At 09:00 AM 12/28/99 -0500, Eric S. Johansson wrote:
>On Mon, 27 Dec 1999, Dan O'Reilly wrote:
>
> > Why is it stupid? Maybe usury, but surely they have a right to charge what
> > they want for their products, don't they? The thing that kills me
> about this
> > whole "free software" thing is that people seem to expect me, as a business
> > owner, to invest literally hundreds of thousands (or even millions) of
> $$$ to
> > develop software, betting my company on it, then give it away just because
> > people don't want to pay. Where's the incentive, then, to develop quality
> > software, if people are to force you to give it up with no hope of
> recovering
> > your development costs?
>
>I don't know about other people but I certainly don't expect that. I
>view open source as being more akin to the stone soup model of
>development. Everyone contributes pieces and someone integrates it
>into the whole. for example, I'm developing significant IP behind
>speech driven programming techniques. Programmers that haven't been
>injured (and some that are) are helping me implement the IP. we all
>contribute something and we can all benefit from the whole. I'm
>giving all my IP away because a good speech user interface is too
>important for anyone company to play "dog in the manger". we're giving away
>the code again because it's the right thing to do.
In your specific case, it's a) something you've chosen to do ("right thing"
is pretty subjective, although I agree with your views on it); and b) it's
the "right thing" because of your circumstances. That doesn't mean that the
"right thing" for RSA to do (or any other company, for that matter) is to
follow suit and give their stuff away. On the contrary; the "right thing"
for RSA is to make a profit. That's why they're in business.
>from economic side, if you're looking at solely recovering development
>costs, there are other models. For example, customers could pay for
>"their share" of the development costs or they could pay for specific
>bug fixes if they needed that fixed done now. Auction techniques
>could be used to determine what feature should be added for the next
>release. If you don't like auctions, you can use the PBS begging
>model. One can sell additional services (support, consulting,
>training, etc.) to pay for essential development as well.
But the bottom line is, RSA has chosen not to do business that way.
>when you leverage to product development off of existing code bases or
>Toolkits, you'll find your costs of development drop significantly.
>If you're forced to recreate or re-implement useful tools, development
>costs go through the roof.
True, and that's why you go out an license technologies to use in your
software. A $100k fee for one-time use may be high - unless you're using
it in $20m of software. On the other hand, a $100k fee dispersed over sales
of 10,000 pieces of software into which it is incorporate substantially
lowers the per-unit cost, and hence, $100k isn't as "big" as it appeared.
> > The other argument, that of "it's just mathematical equations", is equally
> > as farcical. Books are just collections of words, yet people copyright
> them.
> > Surely words should be in the public domain, right? If it's that easy,
> then
> > expend your own resources and write your own algorithms. If you build a
> > better mousetrap...
>
>not exactly. Using your analogy, if books are just collections of
>words, what would a world be like if people patented books. They'd be
>a single author for mysteries, science-fiction, technical books,
>etc. and they can use patents to keep other authors from expressing
>their views and publishing their books.
Not really. A copyrighted book addresses a specific use of a combination of
words, just as a patented piece of software addresses a specific use of an
algorithm.
>if on the other hand you truly believe that patents are appropriate to
>protecting software, you might consider doing a patent search if you
>use linked lists in your code. you're probably in violation of one of
>the over 400 patents featuring linked lists. check the IBM patent
>search engine. Try looking up your favorite data structure there.
True enough - except for a couple things. First, in many cases, patents
are granted for specific and unique applications of existing technologies.
Second, even if the patents do address commonly-used data structures, as
long as IBM allows their use, then a patent is not being infringed upon.
On the other hand, if IBM were to change their mind, certainly they would
have a case to be made for forcing royalties to be paid, whatever (speaking
hypothetically, of course).
------
+-------------------------------+---------------------------------------+
| Dan O'Reilly | |
| Principal Engineer | "Time flies like an arrow. Fruit |
| Process Software Corporation | flies like a banana." |
| http://www.process.com | -- Groucho Marx |
+-------------------------------+---------------------------------------+
______________________________________________________________________
OpenSSL Project http://www.openssl.org
User Support Mailing List [EMAIL PROTECTED]
Automated List Manager [EMAIL PROTECTED]
