While MD5 isn't really secure against intentional attacks any more, the probability of an random collision is still negligible.
I originally started using MD5 for this purpose because in a test I did many years ago one some machine, md5sum actually ran faster than cksum. The shorter cksum data also does have a chance to generate hash collisions on reasonable sized data sets, although that probably doesn't matter too much for just disk error checking. I don't use the newer algorithms because they're overkill and their hash strings just look too long.
I never archive any significant amount of data without first running this script at the top:
find -type f -not -name md5sum.txt -print0|xargs -0 md5sum >> md5sum.txt
It's always good to run md5sum --check right after copying or burning the data. In the past, at least a couple of percent of all the DVDs that I've burned had some kind of immediate data error
(A while back, I rescanned a couple of hundred old DVDs that I burned ranging up to 10 years old, and I didn't find a single additional data error. I think that a lot of cases where people report that DVDs deteriorate over time, they never had good data on them in the first place and only discover it later.)
In the context of FAT file systems, "short file names" has a very specific technical meaning. It's a name of exactly 8.3 upper case ASCII letters which is baked into the file system format.
We're not talking about just using short strings for the arbitrary file names supported by modern file systems. Your data set wouldn't even begin to fit on an old DOS machine that only understands short FAT filenames.
You typically need cooling water to efficiently generate electricity, no matter what source of heat is used to drive the boilers. You have to be able to condense the steam coming from the turbines to create a near vacuum, which requires a vast heat sink. That's why coal-fired stations are also often put next to rivers or lakes.
You wouldn't forego the interoperability. It's fine to continue using FAT.
I was simply saying that the patent system should be changed so that the patent on the short file name feature should have been revoked long ago (even if it had not been found to be obvious), due to the fact that nobody needs to interoperate with DOS machines any more, which was the original point of the invention. Having to use the patented feature so that FAT can interoperate with FAT is just a tautology that provides no benefit to a patent licensee. Thus, people should be free to use FAT today without worrying about this patent.
Sorry, but I'm unable to parse your question in its entirety, nor do I see how any of its fragments are relevant to the discussion.
I would like to pretend that DOS is not still with us today, but that would be false. There's still people out there using short file names today. They're useful to someone.
Ok, so there are a few embedded industrial controllers still running DOS, and a few dorks running retro games in their moms' basements.
That doesn't mean that any of them are loading media files off of today's hardware gadgets into these relics. (In fact, a single one of these media files typically exceeds the entire addressable storage capacity of a DOS machine.) Yet Microsoft extracts royalties from the gadget manufacturers as if compatibility with short file names were an essential feature of the media libraries of everybody on the planet.
Whether it's relevant any more or not is neither here nor there.
I was arguing that it should be.
If I invent a way to make a clockwork mechanism work more efficiently, that's still an invention, still patentable.
And it's useful to someone. People still buy mechanical timekeeping devices, often at a very high price premium.
Short file names aren't useful to anyone in this day and age.
The network effect is similar to begging the question.
Something is popular because it's popular.
One of the important requirements for a patentable invention is that it must be "useful".
This patent originally covered a way to provide compatibility between short and long file names. But nobody has used short file names in decades.
So now, the "feature" continues to be necessary only so that FAT can provide compatibility with itself. That's like begging the question. The feature no longer has any intrinsic usefulness, and in fact just serves to make the file system format more convoluted and less efficient.
The patent system ought to be changed so that any patent should be revoked once it is no longer useful for its intended purpose. This particular patent has recently been "useful" solely as a way to give Microsoft leverage in the media device market. The covered feature provides zero benefit to end users.