[Gzz-commits] manuscripts/storm short-paper.rst

[Tuomas J. Lukka]

CVSROOT:        /cvsroot/gzz
Module name:    manuscripts
Changes by:     Tuomas J. Lukka <address@hidden>        03/05/30 13:40:20

Modified files:
        storm          : short-paper.rst 

Log message:
        muddling

CVSWeb URLs:
http://savannah.gnu.org/cgi-bin/viewcvs/gzz/manuscripts/storm/short-paper.rst.diff?tr1=1.30&tr2=1.31&r1=text&r2=text

Patches:
Index: manuscripts/storm/short-paper.rst
diff -u manuscripts/storm/short-paper.rst:1.30 
manuscripts/storm/short-paper.rst:1.31
--- manuscripts/storm/short-paper.rst:1.30      Fri May 30 13:14:33 2003
+++ manuscripts/storm/short-paper.rst   Fri May 30 13:40:20 2003
@@ -47,7 +47,7 @@
    \terms{Design, Reliability, Performance}
 
    \keywords{dangling links, 
-   peer-to-peer, location-independent identifiers}
+   peer-to-peer, location-independent identifiers, content addressable 
networks}
 
 
 Introduction
@@ -91,7 +91,8 @@
 .. can be used to implement external links, backlinks &c:
 
 However, recent developments in peer-to-peer systems have
-rendered this assumption obsolete. Distributed hashtables (DHTs)
+rendered this assumption obsolete. 
+Distributed hashtables (DHTs) and other content-addressable network algorithms
 
[malkhi02viceroy-andalso-maymounkov02kademlia-andalso-ratnasamy01can-andalso-stoica01chord-andalso-zhao01tapestry]_
 
 allow location-independent identifiers
 to be resolved on a global scale. 
@@ -105,6 +106,17 @@
 all published external links to a document can be found using
 a DHT.
 
+.. Freenet's cryptographical identifiers:
+
+In Freenet [freenet-ieee]_, an anonymous peer-to-peer
+publication system, *cryptographic hashes* 
+(see, e.g., [schneier96appliedcryptography]_)
+are used for location-independent identifiers of 
+documents versions. Hashes
+are short (often 20 bytes), yet it is practically impossible
+that two different documents will ever have the same hash,
+making them ideal unique identifiers on the Web.
+
 .. [#] Domain Name System (DNS) [rfc1101]_ is a widely used RRS 
    system in the Internet.
 
@@ -127,17 +139,6 @@
 or on the
 public Internet.
 
-.. Freenet's cryptographical identifiers:
-
-Location-independent identifiers are not a new idea.
-In Freenet [freenet-ieee]_, an anonymous peer-to-peer
-publication system, *cryptographic hashes* 
-(see, e.g., [schneier96appliedcryptography]_)
-are used to identify versions of documents. Hashes
-are short (often 20 bytes), yet it is practically impossible
-that two different documents will ever have the same hash,
-making them ideal unique identifiers on the Web.
-
 .. Hashes allow reference to a single version of a
    document. Documents are refered to by public keys [XXX].
    The private key corresponding to a document's public key
@@ -184,9 +185,9 @@
 One of the key innovations in storm is the use 
 of *pointer blocks*
 to build mutable documents on top of
-the lowest-layer append-and-delete-only model.
+the lowest layer append-and-delete-only model.
 A pointer block is simply a storm block which says
-"The identifier X points to document Id, overriding 
+"The identifier X points to document Id, and this block overrides 
 the previous pointer blocks a, b" and contains
 a digital signature of the above information. The pointer Id
 contains the public key used for signatures --- this allows
@@ -194,33 +195,34 @@
 A suitable user interface around pointers allows the user to
 still see old-style file names, but with the file names attached
 to his/her person, not to a particular computer.
+There is also a scheme to compress blocks using differences and reverse
+differences between consecutive blocks but there is no space to explain it 
here.
 
 
 ..          storing
            documents on the desktop and publishing and retrieving
            documents from the network.
 
-.. Pointers
+..  Our system implements append-and-delete-only
+    storage; past versions of documents are stored
+    unless deleted explicitly. It is possible to
+    store only the differences between versions.
+
+Interfacing
+===========
 
 .. part of the web -- URN scheme (so far experimental, 
    targetting registration):
 
 
-We are using an experimental URN namespace
+Storm uses an experimental URN namespace
 (Uniform Resource Name, a persistent kind of URI)
-and are preparing the registration
+and we are preparing the registration
 of a real namespace with IANA. 
-The URN is constructed analogously to a DATA URI (RFC XXX)
-from the hash of the block content and the MIME content-type,
-for example ``urn-...XXX``.
-
-..  Our system implements append-and-delete-only
-    storage; past versions of documents are stored
-    unless deleted explicitly. It is possible to
-    store only the differences between versions.
+The URN is constructed analogously to a data URL (RFC 2397)
+from the MIME content-type, using the hash of the data instead
+of the data itself.
 
-Interfacing
-===========
 
 ..  On the user level, 
 ..  Even without user interface changes,