[Gzz-commits] manuscripts/pointers article.rst

[Benja Fallenstein]

CVSROOT:        /cvsroot/gzz
Module name:    manuscripts
Branch:         
Changes by:     Benja Fallenstein <address@hidden>      03/11/10 05:30:46

Modified files:
        pointers       : article.rst 

Log message:
        tighten abstract

CVSWeb URLs:
http://savannah.gnu.org/cgi-bin/viewcvs/gzz/manuscripts/pointers/article.rst.diff?tr1=1.207&tr2=1.208&r1=text&r2=text

Patches:
Index: manuscripts/pointers/article.rst
diff -u manuscripts/pointers/article.rst:1.207 
manuscripts/pointers/article.rst:1.208
--- manuscripts/pointers/article.rst:1.207      Mon Nov 10 05:07:55 2003
+++ manuscripts/pointers/article.rst    Mon Nov 10 05:30:46 2003
@@ -41,12 +41,18 @@
 The Web could benefit greatly from the bandwith savings,
 increased availability, and reduced server costs offered
 by filesharing systems.
-However, filesharing systems don't allow documents
-to be updated, and
-in P2P, updating has so far only been investigated
-in systems that do not allow files to be obtained
-from any user who has a downloaded a copy,
-but only from locations designated by the system.
+However, current update mechanisms for P2P networks do not
+offer the simplicity of filesharing systems, as versioning
+information has to be kept as special internal state in the network.
+
+
+.. filesharing systems don't allow documents
+   to be updated.
+
+.. In P2P, updating has so far only been investigated
+   in systems that do not allow files to be obtained
+   from any user who has a downloaded a copy,
+   but only from locations designated by the system.
 
 .. thus making them disappear once the original publisher loses interest.
 
@@ -56,21 +62,23 @@
     because 
 
 To allow versions to be obtained from any host that has a copy,
-we propose to use *pointer records*, 
-signed files associating a document's id, 
-its current version's hash, and a timestamp.
-Pointer records can be shared and sought for
+we propose to use *pointer records*,
+which can be shared and sought for
 through the network as ordinary files.
+Similar to
+OceanStore's heartbeats, they are signed files containing a document's id, 
+its current version's hash, and a timestamp.
 We also present the Storm data model, 
-a simple API for hash-based storage, such as P2P networks.
-Using pointer records layered on the Storm
-data model, several P2P networks can be used interoperably
-as a P2P Web (except, naturally, for dynamically generated pages).
-All versions of a web page
-can be found using the same URI 
-while residing on any host on any network.
-Past versions of a web page 
-remain accessible as long as at least only one person keeps a copy.
+a simple API for hash-based storage systems, such as P2P networks.
+Using pointer records layered on the Storm data model, 
+all versions of a document, stored on any host in any P2P network,
+can be found using the same URI. 
+
+.. Past versions of a web page 
+   remain accessible as long as at least only one person keeps a copy.
+
+.. several P2P networks can be used interoperably
+   as a P2P Web (except, naturally, for dynamically generated pages).
 
 We discuss our preliminary implementation
 of pointer records