Over at samizdata, Perry Metzger (not De Havilland) has a few trenchant observations about stupid people who don’t use condoms when they have unprotected Internet intercourse, or something. (For those who don’t know him, Perry’s writing style is often blunt and dismissive, which is one of the reasons I enjoy reading his stuff. Go figure.) Read it all, including the comments, because a lot of what I say from here may be otherwise incomprehensible.
I’m not going to argue with Perry’s main point about the need to upgrade your computer’s software regularly. From a security standpoint, it makes sense to install the patches which cover the gaping holes in the thing. I also understand that the software companies don’t care to maintain elderly platforms, for the same reasons that Ford no longer maintains Model Ts. (The fact that software changes occur at an exponentially-quicker rate than automotive ones is just the nature of the beast.)
The problem, as noted the the Comments, is that system “upgrades” are not devoted exclusively to security patches anymore. Instead, all sorts of crap is included which at best causes irritating changes in functionality, and at worst undoes a lot of the learning and experience that one has accumulated. I understand why this occurs, but that doesn’t mean I’m at all happy about it. And so far, Microsoft has accommodated us Old Farts by including a “traditional” desktop view for all new Windows operating system versions, so I don’t have to memorize all the silly new pictograms in Windows 7 – infinity. (Note to MS: remove that feature and I’m gone.)
And this is the point. One of the commenters at samizdata made the excellent point that Microsoft (and all software developers, as far as I can see) are more interested in getting new customers, who would be more comfortable with apps, pictograms, symbols and what have you than they would be with the old icons or, gawd forbid, text (all those words and stuff? dude!). That’s stupid, for all sorts of reasons, and here’s why.
I might not be worth much to Microsoft as an existing customer right now; but there was a time when people like me — the early personal-computer adopters — helped build Microsoft into what it is today. When you have a person who like myself has been through all the hardware iterations of the PC, XT, AT, 386x all the way through to the current whatever-it-is-I’m-writing-on-right now, and has likewise been through all the software iterations of DOS 2.0 through Windows 7/8/not-9 [ahem] and 10; when you have a longtime customer group like that, then surely I, and all the countless millions of people like me, deserve just a little accommodation in the Grand Microsoft Marketing Plan? (Okay, you can stop laughing now.)
I know, everything these days falls into the “But what have you done for me lately?” category, but it’s still a basic truism of marketing that it’s ten times easier to get an existing customer to stay with your product than it is to lure a new one away from a competitor. But if you persist in changing your product so that it not only becomes a purely new-customer attractant, but also an existing-customer repellent, then I would suggest that someone in Marketing needs to go back to business school and/or get a swift kick in the teeth to adjust their thinking.
I know that it’s expensive and resource/time-consuming to maintain old products. Of course it is. But I would suggest that it’s also a lot easier than new product development — we old-timers don’t ask for much, because we’re used to working with, by today’s standards, relatively unsophisticated products.
Using the automotive industry one more time: Ford, GM and Chrysler have discovered that there is a huge market for nostalgia models such as the Dodge Charger/Challenger, Ford Shelby Mustang and the like. These new iterations of the venerable hot rods of yore have been improved, of course: better brakes, suspension and so on; they’re still simple and unsophisticated by modern whizzbang standards, but their manufacturers can’t make them quickly enough. Let’s go exotic: a new 2016 LaFerrari costs about $1.5 million; a 1966 275 GTB recently sold for $2.1 million. (I know, that’s mostly a factor of scarcity; at the same time, however, the 2016 model is a hundred times better than its 50-year-old counterpart — and still, someone was prepared to pay good money for it.)
Somewhere is all the above rambling is the seed of an idea for Microsoft. Or maybe, for someone not in Microsoft who can see a niche in the PC market which is similar to the automotive restoration market.
Or maybe I’m just an old fart “shaking his fist at heaven”, as Perry Metzger suggests. Still, I’m pretty sure I’m not the only one who feels this way — in fact, there may be more of us than of them. Software developers — or to be more accurate, software maintainers — might want to take a look at that.