That's been my experience too. Google discontinuation notices come up on YC regularly, and I've heard of maybe half the products. The other half I've heard bad things about.
Except Google Code Search. I liked Google Code Search. But there's no useful way to hang ads on it. They could have used it for recruiting. People doing searches for code in difficult areas might be worth talking to. Google has tried that. At one time, if you searched for "proof of correctness", you got a Google recruiting ad. I was once contacted because I posted something in comp.lang.c++. (Unfortunately the person who contacted me was so clueless I told him to put me on their do-not-call list.)
Google Code Search: 2006-2013.[1] It was from the era when Google claimed their mission was to organize all the world's information, and they had book-scanning and newspaper-scanning projects. But that content generated no ad revenue.
Of course, many of us have stopped promoting new Google services because we've seen the fate of too many. Among other things, we don't want to have Google's increasingly bad reputation for this rub off on us.
Is that really any different that promoting a startup's product? Somebody has to try new things and shut down a large percentage of them. I'm surprised HN is so against it. Perhaps google should distinguish services that have long term commitments from those that don't?
I think your second sentence has got it: Google is not a startup, and it should distinguish between startup like experiments and things they are making a long term commitment to. And even then G+ shows they won't be perfect, but they can at least try.
I don't think they feel the need. Perhaps Google believes if it can't gain network effect without advertising, it's not a product that can ever attain Minimum Viable Ubiquity.
I think the last time I'd actually heard of the product in question was Google Reader. But that goes for virtually all of the "X is shutting down" stories I see on HN.
