This again. Not a week goes by without someone mentioning this. When a product is used by 0.001% of your user base you have to think about putting engineering r

This again. Not a week goes by without someone mentioning this. When a product is used by 0.001% of your user base you have to think about putting engineering resources towards that service. It just so happens that non-core products are often used by a vocal early adopter tech crowd like people on HN. Also 0.001% of Googles User base might still be 20,000 people. The other thing is Google's business practices. I always think of Google as hundreds or thousands of startups. Small agile team builds up a new service. Where are we within 3 to 5 years. Over a billion users yet? No? Ok scrap or integrate it somehow. One big problem of big slow boring corporate entities is that they always miss the next big thing that's gonna disrupt them. Google did it to many incumbents. So it makes sense that they might be worried to be next to be disrupted. Also I think it's important to differentiate between consumer and commercial offerings with SLA's. There is no difference between the service discontinuation policies in GCP vs AWS and Azure. And also, gaming is way to important to abandon. Plus they have first mover advantage now as opposed to G+.