/[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 1.33 by benja, Fri Jan 24 14:23:50 2003 UTC revision 1.34 by benja, Fri Jan 24 14:57:16 2003 UTC
# Line 5  Don't fence me in: Supporting data mobil Line 5  Don't fence me in: Supporting data mobil
5  1. Introduction  1. Introduction
6  ===============  ===============
7    
8    [Note: The following are my notes for what should be written,
9    not final text! --benja]
10    
11    The web: servers. {WebDAV}? Links break! Microcosm, alike;
12    Xanadu, alike summat. (Tho no link breakage, given permanent ids.)
13    [Read about Open Hypermedia work on permanent identifiers.]
14    
15    However, users do not live in a world of servers.
16    Real-live data mobility. Server centricity not suited to this.
17    It's well recognized that references should not be by location [ref URN].
18    
19    (To explain data mobility:
20    Data moves like this and that. The server/location paradigm
21    is not suited to this: To support hypermedia functionality correctly,
22    we need to recognize two copies of the *same* document.)
23    
24    Server centricity is what made the web scalable. Backlinks rejected
25    for this reason [ref TBL]. However, recent innovations in P2P have made
26    scalable hashtables possible.
27    
28    {If standards could be agreed on, web servers should be able to
29    self-organize into a DHT implementing bidi links. There has been
30    interest in p2p hypermedia [ref Bouvin]. This would not, however,
31    solve data mobility on disconnected clients.}?
32    
33    A second issue arising from data mobility is version consolidation
34    (as well as simply keeping track of alt. versions). [ref HTML
35    version format proposal] Alternate versions important for
36    authoring process [search refs]. (Note: Keeping track of versions
37    structure is also \*hyper*media. Refs?) (WebDAV!)
38    
39    Ultimately, we need a system able to keep track of *all*
40    alternative versions generated in the authoring process.
41    (Even if distributed)
42    
43    Another challenging area of data mobility is the support for
44    mobile and nomadic users [define]. Mobile users often have
45    different machines, among which data must be Notes-replicated
46    [ref Lotus Notes]. Also, caching of data for offline use.
47    This also needs to be addressed for dialup users. Finally,
48    train collaboration; this raises caching (local storage)
49    as well as serverless versioning (like e-mail collaboration).
50    
51    We have raised a small set of issues that need to be addressed
52    to support data mobility in hypermedia across a number of use cases:
53    
54    - [bulleted list]
55    
56    These use cases are not very visionary.
57    All of these use cases except train collaboration are in use *today*.
58    To support hypermedia functionality well in these cases, we need to
59    address the issues above.
60    
61    We present a design which addresses the issues we have raised.
62    We are currently in prototype stage, which we describe...
63    
64    This paper is structured as follows.
65    
66    
67  2. Block storage  2. Block storage
68  ================  ================
69    
# Line 29  Don't fence me in: Supporting data mobil Line 88  Don't fence me in: Supporting data mobil
88  7. Experience and future directions  7. Experience and future directions
89  ===================================  ===================================
90    
91    Open issue: Tree hashing, also possibly needed for (OceanStore-like?)
92    distribution of shares
93    
94  8. Conclusions  8. Conclusions
95  ==============  ==============
96    

Legend:
Removed from v.1.33  
changed lines
  Added in v.1.34

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