Compression Techniques and GIS Systems

[Simon Cox]

Thought grasshoppers might not want to miss this one!

In article 10112 at netnews.whoi.edu, gie at charon.er.usgs.gov (Gerald I. Evenden)
writes:
>> Newsgroups: comp.infosystems.gis
>> >From: sidell at swine.cs.uiuc.edu (Jeff Sidell)
>> Subject: Compression Techniques and GIS Systems
>> Date: Sat, 6 Feb 1993 17:18:12 GMT
>>
>> I would like to know what, if any, data compression techniques are
>> being used in GIS systems.  Specifically, I'm interested in whether
>> or not any systems represent geographic data as point sets, and
>> whether or not they compress the data.  Please post or e-mail replies.
>>
>> Jeff Sidell
>> --
>> ______________________________________________________________________
>> Jeffrey P. Sidell                          | "Computers are useless. |
>> Database Group                             |  All the can do is give |
>> Department of Computer Science             |  you answers."          |
>> University of Illinois at Urbana-Champaign |      - Pablo Picasso    |
>> ----------------------------------------------------------------------
>
>From what I have seen (mostly governmental data types): NONE!
>
>The reason I use an emphatic answer is that I personally believe that
>far too little concern has been given to the use of even common techniques
>to minimize the size of the physical storage required for gis information.
>
>Too often I have been able to achieve 10-1 compression of the
>information with the most primative of techiques.  And still make the
>data machine independent and transportable.  The downside is that
>a special procedure must be used to reconstitude the data, but this
>is normally fast and, given the slow access of CD-ROMs, can probably
>keep up with the input data rate.  It possible that compressed data
>may actually be more rapidly retrieved from a CD-ROM than raw
>information.
>
>For example, the usable information on the USGS 1:2,000,000 vector
>map of the USGS requires only ~15Mb versus the ~300Mb on the CD-ROM.
>Of course this is misleading, because the distribution contains
>triple redunancy and a lot of non-essential fluff.
>
>I have the feeling that most people who are now creating CD-ROMs
>have the attitude of "Gee! 600Mb! All I have to do is just dump my
>data to the the ROM.  I wonder how many discs the TIGER data would
>taken if SOME thought had been given to data compression.
>
>There is so much gis information that we are going to be buried up
>to our eyeballs if attention to data compression is not forthcoming!


__________________________________________________________________
				Dr Simon Cox
	 __  L
      ,~'  L_|\                 Department of Earth Sciences
   ,-'         \                Monash University
   (            \               Clayton  Vic  3168  Australia
   \    ___     /
    L,~'   "\_x/                Phone +61 3 565 5762
	      u                 Fax   +61 3 565 5062
				simon at cerberus.earth.monash.edu.au
__________________________________________________________________

----
__________________________________________________________________
				Dr Simon Cox
         __  L				
      ,~'  L_|\            	Department of Earth Sciences       
   ,-'         \         	Monash University    
   (            \		Clayton  Vic  3168  Australia
   \    ___     /	
    L,~'   "\_x/		Phone +61 3 565 5762
              u   		Fax   +61 3 565 5062
				simon at cerberus.earth.monash.edu.au
__________________________________________________________________