/[fenfire]/journals/plans/SA_2003.rst
ViewVC logotype

Diff of /journals/plans/SA_2003.rst

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

revision 1.111 by hemppah, Wed May 14 08:04:40 2003 UTC revision 1.112 by tjl, Wed May 14 08:18:47 2003 UTC
# Line 78  The Fenfire project Line 78  The Fenfire project
78    
79  TJL  TJL
80    
81  The Fenfire project is a free software effort to build  The Fenfire project is a free software research effort to build
82  a location transparent, hyperstructured desktop  a location transparent, hyperstructured computer operating
83  environment.  environment.
84  The work leading to the current Fenfire project (formerly known as GZigZag and Gzz)  On an abstract level, the Fenfire differs from conventional desktop environments
85  was started in 1999.  by bringing together
86    three components:
 Fenfire consists of three components:  
87    
88  Hyperstructured data model  Hyperstructured data model
89      ... RDF + Xu      As opposed to the conventional file system trees and location-based pointing
90        within them, the data model in Fenfire is more database-like, allowing
91  Novel focus+context user interfaces      anything to be connected permanently to anything else.
92      The bidirectionally linked hyperstructural approach to information opens new      The data model consists of two orthogonal pieces,.
93      possibilities to user interfaces.      The Xanalogical media model[ted-xanalogical-structure-needed]_
94        endows to the smallest units of media with identity and as a consequence
95        changes
96        the cutting and pasting operation from copying pieces of media to actually
97        creating a structure between the instance locations since the identities can be tracked.
98        On a higher level, RDF[w3rdfurl]_ graphs are used to create dynamic structure
99        between instances of xanalogical media.
100        In this data model, all connections are fundamentally bidirectional, traversable
101        in both directions.
102      The user can connect any two items which      The user can connect any two items which
103      will then be near each other in hyperspace, easily visible and      will then be near each other in hyperspace, easily visible and
104      accessible from one to the other.      accessible from one to the other.
105    
106  P2P information sharing  Novel focus+context user interfaces
107      To support data mobility, we have developed a software library, named as Storm      The bidirectionally linked hyperstructural approach to information opens new
108      (STORage Module), for storing and retrieving data as immutable byte sequences, which      possibilities to user interfaces, since each piece of information has
109      are identified by cryptographic content hashes. With Storm, we plan to use globally      a richer context available to the computer than in conventional filesystems.
110      unique, semantic-free references for locating data in Peer-to-Peer environment.      The Fenfire project includes libvob, a library for creating flexible, animated
111      Location-independent identifiers allows a peer-to-peer lookup system find documents      user interfaces easily, allowing nonlinear fisheye coordinate systems[carpendale01presspace]
112      wherever they are moved. We believe that the properties of Peer-to-Peer technology,      and smooth text rotation and scaling.
113      such as the ad hoc connectivity and the lack of central authority, are prerequisites      The user interfaces currently used in fenfire are mostly
114      while mobilizing our system. Additionally, Storm allows us to study two important      focus+context[fc-fisheye]_, showing one location in the hyperstructured database
115      issues *of data mobility*: dangling links and version management.      as the central item and peripherally the connected locations.
116      We study how location-independent identifiers can      Several of the new technologies in fenfire are related to user interfaces,
117      support *data mobility*. For instance, documents are often transferred quite freely      in particular to showing the context visibly yet nonintrusively.
118      between computers: they are sent as e-mail attachments, carried around on disks and  
119      published on the web. We use 'data mobility' as a collective  Permanent data and peer-to-peer (P2P) information sharing
120      term for the movement of documents between computers (or folders!),      Collaboration with even only
121      and movement of content between documents (through copy&paste).      conventional, rigidly structured data leads to versioning problems
122        unless a version control system such as CVS is used.
123        In Fenfire, we have developed a group of technologies collectively known
124        as Storm (STORage Module) to address this problem.
125        Storm starts from a fairly unconventional idea:
126        instead of mutable files or databases, all information is stored
127        in immutable byte sequences, blocks[lukka02guids].
128        Blocks are accessed through their permanent globally unique identifiers (GUIDs).
129        To allow mutable data, a system of pointer blocks is built on top of the
130        block system.
131        This already helps versioning: unless explicitly requested, old versions
132        are never deleted; new and old and alternate versions are accessible on a computer
133        without contacting a central server.
134        Location-independent identifiers allow a P2P lookup system find documents
135        from any source, as opposed to URLs which specify the explicit location for
136        fetching a document.
137    
138    There is a strong interaction between these components: for example,
139    the data model *enables* the novel user interfaces, but on the other hand,
140    without new types of user interfaces, the richer data model would be of limited use.
141    
142    The work leading to the current Fenfire project (formerly known as GZigZag and Gzz)
143    was started in 1999 in collaboration with Ted Nelson, who has since discontinued
144    the collaboration and due to patent and trademark reasons, the Fenfire project
145    cannot currently use the name GZigZag nor the ZigZag structure.
146    However, the fundamental ideas embodied in the ZigZag structure were not affected
147    by the particular implementation and in spring 2003 we were able to convert
148    our system to using RDF.
149    
150    Currently, Fenfire is nearing the public release of its first subgoal: the FenPDF
151    system for reading, annotating and linking electronic publications.
152    
153  .. raw:: latex  .. raw:: latex
154    
# Line 144  P2P information sharing Line 175  P2P information sharing
175    
176    
177    
178    Previous activity and publications
179    ----------------------------------
180    
181  .. previous activity pertaining to the scope of the topic of the  .. previous activity pertaining to the scope of the topic of the
182     research (list of the research team's five most important     research (list of the research team's five most important
183     publications in terms of the topic)     publications in terms of the topic)
184    
185  In [lukka02guids]_ we discuss the possibility of implementing    The most important publications related to this project are:
 the xanalogical storage model in Peer-to-Peer environment with globally unique identifiers.  
   
