4 ms·

Nothing wrong with it. People make errors, distributed systems aren't easy. More frequent changes - more likely a bug was introduced. My post is just a specula

Nothing wrong with it. People make errors, distributed systems aren't easy. More frequent changes - more likely a bug was introduced. My post is just a speculation, let's wait for the actual technical details doc from Google.

oblio 5y ago
Well, there are various whitepapers that reformulate this exact truth you've highlighted: issues tend to happen more when changes are made. Regarding the technical details doc, Google will never state that outright in individual postmortems. And they will definitely not draw this to the logical conclusion regarding the spiky yearly activity.
- yunohn 5y ago
  > logical conclusion regarding the spiky yearly activity Why is it logical? There’s tons of changes being deployed at all times, at all large companies. Across products, verticals, everything - hundreds of changes at any given point. Some of these changes can introduce hard-to-predict bugs in globally distributed systems. Most of the time, external users don’t even notice before they’re fixed. Like another commenter said, performance reviews do not coincide with with the year-end at Google and other companies.
  - oblio 5y ago
    > Like another commenter said, performance reviews do not coincide with with the year-end at Google and other companies. When are they? My bet is that they're roughly around the same time period, even if they don't start or close on December 31st.
  - michaelt 5y ago
    Hypothetically speaking? 11 months a year there's no incentive to cut corners and only do 2 weeks of testing on something that really needs 3. If one month a year rushing things out with a bit less testing is rewarded, I can believe some people would respond to that. Of course, I wouldn't go as far as to call this a "logical conclusion" as the evidence I've seen is very slim.
    - fragmede 5y ago
      Except that there are two a year at Google., so it's 10 months of the year, 5 and 5, and Covid's made everything weird.
- UncleMeat 5y ago
  It can’t be logical because it is based on bad facts. We wrote evals in August. Anybody racing to get things launched before perf did so months ago. The timing described in the top post is just false. We write the next one in February. We are almost exactly half way between perf cycles. Your hunch that this lines up with the end of the year is false.

oblio 5y ago

Well, there are various whitepapers that reformulate this exact truth you've highlighted: issues tend to happen more when changes are made. Regarding the technical details doc, Google will never state that outright in individual postmortems. And they will definitely not draw this to the logical conclusion regarding the spiky yearly activity.

yunohn 5y ago
> logical conclusion regarding the spiky yearly activity Why is it logical? There’s tons of changes being deployed at all times, at all large companies. Across products, verticals, everything - hundreds of changes at any given point. Some of these changes can introduce hard-to-predict bugs in globally distributed systems. Most of the time, external users don’t even notice before they’re fixed. Like another commenter said, performance reviews do not coincide with with the year-end at Google and other companies.
- oblio 5y ago
  > Like another commenter said, performance reviews do not coincide with with the year-end at Google and other companies. When are they? My bet is that they're roughly around the same time period, even if they don't start or close on December 31st.
- michaelt 5y ago
  Hypothetically speaking? 11 months a year there's no incentive to cut corners and only do 2 weeks of testing on something that really needs 3. If one month a year rushing things out with a bit less testing is rewarded, I can believe some people would respond to that. Of course, I wouldn't go as far as to call this a "logical conclusion" as the evidence I've seen is very slim.
  - fragmede 5y ago
    Except that there are two a year at Google., so it's 10 months of the year, 5 and 5, and Covid's made everything weird.
UncleMeat 5y ago
It can’t be logical because it is based on bad facts. We wrote evals in August. Anybody racing to get things launched before perf did so months ago. The timing described in the top post is just false. We write the next one in February. We are almost exactly half way between perf cycles. Your hunch that this lines up with the end of the year is false.