-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Je Merkredo 18 Aprilo 2001 12:08, vi skribis:
> Catalogs are a good way to find information. Yahoo and DMOZ are both
> catalogs. They are also a good way to do keyword and other metadata
> searches.
But there are inherant problems with catalogs:
1) They are fairly static
2) They are abusable, so you generate extra work for yourself to find ways
to reduce the exploitability, through some sort of anonymous public key
structure, or some other method that renders them even less dynamic.
3) The infrastructure to make catalogs work is much more complex than a
simple dynamic, distributed key-matching system, and provides very few
advantages. I don't know about anybody else out there, but I like simple
designs. Sure, the *concept* of the catalogs seem simple, but the
implementation raises a number of issues: How to protect against bogus
catalogs, how to "rate" catalogs, how to timestamp catalogs, and so on and so
on.
4) They introduce weakness into the system.
5) They require some sort of authorization
Yahoo and DMOZ are reasonable catalogs because there is a central owner of
the catalog, but even so, Yahoo's database responds EXTREMELY slowly to the
presence of new information on the web.
> Yes, although this will actually be done by a client and be hidden behind
> a user interface.
This seems like sort of a hack, although I'll have to think more about the
implications to give you specifics as to why.
> > The only thing hindering the implementation of a scheme like this is the
> > combination of factors:
>
> Catalogs *are* an implementation of such a scheme.
I have to disagree here. Consider the differences (let's call the system I
described a hashtabled key system) : a hashtabled key system doesn't require
the existence of an "authority" to validate the contents of a catalog or
entries into the catalog. A hashtabled key system responds *immediately* to
data inserts. A hashtabled key system is completely secure, given the nature
of one-way hashes.
Most importantly, a hashtabled system increases the legal security of the
data requestor, in that the file requestors aren't asking for a specific file
which has been clearly identified as to content in a file somewhere. I feel
that a strong argument *against* a catalog system is that such a system would
provide an additional tool for prosecuters to attach content holders. If a
file's content is clearly described in a catalog file, and you can identify
the file from the catalog, then you can (more easily) prosecute data stores
using the catalog. If you can't clearly identify a file with the catalog,
what use is the catalog?
The original Freenet lookup mechanism was *good*: it was simple, efficient,
dynamic, and uncoupled; it just suffered from design limitations.
=== SER Deutsch|Esperanto|Francaise|Linux|Java|Aikido|Dirigibles|GPG
=== http://www.germane-software.com/~ser jabber.com:ser ICQ:83578737
Evil originates when people begin viewing other people as objects.
-- Granny Weatherwax
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.2 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE63fQDP0KxygnleI8RAo1iAKCuRtHvohkxa5xYy7g3PVSoNZ0EIACfW5hv
B8tyu0DtfIuhSUjprFrvWR4=
=HP6L
-----END PGP SIGNATURE-----
_______________________________________________
Devl mailing list
[EMAIL PROTECTED]
http://lists.freenetproject.org/mailman/listinfo/devl
