I gotta be frank here: If you expect anyone in the tech world, Google or otherwise, to support a non-paid product past the point it's obviously declining in user base (IE obvious to not just internal users, but to external ones), for more than a few years, you are just not going to use anyone's products ;)

Google code is mostly irrelevant to this calculus, because it was not ever meant to make money for anyone, but i think you are a little harsh here.

Don't get me wrong, as a user, I would love to have the stuff i want to use supported forever. But it's unrealistic.

The problem is that Google represented itself as an altruistic supporter of open-source. People trusted it because it deliberately cultivated the impression of being more responsible than the default 'duh capitalism' mentality it now represents.

Google is suffering the reputational damage it deserves - not because it is now behaving any differently from other corporations but because it has betrayed the promise that it would.

> Google just needs to be up front about it (and maybe they are)

Nobody launches a new product and guarantees "If this doesn't work out, we'll keep it around indefinitely on life support". You don't start a new effort expecting failure. If it doesn't end up working out, it'll be closed down. Google's failures are just a lot more public than whatever side business many of us may take a stab at.

This isn't unique to Google. Failures need to be learned from and culled. We as a small business/startup oriented crowd should understand that more than most. I've had to kill off more projects than I'd like to admit because I didn't want to spend the time/money on what I perceived to be failures.

Not that failure is bad. Like Google, I've learned a ton from my unsuccessful projects.

In fact, many vendors, especially in the enterprise market and physical goods, stand behind what they sell through their products' lifetimes. As I mentioned elsewhere, IBM supported OS/2, a marketplace failure, for ~16 years. Buy a washer from Sears; even if that model doesn't sell well, they will support yours.

Consumer and 'free' software is the exception, and it's a recent exception with rapid release schedules. It's just shifting the costs from the vendor (maintaining old products) to the consumer (changing/upgrading).

> Consumer and 'free' software is the exception, and it's a recent exception with rapid release schedules. It's just shifting the costs from the vendor (maintaining old products) to the consumer (changing/upgrading).

It's important to note that in many cases, the vendor pays for what is otherwise a free service (Reader, Wave, G+). In the case of some services, they may gather some data from which they can profit from, but that doesn't always pay the bills.

So yes, the customer may have to switch services or upgrade, but unlike enterprise software, they aren't paying four to six figures to use said services.