tech-install archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: [Feature Request] UUID in fstab.



>If you have 32 hex-numbers, you have
>340282366920938463463374607431768211456 options. If you use 4 random
>words out of /usr/share/dict/words, you only get
>3100519963322993491281 variety, but if you really are worried that's
>not enough you can use 8 words and get
>9613224042964416904375938680615889423020961 options.
>This assertion needs proof.
>
>12345678-abcd-abcd-abcd-12345678abcd
>a-dog-ate-five-penny-worth-pork-meat
>
>consider that one takes its positions from [0-9a-f] and the other
>from [0-9a-z]
Dumber than the dumbest I could ever imagine. You always consider the 
total length of the string when comparing them. Although in this case 
they do have equal lengths, you'd be a fool if you try to maintain that 
length, given how many words are present in a dictionary. It'd be a 
complete waste of computing resources and time. Your algorithm will 
certainly be very slow.
Secondly, can you imagine how fucked up your /etc/fstab file can look 
if you use 8 words. The longest published word (I don't know about the 
dictionary you just mentioned), is 1909 characters long. So, if you get 
8 of them, your sequence would be 15272 characters long (without hyphens
) and the next entry could be only 8 characters long. I'm sure that's 
not human readable.
Thirdly, I still emphasize on the point that labels are designed to be 
chosen by humans (let them have the previlege). Don't enforce machine 
generated labels on them. UUIDs are what machines should assign.
Fourthly, humans are much more prone to making mistakes. It's much more 
likely for a human to label two partitions by the same name, or just 
fail to change the generic label. (Since you'll only be relabelling them
if they don't already have a label.)
>The average NetBSD user is not a complete dummy.
Maybe that's the case now, but if you really wan't mass adoption (which 
you don't seem to), you will have to assume that.
>Again, you take what I wrote and galopp off into the blue with it.
As if you were doing something different when accusing me of copying 
linux.
As for your SUSE, Red Hat, NetBSD and Linux dialogue, I think it's 
completely dumb to try and distinguish your OS with one single field in 
a config file that is almost the same as that of linux.
Given how stubborn you are, I think it's the right time for me to leave 
the conversation right here.


Home | Main Index | Thread Index | Old Index