Hacker News new | ask | show | jobs
by bitpush 2816 days ago
I'm sorry but this joke is getting really old.

HN loves to point fingers when company stands still. Heck, we made a teaching moment out of Kodak and others who failed to innovate and relied too much on their cashcows.

On the other hand, we proclaim nice-sounding ideas such as being aware of sunk cost fallacy and how everybody should say no more. We talk about being data-driven and dont go by what gut says. Beware of the vocal minority. Dont be afraid to pivot or let go. We tell others dont wait for your products to be perfect to launch. Get it out there, test the market and iterate.

HN has their biases (I can build that in a weekend!) but the recent discourse of "OMG! they are going to desecrate it" is very silly.

2 comments

>I'm sorry but this joke is getting really old. HN loves to point fingers when company stands still.

Joke? There's no need for the false dichotomy.

Google in particular has the organizational attention span of a goldfish on meth, and many of us have been burned one, two, or three times by their recurrent habits of setting something awesome up and then killing it off a short time later with no sane alternatives in place.

This reputation is entirely earned[1]. The safe thing to do if you don't want to sign up for a sudden migration headache in the future is assume that any new Google product is another symptom of their "throw everything at the wall and see what sticks" mentality, and that it won't last for more than a year or two.

[1]: https://en.wikipedia.org/wiki/List_of_Google_products#Discon...

(For bonus points, do a mental exercise and note how many of the "absorptions" lack the features of their standalone counterparts)

All of this would be resolved if Google called these projects experimental.

This is a significant distinction. It's also a massive deterrent. Who builds on experimental? Google isn't incentivized to do this. So, everything new Google puts out is hence, experimental.

It's not an old joke. You're just getting tired of it.

You're basically invalidating all of these engineers who believe in a thing called "trust," and undervaluing it with a statement like this.

They're calling it a "technical test", how is that not equivalent to experimental?