[Geotiff] Geotiff API's

[Frank Warmerdam]

Daniel Denk wrote:
> Hello, Geotiff,
>  
> Is anyone aware of existing free or Open Source graphical API's 
> installable to Windows (and other typical home-use platforms) that allow 
> for a simpler, more intuative approach to say dumping metadata or 
> restuffing the tags to Geotiff [ie: listgeo, geotifcp, etc]?

Daniel,

There is a program called GeoTIFFExaminer that might be helpful in
this regard.  It used to be available from Mentor Software.  I think
a google search would likely turn up a link.  I haven't used it myself.

> Not only would this be useful for those of us who are command-line 
> retarded (as my colleagues might say), but it would have potential to 
> aid in awareness of capability for the home-user who will one day be 
> exposed to the wonderful world of GIS / Remote Sensing via mobile 
> applications, etc (dot-com number 2?). It would also be a productive 
> tool for those who are impatient with command-line utilities.

Well, I am personally somewaht doubtful about the usefulness, but I am
a bit GUI retarded myself.  It would certainly be nice to have a GUI
app that had quite a bit of smarts and would help people author
compliant geotiff metadata, pick from the EPSG lists and so forth.

> I have spoken with Chris Russ of Reindeer Graphics, who seems to think 
> that the Geotiff format itself - or at least the Tiff spec itself - has 
> inherent limitations in allowing for such functionality due to its bit 
> structure. The question that was instantly raised in my brain when 
> hearing this, was 'Well then, how is it that COTs packages are able to 
> incorporate it with little to no limitations?' He had also briefly 
> mentioned that it would be possible to work-around the 2GB / 30,000 line 
> limit within the newly developed .PSD format, which Adobe claims will 
> work around the bit structure and memory limitations - (according to 
> Chris - provided with the SDK for CS?) I'm not certain of this, or if 
> it's possible. It would also propose the concept of creating a 
> geo-spatial format in .PSD, thus causing all customers to adhere to a 
> newer, more proprietary format, from a vendor source outputting in this 
> format.

TIFF (and GeoTIFF) are essentially limited to 4GB maximum file size due
to the use of 32bit file offsets.  There is work underway in the TIFF
community to definea BigTIFF variation on TIFF that would support large
files.  I have no idea why you would run into a 100MB limit with apps,
though many apps only support files up to 2GB due to use of the old
file IO API that uses signed 32bit ints for offsets in calls like fseek.

> What that idea might introduce however, is the capabilities for 
> unlimited bands via channels with no limitations. Is this valueable? I'm 
> not sure, as I'm unaware of any sensor or potential product that would 
> benefit from this - at this time. If any of you are aware - by all means 
> share your insight?

I think I missed a connection here.  However, TIFF does support potentially
large numbers of bands (ie. samples) in an image.  I have certainly worked
with hyperspectral sensor data in TIFF format with 225 or more samples
per pixel (bands).

> If anyone would like me to compress the PS Geotiff plugin and send it 
> via the mailing list, or to specific Email addresses, please feel free 
> to reply.

Please don't send binaries to the mailing list unless they are tiny.
It also seems likely that the plugin would be copyrighted software and
that you would not have permissions to redistribute.  I do not approve
of copyright violation, even though I would like to see a more liberal
"IP" regime.

Best regards,
-- 
---------------------------------------+--------------------------------------
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | Geospatial Programmer for Rent