RE: [OSGeo-Discuss] Open File FormatsandProprietaryAlgorithms[SEC=UNCLASSIFIED]

Lucena, Ivan ivan.lucena at pmldnet.com
Sat Aug 22 06:43:05 PDT 2009


That is great. Thanks!

-isl


>  -------Original Message-------
>  From: Michael P. Gerlek <mpg at lizardtech.com>
>  Subject: RE: [OSGeo-Discuss] Open File	FormatsandProprietaryAlgorithms[SEC=UNCLASSIFIED]
>  Sent: Aug 21 '09 13:59
>  
>  Yes, JP2 supports signed and unsigned types of up to ~24 bits.  And lots of channels (bands).  And alpha masking.  And arbitrary metadata blobs (geospatial and otherwise).
>  
>  -mpg
>  
>  
>  -----Original Message-----
>  From: discuss-bounces at lists.osgeo.org [mailto:discuss-bounces at lists.osgeo.org] On Behalf Of Lucena, Ivan
>  Sent: Friday, August 21, 2009 12:22 PM
>  To: OSGeo Discussions; OSGeo Discussions
>  Subject: RE: [OSGeo-Discuss] Open File FormatsandProprietaryAlgorithms[SEC=UNCLASSIFIED]
>  
>  But you can't compress data types other than byte in JPG. Can you do that in JP2K?
>  
>  
>  >  -------Original Message-------
>  >  From: Landon Blake <lblake at ksninc.com>
>  >  Subject: RE: [OSGeo-Discuss] Open File	 FormatsandProprietaryAlgorithms[SEC=UNCLASSIFIED]
>  >  Sent: Aug 21 '09 12:42
>  >  
>  >  Paul,
>  >  
>  >  I was wondering the same thing.
>  >  
>  >  It seems a little like choosing to drive a Honda Accord, or a Ferrari.
>  >  The Ferrari is a lot faster and comes with a better looking trophy wife
>  >  (or husband), but the Honda is a lot easier to fix. (Try finding an
>  >  affordable Ferrari mechanic in Stockton, California.)
>  >  
>  >  To tie this back into our original discussion, it seems like the
>  >  government should be choosing to drive a Honda Accord when it can,
>  >  instead of the Ferrari.
>  >  
>  >  I guess you'd really have to crunch the numbers and see if the savings
>  >  in bandwidth/disk space costs were really worth the compression savings
>  >  that result from a proprietary compression scheme ("wavelet black
>  >  magic").
>  >  
>  >  The problem with this is a lot of the benefits that come from the Honda
>  >  Accord (open image format + open compression algorithm) aren't easily
>  >  calculated in dollars and cents.
>  >  
>  >  Still, this speaks to an important truth I have discovered in open
>  >  source development: Simple is better, even when it isn't necessarily
>  >  faster and smaller.
>  >  
>  >  I'd rather have code that I can understand, or a file format that a
>  >  programmer in 20 years will understand, than a Ferrari you can't drive
>  >  unless you have a PHD and did a thesis on wavelet compression. :]
>  >  
>  >  Landon
>  >  Office Phone Number: (209) 946-0268
>  >  Cell Phone Number: (209) 992-0658
>  >  
>  >  
>  >  
>  >  -----Original Message-----
>  >  From: discuss-bounces at lists.osgeo.org
>  >  [mailto:discuss-bounces at lists.osgeo.org] On Behalf Of Paul Ramsey
>  >  Sent: Friday, August 21, 2009 10:36 AM
>  >  To: OSGeo Discussions
>  >  Subject: Re: [OSGeo-Discuss] Open File
>  >  FormatsandProprietaryAlgorithms[SEC=UNCLASSIFIED]
>  >  
>  >  So hung up on wavelets, we are.
>  >  
>  >  Internally tiled TIFF with JPEG compression and similarly formatted
>  >  internal overviews can achieve 10:1 compression rates without
>  >  noticeable image quality reductions, and as an added bonus can be
>  >  decompressed a heck of a lot faster than wavelet-based formats. The
>  >  wavelet stuff is k00l, in that there is no need for an overview
>  >  pyramid (it's implicit in the compression math) and much higher
>  >  compression rates can be achieved. But operationally, you can go a
>  >  long way with the more primitive (open image format + open compression
>  >  algorithm) approach.
>  >  
>  >  P.
>  >  _______________________________________________
>  >  Discuss mailing list
>  >  Discuss at lists.osgeo.org
>  >  http://lists.osgeo.org/mailman/listinfo/discuss
>  >  
>  >  
>  >  Warning:
>  >  Information provided via electronic media is not guaranteed against defects including translation and transmission
>  errors. If the reader is not the intended recipient, you are hereby notified that any dissemination, distribution or
>  copying of this communication is strictly prohibited. If you have received this information in error, please notify the
>  sender immediately.
>  >  _______________________________________________
>  >  Discuss mailing list
>  >  Discuss at lists.osgeo.org
>  >  http://lists.osgeo.org/mailman/listinfo/discuss
>  >  
>  _______________________________________________
>  Discuss mailing list
>  Discuss at lists.osgeo.org
>  http://lists.osgeo.org/mailman/listinfo/discuss
>  



More information about the Discuss mailing list