4 ms·

No, subscriptions introduce the perverse incentive to release unfinished products and slowly drip-feed fixes. Having to test extensively before a one-and-only r

by bsanr2 6y ago

No, subscriptions introduce the perverse incentive to release unfinished products and slowly drip-feed fixes. Having to test extensively before a one-and-only release may not drive nearly as much revenue, nor provide a running deliverable stream for a given engineer/product manager's CV, but it is clearly the better user experience.

leetcrew 6y ago
you say this, but given a choice between a "finished" product and a competitor with more bugs and more wanted/needed features, users will almost always pick the latter unless it catastrophically affects their workflow. engineers care about quality; customers care about the fastest way of solving their problem.
- _qulr 6y ago
  > engineers care about quality; customers care about the fastest way of solving their problem Almost all customers care about quality. The problem is that many customers have only very limited information about products, so they have a hard time judging quality vs. competitors before (or even after) purchase. This reveals a general problem with the market: it doesn't select for quality. Otherwise we wouldn't be having this conversation. The market is really good at producing cheap crap. So the truth is, yes, engineers have to care about quality. The motivation for quality has to come from pride in your own work, not from outside market forces. If you care about quality, then you have to strive for that over quantity, and also charge sustainable prices instead of trying to lowball. You may not be the market leader, but there are many profitable niches. Some customers are definitely willing to pay for quality.
  - leetcrew 6y ago
    > Almost all customers care about quality. yes, but not to the exclusion of features. I work on a B2B product where our customers bill their customers by the hour, so they tend to have a pretty good idea of how much time a feature saves them. if a competitor adds a feature that cuts the time needed for a project in half (not unrealistic) but crashes and forces them to start over a quarter of the time, the customers will still buy their product instead of ours. they'll complain incessantly on the competitor's forums about the crashes and threaten to switch back to our product, but they won't actually do it unless we come up with something new that saves them even more time. customers care about saving time and/or money; they only care about quality to the extent that it furthers that fundamental goal. if there is some bug-ridden alternative that solves their problem faster, they will do their best to find it and purchase it. edit: to be clear, I mean "reliability" when I say "quality"; their are many other "qualities" a product can have, one of them being "cheap".
    - _qulr 6y ago
      There's a big difference between not caring and making a tradeoff. You can care about A and B but decide B is more important than A. But if you don't care about A, then there's no tradeoff, you just choose B no matter what. The question is, why do we make consumers make that tradeoff? Why are we shipping junk at all? There shouldn't be a reliability tradeoff. All products should be reliable. It ought to be a bare minimum standard.
      - leetcrew 6y ago
        > There's a big difference between not caring and making a tradeoff. You can care about A and B but decide B is more important than A. But if you don't care about A, then there's no tradeoff, you just choose B no matter what. fair enough, I probably overstated my point with some of the wording. > The question is, why do we make consumers make that tradeoff? Why are we shipping junk at all? There shouldn't be a reliability tradeoff. All products should be reliable. It ought to be a bare minimum standard. everything in life is a tradeoff. we could make software more reliable, but then we would have to spend less time adding new features, or we would have to hire more/better engineers and charge more. maybe we even get to pay down tech debt and gain the ability to add features faster in the long run. doesn't really matter if someone else dumps a bunch of buggy new features in the meantime, converts our customers, and forces us out of business. in the absence of some industry-wide gentleman's agreement or regulation, we have to observe the behavior of customers and do what their behavior (not words!) indicates they want.
        _qulr 6y ago
        > doesn't really matter if someone else dumps a bunch of buggy new features in the meantime, converts our customers, and forces us out of business This is always presented as the doomsday scenario, but how often does it actually happen? The story of Apple in the Tim Cook era is unrelenting annual releases, more and more "subscriptions", massive return of cash to AAPL shareholders, but decreasing product quality. Did Apple make that tradeoff because they were scared of going out of business? No, they were doing very well before Cook took over. Cook simply has lower standards than Jobs, there's no other reason. He's been great for investors, not so great for customers.
- bsanr2 6y ago
  You state that "unless" as if it weren't generally the common state of "competitor with more bugs." No, users pick a finished product.

by bsanr2 6y ago

leetcrew 6y ago

you say this, but given a choice between a "finished" product and a competitor with more bugs and more wanted/needed features, users will almost always pick the latter unless it catastrophically affects their workflow. engineers care about quality; customers care about the fastest way of solving their problem.

