/[gzz]/manuscripts/storm/article.rst
ViewVC logotype

Diff of /manuscripts/storm/article.rst

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 1.87 by benja, Wed Feb 5 21:13:09 2003 UTC revision 1.88 by benja, Wed Feb 5 21:32:17 2003 UTC
# Line 201  However, CFS *splits* files into several Line 201  However, CFS *splits* files into several
201  available CFS servers. Freenet [ref] and PAST [ref, pastry ref] doesn't split  available CFS servers. Freenet [ref] and PAST [ref, pastry ref] doesn't split
202  files into blocks, since they store data as whole files. All previously mentioned  files into blocks, since they store data as whole files. All previously mentioned
203  systems lack of the immutable property which is used in Storm blocks.  systems lack of the immutable property which is used in Storm blocks.
   
204        
205  Related work: we need something about p2p hypermedia:  Related work: we need something about p2p hypermedia:
206  [ref Bouvin, Wiil ("Peer-to-Peer Hypertext")]  [ref Bouvin, Wiil ("Peer-to-Peer Hypertext")]
207    
 [Note: The following are my notes for what should be written,  
 not final text! --benja .. adding comments in the middle --antont]  
   
208  It's well recognized that references should not be by location [ref URN].  It's well recognized that references should not be by location [ref URN].
209    
 (To explain ad hoc *data mobility*:  
 Data moves like this and that. The server/location paradigm  
 is not suited to this: To support hypermedia functionality correctly,  
 we need to recognize two copies of the *same* document.  
 Only then can we merge the fragmented worlds of desktop computing  
 and online publishing into one.)  
   
 Server centricity is what made the web scalable. Backlinks rejected  
 for this reason [ref TBL -- XXX note: not true like this; the ref I had  
 in mind is http://www.w3.org/DesignIssues/NameMyth.html, and  
 it's not about back links]. However, recent innovations in P2P have made  
 scalable location independent routing/hashtables possible.  
   
 ->  
 Binding documents to servers has been necessary to make the Web scalable,  
 as no central index of documents was needed.  
 Berners-Lee [ref NameMyth] argues that identifying documents  
 by unique, completely random identifiers may be desirable because  
 there is never a need to change such an identifier because e.g. a server  
 moves to a different organization, but that it is not possible to  
 resolve  
   
   
210  {If standards could be agreed on, web servers should be able to  {If standards could be agreed on, web servers should be able to
211  self-organize into a DHT implementing bidi links. There has been  self-organize into a DHT implementing bidi links. There has been
212  interest in p2p hypermedia [ref Bouvin, Wiil ("Peer-to-Peer Hypertext")].  interest in p2p hypermedia [ref Bouvin, Wiil ("Peer-to-Peer Hypertext")].
# Line 294  All of these use cases except train coll Line 267  All of these use cases except train coll
267  To support hypermedia functionality well in these cases, we need to  To support hypermedia functionality well in these cases, we need to
268  address the issues above.  address the issues above.
269    
 We present a design which addresses the issues we have raised.  
 We are currently in prototype stage, which we describe...  
 Along with the design, we illustrate how it supports various situations in  
 the use cases.  
   
 ->  
 We present Storm (for *storage module*), a design which addresses the issues  
 we have described.  
   
 ->  
 We describe  
 the partially implemented Storm design,  
 a storage system designed to cope with two problems  
 raised by moving documents: Dangling links [refs stating problem],  
 and keeping track of alternative versions [refs stating problem].  
   
   
   
   
270    
271  3. Block storage  3. Block storage
272  ================  ================

Legend:
Removed from v.1.87  
changed lines
  Added in v.1.88

savannah-hackers-public@gnu.org
ViewVC Help
Powered by ViewVC 1.1.26