Two Slashes

For every useful piece of technology, we’re left with some crock of crap that somehow manages to become a reigning champion and hang around for a while with no improvement. No, as much as I’d love to, I’m not actually talking about Microsoft Office today. This little blurb is about something much more common than that.

It’s on the very hard drive you’re probably using now, and chances are it’s probably alerted you to some problem at one point or another. Now, if that problem ever really existed, nobody knows, but you probably took the warning’s advice and replaced the drive pretty quick. Still not catching on? I’m talking about S.M.A.R.T. – the “drive health monitoring system” that’s been standard in every drive manufactured in the past several years.

Standardization is good. It’s why we can swap things around and still expect them to work without a lot of jerry-rigging or modification. It’s also the cause of plenty of things ranging from why Windows is the dominant operating system of this point in time to why the power connector on most internal devices is interchangable.

Still – why S.M.A.R.T.? Well, that’s a simple answer. The very “technology” designed to monitor the health of your hard drive and alert you when it might be dying has a very poor track record with me, and it’s about to get worse. Every S.M.A.R.T.-enabled drive I’ve used that has died has done so without so much as a single S.M.A.R.T. alert. That’s right, nothing but the dreaded “click of death” to tell you your data’s on its way to the curb with your useless hunk of metal. And in my case, that’s not exactly something to be celebrating or claiming that the drives in question failed suddenly or were abused – not at numbers like mine. These are drives that have essentially been babied from the get-go. And interestingly enough…the ones I get warnings on? False alarms. Yup, they keep churning away, while the drives that don’t give so much as a warning die off every once in a while.

Now, it’s been a while since I’ve had a drive fail…maybe a few months. (Though the fact that I’m now away from 90% of my equipment and unable to mess with it helps that longetivity score just a bit.) But I brought a desktop and my laptop with me to school.

My usage habits might raise some doubt in why I’m posting this, but it’s interesting all the same. My laptop, for instance, usually rides around with me in an unprotected backpack alongside my schoolbooks, sometimes being squeezed, pressed, crushed, dropped, smacked, banged, hit…in short, it takes quite a beating just being carried with me. I know I could protect it more, but it’s a cheap machine, it’s long outlived the lifespan I had expected for it, and in general it just goes to show that treatment isn’t always the primary reason for failure. Not to mention that I’ve pulled the thing apart several times, flung it around onto beds, tables, desks, chairs, and any other matter of instant-computing surface…and it works. For three years now it’s gone through treatment like that, and here it is, still churning away and letting me write up my post. In fact, the only new hardware the thing has seen is a stick of RAM. That’s it.

My “school” desktop, on the other hand, is a different story. For the most part, the machine’s been coddled like one would coddle a newborn. At least, since I got some of the parts. (I got them from a friend; long story, but they still work and they saved me a bit of cash.) The hard drive in there’s been mine since day one though. And even after suffering a nasty PSU incident about a year into its life (it’s now about 4-5 years old), it’s served me well. After that aforementioned incident, the drive developed a small clicking problem. And despite the clicking, I continued to use it. I mean, what’s the point of status monitoring if it tells you the drive is in better condition post-trauma than it is beforehand? Anyway, it managed to make it into my school machine (at which point the clicking had subsided). It’s been between a month and a month and a half since that…and the clicking’s back, but it’s no *click* this time. It’s almost more of a screech. In the signs of hard drive health, it’s not a good sign when the thing is screeching.

Anyway, even I can tell by now that the thing is about to kick the bucket…yet where’s the “monitoring” that was supposed to have alerted me to the problem even before it occured? I wouldn’t expect it to warn me about my PSU blowing up, but it certainly should be able to predict the drive’s death based on the fact that it spends more time attempting to perform disk activity than actually doing it.

Now, the concept of drives monitoring their health is novel, and bloody useful, especially in a corporate world, and if it weren’t for the problem of flash-based internal media coming to the consumer world in the next few years, I’d be wondering more why there haven’t been any recent announcements concerning a technology that may very well be on its way to redundancy. But based upon its standardization and wide acceptance, and (lack of) monitoring prowess, I’m left to wonder why nobody ever bothered to try and improve it considering it fails in every way, shape, and form imaginable.

So, this brings me back to my original point. Why include something that doesn’t work, and charge the consumer to implement it? Certainly there can be a middleground where the technology either is not implemented or actually functions well enough that the data on the drive can be backed up well before any audible clicking can be heard? It’s not smart, and it’s not fair to extort consumers for random “features” on that bulleted list that don’t work remotely near as advertised.  I realize that drive death prediction is really hard to do…but seriously…can it actually predict something?