Bandwidth is getting better more quickly than latency is. TCP slow start requires a round trip to speed up, and if you want that to improve, you need to reduce latency. Increasing bandwidth won’t help.
2,413 karma · joined July 3, 2015
Context-sensitive buttons are less awesome. They might be superior, if you memorize the combination, but they're decidedly inferior to touchscreens for the long tail of infrequently used features.
The study linked here focused on "simple" tasks, and the top performing cars probably have dedicated buttons for everything measured. The story would likely be very different for context sensitive buttons, and subjective reporting would likely be unreliable, if studies on the speed of mousing vs keyboarding are anything to go on [1].
[1]: https://ux.stackexchange.com/questions/30682/are-there-any-r...
I don't think "just stop picking a side regarding abortion, gun ownership, or gay marriage" is a reasonable solution. These are political wedge issues, but they are also legal questions with answers that can affect your daily life. Of course you want it to go a certain way!
I claim writing the name first is much more readable. Compare
integer five = 5
five: integer = 5This one makes no sense. Wouldn't 99.9% of power supply glitches be some sort of accident, and something that the end user probably doesn't want?
The managers who want remote attestation aren't the people implementing it. They either pay someone else to do it, or they pay someone else to do it. The difference between paying a third-party company and an employee is that employees are more expensive, because the costs aren't amortized over other customers who want the same stuff. Why would they be more trustworthy? Why would they be better at it? Why would it be any less likely to be hacked if you did it at your company than if you outsourced it?
Most vaccines don't provide sterilizing immunity [1] [2], but what's a little double standard compared to political clout, right?
[1]: https://www.theatlantic.com/science/archive/2021/09/steriliz...
[2]: https://www.pbs.org/newshour/health/can-people-vaccinated-ag...
But this one is definitely wealth inequality. They're probably making over half their revenue on "whales" that make up only 10% of their audience, but they don't want to ignore the rest of their audience because (1) they need to continuously convert casuals into whales or their audience will eventually just die off (2) word-of-mouth marketing is the most effective form of marketing, and there just aren't enough whales out there to keep something like MLB going all on their own.
As for what's wrong with `Type name(constructor, args)`? A lot of tooling wants to be able to parse "mostly-valid C++", like IDEs and compiler diagnostics. Sure, once clang's type inference is finished, the lexer hack and most vexing parse aren't problems, but when the program isn't complete, parsing isolated fragments is impossible, and that limits the amount of useful tooling the language can have.
2. Inversed cause and effect. Even if you aren't Christian, there's a good chance that the morals you were taught as a child are influenced by it. The Christian Bible is literally the world's best selling book.
Hacker News is almost entirely influencer marketing, and has been from the very beginning. The whole point is to push Y Combinator the angel investment company.
Lisp ain't gonna help with any of that.
I don't think that's actually true. Mining earns money through transaction fees and the mining block reward.
The block reward is directly set by an algorithm which lowers the amount over time. Transaction volume is totally irrelevant.
Miners and users set the transaction fee by choosing what they'll accept, so it should respond directly to supply and demand. If the Jevons Paradox [1] holds, making transactions cheaper means that there will be more of them, so even though on-chain Bitcoin transactions now account for less than 100% of transactions denominated in BTC, that doesn't necessarily mean that there will be fewer total on-chain Bitcoin transactions. So transaction fees paid to miners might not actually go down, either.
The vast majority of energy consumed by the bitcoin network is in mining, and that's not impacted by any of this.
The downside with hired testers is that they are unlikely to be a statistically representative sample of your target market. I don’t think this is actually a problem for most startups or new products being launched by major players, since a lot of UX issues early in product development will be obvious to anyone who isn’t the original developers, and the number of actual users you have won’t be big enough to serve as a statistically representative sample of your target market anyway.
It totally does, though. If I write an emoji thumbsup, and the recipient only sees a tofu box, they won't know if I'm agreeing, disagreeing, or saying something else.
If I'm participating in a heterogeneous environment, I need to refrain from including any mission-critical information in a form that can't be read by all of the participants. And since writing the same information twice is usually too much work, the extended functionality winds up completely unused, or at least only gets used for low-value spam.
> Or do you worry that the person you are talking to on slack might just be connected via matterircd via IRC
Yes, I do worry about that. If I find out that one of my coworkers is using a Slack client that doesn't implement all the functionality in the reference client, and can't just talk them out of it, then I'll have to make sure not to use anything that they can't see.
I might even start doing an IRC bridge myself, in fact, since at that point there's really no actual purpose in running a chat client with a bunch of functionality that I cannot use.
A major part of Slack's value is that this is very rare. Almost everyone just uses the reference implementation.