[Gzz-commits] manuscripts/pointers article.rst

[Benja Fallenstein]

CVSROOT:        /cvsroot/gzz
Module name:    manuscripts
Branch:         
Changes by:     Benja Fallenstein <address@hidden>      03/10/27 15:58:44

Modified files:
        pointers       : article.rst 

Log message:
        intro mostly there in first draft stage

CVSWeb URLs:
http://savannah.gnu.org/cgi-bin/viewcvs/gzz/manuscripts/pointers/article.rst.diff?tr1=1.28&tr2=1.29&r1=text&r2=text

Patches:
Index: manuscripts/pointers/article.rst
diff -u manuscripts/pointers/article.rst:1.28 
manuscripts/pointers/article.rst:1.29
--- manuscripts/pointers/article.rst:1.28       Mon Oct 27 14:41:04 2003
+++ manuscripts/pointers/article.rst    Mon Oct 27 15:58:44 2003
@@ -48,7 +48,7 @@
    have lost interest.>>>
 
 We don't propose that every byte of information ever published
-on the Web should have to be kept around forever. However,
+on the Web has to be kept around forever. However,
 we do believe that as long as someone does keep a copy,
 data should remain accessible, and links should continue
 to work.
@@ -65,7 +65,7 @@
 like cryptographic hashes. 
 A file identified by a cryptographic hash
 can be downloaded from any server that has a copy; the client
-can then check it is the correct file by checking the hash.
+can then check the integrity of the file by checking it against the hash.
 
 .. <<<Self-verifying means that given the identifier and a file,
    a client can check that the file matches the identifier.>>>
@@ -78,24 +78,38 @@
 is not viable on a global scale:
 "[I]f you put information in a name, it decreases its longevity; 
 if you don't you can't dereference it to a resource."
-However, as we have observed in [fallenstein03storm]_, 
-with the advent of 
-efficient peer-to-peer lookup mechanisms such as 
+
+However, as observed in [fallenstein03storm]_, 
+with the advent of efficient peer-to-peer lookup mechanisms such as 
 distributed hashtables (DHTs), this observation
-is no longer true.
+is no longer true. A DHT is quite able to resolve a
+hash-based identifier on a global scale,
+as evidenced by applications like the Cooperative
+File System (CFS, [dabek01widearea]_) and 
+the Overnet file sharing client [overneturl]_.
+
+.. SFR (semantic-free referencing) not all that close, 
+   though semantic-free idea shared (SFR takes along 
+   many problems of the Web)
 
-Using DHTs
-to resolve location-independent identifiers on the Web
-has been proposed by Balakrishnan et.al. 
-[balakrishnan03semanticfree-andalso-walfish03dns]_.
+.. <<<(Using DHTs
+   to resolve location-independent identifiers on the Web
+   has been proposed by Balakrishnan et.al. 
+   [balakrishnan03semanticfree-andalso-walfish03dns]_.
+   However, in their work, the location-independent identifier
+   merely points to a Web server administered by the publisher
+   of a Web page; if the original publisher discontinues
+   maintenance of the page, it would still drop off the Web.)>>>
 
-.. Proposal: A location-independent Web (closest thing is Freenet (ref))
+   XXX move to related work?
+
+.. Proposal: A location-independent Web <<<(closest thing is Freenet (ref))>>>
 
 We propose, then, to build a location-independent Web
 based on self-verifying identifiers.
 
