The answer is called "archives." It's different from backups. I'm working on a script to use xorriso to write only my changed files to BD-R[E], after an initial full write of all my important data (self-created data, financial records, important email dirs, all amounting to only 4-5 GB).
I can even run this several times per hour when doing high value work, such as electronics design/embedded software engineering. The overhead is small, a few MB per session, just to write out a few changed files and a new version of the directory tree. What the disk looks like is a list of directories in the root named according to date+time, plus one named "current." I can swap disks ever day or so, then get interleaved archives, so that even if one disk goes bad, I loose at most one day.
If anything unexpected happened to my files, like being changed without authorization, then I'd see an unexpectedly large write being prepared. I could stop there, and I'd still have everything since the last run.
I strive to make archiving and backups take so little effort, that they just become a habit. There are many strategies one can apply. This was just one example.
If you are astute you should know that Windows crash reports have been intercepted off the net and used to reveal vulnerabilities about particular machines.
A computer should do nothing without the explicit command or permission of the owner.
No, I'm not that worried about M$ per se. And yes, I assume that the
The point is that when the OS is uncontrollably doing unknown things, then it is impossible for me to even estimate how well I am protected from the common computer crooks. Whether they be trying to attack me from the wire, or from inside M$, or by attacking M$, etc.
Don't even try to tell me that this is paranoia. My mother was already scammed by crooks who impersonated me after getting hold of the
The more data and control of machines holding that data is out of the direct control of the owner, the worse off we will be.
I'd rather have two of the $130 drives in RAID1, vs. the $264 drive. Then buy a cup of coffee.
I just experienced the first drive failure ever, since getting my first 20MB Winchester hard drive in the 80s. Fortunately it happened while testing before using as temp. storage to allow repartitioning another drive. Granted, my drives have only ever seen "desktop" workloads.
No, it's not about policing. The point is that airlines don't want to loose planes and have their customers get killed and the families sue them. They have an incentive to maintain effective security. The government doesn't. The government has an incentive to grow powerful.
The Feds absolved the airlines of liability for 911. What do you think would have happened if the airlines involved had instead been sued into liquidation? The other airlines would have learned a lesson! We'd have less inconvenience and better security than with the TSA.
Why would anyone trust the very entity that is largely responsible for the conditions which motivated the terrorists, and who dropped the ball on every clue that something was brewing beforehand? Yet, we must have the government protect us, because, well, it's the government! Our new god.
Our system of constraining debate to within the bounds permissible to the authorities is working perfectly once again. There seems to be almost no discussion of the fact that:
Drug tests, besides being an invasion of privacy and likely to produce data which will be abused, do not correlate well with *impairment*.
Whereas *impairment* correlates 100% with *impairment*.
I'm supposed to believe that drug testing, when there are about 100x more substances than tests, is the only thing we can possibly think of to do in order to evaluate impairment? When there are also countless causes of impairment besides drugs?
Or do we really not give a fuck about impairment at all, but just want more reasons to probe people up the ass until the common citizen is totally conditioned to accept random blood draws, anal probes, piss tests, by their employers, the cops, etc.? Then what? Cameras in our bedrooms to be sure we sleep enough, etc.?
Why aren't we talking about how to test for actual impairment, which would raise a lot less fuss, and actually have a chance of giving much more reliable information in cases where it actually matters?
As of next Tuesday, C will be flushed in favor of COBOL. Please update your programs.