home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.theory
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!rpi!batcomputer!cornell!uw-beaver!news.u.washington.edu!serval!moscow.uidaho.edu!jed.cs.uidaho.edu!foster
- From: foster@jed.cs.uidaho.edu ()
- Subject: Re: solution or information needed
- Nntp-Posting-Host: jed.cs.uidaho.edu
- References: <1993Jan19.151627.21525@matrix.com> <1993Jan21.153028.7679@cs.yale.edu>
- Sender: news@moscow.uidaho.edu
- Organization: University of Idaho CS Dept.
- Date: Fri, 22 Jan 1993 16:25:43 GMT
- Message-ID: <C19KAw.GG2@moscow.uidaho.edu>
- Distribution: usa
- Lines: 33
-
- To see why "copy languages" are non-regular, learn a pumping lemma for CFLs,
- as was just suggested.
-
- I recently read an interesting explanation of why copy languages are non regular.
- Think about "dependencies" in a string, which are produced by the process of
- derivation. Rather than be formal, I'll give an example. For palendromes, the
- dependencies are nested. For example:
-
- a b b a a a b b a
- | | | | | | | |
- | | | +-------+ | | |
- | | +--------------+ | |
- | +----------------------+ |
- +-----------------------------+
-
- This property holds for all context free languages. However, consider the
- dependencies for non CF languages. For example, one from a copy language:
-
- a b b a a b b a
- | | | | | | | |
- | | | +-------^---^---+
- | | +-----------^---+
- | +---------------+
- +---------------+
- The dependencies cross. This is evidence that a more complicated process
- took place when they were generated, one where the productions had to "be aware"
- of other productions "in context".
-
- I thought this was cute. Incidentally, I read about it in "The Linguistics of DNA",
- Davide B. Searls, American Scientist, Vol 80, Nov-Dec 1992, pp 579--591.
- --
- James A. Foster foster@cs.uidaho.edu
- Universty of Idaho Dept. of Computer Science
-