skj

72 ms·

871.
by skj 13y ago
When you say "You are wrong.", that indicates that you disagree. But the rest of your comment indicates that you agree. Could you clarify?
872.
by skj 13y ago
When Go was released, back in 2009, I decided I wanted to use it for my machine learning experimentation software (I was year 4 of a PhD program). There was no matrix library, so I created one: https://github.com/skelterjohn
873.
by skj 13y ago
Google's performance reviews are every 6 months. No bucketing. Well, you get to know if you are "needs improvement", "meets expectations", or "exceeds expectations", and you can pretend those are buckets i
874.
by skj 13y ago
Not so. Search is specifically sequestered.
875.
by skj 13y ago
You are correct for the search codebase - it is separate from the main tree, as is a lot of the core ad placement code. If you aren't working on those projects, you don't have access.
876.
by skj 13y ago
So you can have a maintainable codebase.
877.
by skj 13y ago
I think it's important for example code to also be correct. If it needs to gloss over some details, put them in a function call that isn't defined. Using an unprotected map here is incorrect.
878.
by skj 13y ago
There is trusting your employees and not feeling the need to scan everything they say to each other looking for boogiemen... and then there is not trusting employees who have access to your conversations to also not feel the need to scan ev
879.
by skj 13y ago
Being a Google engineer is pretty nice.
880.
by skj 13y ago
Thanks, my thoughts exactly. Coming from a machine learning background, I am often shocked at the quality of experiments in psychology and even economics. I think the reason may be because in machine learning, you describe your algorithm an
881.
by skj 13y ago
Sure. I was only addressing the concern of accessing source.
882.
by skj 13y ago
The source is not generally accessible from prod servers - only binaries and supporting data, and only the ones running on that computer. I guess it's possible you could find a computer that hosted both search and the codebase. But, si
883.
by skj 13y ago
But I heard that discontinuing unstaffed projects was evil?
884.
by skj 13y ago
Ruh roh - Ubuntu discontinued a service. Ubuntu, why are you being evil all of a sudden?
885.
by skj 13y ago
No, the action of donating money to helping the suppression.
886.
by skj 13y ago
Allowing users to set which of any possible 2^N subsets of these N types of permission would cause a lot of developer complication and errors, and people complaining about buggy apps, making the android platform "seem" buggy all t
887.
by skj 13y ago
Shareholders are kind of dumb.
888.
by skj 13y ago
Well, not quite - the 51% attack is not free. There is some value that makes the 51% attack too expensive, and game theory, yadayada.
889.
by skj 13y ago
It may be colluding, but the comment said it wasn't "about" colluding, which may also be true.
890.
by skj 13y ago
So, the advice given here is certainly practical and useful, but it's also wrong. Not wrong as in incorrect, but wrong as it's effectively steeling an opportunity from an individual (and you!) in order to appease a relationship wi
891.
by skj 13y ago
Downvoted for stating an irrelevant truth.
892.
by skj 13y ago
It really depends on what you mean. It is a private git repo, so in that sense yes.
893.
by skj 13y ago
This guy will probably avoid traveling to any country where people want to prosecute him.
894.
by skj 13y ago
I read schmoe as meaning "any old person with the inclination", which is not the case.
895.
by skj 13y ago
The private git repos rolled out a while ago. https://source.developers.google.com/p/PROJECT/r/default You might have to turn it on in your developer console, not sure. At the moment, pushing to its "mas
896.
by skj 13y ago
It's not possible to manage the data if no one has access. The SRE in question didn't "hack" into the data - he had access. The access was audited, so when it was abused, the SRE was fired with cause, and I think that th
897.
by skj 13y ago
That's correct. There are thousands and thousands of SREs, but gmail data is considered to be basically the most private thing in the data centers - its access is heavily gated and audited.
898.
by skj 13y ago
No, for improved security. The http/https question is completely irrelevant to the PRISM-complicity question. I'm not sure why you brought it up.
899.
by skj 13y ago
Not some schmoe. I am a Google engineer (and a bit schmoey) and there's just no way I could do it. There are people who can, though that number is small. However even then, the access is highly audited. The only way around it would be
900.
by skj 13y ago
Such access is highly audited in Google. That audit trail was used as evidence against the person in question here in a criminal trial. (That link isn't working, for whatever reason, so I'm just assuming it's the same case th

More ›