rust-analyzer, the Rust LSP used in e.g. VSCode, can expand declarative and proc macros recursively.
it isn't too bad, although the fewer proc macros in a code base, the better. declarative macros are slightly easier to grok, but much easier to maintain and test. (i feel the same way about opaque codegen in other languages.)
I have a similar experience. Was drawn to Rust not because of performance or safety (although it's a big bonus), but because of the tooling and type system. Eventually, it does get easier. I do think that's a poor argument, kind of like a TV show that gets better in season 2. But I can't discount that it's been much nicer to maintain these tools compared to Python. Dependency version updates are much less scary due to actual type checking.
I don't really understand this argument, and it isn't the first time I've heard it. What problem other than name squatting does it solve?
How does a Java style com.foo.bar or Golang style URL help e.g. mitigate supply chain attacks? For Golang, if you search pkg.go.dev for "jwt" there's 8 packages named that. I'm not sure how they are sorted; it doesn't seem to be by import count. Yes, you can see the URL directly, but crates.io also shows the maintainers. Is "github.com/golang-jwt/jwt/v5" "better" than "golang.org/x/oauth2/jwt"? Hard to say at a glance.
On the flip side, there have been several instances where Cargo packages were started by an individual, but later moved to a team or adopted. The GitHub project may be transferred, but the name stays the same. This generally seems good.
I honestly can't quite see what the issue is, but I have been wrong many a time before.
It sure seems that things like concerts, festivals, theme parks, even movies are more expensive than e.g. 20 years ago - though I don't have inflation-adjusted data on hand. If true, people are willing to pay more for these kinds of "experiences", and I'd have to agree with parent. Although I'm not sure if it really is a generational thing.
This doesn't mean what Ticketmaster is doing is right, but they are hardly the only ones. E.g. cinemas or theme parks don't use them.
I’m specifically curious about how “experience are more valuable to generations today.” I don’t think there’s much difference between today’s generation and past generations in terms of their enjoyment of concerts, festivals, and other live events. Surely Michael Jackson fans in the 90s were as excited to see him as Taylor Swift fans are today, maybe even more so. Michael Jackson, the Stones, U2, Bruce Springsteen, heck, The Beatles: all of these acts’ had super-fans and their tickets had nearly inelastic demand. In their heydays, they could have charged any price and filled stadiums.
People in those days demanded “experiences” just as much as people today.
Wasn't alive in the 60s and don't have a US perspective only a UK/EU one. My parents did see the Stones in the day and I distinctly remember them struggling to decide whether they should go see them in the 00s because tickets were 100€+ (seems cheap now!). But they are not "super-fans" I guess.
I also remember festivals getting more expensive 15-ish years ago, so my friends and me often decided to go hiking instead. IDK if that's a typical case but there are alternative "experiences". With the rise of social media I could understand there's additional incentive to go to more well-known events for the photos, over smaller live-music venues. At least it seems like a lot of photos are taken.
This is all anecdotes, so makes for an unsatisfying answer. But I think it's possible (certain) "experiences are more valuable to generations today". It's at least interesting to consider.
It’s probably true that there were fewer of the eye watering ticket prices that top stars demand today in many cases. Someone told me they were actually flying from the US to Brazil to see Taylor Swift because it was cheaper than getting a local ticket. My sense is highly in-demand concerts was more of a lottery than an auction in the past.
It's pretty crazy how quickly it happened/happens. The Deppenapostroph is maybe less problematic; I see it more as a simplification just like the dative replacing the genitive. But Denglish really just makes everything harder to understand; even if you are fluent in both English and German the "switching" is tiresome. Still, maybe we should get rid of "handy" and "beamer" first...
Ironically, even British English has the issue of Americanisms sneaking in, see e.g. the IT Crowd episode: "How hard is it to remember 911?" "You mean 999? That's the American one".
Had, until July 2011. This isn't unique. Austria, Denmark, Finland, Estonia, Greece, Norway, Sweden, Switzerland, and others have something similar, varying in details. (I think it's more commonly called military service or equivalent.)
Not just you. Best case the tracing ecosystem is weird and the documentation is lacking, but all of the weirdness makes it very performant. Worst case it is weird and badly documented and over-engineered.
Since rustc uses tracing I really hope it's the first...
This is incorrect, it's "for limited periods of time under certain circumstances" [0] (and citizenship is not required).
> they can always keep their current insurance through COBRA. [...] needs to pay the full premium
This is true but vastly underplays the cost. It isn't unusual for "the full premium" to be over a thousand dollars per month [1] (!). They even get to charge you an "administrative fee". Oh, and there's also tax implications.
> For people who don't want to stay on their previous employer's plan, the loss of the job is a "qualifying event" which allows them to immediately do things like join their spouse's plan or sign up for an individual "Obamacare" plan through healthcare.gov.
More likely for people who can't stay on their previous plan because it would bankrupt them. Or if their spouse is not working. What happens if the employee has children/dependents on the plan? Etc, etc.
None of this is easy. Switching insurance changes which doctors are "in network", so at minimum it's significant hassle. At worst, their or a family member's current treatment is only partially or not at all covered.
So, in practice, parent's description is accurate enough.
it isn't too bad, although the fewer proc macros in a code base, the better. declarative macros are slightly easier to grok, but much easier to maintain and test. (i feel the same way about opaque codegen in other languages.)
reply