[Date Prev][Date Next][Thread Prev][Thread Next][Author Index][Date Index][Thread Index]

Hold the baloney (Re: a portable Transcopyright browser)

Writes David Phillip Oster <oster@xxxxxxxxxx>

>    How much of the browser needs to be "our browser"? [....] 
>    a large portion of a Transcopyright-aware browser should be
>    prototype-able inside any Java aware browser.  This lets us
>    concentrate, right from the start, on interesting issues, 
>    without needing to dive in and do a whole browser. 

  That is entirely correct, but unfortunately also a minor aspect 
  of any Transcopyright browser ("problem"?), as the whole of that
  latest incarnation of the Xanadu, is primarily, secondarily and
  lastly a back-end, server issue.  In any event I wouldn't start
  any Java-work in earnest until there is a defined, correct set 
  of Transcopyright-serving commands AND an understanding, if not
  formal protocol, of how such is to cooperate with/ besides/ per-
  chance atop any HTTP server or equivalent, payment-accounting 
  routines and other back-end tools. 

  The Transcopyright server's function would after all not be solely
  to resolve which of the requested streams are to be handled in its
  native, accounted-for, fashion (while others "merely" as default
  HTTP and other subprotocol streams), but to stitch together "view
  objects" out of disjoined fragments, some of which will NOT be
  available at the moment no matter what. And then do it in a trans-
  parent and speedy enough fashion for the browsee not to notice the
  difference.  Solving of which will not be a trivial issue, as any
  of us who have read Gary Wolf's in-depth account of the Xanadu
  tunnel vision in WiReD in June 1995 ( also available, sort of at:
  http://www.hotwired.com/Coin/Ecash/index.html ) can attest.  Or,
  should there be any "better," enriching text of like information
  granularity about it, I'd be glad to for a pointer to it. 

>    Since you can link C code into Java, for those environments that
>    support irrregularly shaped windows, which may also float above
>    any application, this allows cross-window arrows to be implemented
>    as a special kind of arrow-shaped window, that moves in response
>    to inter-task communication messages sent to it by the applets
>    that are managing space in browser windows.  In particular, I can
>    do this on a Macintosh. 

  Instinctively speaking, without too much reflection, I'd say an
  interesting implementation of a visual navigation scheme in a by
  definition fragmented reality, but the back-end (well, _my_ inner
  back-end) of the dual bugaboos of knowledge and experience tells
  me that such in all their logical vision, er, logicKal solutions
  usually fall flat in the face of the very first HCI studies, even
  such conducted with paper protoypes.  For a moment's reflection
  on complex-software design issues and methods/ processes, consult:

<dd><h2>          A few pointers of ergonomic interest . . .
</h2><!--         __________________________________________
<a href=     http://www.sun.com/sun-on-net/uidesign/
><b>         http://www.sun.com/sun-on-net/uidesign/
</a><dd>          A detailed, illustrated account of a nine-step ergonomic
                  design/ redesign process of Sun's internal Web pages
                  (rich in Java, but only non-Java screenshots posted in 
                  this presentation); fairly bandwidth- intensive, but worth
                  every byte of it. Contains a pointer to a rare on the Web,
                  well-illustrated paper on the _methodology_ of that, er, 
<a href=          http://www.sun.com/sun-on-net/uidesign/sunweb/sunweb.html
>                 paper-prototyping process
<a href=     http://www.sun.com:80/current/columns/alertbox/
><b>         http://www.sun.com:80/current/columns/alertbox/
</a><dd>          Jacob Nielsen, SunSoft Distinguished Engineer and the 
                  author of the above "private" column containing a wealth
                  of well-put, intriguing issues and a must-read for anyone
                  whose software-writing ambitions stretch beyond staying 
                  within a formal budget. Cannot be recommended strongly 
                  enough; Nielsen should be given a Nobel Prize in Tech-
                  Writing Lucidity. A low-bandwidth fount of wisdom. Of
                  for this Xanadudlian forum particular interest should 
                  be Jacob's pitfalls of Java borderless adoration 
<a href=          http://www.sun.com/950523/columns/alertbox/jav.html
><b>              a l e r t !   a l e r t !   a l e r t ! . . .
<a href=     http://www.acm.org/~baychi/meetings/archive.html
><b>         http://www.acm.org/~baychi/meetings/archive.html
</a><dd>          Archive of past presentations at the Sillivalley/ Bay
                  Computer- Human Interaction SIG chapter of the ACM,
                  which contains many interesting (text) pointers. Will 
                  soon move to its own domain of "baychi.org" I believe.
                  Low-bandwidth resource.
<a href=     http://www.cs.unm.edu/pad%2b%2b/begin.html
><b>         http://www.cs.unm.edu/pad%2b%2b/begin.html
</a><dd>          An eye-opener both for presentation, content AND 
                  Xanadu/ Transcopyright context: an innovative infinite-
                  zoom interface which *MAY* be closer in spirit to any
                  data-fragment front-end than any other (non-ideavapor-
                  warish) client model that I know of.
<p><a href=       mailto:ianf@xxxxxxxxx

__Ian </a>        THE Missing Link Between Natural St
                  upidity and Artificial Intelligence