 In [lukka02fillets]_ we introduced the use of fillets ---  
 smoothing of sharp angles at the joints ---  
 for rendering on-screen connections in a way that makes them easier to  
 perceive even with several overlapping objects.    
   
 We have developed a perceptually designed hardware-accelerated  
 algorithm for generating unique background textures  
 [kujala03paper]_ for distinguishing documents.  
   
 Additionally, the methods developed and used in [lukka02gomcmc-andalso-kujala02solutions]_  
 are relevant to the modeling problems in our user interface research.  
   
 List of the most important publications  
186    
187      - Tuomas J. Lukka and Benja Fallenstein:      - Tuomas J. Lukka and Benja Fallenstein:
188        Freenet-like GUIDs for implementing xanalogical hypertext.        Freenet-like GUIDs for implementing xanalogical hypertext.
189        *Proc. Hypertext 2002* 194-195, 2002        *Proc. Hypertext 2002* 194-195, 2002.
190          This article discusses the possibility of implementing  
191          the xanalogical storage model in Peer-to-Peer environment
192          using globally unique identifiers.
193    
194      - Tuomas J. Lukka, Janne V. Kujala, and Marketta Niemelä,      - Tuomas J. Lukka, Janne V. Kujala, and Marketta Niemelä,
195        Fillets: Cues for Connections in Focus+Context Views of Graph-like        Fillets: Cues for Connections in Focus+Context Views of Graph-like
196        Diagrams. Proceedings of the        Diagrams. Proceedings of the
197        *6th International Conference on Information Visualisation*,        *6th International Conference on Information Visualisation*,
198        pp. 557--562, London, July 2002.        pp. 557--562, London, July 2002.
199          This article introduces the use of fillets ---
200          smoothing of sharp angles at the joints ---
201          for rendering on-screen connections in a way that makes them easier to
202          perceive even with several overlapping objects.  
203    
204      - Janne V. Kujala and Tuomas J. Lukka:      - Janne V. Kujala and Tuomas J. Lukka:
205        Rendering recognizably unique textures.        Rendering recognizably unique textures.
206        To be published in the proceedings of the        To be published in the proceedings of the
207        *7th International Conference on Information Visualisation*,        *7th International Conference on Information Visualisation*,
208        London, July 2003.        London, July 2003.
209          This article introduces
210          a perceptually designed hardware-accelerated
211          algorithm for generating unique background textures
212          for distinguishing documents.
213    
214      - Tuomas J. Lukka and Janne V. Kujala:      - Tuomas J. Lukka and Janne V. Kujala:
215        Using Genetic Operators to Speed up Markov Chain Monte Carlo        Using Genetic Operators to Speed up Markov Chain Monte Carlo
216        Integration.        Integration.
217        *Monte Carlo Methods \& Applications*, 8(1), 51--71, 2002        *Monte Carlo Methods \& Applications*, 8(1), 51--71, 2002.
218    
219    
220      - Janne V. Kujala and Tuomas J. Lukka:      - Janne V. Kujala and Tuomas J. Lukka:
221        Solutions for certain number-conserving deterministic cellular automata.        Solutions for certain number-conserving deterministic cellular automata.
222        *Physical Review E* 65, 026115, 2002.        *Physical Review E* 65, 026115, 2002.
223          The simple cellular automaton
224          rules introduced in [JVKXXXFUCHSBOCCARAREF] produced an unexpected result in simulations.
225          We were able to solve the problem exactly starting from first principles,
226          by using generating functions in a rather twisted way.
227    
228    
229  Of these, the two last ones contain some mathematical methods which we hope will  Of these, the two last ones contain some mathematical methods which we hope will
230  be applicable to e.g. filtering in the text rendering problem.  be applicable to e.g. filtering in the text rendering problem.
231    
232    Additionally, the methods developed and used
233    are relevant to the modeling problems in our user interface research.
234    
235    In addition, several manuscripts about various aspects of Fenfire are in preparation.
236    
237    
238  - postgraduate studies performed in the research team  - postgraduate studies performed in the research team
239    
240    Janne Kujala    Janne Kujala
# Line 200  Significance of the research Line 246  Significance of the research
246    
247  TJL  TJL
248    
249  The current proposal  The proposed research consists of two parts: continuing research into questions
250    opened by techniques invented for Fenfire, and theoretical questions whose
251    solutions are relevant to developing algorithms used in Fenfire.
252    
253    The questions opened by the inventions in Fenfire look at some phenomena in new light:
254    for example, we do not believe the questions in [kujala03paper]_ on how
255    to generate recognizably memorizable textures were asked before this; so far,
256    most research has concentrated on texture *discrimination*, i.e. under which
