home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.infosystems.gopher
- Path: sparky!uunet!charon.amdahl.com!amdahl!rtech!sgiblab!zaphod.mps.ohio-state.edu!van-bc!cs.ubc.ca!news.UVic.CA!fester!mmansell
- From: mmansell@fester.UVic.CA (Mike Mansell)
- Subject: Re: Retrieving Gopher documents chunk by chunk
- Message-ID: <1993Jan26.234056.13234@sol.UVic.CA>
- Sender: mmansell@nero.UVic.CA (Michael C. Mansell)
- Nntp-Posting-Host: fester.uvic.ca
- Organization: University of Victoria, Victoria, BC, Canada
- References: <mtm.727513150@CAMIS> <1jjuqnINN5mn@msuinfo.cl.msu.edu>
- Distribution: world
- Date: Tue, 26 Jan 93 23:40:56 GMT
- Lines: 28
-
- What is needed is a standard format for all documents of a specific type.
- Ie. When a "page" is requested, the server converts the document to
- a type (ie. RTF) that it "knows" about. Calculates the page length
- and sends the chunk. Obviously this is slow, but several things
- could speed it up, like caching page sizes for a document, or
- learning about other documents.
-
- The only problem is that some documents really don't like being transmitted
- in chunks. You couldn't do this in Word, unless the server:
-
- 1. Converted the Word document to RTF
- 2. Calculated the page requested.
- 3. Converted the data on that page only back to a new Word document
- 4. Send that to the client.
-
- The same idea can be done with other formats, like a common sound
- format or image format.
-
- Conversion programs are the only problem, but they are appearing rapidly
- as different companies want to keep up in this interchangeable world.
-
- This, to me, is the second major hurdle to be overcome (the first being
- locating the resource in the first place) since many clients/servers are
- on slow links to the rest of the gopherspace.
-
- Michael Mansell
- mmansell@nero.UVic.CA
- Computer Engineering
-