_qulr 6y ago
> engineers care about quality; customers care about the fastest way of solving their problem Almost all customers care about quality. The problem is that many customers have only very limited information about products, so they have a hard time judging quality vs. competitors before (or even after) purchase. This reveals a general problem with the market: it doesn't select for quality. Otherwise we wouldn't be having this conversation. The market is really good at producing cheap crap. So the truth is, yes, engineers have to care about quality. The motivation for quality has to come from pride in your own work, not from outside market forces. If you care about quality, then you have to strive for that over quantity, and also charge sustainable prices instead of trying to lowball. You may not be the market leader, but there are many profitable niches. Some customers are definitely willing to pay for quality.
- leetcrew 6y ago
  > Almost all customers care about quality. yes, but not to the exclusion of features. I work on a B2B product where our customers bill their customers by the hour, so they tend to have a pretty good idea of how much time a feature saves them. if a competitor adds a feature that cuts the time needed for a project in half (not unrealistic) but crashes and forces them to start over a quarter of the time, the customers will still buy their product instead of ours. they'll complain incessantly on the competitor's forums about the crashes and threaten to switch back to our product, but they won't actually do it unless we come up with something new that saves them even more time. customers care about saving time and/or money; they only care about quality to the extent that it furthers that fundamental goal. if there is some bug-ridden alternative that solves their problem faster, they will do their best to find it and purchase it. edit: to be clear, I mean "reliability" when I say "quality"; their are many other "qualities" a product can have, one of them being "cheap".
  - _qulr 6y ago
    There's a big difference between not caring and making a tradeoff. You can care about A and B but decide B is more important than A. But if you don't care about A, then there's no tradeoff, you just choose B no matter what. The question is, why do we make consumers make that tradeoff? Why are we shipping junk at all? There shouldn't be a reliability tradeoff. All products should be reliable. It ought to be a bare minimum standard.
    - leetcrew 6y ago
      > There's a big difference between not caring and making a tradeoff. You can care about A and B but decide B is more important than A. But if you don't care about A, then there's no tradeoff, you just choose B no matter what. fair enough, I probably overstated my point with some of the wording. > The question is, why do we make consumers make that tradeoff? Why are we shipping junk at all? There shouldn't be a reliability tradeoff. All products should be reliable. It ought to be a bare minimum standard. everything in life is a tradeoff. we could make software more reliable, but then we would have to spend less time adding new features, or we would have to hire more/better engineers and charge more. maybe we even get to pay down tech debt and gain the ability to add features faster in the long run. doesn't really matter if someone else dumps a bunch of buggy new features in the meantime, converts our customers, and forces us out of business. in the absence of some industry-wide gentleman's agreement or regulation, we have to observe the behavior of customers and do what their behavior (not words!) indicates they want.
      - _qulr 6y ago
        > doesn't really matter if someone else dumps a bunch of buggy new features in the meantime, converts our customers, and forces us out of business This is always presented as the doomsday scenario, but how often does it actually happen? The story of Apple in the Tim Cook era is unrelenting annual releases, more and more "subscriptions", massive return of cash to AAPL shareholders, but decreasing product quality. Did Apple make that tradeoff because they were scared of going out of business? No, they were doing very well before Cook took over. Cook simply has lower standards than Jobs, there's no other reason. He's been great for investors, not so great for customers.
bsanr2 6y ago
You state that "unless" as if it weren't generally the common state of "competitor with more bugs." No, users pick a finished product.