-The project that is currently closest to this goal is Freenet,
-a XXX
+.. <<<The project that is currently closest to this goal is Freenet,
+   a XXX>>>
 
 .. Benefits of hash-based addressing:
    - Pages easily movable between servers
@@ -106,18 +120,26 @@
    - Verifiable
    - Same namespace for local and for non-local data
 
-.. SFR (semantic-free referencing) not all that close, 
-   though semantic-free idea shared (SFR takes along 
-   many problems of the Web)
+Using hash-based addressing in such a scheme has several benefits:
 
-.. Other projects exploit some of the advantages of hash-based
+- Pages can be moved between servers without breaking links.
+- Links continue to work as long as *anybody* keeps a copy
+  of their targets, even if the original publisher does not
+  maintain the Web page any longer.
+- Downloaded files can keep their global identifiers.
+  Links between two downloaded Web pages automatically work.
+- It is possible to implement load-balancing schemes in which
+  a user downloads a page from anybody who happens to have a copy.
+- Different networks can be used to download a file. XXX
+
+.. <<<Other projects exploit some of the advantages of hash-based
    (storage systems: CFS, PAST; web caching: Squirrel),
-   but don't address the Web.
+   but don't address the Web.>>>
 
-.. The infrastructure behind CFS, PAST and Squirrel: Peer-to-Peer
+.. <<<The infrastructure behind CFS, PAST and Squirrel: Peer-to-Peer>>>
 
-.. Quite recently, several Peer-to-Peer architectures have been
-   proposed that use hash-based, loc.ind. ids
+.. <<<Quite recently, several Peer-to-Peer architectures have been
+   proposed that use hash-based, loc.ind. ids>>>
 
 .. Possibility of desktop integration in ways that the location-dependent
    Web cannot archieve, through the novel combination of
@@ -126,14 +148,61 @@
 .. However, there's a problem with this: versioning -----
    Basic problem: Hash-based addressing allows no updates
 
+However, a document refered to by a cryptographic hash
+can obviously not be changed (without making it a different document,
+refered to by a different hash). For most Web pages, this is not
+acceptable [#academic-articles-hashlinking]_. In this paper,
+we examine solutions to this problem that keep the desirable properties
+of hash-based addressing.
+
 .. Contributions; structure of this paper ----
 
 .. Main contrib: Pointer records for implementing updating
 
+The main contribution of our paper are *pointer records*.
+A document is identified by a *pointer*, the hash of a
+public key. A pointer record is a file signed by the
+corresponding private key, containing the pointer's identity,
+the hash of a version of that document, and a timestamp.
+The newest version of a document is obtained by searching
+the peer-to-peer network for all pointer records associated
+with this pointer, and picking the newest one.
+
+This scheme is quite close to OceanStore's concept of
+*heartbeats*. However, the difference is that in OceanStore,
+heartbeats are controlled by an inner circle of servers,
+the *primary replica*, chosen by the owner of the document.
+In this scheme, when a document ceases to be maintained
+by its primary replica, it is impossible/becomes harder/*what*?XXX
+to retrieve the newest version of the document.
+
+(XXX IS THIS TRUE?)
+
 .. Other contribs:
    - The idea of a location-independent Web including
      location-independent version management
    - Diffs
+   - Permanent signature scheme sketch
+
+In addition to pointer records, this paper XXX
+
+.. Structure of this paper
+
+The remainder of this paper is structured as follows.
+In Section 2, we review related work in version management,
+and find that existing peer-to-peer versioning systems 
+do not provide the benefits of hash-based addressing.
+In Section 3, we introduce our fundamental, hash-based data model
+for a location-independent Web. In Section 4, we propose
+a versioning system built on top of this data model.
+In Section 5, we introduce a scheme for storing only
+the differences between versions that works with our
+basic data model, making the storage of past versions
+more economical. In Section 6, we outline solutions
+to the permanency issues raised by the use of 
+cryptographic keys (which may be stolen or lost).
+Section 7 concludes.
+
 
 
 Related work
@@ -305,3 +374,8 @@
 .. [#rtg-links] ``http://www.seds.org/spaceviews/cassini/rtgpages.html``.
    All links dereferenced on October 27th, 2003.
 
+.. [#academic-articles-can-use-hashlinking] A notable exception
+   may be academic articles, in which references to other articles
+   are ideally to an immutable version, and in which references
+   are generally not circular, making this domain map neatly
+   to the properties of hash-based referencing.
\ No newline at end of file