From: isw on
In article <jollyroger-98460E.12480418122009(a)news.individual.net>,
Jolly Roger <jollyroger(a)pobox.com> wrote:

> In article <isw-AC315E.10332418122009@[216.168.3.50]>,
> isw <isw(a)witzend.com> wrote:
>
> > Not with Time Machine, but with "backuplist+", which does them once a
> > week.
> >
> > Pros/cons?
>
> What if you lose it? Perhaps <https://www.ironkey.com> is the answer.

Same as if I lose the hard drive I'd be replacing it with...

Isaac
From: isw on
In article <fmoore-9670B1.16410918122009(a)feeder.eternal-september.org>,
Fred Moore <fmoore(a)gcfn.org> wrote:

> In article <isw-AC315E.10332418122009@[216.168.3.50]>,
> isw <isw(a)witzend.com> wrote:
>
> > Not with Time Machine, but with "backuplist+", which does them once a
> > week.
> >
> > Pros/cons?
>
> NO! NO! NO! NO! NO! NO! NO! NO! NO! NO! NO! NO! NO! NO! NO! NO! NO!...
>
> ...if by 'backup' you mean a REAL backup in the sense that we usually
> use it here (secure copy of data from which the data can be recovered
> _uncorrupted_ at some time months or even years in the future), and not
> just an extra convenience copy or a copy to sneakernet to another
> machine.
>
> RAM chips are FRAGILE! Static electricity either by a single jolt or
> through accumulated smaller zaps will KILL integrated circuits. The
> microscopic connections on integrated circuits are TINY (think
> nanometers). Every exposure to high voltage (and the voltage doesn't
> have to be that high) vaporizes a small amount of the metal in one or
> more of the connections. Sooner or later the last molecule of gold or
> copper escapes into the cosmos from which it came, and the RAM chip is
> hosed. Some of the better constructed thumb drives claim to have
> electrical circuit buffering to extend their lives, but they are still
> vulnerable.
>
> Strong magnetic fields and x-rays can also corrupt/destroy integrated
> circuits.
>
> Why aren't the RAM chips inside your computer just as fragile? THEY ARE!
> except that the computer manufacturers take a number of steps to protect
> all of the semiconductors inside the machine.
>
> And none of the above addresses losing a thumb drive because it falls
> out of your pocket when you go for your car keys, or other numerous and
> mundane accidents.
>
> For more info on semiconductors and static electricity, Google can be
> your guide.

Probably a good place to visit for accurate information on thumb drives,
too...

Isaac
From: Tim Murray on
isw wrote:
> Not with Time Machine, but with "backuplist+", which does them once a
> week.
>
> Pros/cons?
>

Flash memory, the kind used on thumb drives, is not the same as the chips
used inside the computer. Flash memory wears out: It is the "flash" -- the
signal that causes the bit to be locked in -- that eventually wears out the
chip. Some thumb drives can detect a bad bit and compensate, but this only
postpones the inevitable (see http://en.wikipedia.org/wiki/Wear_levelling).
Many more expensive SLC-type thumb drives are rated for 300,000 cycles, but
cheaper MLCs are only around 10,000.

Flash has a problem disks don't, in that flash drives move your data around
with every update. Every time a flash drive writes a page, it has to erase
the entire block that page is in.

What happens to the data in the block? It gets read and rewritten along with
the new page to a new location. The map that keeps track of where your data
is rapidly gets very complex and itself is regularly read and rewritten.

If the drive's physical to virtual block map gets corrupted there is no way
to recover data. And that critical map is getting read, updated and rewritten
with every file update.

While I would recommend an external disk for a backup, statistically you're
probably okay with a USB as a backup, but not as a "working" drive. The
reason is that if your computer goes south the chance that your thumb drive
is also bad is small, unless what killed your computer was, say, a lightening
strike that happened while your thumb drive was installed.

From: Tim Murray on
nospam wrote:
>> The microscopic connections on integrated circuits are TINY (think
>> nanometers). Every exposure to high voltage (and the voltage doesn't have
>> to be that high) vaporizes a small amount of the metal in one or more of
>> the connections. Sooner or later the last molecule of gold or copper
>> escapes into the cosmos from which it came, and the RAM chip is hosed.
>> Some of the better constructed thumb drives claim to have electrical
>> circuit buffering to extend their lives, but they are still vulnerable.
>
> no

Every connection causes a spark. Some metal is lost in vaporization (the
metal sputters and is carried away in an aerosol) and some to excitation. The
melting point of the metal is a major factor. Gold has great conductive
properties but a low melting point and a low life-span.

The flashing circuit in a thumb drive is destructive, only for a different
reason.

From: nospam on
In article <0001HW.C751755400171A5BF0182648(a)nntp.charter.net>, Tim
Murray <no-spam(a)thankyou.com> wrote:

> >> The microscopic connections on integrated circuits are TINY (think
> >> nanometers). Every exposure to high voltage (and the voltage doesn't have
> >> to be that high) vaporizes a small amount of the metal in one or more of
> >> the connections. Sooner or later the last molecule of gold or copper
> >> escapes into the cosmos from which it came, and the RAM chip is hosed.
> >> Some of the better constructed thumb drives claim to have electrical
> >> circuit buffering to extend their lives, but they are still vulnerable.
> >
> > no
>
> Every connection causes a spark. Some metal is lost in vaporization (the
> metal sputters and is carried away in an aerosol) and some to excitation. The
> melting point of the metal is a major factor. Gold has great conductive
> properties but a low melting point and a low life-span.

um, no.

> The flashing circuit in a thumb drive is destructive, only for a different
> reason.

flash memory does have a limited life but it's far higher than what
someone is realistically going to reach. they'll probably want a larger
capacity device before it fails (or lose it).