This post originated from an RSS feed registered with Web Buzz
by Mark Masterson.
Original Post: RIP, Google Reader. I'm glad you're dead.
Feed Title: Process Perfection
Feed URL: http://www.jroller.com/MasterMark/feed/entries/rss
Feed Description: Life, the Universe, workflow, BPM, Java, Ruby, functional/generative/meta programming, pi calculus, patterns, the Grid, agents, software architecture and the kitchen sink. :)
<rant>I gotta say... <tl:dr>All you people whinging about the death of Google Reader are behaving like idiots. You're wrong, and the error you're making is the single most important problem in software architecture and design</tl:dr>
Look, in the real world, things die. This turns out to actually be quite important. We've all seen the sci-fi movie where humanity figures out how to be immortal: it's always really, really bad news.
The problem is that when a species achieves some form of immortality, evolution stops. Dead in its tracks. Because death is a big part of the engine that keeps it going.
When you buy a physical product, you don't expect it to last forever: indeed, you know well that it can't. It's absolutely correct to whinge about plannedobsolescence -- those infernal products that somehow manage to break one day after their warranty expires are the work of Satan. But we have no expectation that physical things be immortal.
Interestingly, in the few cases where they effectively are, we either fail to notice, or are puzzled by it. A few weeks ago, a friend came out of the loo, still rubbing her hands dry, and said "It's weird that toilets don't evolve, isn't it? Think about it: they've been the same basic solution for how many hundreds of years now?" I cocked my head at her and said, "Why is that weird? How often have you had a toilet die (ie. break irreparably)? As a species, the individuals are more or less immortal. Of course they never change."
But when the product is made out of thoughtstuff, when it's immaterial (ie. software), suddenly, expectations change. Suddenly, it's the norm that software should live forever (or at least as long as we do).
And that's doing -- and has been doing -- terrible harm to us all. The software that your insurance policy lives on is probably running on a mainframe, was written in COBOL, and has been alive for >50 years. Same thing (more or less) for the software that your bank account lives inside of, the software that runs your ATM, the software that makes the trains run, the airplanes fly, the power plant work, and countless other such minor, trivial examples.
And that sucks. That software sucks. It's like those terrible, deathless monsters in that sci-fi movie. *It's really bad.*
Of course, civilians often believe that the pace of software evolution is already too fast -- dizzying, overwhelming. That misapprehension is a relative one -- what civilians can't know is how much faster it could have been. If only software could die.
As it should.
It's kind of ironic, on some deep level, that I saw the news of G-Reader's death at more or less the same moment in time that I saw Harry Stamps obit (http://www.legacy.com/obituaries/sunherald/obituary.aspx?n=harry-stamps&pid=163538353&fhid=4025#fbLoggedOut) (HT: +Jessica Masterson ). The contrast could hardly be starker. Of course it's right and proper to grieve the death of a beloved thing. But if you think the solution to that is to prevent the death -- well, then you're well on your way to that sci-fi dystopia. And that's a bad place to live.
Indeed, my larger point is: we've allowed that dystopia to already happen, in the software ecosystem. And it's already a bad place to live: much, much worse than it could have been.
RIP, Google Reader. Many people obviously loved you. I'm glad you're dead.</rant>
Previous Topic
