Subject: Re: Would you like to free knowledge now?
From: GKM (gkmaddressNOgkSPAM@www.spaceinfo-gkm.kiev.ua.invalid)
Date: Sat Mar 11 2000 - 20:44:56 MET
Thank you again! You are really not indifferent.
- Your thoughts -
I doubt anyone can produce a completely tight definition of
knowledge but yours is quite good. Your concept of knowledge as
having no owner is also a very desirable thing and it would be
healthy if more people believed in this. Unfortunately, the
current world trend is finding an owner for everything,
especially knowledge. Your disbelief won't be much help unless
you can establish a system of free knowledge that remains
compatible with our current system of laws.
This means that right here and now, you must face the fact that
it is possible for companies to take ownership of large
databases of useful knowledge and use that ownership for
financial gain. This is currently considered the ``normal'' mode
With the threat of software patents hanging over our heads, it
is not possible to ignore the current legal situation. I have
written to my government and signed various petitions but the
software patents are still there. Sadly, it is possible to own a
formula under some circumstance. It is also possible to own
whole swathes of genetic permutations, possibly even a whole
species. This is the reality we must deal with.
Thank you for your kind estimation of our definitions. We see
their main value in their applicability. You may use them
immediately, analyzing any source.
As far as I know, the main trend in legal environment is that
presentation form can be announced as property, but not ideas.
Many cases I have watched in software industry were based on
finding of similar code in the products that belong to
competitors. But many companies, including giants, try to
replicate features found in other firmsâ€™ production without any
permission. Look at X-Windows - MacOS - Windows - Linux Gnome.
Yes, a paper as â€˜dataâ€™ in our definition, may be someoneâ€™s
property - it is right of author to sell it or to distribute
free. Of course, some companies can own large databases of very
valuable printed sources. In no way we should announce them as
our own papers, or try to sell them. But I see no way to decide
whether it should be forbidden to propagate ideas from these
papers - and you can formulate them with your own words! If you
use for your work a formula in other notation - should it be
legal? Who can prove that if one observer told news, others
should not tell about the same? Who can own real world events?
Our suggestion is: â€˜dataâ€™ as presentation form - may be a
property, but â€˜knowledgeâ€™ as set of organized ideas - should
not. I never heard that some physician has tried to restrict
using of description of rear disease discovered by him. And you?
- Your thoughts -
This is why I suggest that a licence such as the GNU GPL is a
good idea for your work. It works with the current copyright
laws and uses them to prevent someone hijacking your work for
their own exclusive ownership while still allowing others to
develop that work in the spirit in which it was intended.
Your suggestion is quite suitable. We are working now on
portable version of code, oriented on GNU compilers. It will
take few months. At the moment some software has my own
copyright notice to prevent hijacking, but I really plan to
transfer it into GPL.
Though I think that software is not the main component of the
project. If you have read our papers - nothing should prevent to
build your own knowledge machine. Hint - donâ€™t try to use LISP,
PROLOG and SQL. The main force is people and methodology. Our
project is people-oriented.
- Your thoughts -
I don't see this as such a big problem. In the far future no
doubt we will have something better than books, but the
traditional book format is pretty good. A well written HTML page
is maybe slightly better than a book but my main concern is less
to advance the technology of publishing and more to make our
existing technology available more freely than it currently is.
Most of us learn to scan quickly for what we are looking for but
just suppose I have a human expert available instead of a book.
They may give me an excellent answer to my question but I cannot
understand the answer because I have not studied the subject
enough to know what the answer means. I might as well sit and
read the book, maybe it will take several books before I can
actually get enough concepts into my head that I can make sense
of a real answer.
I would like to point out that our project is human-oriented.
What we suggest in no way replaces real people. From my point of
view all activity not supporting our mutual survival as humans
Letâ€™s think - can any of us walk as fast as an airplane flies,
or carry a cargo like a ship? Is there a man going to try? Books
were developed to assist people, and you personally like them.
Can you learn everything that contained in books? Is it possible
that human can be used as a replacement of the whole volume of
There are limitations of human brain. Our study shows that
knowledge area that could be learnt by one man usually contains
only several hundred ideas. I would suggest all the people to
verify it in their own area of expertise. Please read our papers
to find what we mean by â€˜ideaâ€™
We have found statistical figures on the former USSR health
ministry - 50% of patients died not because of their disease,
but because of side effects of treatment, 40% of diagnoses were
wrong. Please supply us the figures on your country if possible.
That was the starting point of our work. We could not understand
why physicians hate their patients. But we have found the
following: ordinary physician can recognize near 100 different
diseases, good physician - near 200, genius - near 300. And the
full list of World Health Organization contains tens thousand
possible diseases. Thatâ€™s why specialization is very narrow and
quite ineffective. Physicians treat what they think they
recognize, but not what is your disease. Would you find it
Electronic Knowledge Publishing provides really effective way to
find a knowledge, which regards your unique case. Please take a
look at EMEDICINE Web site (http://www.emedicine.com). You will
find very good descriptions of approximately 1,000 diseases. The
site is well structured and absolutely free. Try to find
knowledge regarding your case - it is not that easy, if possible
- Your thoughts -
Improving our search engines is something that would be very
useful. I would say that this does not require us to rewrite our
books into a new structure but merely improve the way we build a
network of relations between books. This is why I say that your
e-knowledge is attempting to find an automatic way of replacing
something which humans are already good at. Much better to set
your target on something which humans find difficult which is
the searching part.
We describe our activity at our Web site (http://fast.to/gkm)
as â€˜Advanced development of intelligent forefronts to any
knowledge sourcesâ€™. We donâ€™t tell about replacement of sources.
We tell how to make them working for people. If ordinary engines
provide search and quite arbitrary ranking by words, e-knowledge
systems provide search and very precise ranking by ideas. Our
systems provide immediate on-line consulting, if desired, and
ground for adaptive distance learning when you learn only what
really needed. And, you know, it is great entertainment to look
for ideas in texts. You will find that number of ideas does not
depend on length of text, but mainly on author. In principle, it
does not differ from keywords defining for a paper, and many
publishing institutions demand to do it. But if they demanded to
define original ideas - that would be extremely helpful for all
General Knowledge Machine Research Group
to be continued
* Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
The fastest and easiest way to search and participate in Usenet - Free!
This message was posted through the fuzzy mailing list.
(1) To subscribe to this mailing list, send a message body of
"SUB FUZZY-MAIL myFirstName mySurname" to firstname.lastname@example.org
(2) To unsubscribe from this mailing list, send a message body of
"UNSUB FUZZY-MAIL" or "UNSUB FUZZY-MAIL email@example.com"
(3) To reach the human who maintains the list, send mail to
(4) WWW access and other information on Fuzzy Sets and Logic see
(5) WWW archive: http://www.dbai.tuwien.ac.at/marchives/fuzzy-mail/index.html
This archive was generated by hypermail 2b25 : Thu Apr 06 2000 - 15:59:48 MET DST