257    conditions a human notices that two textures placed right next to each other are different.
258    Contrasting results obtained in these two different questions may lead to surprising
259    new insights on the human visual system and memory.  
260    This basic research may of course yield new technologies but that is not the point;
261    the point is to understand and measure the phenomena we have observed in constructing Fenfire.
262    
263    On the other hand, some of the research questions in this proposal are
264    driven by technological requirements. For example,
265    enhancing the readability of the hardware-accelerated rendering of
266    rotated or nonlinearly transformed text is a problem
267    whose solution would benefit Fenfire.
268    Here, however, before being able to tackle that problem, a considerable
269    amount of mathematical framework is required, related to filtering and perception.
270    Creating this framework is a complex task, requiring both mathematics
271    and user experiments.
272    Armed with the framework developed by this research, developing the actual
273    hardware-accelerated filtering algorithms becomes a relatively easy task.
274    
275    All subproblems presented in this proposal are, however, significant in their own right:
276    even if the Fenfire project ceased to exist, the research results would be applicable
277    to other programs and problems.
278    
279    
280  Objectives and methods  Objectives and methods
281  ======================  ======================
282    
283    (reorg: "general (user experiments, math. modeling)", then specific projects,
284     stating for each the objectives *and* methods.)
285    
286  Objectives of the research  Objectives of the research
287  --------------------------  --------------------------
288    
# Line 338  in more general layouts including curved Line 416  in more general layouts including curved
416  and smooth animation.  and smooth animation.
417  Generalizing filleting to three dimensions ...  Generalizing filleting to three dimensions ...
418    
419        - quantitative experimental results from general graphs layout
420    
421  Text filtering  Text filtering
422  ''''''''''''''  ''''''''''''''
423    
# Line 378  Unique background textures Line 458  Unique background textures
458    
459  TJL  TJL
460    
461  The hardware algorithm for rendering the unique background textures  In [kujala03paper]_, we introduced the use of unique background textures
462  currently uses ad hoc distributions in the sampling of the texture  for distinguishing documents.
463  features. Although the distributions (e.g., for choosing the texture  The current hardware-accelerated algorithm for rendering the unique background textures
464    uses *ad hoc* distributions in the sampling of the texture
465    features.
466    Although the distributions (e.g., for choosing the texture
467  colors)  are qualitatively based on  colors)  are qualitatively based on
468  general principles for making the texture distinguishable and  general principles for making the texture distinguishable and
469  recognizable, their exact values are not rigorously justified.  recognizable, their exact values are not rigorously justified.
470  We plan to obtain justification for different aspect of the algorithm  
471  experimentally and to further evaluate the recognizability of the generated  The goals in this area are first to
472  textures.  formulate and test (with user experiments) a general way of adjusting a distribution
473    towards being memorable. This is a nontrivial optimization problem,
474    as the recognizability of a texture depends on the distribution it is
475    presented as a part of.
476    Next, it is important to research how much different aspects of the textures
477    may be optimized separately by the methods developed in the previous step, i.e., how much
478    the different aspects interact.
479    These questions are interesting not only from a practical point of view but also
480    from the point of view of cognitive psychology: for instance,
481    if any data can be obtained on how fast and how the brain adapts
482    to a particular input distribution, it would be an interesting result.
483    On a practical level, these experiments would allow us to get rid of the *ad hoc*
484    distributions and create a well-founded
485    algorithm for generating unique textures.
486    
487        - quantitative model of certain aspects of texture perception and recognition
488    
489  Controllers  Controllers
490  '''''''''''  '''''''''''
491    
492  TJL  TJL
493    
494  Custom controllers can offer new ways to navigate and control the information  In the context of Fenfire,
495  flow on the screen.  we have invented a new type of computer controller which we cannot describe here
496  They are many usability aspects in the configuration of the degrees of freedom  before the (potential) filing of a patent application.
497  provided by the controller that can be theoretically modeled and experimentally  This controller, however, also opens some interesting related questions for
498  tested.  basic research.
   
