You think they are publishing all these papers and open-sourcing TensorFlow for the good of the community solely?
No... if the entire ML community publishes papers with TensorFlow as code, then now Google can sell that many more TPUs and that much more access to their cloud systems to run the models...
I was shocked they gave away the K8s brand. Now that was ballsy and surprised necessary.
Because with Borg they only had an advantage with their own products' operations. That's nice, in that they are able to maintain a decent amount of technical superiority and have their product reliability be leading or tied for first, but it's not a clear line between this and them generating more revenue. They want their cloud offering to compete with AWS and Azure, which is why they want K8s to gain mindshare. GCE has huge potential for growth and significant switching costs; meaning they need to get new products on it sooner rather than later, and developer mindshare is the way to do that.
They were tired of writing something up as a paper and then someone implemented as OSS and then companies and engineers knew the OSS implementation instead of their own.
So instead do a OSS version and then drive what everyone learns.
I found this a bit funny. One solution would have been to not write the papers. But so glad Google does. Spend a lot of time on their research site.
The original papers on Map/Reduce, Bigtable, Dremel, Dapper, Spanner, directly spawned a bunch of successful open source implementation projects as well as companies.
They are giving away everything they learned with Borg and you somehow take it as a negative? Really?
Do think from a business standpoint not sure why Google would give up the K8s brand but it is good for everyone.
I really worry that Google will stop opens sourcing and more important sharing so much through papers when people b*tch so much.
Starting to be hard to see why they even do it any longer. Who else have a huge competitive advantage with something like Borg and just gives it away to the industry?
But I am someone super curious and just probably spend more time on
https://research.google.com/pubs/papers.html
Read most of the paper. But then what they presented at NIPS. My favorite paper right now is
https://arxiv.org/abs/1712.01208
Bigger example is how much they give away in terms of AI/ML.