> would it truly be an Apple mouse if it didn't have at least one questionable design decision?
I don’t know if you consider a single button a questionable decision, but the ADB II mouse was as close to perfect as the technology was able to support when it was released.
their classic mice weren't horrible, but from the puck on the original imacs (so like 98ish?) and everything after has been absolutely horrible on ergonomics.
It's super clear that they don't view the traditional mouse as the intended interface to their system and expect you to use the trackpad.
Still quite common in NYC. There’s a bit of a social distinction between doorman buildings and elevator-man buildings, and some people prefer the latter.
A memory safe language is one for which (a) there is a subset of programs which can be statically proven to not perform unsafe memory operations at runtime and (b) no programs outside of this subset will be accepted. The set of operations that are considered to be unsafe can vary, but always includes writes to unowned memory, and often includes reads from unowned and/or uninitialized memory.
I make a habit of asking support whether things I’m buying are commingled. First-level support used to be able to tell me pretty quickly (and I got both “yes” and “no” answers), but these days I’ve had to escalate most of the time. Kind of nuts to me that it’s worth ten minutes of support time and twenty minutes of supervisor time for them to sell me a $3 usb cable, but they’re the ones paying, so…
> Is this limited to lockstep between softcores on a die - so good for low level error failures like soft error, but no good if the package dies? (Still very neatly done)
Depends on what you mean by "good for." The intent of lockstep is to convert essentially all undetectable errors to detectable errors, usually to allow fail-silent behavior, rather than to eliminate detectable errors. This property that all failures have defined failure modes is then used at the system level to build robust systems; for example downstream actuators can receive multiple command streams from multiple lockstep systems, and, relying on the invariant that a correctly received message came from a correctly operating system, can safely act on any of them, rather than needing to vote on the received messages. A package failure should be very unlikely to introduce an undetectable error in this context.
Sure, it was down 60%. But the real question is whether it outperformed Intel as a whole, and outperformed other internal investments Intel could make. I certainly wouldn't think that a 2015 dollar anywhere else within Intel is worth more than 40¢ today, given how they've been running.
> If you include the time spent arriving early, going through security, etc, it probably comes out about even.
Yeah, but that only lasts until someone figures out a clever way to use a train as a weapon, in which case you get to add the same security time to the front of the trip -- or even longer, since we've actually managed to get the airport security time down a bit over the past two decades. It would seem optimistic to try to scale usage of rail without accounting for the (time, safety, etc) costs that come with increased usage of rail...
Selling shoes that you purchase wholesale for $75 has costs that go beyond selling shoes that you purchase wholesale for $50. There's the cost of money to buy the inventory, the cost of holding the inventory (and insuring it), the cost of shrinkage, the risk of being unable to sell some of that inventory. Most of those costs scale with the wholesale cost of the product being sold, although not necessarily fully linearly. As a result, a top line $50 margin on a $75 product gains you less than a $50 margin on a $50 product -- in a world with cheap capital. If you're restricted to holding $N of inventory due to cost of capital, this becomes even worse -- not only are your bottom line margins going down as much as 15%, but you're able to do it on only two thirds as much inventory, which (depending on turnover rates, etc) can drive you even lower.
I don’t know if you consider a single button a questionable decision, but the ADB II mouse was as close to perfect as the technology was able to support when it was released.
reply