[OSGeo-Discuss] Open File Formats and Proprietary Algorithms

Michael P. Gerlek mpg at lizardtech.com
Thu Aug 20 17:45:30 EDT 2009


I'll mention too the question of patents and JP2, since this thread is bound to get into THAT issue too before long :-)



Some of the algorithms within the JP2 standard (from ISO) are patented.  However, the companies in question have agreed to not exercise their rights on those patents for any implementation of the standard.  That is, if you write a ISO-compliant JP2 encoder, Company X won't come after you.  This is a good thing, and is not uncommon practice for some standards groups.  It's better for us than the RAND ("reasonable and non-discriminatory") clauses that get used by some groups.



However, there is an interesting philosophical consideration for the open source community here.



Let's say I write a nice, compliant MpgJp2 library on Monday and open source it.  Landon looks at my code and, smart cookie that he is, realizes that he could improve the overall compression ratio by tweaking one of the core algorithms.  He forks my code, makes the change, and posts the SunburnedJp2 library to the web on Tuesday night.  Cool.  We like that.  Open source in action.



But wait -- Wednesday morning, he finds an email from Company X's lawyers in his inbox: he is now in violation of X's patent, because he is not using the patent within the bounds of a "compliant JP2 encoder".  He broke the file format.  ["You break it, you buy it"?]  It's not a "JPEG 2000" library anymore.



Some open source partisans may therefore consider the JP2 standard to not be truly open enough.



I'm sure there are other standards with this same problem, although I don't know of any offhand.



-mpg





From: discuss-bounces at lists.osgeo.org [mailto:discuss-bounces at lists.osgeo.org] On Behalf Of Landon Blake
Sent: Thursday, August 20, 2009 2:57 PM
To: OSGeo Discussions
Subject: RE: [OSGeo-Discuss] Open File Formats and Proprietary Algorithms

MPG:

Thanks for the clarification.

When you said "there is today no open source implementation of JP2 that is suitable for geo work" do you mean that there is no open source library that can read and write JP2? If so, who is using the format?

Do you know why there hasn't been a broader adoption of JP2?

(I should also add the MPG helped me publish a short article in support for open file formats, so I know he is on our side.)  :]

Landon
Office Phone Number: (209) 946-0268
Cell Phone Number: (209) 992-0658


________________________________
From: discuss-bounces at lists.osgeo.org [mailto:discuss-bounces at lists.osgeo.org] On Behalf Of Michael P. Gerlek
Sent: Thursday, August 20, 2009 1:55 PM
To: OSGeo Discussions
Subject: RE: [OSGeo-Discuss] Open File Formats and Proprietary Algorithms

Some clarifications:

- MrSID has both lossy and lossless modes
- MrSID is not fractal based; it uses wavelets (and arithmetic encoding)
- you can't copyright algorithms; the MrSID source code certainly is, however
- MrSID relies on a number of patents, not all of which are owned by LizardTech
- reading MrSID does not require any fees; we have libraries you can download, although they are not open source

That said, some editorial comments (although I'm now wishing I hadn't been so quick to rise to Landon's bait :-)

- Some of you know the history of trying to open source MrSID; I won't go into that here, except to say that LizardTech doesn't own all of the required IP needed to make that happen.
- If we are speaking of the NAIP data, then no, it is not exclusively available in MrSID format; it is also shipped as GeoTIFFs.
- JPEG 2000 is a very robust open standard alternative to MrSID, and a number of players already support it (including LizardTech), but not enough to make it viable for certain domains like NAIP.
- some of you also know the history on open JP2 support: there is today no open source implementation of JP2 that is suitable for geo work.  Alas.

-mpg


From: discuss-bounces at lists.osgeo.org [mailto:discuss-bounces at lists.osgeo.org] On Behalf Of Eric Wolf
Sent: Thursday, August 20, 2009 2:15 PM
To: OSGeo Discussions
Subject: Re: [OSGeo-Discuss] Open File Formats and Proprietary Algorithms

The MRSID format is a very special case - and perhaps an opportunity for a new FOSS file format. MRSID is a lossless, fractal-based, multi-scale raster compression format. LizardTech has the algorithms to encode and decode MRSID locked up in copyrights, and I believe, patents. Even companies like ESRI shell out big bucks to LizardTech to be able to read and write the MRSID format.

I guess I missed the context of the discussion. Is the government releasing certain data exclusively in this format? If so, I think the argument can be made against this practice. The different in compression between MRSID and gziped TIFFs isn't really that great in this day of cheap disks and fat pipes.

-Eric

-=--=---=----=----=---=--=-=--=---=----=---=--=-=-
Eric B. Wolf                    New! 720-334-7734
USGS Geographer
Center of Excellence in GIScience
PhD Student
CU-Boulder - Geography



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.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.osgeo.org/pipermail/discuss/attachments/20090820/15a2f909/attachment.html


More information about the Discuss mailing list