What does "GELI" stand for (not what does it do)?

L

listentoreason

So I'm busily working at figuring out the best geli implementation on my system, piles of notes, some early benchmarks. It's silly, but it's been irritating me that I don't know what g-e-l-i means. I know that "g" is GEOM (so I've ruled out "Georgia Early Learning Initiative"), but what about "eli"? I've seen it written in all caps, so I presume it's an acronym, but maybe "eli" is Dutch or Catalan for "crypto" or some such? The closest Google result I found is here which ends cryptically (ha!) with "It is secred[sic]".
 
The "G" stands for "GEOM" I'm pretty sure. The "E" and "L" may be for words "Encryption" and "Layer". No idea though what the "I" would mean then.
 
listentoreason said:
but maybe "eli" is Dutch or Catalan for "crypto" or some such?
Click to expand...
For what it's worth: "eli" is not a word in Dutch ;) Perhaps this is something you might want to ask the author of geli about; his e-mail address can be found at the end of the geli(8) man page.
 
fonz said:
For what it's worth: "eli" is not a word in Dutch ;) Perhaps this is something you might want to ask the author of geli about; his e-mail address can be found at the end of the geli(8) man page.
Click to expand...

Good point; my brain is apparently single-threaded when it comes to communication mode and when in "forum" forgets about "email". I've mailed a request!

For what it's worth, you corrected out a literal quote in the initial message. The quote did in fact read "It is secred"; I should have added [sic] to it. I actually found it an interesting misspelling, since there are two context-appropriate single-character corrections to it:

It is secured (insertion-deletion mutation)
It is secret (substitution mutation)

AND the quote is attributed to Pawel Jakub Dawidek, geli's author. Wheels within wheels... Perhaps there's a seventh self-encrypting algorithm hiding somewhere in the source? ;)
 
Official Answer

I have heard back from Pawel Jakub Dawidek! He responds:
Unfortunately I cannot reveal what it means.
Click to expand...

Normally I consider any encryption framework that includes hidden components more than suspect. However, after careful consideration and a cup of tea, I've decided that in this particular scenario the uncertainty is tolerable. Moving forward, where an explicit value is necessary for a particular implementation, I shall substitute the reasonably estimated value of "GEOM Encryption Layer Interface" suggested by readers above.

Thanks again to Pawel and the rest of FreeBSD for geli!
 
You must log in or register to reply here.
Back
Top