Perhaps the greatest thing about TIFF, but also the most horrible things, and probably why TIFF is mostly historical. It was so extendable that no two programs ever accepted the exact same TIFF extensions. (omitting the war story)
edit: forgot about byte order...
But most of these variations were part of the spec (endianness with II or MM, later magic 43 for bigTIFF 64bit extension). I work with tiff and tiff-derived formats in digital microscopy where its very much not historical. And the alternatives (DICOM supp 145, vendor-specific garbage ... and thats it) are worse.
I quite like the format, the only thing I would change is to have the option not to store directory information in a linked list spread throughout the file but in a simple array. Duplicate it at the beginning and end of the file and you've got resilience too (important in the age of floppies)
TIFF is still very much alive in certain circles, see for example https://cogeo.org/ The format is basically a TIFF file with attached georeferencing information and with the data organized by geographical sector, enabling fast downloads of regional subsets.
The Adobe DNG standard for raw camera images is based on TIFF as well. DNG is used in lots of places, including the raw capture support built into all modern iOS and Android smartphones.
I’ve been using both TIFF and DNG this very week in my work (https://filmlabapp.com), so I was happy to read this post and learn about Steve Carlsen aka Mr. TIFF, whose work we’re still building on 39 years later.
I'm not super knowledgeable about this stuff.. but out of curiosity, what advantage does it have over NetCDF?
I wrote a program processing GeoTIFF data. When I had started this project I chose GeoTIFF mostly b/c i wanted something simple. And I could load them in to Java's BufferImage class and manipulate them that way. But it seems all the pros exclusively use NetCDF and GeoTiffs are for noobs (working with atmospheric science data here)
GeoTIFF does extend "images" to cover more usecases, but a lot of stuff doesn't fit (like say a wind vector) and then you need some other container or metadata b/c you generally have many images. So I get the sense the complexity just ends up being moved elsewhere.
Generally speaking I would classify TIFF and its variants as imaging formats (or for very simple numerical datasets), and NetCDF as more suited for raw data, in particular multi-dimensional data with time series, etc.
For forecast and climatological data I find NetCDF is vastly superior, but also much more complicated to work with due to the capabilities and how open the format is. Just have a look at the complexity of the CF Conventions to see what I mean: https://cfconventions.org/cf-conventions/release/v1.12.0/cf-...
For visualizing orthophotos and the like, I would choose GeoTIFF any day of the week, as they're easy to visualize across platforms using existing libraries. Using COGs you also get the functionality of a spatial index within each GeoTIFF file, meaning that you can stream subsets of GeoTIFF files without having to scan through the entire file for each request.
We have hundreds of thousands of TIFF files where I work which are scans of questionnaires filled out by clinical trial participants. The one annoying thing is that web browsers don’t natively display them. I did some incredibly inefficient JavaScript bs to decode the pixel data, plop it in a canvas, get a PNG data url from it, and set that as the src for an img element xD (why not just display the canvas? because I was too lazy to manually handle resizes…) good times