concede_pluto

Everyone has a test environment; some people happen to send prod traffic to it.

$x == $y should guarantee that $a[$x] == $a[$y]. In PHP, it doesn't. That's not different, that's just broken, and the language design makes dozens of elementary mistakes like that where getting it right wouldn't have sacrificed the only strength (streamlined deployable server integration) in any way.

"Welcoming" is not a good thing. The worst thing you can do with a bunch of clueless people is make them prolific at writing code they don't realize is unreliable. We should be teaching the minority who have the aptitude and turning away the rest.

If your published code is bad I can quickly learn whether I'm wasting my time, but if it's good I still need to see you code in person. Neither GitHub nor homework proves you did the work.

I once got a free "upgrade" to economy plus. They seated me at a bulkhead and (because no under-seat space) required my carry-on go in the bins where I couldn't see whether my company laptop was being stolen.

If instead I could get elbow room and still have something to read, that might be worthwhile.

The ad business requires Google to hire a hell of a lot of developers who are not working on the search engine. A pay service would be viable on less revenue. And nobody knows who many people are actually willing to pay for Google search because we've never had to.

I'd use Emacs and the same X11 apps I'd use on a Linux cluster. There are X servers for Windows, though WSL strangely omits one.

If you like Windows apps, why run WSL?

"Cool URIs don't change." Link rot is the biggest threat to the World-Wide Web. In my view it's irresponsible to issue a URL without being confident I can keep it available at least as long as the document exists. I don't have that kind of confidence in the government of Libya or Tonga decades from now.

I'm also not a big fan of redirections that cross failure domains, just because they make the web slower and less reliable.

The goal of an init system is to start daemons. It shouldn't have any opinions about which daemons you run, and it has no business doing random tampering like intercepting logins and DNS requests and renaming device nodes. Breaking "nohup foo & logout" is shockingly inappropriate; end users should never ever care how an admin may have booted the system much less have to know how to interact with it!

They should have created an annotation or something saying "this file uses the python3 dialect", made python3 run all python2 code unmodified (the only way to not require both being on $PATH), and halted python2 releases except critical security bugs.

Because python3 and python2 are incompatible dialects, you have to say which one you want, and there's no reliable way to use just "python" and no useful reason to have it on $PATH. Any code needs to declare all its dependencies, including either python3 or python2. If I didn't want any specific dependencies, I'd be targeting POSIX and writing in Bourne shell.

I seem to remember reading that NASA would use short sacrificial extension cables so that ground tests would certify the jacks they'd be flying but not wear them out.

It is when you know you can't persuade them you have a good-enough reason to need it, so instead you don't even try.

It's worth noting that the options for hosting snopes.com were much more limited in 1994.

perma.cc? Wow. Thank you for not trying to establish a permanent namespace in a vanity ccTLD with a tenuous relationship to some privately-held company.

Old buildings stand until they decay or we run out of space nearby and demolish them. HTML doesn't decay and we can't run out of URLs; web pages only disappear because our hosting infrastructure needlessly demands ongoing payment from (only!) the author.

I say "needlessly" because the actual cost of storage, CPU, and networks are dropping so quickly that hosting a page forever should asymptotically approach a fixed cost that isn't much more than the first few years.

"Don't hire the average developer" tops the list of best practices.

https://godoc.org/golang.org/x/sync/syncmap leaps to mind as a type whose users should demand and get static type safety, but it can't happen until the language definition adds yet another special case blessing it as a properly supported collection type.

And if I don't buy a Fiat 500e because it only has a range of 84 miles, do you think "you can go different places!" would be a persuasive argument?

It's rare and depressing to see a page so fatally infected with javascript overengineering that not even Google's cache managed to rescue the content.

http://webcache.googleusercontent.com/search?q=cache:u3lFxVi...

Ouch. Is Xwayland still the only way to do remoting? https://wayland.freedesktop.org/faq.html#heading_toc_j_8 just punts on it and speculates about what could theoretically happen someday, but I don't even know how or why I'd start using a backend that can only run on one computer.

Liquidity is willingness to make a trade others aren't. To be useful, it has to linger in the order book for a long time. If you're winning a race by milliseconds, you are trying to interpose yourself into a trade that was already going to happen that day.

Laziness, impatience, and hubris.

  # zcat > /dev/kmem

When we had a frontier, we had a surplus of habitable land and resources and immigration was a quick way to make the most use of it. That's over now, and we're spreading the same wealth among more and more people.

An migrant worker creates less demand than a native doing the same job, because they're trying to live austerely and remit as much of their income as they possibly can out of our economy and back to their family's. They aren't building out water and power homesteading our uninhabitable scrub, they're living ten to an apartment near the few jobs.

Real estate has been zero sum ever since we ran out of frontier. Employment is zero sum when immigrants (of any race) try to add a lot of labor supply and minimal demand to an economy that already had a severe job shortage.

Google is infamously bad at resolving or even acknowledging customer service problems.

I don't mean to disparage twitch games, that's my go-to example of one of the last domains where it's cost-effective to get really good at living within customers' hardware constraints. But when prod is a growing distributed system, it's natural for tests to assume the same distributed system, and forcing those tests to sort-of run on a single box with the wrong kernel/fs/network config just trades in cheap hardware for expensive engineers doing work that doesn't make prod better.

When I write java, I can't run maven in prod and expect it to get native libraries into /usr/lib64 and the sysadmins' Python and Go plumbing and config files into ... wherever the hell that may live. So together we tweak a .spec file that not only provisions the entire machine correctly but answers questions about whether the entire machine is provisioned correctly, not just the java half. (We probably could make maven do all that, but the result would be worse in every conceivable way, and in most languages it's not even an option.)

If you want a rolling release distro, use one. Godspeed. But nobody's going to sell a support contract that covers random alpha builds published overnight. "We can't upgrade the distro and get the code we need, so we're smuggling in code that the distro doesn't trust yet" is just devs and sysadmins playing chicken over the fate of the project. The bleeding-edge vs supported argument should have been settled before going live.

> brown shoes in town

What's the story here? All I know about dress shoes is that they're hideously uncomfortable and generally not blue suede.

I never met any of our professors of journalism or dance. Everyone at college is paying for a bundle of credentials and access to experts in many unrelated fields, and nobody uses all of it.

MIME was using "application" in media type names in 1996. I think macOS may have started it because they needed to distinguish them from desk accessories.