5 ms·

I concur. But then I tried to classify Google projects in 3 categories and find for "obviously useless things that only exist so engineers can have fun", and I

by tajen 10y ago

I concur. But then I tried to classify Google projects in 3 categories and find for "obviously useless things that only exist so engineers can have fun", and I didn't find so many obvious things: - Obviously productive ones: Gmail, Google Analytics, Android, Android TV, Maps, Youtube, OpenSocial, Chrome, because they all participate to "universally trackable users", - Skeptical projects: Fiber, Flights, Google Scholar, TensorFlow, Chrome OS, GDrive, GApps for Business, - Pet projects that may be fun but don't seem to serve as leverage: GCE (yes, their cloud solution), Google+, Go, Google Earth, Google Play/Music, Google Wallet, Google Glass, GGroups, GFinance, GTranslate, Google Fonts, Google Classroom, Inbox, Wave, Hangouts, Google News, Google Currents, Google Keep, Google Now, Wave... Therefore... If they wanted to stay as lean as possible, would they have come up with the right list straight away? What makes products "obviously" part of the first list? [1] https://en.wikipedia.org/wiki/List_of_Google_products https://en.wikipedia.org/wiki/List_of_Google_products

dx034 10y ago
Even the skeptical ones are probably productive: - Flights will likely make money, it operates on a comission basis. It's just not big compared to other projects - TensorFlow is part of the machine learning programme that is used heavily to make more money in other areas. Sharing knowledge helps to get talent in that area - Chrome OS is like Android for laptops. It ensures that you use Chrome and Google. The adaption rate is not huge, but I could imagine that it still pays off
Eridrus 10y ago
> - Skeptical projects: TensorFlow I have to disagree. TF is Google's internal ML system, so the waste here is minimal, on top of that the luck they had in adoption means that lots of people they hire are already quite familiar with their internal systems and their internal system is now very common for deep learning research, and plenty of people are writing systems compatible with their internal tech.

by tajen 10y ago

dx034 10y ago

Even the skeptical ones are probably productive: - Flights will likely make money, it operates on a comission basis. It's just not big compared to other projects - TensorFlow is part of the machine learning programme that is used heavily to make more money in other areas. Sharing knowledge helps to get talent in that area - Chrome OS is like Android for laptops. It ensures that you use Chrome and Google. The adaption rate is not huge, but I could imagine that it still pays off

Eridrus 10y ago

> - Skeptical projects: TensorFlow I have to disagree. TF is Google's internal ML system, so the waste here is minimal, on top of that the luck they had in adoption means that lots of people they hire are already quite familiar with their internal systems and their internal system is now very common for deep learning research, and plenty of people are writing systems compatible with their internal tech.