright law. however i see it
in my interest for others to take the ideas, copy and use them.
otherwise they'll simply die.
the ip is broad - a database language contemporary with sql that
understands data relationships and can express far more complex
mathematics than the simple functions we seem to discuss in this group,
b+ tree algorithms, tree pruning, optimistic table locking, and
commercial apps.
right now i want to build xml functionality deep into the database, as
an updated method of expressing our totally declarative programming - ie
a slightly more flexible approach. we also use a lot of interprocess and
intersystem communication and xml seems like a good way to do that.
if i can ever get to the bottom of web services that should fit nicely
too, but they will be the subject of a new thread.
so for me at least all this is about our work personally. if the
corporation is happy buying based on the size and security of the
organisation and its ip, that's fine, we all get what we buy.
|
