Right. Looking again, sorry I misread your post.
Right. Looking again, sorry I misread your post.
I don't understand. The "business" can buy a printer exactly like yours, get better deals on materials, and beat you in price.
Which is also stupid. Because the "business" can buy a much much better printer at 100 times the price you paid. Possibly with features specifically required for their kind of "printing". And beat you in price, quality, durability, customer-service, aesthetics, ergonomics, safety.
Well, I used to not enable pin/pattern/password security on my phone because of the high hassle to security ratio, for an admittedly less sensitive device. Motorola Atrix 4G's fingerprint feature made me use it for the low hassle to security ratio.
The "fancy"ness of the security system lies in hassle to security ratio. Whether it is used or not depends on sensitivity of the item. E.g. a very good security feature on my burger where no one but I can eat it may not get used, even though it is trivial to use and incredibly secure. For industrial security, one is ready to tolerate quite a bit of hassle for some real security. Most consumer level electronic items lie somewhere in between.
Yeah, it's so much more complicated to
That is where most of the postponement happens. That it costs money, and a realization of ongoing expenditure to maintain a backup, helps the postponement.
It's actually fairly common for disks of the same make and model to fail near the same time, anyway
And why would you get disks of the same make?
But the point is that online backups and local backups are like comparing apples and oranges
because in order to give the same benefits as online backup, you'd have to spend 4-5 times as much for local backup
And I showed that at much much lower cost, local backups can protect against events that can reasonably be expected in a lifetime by everyone. Lots of people have spent lifetimes without getting their houses burgled or burnt. Reaching adulthood without at least a hard drive failure / accidental file deletion / virus corruption is nearly impossible (assuming data storage from birth, of course).
So the same benefits are NOT NECESSARY to START backing up. Which is what my original post in this thread was about, parent post of which was suggesting to not even START backing up until you have an all comprehensive backup protecting against once (100,1000) year events.
I showed that online backup is more cost-effective.
By dishonestly using bad practices (all drives of the same make? WTF?). By cunningly changing the objective (3 disk failure in the same month is once in 3888 year event, given an average drive fails in 3 years) to saving against once in millenium events rather than reasonably expected events.
And I was arguing about STARTING with local even if remote appears expensive.
The analyst in me is quick to point out that implies failures in ZFS itself, beyond just the disks and "bit rot", but
I think it is that your disks are not giving any bit rot, but memory. Frequently written data passes through memory more times, and can get corrupted in case of memory errors uncatchable by ECC.
Processors overclocked and overvolted to hell also cause data errors, but I don't think that is your case.
While I agree with bcache doing one thing and doing it well, clubbing of volume management and filesystem has lot of advantages. Removing the famous RAID write-hole, for one.
Putting cache and encryption in FS doesn't make sense to me.
I was thinking of "upgrading" btrfs to zfs, but looks like it will be a downgrade given my variety of disks. Btrfs, while not supporting RAID 5 , handles variety of disks very efficiently. IIRC there is already the problem with zfs that you can't shrink a dataset.
Thanks, I need to do more research.
1. This "first" fallacy. As we discussed, online backups sometimes take a while to complete the initial backup. You can do this while continuing to make local backups to external media, and you should. So this whole point is silly and moot.
One has to setup the backup systems. Sign up for dropbox. Choose plans. Directories to backup. Schedule, if not automatic, depending on bandwidth costs at particular times.
Or setup the local backup - get local hard drives. Choose and configure backup program.
This setup is what causes postponing for people. And the majority of costs. And this is not what can be done in parallel - unless you can delegate it to minions. In which case it ceases to be "home" use for most people.
So you can misinterpret the post and call it a fallacy. Or you can put effort in understanding what I posted. Your wish.
Dropbox and CrashPlan, just two examples I'm very familiar with, have redundancy. So let's buy 3 of those external drives. That's $300
It is a BACKUP. To be used when original data is lost. By using 3 disks for backup, you are again preparing for a situation when 4 disks crash SIMULTANEOUSLY. Again once a century , if not a millenium, event.
And I already said local backup while cheaper saves from less vectors but being much much more cost effective should be done first, and will not encourage postponement. So rest of your post is based on a flawed reading of my posts.
This might help.
I would be able to get more usable space from the drives by running raidz instead of mirrored pairs - but only if you upgrade drives all at once
I thought the "zpool replace" is for that purpose? I am planning a ZFS setup, but all I can afford is a motley collection of dissimilar drives.
Can't I replace one disk at a time when upgrading?
People living under rocks in suffocating environments don't always say there aren't alternative ways of living , they just show their preferred way of living. Someone has to point out that it is suboptimal and unhealthy. They may or may not listen.
If all of them are NOT travelling, which is the typical state of a typical person, either B or C is forced to be alone for no fault of theirs.
I don't understand the second one. The website believes that you ran the computation against a passport, but what if you run it against
Being the real Mike, you must know everything about identifying real people
So you have a suspicion this agency will have power/resources comparable to NSA's by any stretch of imagination? Or access to anything comparable to national security letters ?
If not, I don't see why one snooping can't be much worse than the other.
Students are vertices of a graph. Each vertice can have an edge to any other vertice
No, classes are cliques.
The limit will be used up if all students choose to be alone
No, it will be used when no set of cliques the students can be arranged into. Other than the trivial set of cliques with each child a separate clique into oneself, of course.
This can happen in remarkably more ways than if each child were a loner. It also leads to injustices like:
A wants to be with B, C.
C wants to be with A, not B.
B wants to be with A, not C.
Now If you choose sets (A,B) (C), C is punished and B is rewarded for no reason. Similar injustice with (A,C) (B).
You have to take into consideration that NOT wanting to be with someone is to be prioritized over wanting to be with someone? At the expense of loneliness?
Apt and Yum plugins for btrfs snapshotting before updates have been around for more than 3 years. Come out from under the rock and smell some fresh air.
A bug in the code is worth two in the documentation.