[Geotiff] Geotiff API's

[Daniel Denk]

Frank,
Thanks for such a quick reply once again. And thanks for the help!

I suppose I'm from a slightly different perspective in this - having
observed many of the opportunities manifest, but over time begin to come
together like a jig-saw puzzle. Currently, the main focus that's coming to
fruition in Web based design is that of full compliance via the W3C spec for
XHTML/CSS (separation of content and style). This was the original proposal
for the approach in creating web pages and web interface design. The
majority of designers (graphic art) are however odd creatures, in that if
there is a hack in the technology they will exploit it to gain the effect
that they are after visually and in experience for the end-user audience
they target. This is how HTML table layouts inevitably came into being. In
other words, much of what's seen on the web is in essence a hack for effect.
It's bloated and only creates the illusion of elegant functionality. As time
has gone by, and since the bubble broke, development initiatives have taken
the fore - and designers are beginning to seek opportunities that are more
practical and reliable, and perhaps serve the greater good of society
overall. This is why we're seeing more XHTML/CSS conversions, as well as the
implementation of XML/XSLT and other front-end languages that create the
umbrella of true functionality - as well as integrate elegantly or
seamlessly into any infrastructure.

As mobile technology's becoming widespread and accessible - and as the
pieces are all laid-out in that area - designers and companies are beginning
to see the practicality of use for mobile, pocket-sized computing as being
the truer form of practical functionality for any end-user (every designer's
dream is to serve society in functionality - and hopefully for the greater
good). This is what generally separates the designer from the graphics
geeks. The graphics geeks are all about the image - the eye-candy. The
designer is in it for change - progression - shaping a new reality. The
perspective is divided, of course. It all depends on who one talks to, and
always some convoluted philosophical debate.

The missing component is simple, but not far off into the future. I've been
watching this particular technology emerge, and I feel that it's the key
that's going to help to blow the roof off the current cap on mobile
technologies. The 'Wireless Cloud'. With exception to localized hot-spots
and experimentation, it hasn't been installed for mass use. This is likely
to change in the near future (within only a year to a couple). This is the
most interesting of the companies to keep an eye on at this time, but there
are initiatives in play with companies such as Motorola and the like:

http://www.sanswire.com/ (Note: Stratellites)

So... In a nutshell - the technology is there, and in pieces that can be
applied readily within a relatively short amount of time, logistically
speaking. Once the cloud is in place, I foresee applications being focused
toward the mobile markets and released, if at least as Beta releases to the
public and commercial entities the following year. Some companies, such as
Google, are already progressively (and aggressively, although under the
public radar) aligning themselves, or buying-up companies that will factor
into solutions. http://www.keyhole.com/ is one, and a more obvious example
of this.

Note: Google owns and operates, and is the largest buyer of DRAM. They're
entire infrastructure is based on the DRAM Cluster model - truly brilliant.
I had the brief opportunity of participating in a Beta test of their Google
Search Appliance when it was in the final stages of development. We had
learned a great deal about the backbone of Google and how their
infrastructure is built, and became fascinated with the possibilities of
what the DRAM clusters that they operate represented in terms of serving
large amounts of data to the public, via the Net, and to devices in
real-time.

Another known development in technology is that of Video Registration.
Companies are currently 'racing' toward a solution. How exactly this will be
applied is a weird range of debate. Targeting is the obvious solution. But -
what if video-reg is applied to real-time navigation systems via mobile
device capabilities? I ponder the possibilities in functionality for the
visual public that cries-out for elegant functionality.

Anyway. Just some added notes as to why I'm so eager to pursue these kinds
of ideas with those who can make them possible. I don't assume to know what
others here know, although I can contribute something in way of idea
manufacturing that might lead to some hint of influence in direction or
functionality.

Thanks again,
Daniel




-----Original Message-----
From: Frank Warmerdam [mailto:warmerdam at pobox.com] 
Sent: Saturday, November 13, 2004 6:13 PM
To: Daniel Denk
Cc: 'Geotiff'
Subject: Re: [Geotiff] Geotiff API's


Daniel Denk wrote:
> Thank you for your suggestions and insights, Frank.
> 
> I do have these apps, but haven't used them due to their own claims of
> limitation:
> 
> Geotiff Examiner http://www.mentorsoftwareinc.com/freebie/FREE0699.HTM
>
http://ioc.unesco.org/oceanteacher/resourcekit/M3/Toolbox/GeoTIFFExaminer.ht
> m
> 
> Geotiff4
> http://www.terrainmap.com/rm22.html
> 
> I should have mentioned that I was looking for another solution that 
> doesn't claim limitations, unfortunately I had a head-full when I 
> typed away!

Daniel,

OK, now looking at it again I remember what I thought was the fundamental
limitation of the GeoTIFFExaminer.  It completely ignores the coordinate
system definition!  Apparently it was, and still is just a  way of handling
the world file conversion to/from a tiepoint and pixel size in a GeoTIFF.

It also ignores the rotation case, as you noted, and lacks support for a
list of tiepoinst.

I tried to skim the GEOTIFF4 page you reference, but kept getting distracted
by thoughts of supper and gave up before getting into any substantial
detail. What does that say about me?

Anyways, it seems targetted to specific issues of translating ASTER HDF and
GeoTIFF into something slightly more usable.  I don't know how general it
is.

I would add that the OpenEV application includes support for converting
GeoTIFFs between worldfile+TIFF and GeoTIFF.  It includes the rotation and
has some limited support for editing the coordinate system.  In fact, I
believe it was a contract with Ned's organization that funded some of those
improvements.  However, OpenEV does quite a bit more than just this, and so
it isn't what I would call a "focused" GeoTIFF translator. In particular you
generate a TIFF+world file I believe you need to set an obscure option in
the export tool.  Also, it presents and to some extent allows editing the
coordinate system in OpenGIS WKT foramt, which is a step or two removed from
the form it is kept in GeoTIFF.

If you were interested you could look at:

   http://fwtools.maptools.org/

I fear that package is somewhat bulky (10MB) and includes lots of
-- yikes -- commandline tools, along with the GUI interface.

Interestingly, it would be relatively easy to prepare a GUI using the same
framework as OpenEV that is specifically focused on supporting GeoTIFF if
that were seen as important.

However, OpenEV is an excellent tool for visualizing and exploiting a wide
variety of GIS/RS data in GeoTIFF and other geospatial formats.  It's no
photoshop of course.

To that end, I don't have much advice on the photoshop, or FITS Liberator
front.  It would be nice if Photoshop could work smoothly with GeoTIFF data.
I think it would be a nice "finishing" application for raster maps.

Also, I have to admit I remain somewhat dubious about those who look forward
to GIS and Remote Sensing become ubiquitous on the desktop of normal folks.
The data is definately getting wider and wider exposure as the years
progress, but I don't personally see a internet style popularity bubble
ahead even though I have been hearing that suggestion for well over a
decade.  Nevertheless, I am still keen on making geospatial data easier to
get at and use for anyone.

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