4 ms·

LOL, you're assuming anyone gets promoted by fixing issues at Google. Wrong. You build some new shit and switch teams. Someone pitches a new project and builds

LOL, you're assuming anyone gets promoted by fixing issues at Google. Wrong. You build some new shit and switch teams. Someone pitches a new project and builds from scratch. Rinse and repeat. This particular issue happens (at least that's my recollection a few years ago) frequently enough at Google that there were memes about it internally. Not sure giving additional feedback would help.

hn_throwaway_99 6y ago
This is so glaringly apparent as a user. I think Google makes some fantastic products, but the number of "This has to be so exceedingly easy to fix, but Google just doesn't give a damn" issues I've hit is crazy. Here's one simple one: the number of volume steps in Android is only 16: https://issuetracker.google.com/issues/37035441 https://issuetracker.google.com/issues/37035441 . This means that the difference between off and the lowest level volume can be pretty big if you're trying to be quiet, e.g. avoid waking someone in the same room. I almost wish they would just close issues with a "We don't give a shit" resolution status.
- mehrdada 6y ago
  I don't think it is as simple as a singular person or a single room saying "I don't give a shit". It's a matter of prioritization. (In this particular case, even from a UX perspective alone, you could argue that increasing friction in sharing could net-net hurt the average user.) I don't think Google (if we can anthropomorphize it at all) is blind to this general phenomenon though--it's a trade-off you have to make about what company you want to be. Is it better for you to invest resources into incremental improvements to polish the final 2% or try again and maybe make a breakthrough. You have to look it from multiple angles too: are you going to attract the most talented people giving them such gruntwork or will they quit? Is there any meaningful incremental business you could get from polishing Android (remember, polished Android is more expensive and has to compete head-on with a company whose focus is to polish and is best set up to do so). It's complicated. There's a place for multiple models.
  - wolco 6y ago
    Who cares how much talent you have if nothing works. Perhaps they have too much talent and not enough meat and potatoes.
    - mehrdada 6y ago
      "Nothing works" is certainly not an accurate description of the reality (It's more like 80-90% works just great). As for who cares, your colleagues certainly do (there are limits on how much "potato and meat" you can mix with talent before messing the recipe up). Also note that the companies we are talking about here are not in the technology business, they are in the empire business, so they certainly care about who gets to help control the empire and who defects to a competitor.
    - qppo 6y ago
      I mean they're printing money from their ad services so something must be working
  - pavel_lishin 6y ago
    > Is it better for you to invest resources into incremental improvements to polish the final 2% or try again and maybe make a breakthrough. I would argue that maybe they should try Option 1, instead of sunsetting their Nth chat app, or migrating everyone from Google Music (which would very much benefit from some polish) onto Google Youtube Music (which I don't understand the value of for me as a listener.)
  - hn_throwaway_99 6y ago
    I could buy the "either/or" proposition with other companies, but not with Google. They have so much money and so many people they could easily (a) fix lots of these things that drive users nuts and (b) make a "breakthrough" (but as other commenters have pointed out, if their "breakthrough" is another chat app, maybe they should pull back). They quite simply don't incentivize teams to fix this shit.
    - mehrdada 6y ago
- discordance 6y ago
  iOS also has 16 volume steps
29athrowaway 6y ago
Many organizations work like this. People get credit for orphaned projects they no longer work on. Then, more responsible developers have to waste their lives maintaining code that was designed to be orphaned from the start.
UncleMeat 6y ago
Googler here promoted twice by taking over a team/project, simplifying it, and making it hum. This is a common narrative about Google. There is some truth to it but the blanket "nobody gives a shit about anything other than launches" claim is bogus.
- mehrdada 6y ago
  Yes, obviously there's hyperbole in that statement, but there's no denying the fundamental incentive structure is at heart of most of polish issues of this nature; that said, as I argue in my followup comments, I personally don't see the bias towards launches is a bad choice for the organization at large, especially since personal craftsmanship from some individual employees and their own values more often than not covers up the shortcoming in corporate incentives.

hn_throwaway_99 6y ago

This is so glaringly apparent as a user. I think Google makes some fantastic products, but the number of "This has to be so exceedingly easy to fix, but Google just doesn't give a damn" issues I've hit is crazy. Here's one simple one: the number of volume steps in Android is only 16: https://issuetracker.google.com/issues/37035441 https://issuetracker.google.com/issues/37035441 . This means that the difference between off and the lowest level volume can be pretty big if you're trying to be quiet, e.g. avoid waking someone in the same room. I almost wish they would just close issues with a "We don't give a shit" resolution status.

mehrdada 6y ago
I don't think it is as simple as a singular person or a single room saying "I don't give a shit". It's a matter of prioritization. (In this particular case, even from a UX perspective alone, you could argue that increasing friction in sharing could net-net hurt the average user.) I don't think Google (if we can anthropomorphize it at all) is blind to this general phenomenon though--it's a trade-off you have to make about what company you want to be. Is it better for you to invest resources into incremental improvements to polish the final 2% or try again and maybe make a breakthrough. You have to look it from multiple angles too: are you going to attract the most talented people giving them such gruntwork or will they quit? Is there any meaningful incremental business you could get from polishing Android (remember, polished Android is more expensive and has to compete head-on with a company whose focus is to polish and is best set up to do so). It's complicated. There's a place for multiple models.
- wolco 6y ago
  Who cares how much talent you have if nothing works. Perhaps they have too much talent and not enough meat and potatoes.
  - mehrdada 6y ago
    "Nothing works" is certainly not an accurate description of the reality (It's more like 80-90% works just great). As for who cares, your colleagues certainly do (there are limits on how much "potato and meat" you can mix with talent before messing the recipe up). Also note that the companies we are talking about here are not in the technology business, they are in the empire business, so they certainly care about who gets to help control the empire and who defects to a competitor.
  - qppo 6y ago
    I mean they're printing money from their ad services so something must be working
- pavel_lishin 6y ago
  > Is it better for you to invest resources into incremental improvements to polish the final 2% or try again and maybe make a breakthrough. I would argue that maybe they should try Option 1, instead of sunsetting their Nth chat app, or migrating everyone from Google Music (which would very much benefit from some polish) onto Google Youtube Music (which I don't understand the value of for me as a listener.)
- hn_throwaway_99 6y ago
  I could buy the "either/or" proposition with other companies, but not with Google. They have so much money and so many people they could easily (a) fix lots of these things that drive users nuts and (b) make a "breakthrough" (but as other commenters have pointed out, if their "breakthrough" is another chat app, maybe they should pull back). They quite simply don't incentivize teams to fix this shit.
  - mehrdada 6y ago
discordance 6y ago
iOS also has 16 volume steps

29athrowaway 6y ago

Many organizations work like this. People get credit for orphaned projects they no longer work on. Then, more responsible developers have to waste their lives maintaining code that was designed to be orphaned from the start.

UncleMeat 6y ago

Googler here promoted twice by taking over a team/project, simplifying it, and making it hum. This is a common narrative about Google. There is some truth to it but the blanket "nobody gives a shit about anything other than launches" claim is bogus.

mehrdada 6y ago
Yes, obviously there's hyperbole in that statement, but there's no denying the fundamental incentive structure is at heart of most of polish issues of this nature; that said, as I argue in my followup comments, I personally don't see the bias towards launches is a bad choice for the organization at large, especially since personal craftsmanship from some individual employees and their own values more often than not covers up the shortcoming in corporate incentives.