 We have invented a new type of controller which we cannot describe here  
 before the (potential) filing of a patent application.  
499    
500  Storage  Storage
501  '''''''  '''''''
502    
503  TJL  JVK, HH
504    !!! [tjl] Mikä tässä on KYSYMYS?
505    
506  Research on how location-independent identifiers can be used to find  Research on how location-independent identifiers can be used to find
507  information in a Peer-to-Peer environment. To support our research work, we have  information in a Peer-to-Peer environment. To support our research work, we have
# Line 652  TJL Line 748  TJL
748    
749  .. research results to be expected and their significance  .. research results to be expected and their significance
750    
751  The Fenfire project is largely based on novel ideas on both  As already discussed, this project is a combination of several
752  user interfaces and hyperstructure.  subprojects revolving around specific questions raised by our work in Fenfire.
753  Therefore, the project includes basic research as well as  The expected results of each subproject are listed in the subsection above
754  applying the developed technologies on  describing that subsection.
755  the Fenfire distributed hypermedia system.  The results will be published in international conferences and journals.
     
 The expected results of the research are listed below:  
     
 - paper: procedural background textures for  
   bestowing identity to virtual objects  
756    
757      - quantitative model of certain aspects of texture perception and recognition  The expected results of the research are listed below:
758    
 - fillets: drawing connections between things in a perceptually  
   motivated way resulting far quicker perception  
759    
     - quantitative experimental results from general graphs layout  
760    
761  - experimental justification for the layout geometry and type of animation  - experimental justification for the layout geometry and type of animation
762    (buoys, nadir, irregu, etc.)    (buoys, nadir, irregu, etc.)
# Line 676  The expected results of the research are Line 764  The expected results of the research are
764  .. - buoys: show connected documents and ideas floating in the margin              .. - buoys: show connected documents and ideas floating in the margin            
765  .. - nadir: a natural layout for multiple pieces of data by small rotations  .. - nadir: a natural layout for multiple pieces of data by small rotations
766    
   
 - storm  
   
     - saving data in permanent blocks, requiring  
       less bandwidth, allowing access to earlier  
       versions.      
     - data uniquess and immutability  
     - distributed storage framework  
   
 - urn-5  
   
     - registered URI scheme to our project  
   
 - Object/document identity in user interfaces  
 - P2P networking  
 - Interactive creation of hyperstructure  
 - Bidirectional linking in hyperspace  
 - Xanalogical hypertext model  
 - Relational data/data relationship management, document/data item management  
     
767  .. Used techologies, our implementations:  .. Used techologies, our implementations:
768    
769          - RDF structure          - RDF structure

Legend:
Removed from v.1.111  
changed lines
  Added in v.1.112

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