Infirmware

In days of yore, when men were real men, women were real women, and small furry creatures from Alpha Centauri were, well, a twinkling in Douglas Adams’ eye as he lay drunk in a field in Innsbruck, there was no such thing as firmware upgrades over the Internet.

Once you got your hands on a piece of software (or, more to the point here, to a piece of hardware), it worked as you got it from that moment onwards until you consigned it to the junk heap, and if you had any trouble with it, well, you argued with the manufacturer until you got another, which was slow, expensive and fiddly, especially when gadgetry started having more programmable logic than actual moving parts.

But it was by around that time that manufacturers started splitting into two quite clear camps: those who seized the opportunity to better support their customers by actually fixing issues customers found in the field that they could not reproduce in internal testing, and those who realized they could simply start skimping on internal testing and ship a fix later¹.

That second camp tainted firmware upgrades as the mark of the lazy, of the rushed, of the technically unfit, but, worse still, they created a whole (and, to my eyes, largely worse) new occupation – product obsolescence management, i.e., the folk who decide what will be fixed with firmware upgrades until the next product hardware revision ships.

I have, over the years, been exposed to entirely too many negative examples of these – for instance, every phone I’ve ever tested in the past few years has gone through at least four firmware versions: the one it is sent out with for initial testing, the one we get “all” our critical bugs fixed, the one where they’re really fixed, and, quite often, another, public one six months down the line where I can actually trust the things to work (any geek out there who’s bought a top-tier Nokia device in the past year or so can relate to this, and it’s not hard to find examples).

A few of those devices² reached users in little better shape than they were in beta, but what really ticks me off is when firmware upgrades break perfectly good stuff (and yes, I’m finally getting to the point of this post, bear with me).

As you may well know, Sony has recently made an epic foul-up with their 3.0/3.01 firmware upgrade for the PS3, which has, from what I’ve been able to read, rendered hundreds (thousands?) of older, pre-Slim devices unable to read discs.

Guess what, my PS3 broke with the 3.01 firmware upgrade as well – but, as always, that’s not the whole story.

At the time that happened, I had just sold it to a friend, who wanted it specifically due to its backward compatibility with the PS2, but also due to the fact that I used it so little it was practically new. Thanks to Sony, he upgraded a perfectly good piece of hardware to 3.01 and lost the ability to read any sort of disks – the drive just died, and a thorough exploration of the reset and service options failed to revive it.

Fortunately enough the console still had had two weeks left on its warranty period, so I undid the deal, called the local support line for an RMA number, and sent it in. A week later, the Sony support line called back and told me they not only had run out of 60GB replacement hardware, but that they had no expectations of ever getting others again – so they offered me a brand new 120GB Slim instead.

As a customer service experience, it was great – it was blindingly fast by Portuguese standards (and well above the kind of hardware support I’d get from most other companies these days). The only problem is that my friend’s chance of getting a console able to play both PS2 and PS3 games is utterly gone³.

Oh, and it nearly compensated for Sony’s lack of QA when releasing that firmware. But I won’t flog a dead horse, even if it’s trivial to do so digitally.

The thing is, even the device whose firmware I upgrade most often – i.e., the iPhone – isn’t immune to this. Not by a long shot, for although I haven’t yet had any problems myself, with every OS revision comes a new set of niggles and complaints – where I personally include the failure to fix latent issues, like the infamous “snapback slider” when trying to answer some calls that I’ve experienced since 1.x.

Such things are why I was leery of (after over a decade) buying something as simple as a new TV – all I really need is an LCD panel with the requisite amount of HDMI and SCART inputs. I don’t even need it to have a tuner, since what little TV I watch comes from an IPTV box or a media/disc player of some sort.

And yet, the likes of Samsung, Philips and LG (all of whose TV portfolios I got to know at length over the past few months) are stuffing a bazillion features into their panels that are, for the most case, upgradeable by sticking a USB stick into the requisite port.

I’m not sure why I would want a TV that runs Linux (for that is what most of them are running to some extent), but one of the reasons I made sure I didn’t get one with an “Ethernet”:Wikipedia:Ethernet port was that I wanted to make sure there was no way for it to break in the same way my PS3 did.

Still, it could be worse. At least my kettle doesn’t take firmware upgrades…

—

¹ My personal belief here is that that bunch spent a stint in the software industry around the time Windows Update got started, but that’s besides the point right now.

² Which shall, for my part, remain nameless at least for a few decades longer, until I can tell their tale by the fireside to fledgling young engineers, bright-eyed with enthusiasm at hearing first-hand tales of their equivalent of the “Tin Lizzie”:Wikipedia:Ford_Model_T…

³ Plus, of course, my having to decide what to do about the PS3 all over again. It’s hard to get rid of a brand new piece of hardware, even considering I’m just as likely to (not) use it for gaming as the previous model – the only arguments for keeping it around are my not having another Blu-Ray player and the dodgy status of its warranty, for I can’t really go and re-sell it as one would a regularly purchased used console…