Users want software that is fast, but they will put up with sluggish software if it has the features they need. Faster hardware means it takes less programmer effort to meet the user's performance needs; the surplus effort can then be allocated towards more/better features.
-
-
Show this thread
-
Users also want software that is bug-free, but they will put up with some bugs. Automated testing makes it much easier to write bug-free software, but we don't -- instead, we use it achieve the same degree of bug-free-ness as before, and allocate the surplus towards features.
Show this thread -
Two broader connections come to mind. First is that many people predicted that automation would grant us vastly more leisure time. Instead, we work roughly the same number of hours, and just produce more value per hour.
Show this thread -
Second, when cars became commonplace, people assumed they would shorten commute times. Instead, people just started living further away from the office (and from each other).
Show this thread -
In other words, the effect of a technology depends less on what people *want*, and more on what they will *put up with*. We will put up with a 1hr commute, 8hrs of work, 1 bug per KLOC (or whatever), and 100ms of latency. Anything better than that is, in some sense, wasteful!
Show this thread -
And this has been part of our thinking around Sia from the start: if we make storage 10x cheaper, people won't store the same amount of data for 1/10th the cost. They'll store 10x more data and pay the same. What new applications will that enable? I'm very curious to find out.
Show this thread
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.