h a l f b a k e r y"It would work, if you can find alternatives to each of the steps involved in this process."
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
It is possible to automatically generate all haiku and either own them for 70+ years at no cost, or place them into the public domain, thus preventing others from copyrighting them.
As copyright costs nothing in the United States and lasts "from the moment of [a work's] creation ... for the author's
life plus an additional 70 years ...", this lets you: 1) cheaply own an entire artistic field, and 2) influence future copyright law.
Analysis of automatic haiku generation:
A haiku consists of 3 lines of 17 syllables (5-7-5). Let's restrict the problem to haiku written in Japanese.
There are 46 syllables in Japanese. From combinatorics, we know that given 17 syllables in a haiku with 46 possible values for each syllable means there are at most 46^17 possible haiku.
But this assumes that there are no restrictions on what syllables can appear in what positions. In languages, there are syllables that either do not appear next to each other, or are very unlikely to ever be used in combination.
So let's say through a Markov analysis of Japanese or some similar approach, we can say that for any given point in the haiku, there are 5 most likely syllables.
Now, with only five candidate syllables to consider for each position in the haiku, we have 5^17, or 763 billion haiku.
With today's (2002) hardware, a system to create and store this many haiku would cost US $2500.
Under US copyright law, a document is copyright the moment it is created and fixed in tangible form.
So for $2500, you can own all (likely) unwritten Japanese haiku for as long as you live plus 70 years.
A similar approach would work for English and other languages, as well as other well-constrained copyrightable domains (for example: limericks, sonnets, ultra-short stories, pop-music lyrics, 8+ note songs)
--Pat / zippy@cs.brandeis.edu
Copyrights in computer-generated works
http://www.law.duke...s/2001DLTR0024.html
Computer-generated works raise grave authorship concerns under U.S. copyright law, with arguments in favor of allocating copyrights to the computer user, programmer, the computer itself, or some combination therein. The author discusses the issues and paradoxes. [yppiz, Oct 04 2004]
Please log in.
If you're not logged in,
you can see what this page
looks like, but you will
not be able to add anything.
Annotation:
|
| |
But what about haiku in English? And doesn't this call for unlimited resources (time)? |
|
| |
I restricted the problem to Japanese because the syllable problem seemed simpler to analyze there than in English. But the same approach, including constraints on what syllables can appear in given positions, would apply to English and other languages. |
|
| |
As for time, there are only ~ 1.4 TBytes to generate. On a single-processor PC you could probably generate and store 10MB/sec, with the complete run taking a day and a half. |
|
| |
(You may wish to change your last sentence if you are restricting this to Japanese, btw) |
|
| |
Actually, it's not a document that is copyrighted when fixed, it's a work of authorship. This is why the phone book can't be copyrighted. |
|
| |
It's an interesting case, but I suspect the winning argument would be that exhaustion of all possibilities of a form doesn't constitute authorship of each possibility. |
|
| |
Copyright does not prevent people from copying phone books because they are "arrangements of mere facts" (Feist Publications, Inc. v. Rural Telephone Service Company). |
|
| |
The mere facts argument does not apply to haiku. |
|
| |
The exhaustive generation argument is an interesting one. Clearly, this project is absurd. But there is no clause or case law that I know of that deals with exhaustive or automatic generation. |
|
| |
Further, if someone influences the process, that strengthens their claim on authorship of the set of haiku. For instance, a user could base the Markov analysis of likely syllable combinations on their own writings (emails, other documents). |
|
| |
And here's that exam question: |
|
| |
http://www.law.cornell.edu/ecourse/week12.htm |
|
| |
The beauty of Google is that the moment you search for something, it becomes real. |
|
| |
Anybody can
Write haiku. Just stop at the
Seventeenth syllab.
|
|
| |
Some amount of minimal creativity is
required to receive copyright protection.
Thus, randomly generated haiku are no
more copyrightable than randomly
generated sequences of letters, a few of
which undoubtably resemble copyrighted
works. |
|
| |