romed

46 ms·

151.
by romed 8y ago
Maybe for some people it's a game? I have no idea. The pay isn't as progressive as their recruiters claim. A high-performing Senior SWE makes money pretty close to what a newly-promoted Staff SWE makes.
152.
by romed 8y ago
How about: don't believe in yourself because luck is everything? I walked into Google with 15 years of industry experience. I lucked out being in the early days of the web, lucked out in the dot-com boom, lucked out with a referral t
153.
by romed 8y ago
I don't get hdrhistogram at all. It's really just histogram with a shitload of logarithmic buckets.
154.
by romed 8y ago
Very nice. Your numbers are consistent with my experience, and the presentation is very useful.
155.
by romed 8y ago
The most shocking thing in this release is that amazon internally uses Twiki. Happy hacking, black hats!
156.
by romed 8y ago
I'm pretty sure glassdoor is a conspiracy by FAANG executives to confuse their employees. If you believe these numbers you will not be able to get what you deserve when negotiating comp at these companies.
157.
by romed 8y ago
Congratulations on making Staff after only 2 years. You’d think it would be alarming to Google how often engineers leave immediately after being promoted.
158.
by romed 8y ago
Exactly. Now some other slob with a job has to pay to feed my family just because I can afford to buy a house and he can’t. The American way. The only practical problem with this scheme is the cost of private medical insurance is astronomic
159.
by romed 8y ago
Yeah, oncall bonus.
160.
by romed 8y ago
Retired after 8 years as a Staff SWE at Google. Do I really need a why for retired?
161.
by romed 8y ago
At google there are SRE-SWEs and SRE-SysEng. An SRE-SWE has the same responsibilities as any SWE, but is expected to know SRE stuff. A SysEng is not expected to SWE as much, but they do both write software. The main difference between SWE a
162.
by romed 8y ago
Proto doesn't really have a schema. You can encode a proto of type X and tell your program to decode it as type Y, and it might work. Might not, but might. You can easily imagine messages that are isomorphic on the wire. E.g. messa
163.
by romed 8y ago
The machine does not think about category theory. The machine thinks about numbers. It can add them together! The way the machine thinks about numbers has not meaningfully changed in 40+ years.
164.
by romed 8y ago
Which language are you talking about? C++ proto is built around CodedInputStream which is what it sounds like. You can use it with or without the generated message code. https://developers.google.com/protocol-buffers/d
165.
by romed 8y ago
protos are not necessary for grpc. You can use whatever payload format you want to use.
166.
by romed 8y ago
Sure that makes sense if you can write off C++ as a "legacy" language, which is fine if you just can't get your mind wrapped around the scale at which Google operates. Due to the nature of weighted averages, you can't ju
167.
by romed 8y ago
Scalar primitive fields in proto1/proto2 C++ generated code are not pointers. They are just class member fields returned by accessors (foo()) and their presence is indicated by a presence vector (has_foo()). At least that's what
168.
by romed 8y ago
It's abundantly clear why this author lasted only a year at Google. Aside from using the non-idiomatic "protobuffers" ... "nothing wants to inspect only some bits of a message and then forward it on unchanged" In m
169.
by romed 8y ago
A host country can request a lawful intercept regardless of whatever technical conditions exist on your network.
170.
by romed 8y ago
At the last census the Bay Area was 50.4% female and SF in particular was 49.3% female, so no this does not reflect overall gender ratios.
171.
by romed 8y ago
What's Boulder's deal?
172.
by romed 8y ago
Awesome. I hope SFMTA can learn from this. They are taking proposals to replace Nextbus currently [1]. When I last measured Nextbus reliability, the first prediction for a bus was weakly correlated with arrival, and the second and third pre
173.
by romed 8y ago
This article is wrong and reflects what appears to be a widely-held belief (practically canon here on HN) that a thread has a fixed-size stack. This is not the case on every operating system! Linux, which I've heard is popular, initi
174.
by romed 8y ago
But Facebook is and did take a huge bite out of the total ad market. Facebook has roughly 1/3rd as much ad revenue as Google on a far smaller investment of infrastructure and a narrower range of products.
175.
by romed 8y ago
Yep, the degraded read cost increases a lot! But it only matters if you are losing disks at a rate similar to the time it takes you to do the 1PB of i/o, and also depends on how often you might need to serve from the data in question.
176.
by romed 8y ago
Now multiply by the chance of getting into the collision in the first place.
177.
by romed 8y ago
I guess I would expect something more like massively distributed erasure encodings rather than ZFS directly. For example something like Dropbox’s magic pocket but taken to extremes with, say, 100 data stripes and 20 parity stripes, all on d
178.
by romed 8y ago
Do you imagine that Apple has finite engineering resources? The last I heard was it is the richest company on earth. They’re just satisfied with the status quo in which Google does all of their security work and nobody cares because of deca
179.
by romed 8y ago
It is interesting but be aware that expressing these figures as a fraction of listings necessarily biases the result to weak markets, where the properties are on the market longer and the inventories are greater. In "strong" mark
180.
by romed 8y ago
That sounds like a monitoring problem and nothing to do with microservices vs. monoliths. If you have 10000 replicas of your monolith, how do you know what version they are running? If you can answer that question then you can also answer

More ›