5 ms·

It sounds less like they're being discouraged from using Python, and more like they're being encouraged to think critically about what sorts of projects Python

by iseff 17y ago

It sounds less like they're being discouraged from using Python, and more like they're being encouraged to think critically about what sorts of projects Python would excel at. Put another way, they're being encouraged to use the right tool for the job.

amichail 17y ago
What's wrong with prototyping a service using Python and if it takes off port it to a better language for scalability?
- 8plot 17y ago
  The problem, as pointed out in the article/thread, they want products that will _immediately_ scale if the need arises.
- lacker 17y ago
  The problem is that at Google, "taking off" is often equivalent to "launching". When Google launches stuff, it can have millions of users on day 1.
  - aristus 17y ago
    This. Working for a company with a large audience means that your problem often is no longer getting people to pay attention. Your problem is getting it right, at scale, in multiple languages and locales. This can be alleviated somewhat with internal tests, invite-only alphas, bucket testing, and "labs" features.
    - rbanffy 17y ago
      That and good monitoring. I always try to launch services with very detailed server monitoring - I want to know how much memory is being used with what, how much I/O and how much time the CPUs spend doing non-application stuff. I want to monitor response times, queue and dataset sizes and anything that helps me say if we will need more servers, different servers or what parts of the application we should port to amd64 assembly.
      - voxio 17y ago
        Out of interest.. How do you do that?
        rbanffy 17y ago
        Munin and some custom plug-ins. It does not give me all the data I would like, but we found a bug the other day by looking at some graphs and how one related to the rest of them.
  - blasdel 17y ago
    Amazon has an interesting approach to this for extant products: A/B testing for infrastructure. Just as they test UI tweaks by funneling percentages of live customers through them and comparing conversions, they also send live traffic to both production and development servers (with the user only seeing the production responses), checking to see if they produce the same output after refactoring.

by iseff 17y ago

amichail 17y ago

What's wrong with prototyping a service using Python and if it takes off port it to a better language for scalability?

8plot 17y ago
The problem, as pointed out in the article/thread, they want products that will _immediately_ scale if the need arises.
lacker 17y ago
The problem is that at Google, "taking off" is often equivalent to "launching". When Google launches stuff, it can have millions of users on day 1.
- aristus 17y ago
  This. Working for a company with a large audience means that your problem often is no longer getting people to pay attention. Your problem is getting it right, at scale, in multiple languages and locales. This can be alleviated somewhat with internal tests, invite-only alphas, bucket testing, and "labs" features.
  - rbanffy 17y ago
    That and good monitoring. I always try to launch services with very detailed server monitoring - I want to know how much memory is being used with what, how much I/O and how much time the CPUs spend doing non-application stuff. I want to monitor response times, queue and dataset sizes and anything that helps me say if we will need more servers, different servers or what parts of the application we should port to amd64 assembly.
    - voxio 17y ago
      Out of interest.. How do you do that?
      - rbanffy 17y ago
        Munin and some custom plug-ins. It does not give me all the data I would like, but we found a bug the other day by looking at some graphs and how one related to the rest of them.
- blasdel 17y ago
  Amazon has an interesting approach to this for extant products: A/B testing for infrastructure. Just as they test UI tweaks by funneling percentages of live customers through them and comparing conversions, they also send live traffic to both production and development servers (with the user only seeing the production responses), checking to see if they produce the same output after refactoring.