I am! It's my first HTMX project but I am liking it so far. HTMX has a few quirks which it took me a while to understand, but since then it has worked pretty well -- the amount of HTMX "code" I've had to write is quite small compared to the amount of JS I'd normally have to write.
My main takeaways have been:
- It mostly just works and simplifies things, but:
- Working with page state (in my case, which sheet is loaded) in individual requests (like updating a cell) is a bit awkward -- you can either supply it via `hx-vals` or via setting a global hidden input which is specified via `hx-include`. I ended up going with the second approach due to https://github.com/bigskysoftware/htmx/issues/1119 but it's not perfect -- in particular you need to duplicate specifying the input if you want to include additional inputs for some node.
- Error handling is also a bit awkward, but it's not too bad once you get used to the idea that you should return HTML elements in 400s. I found https://htmx.org/extensions/response-targets/ helpful as well.
- Often HTML structure doesn't allow you to limit your request scope as much as you'd like. For me the big thing is I can't just replace one column since tables are row-major -- I have to replace the whole table if a column is modified.
As I mentioned in a sibling comment, the contention I intend to make is that he agrees more with the racial views of his NRx commenters than he publicly let on. That was the framing of the NYT article -- nobody thinks or would be incensed to learn that SA is a Neo-Monarchist or any of the other NRx beliefs that are orthogonal to modern US political discourse.
If you don't think racial IQ disparities are a significant part of NRx thought, fine, you're probably better positioned to know and I am happy to concede the point. In that case, a better comparison would be to Steve Sailor's views. I only mentioned NRx because that is the context of his emails, and I had not recognized that he was linking to Steve Sailor's blog.
If your contention is that we should more frankly discuss IQ disparities, pretending we're talking about height was a strange way to go about it.
In those emails, SA does not say he thinks these questions deserve more study -- he says they're "probably partially true". Again, in that context he's talking about Steve Sailer's views.
Yes, I'm focusing on a few sentences. Do you think he wrote those by accident? That the words came out wrong and the straightforward reading was not his intent? In the context of the rest of his emails, and his writing on e.g. Albion's Seed, I do not think that is likely.
This is obtuse. I didn't make any argument about SA reading the same stuff NRx read, or even reading NRx stuff. I linked SA's own writing that "Many of their insights seem important" and that their views have "nuggets of absolute gold".
> Human biodiversity is an empirical fact. Are some populations taller than other, on average? Is red hair more prevalent in certain populations? Pretty much nobody actually rejects the idea of human biodiversity.
> The neo reactionary types tend to draw specific conclusions from the idea of human biodiversity, like that racial disparities in IQ are inherent[...]
This is a classic motte-and-bailey. The IQ view is clearly the view that he is endorsing in the emails -- he is linking Steve Sailer's blog under "HBD is probably partially correct", and he even demands the recipient "NEVER TELL ANYONE I SAID THIS" -- obviously he's not talking about height.
I guess I am much less familiar with Moldbug's stated views than you, so I'm happy to concede that point.
Anyway, I think the HBD/NRx relationship is peripheral to the discussion of the Metz article -- I just framed it that way because that's the context of Scott's emails (certainly he think's they're at least associated!). The discourse around the Metz article was that it framed Scott as holding racist views, not specifically NRx views.
If I had mentioned Steve Sailer instead of NRx, would you agree that Scott was more sympathetic to those views than he publicly let on?
If you view some races as generally "less useful" than others, most people would consider that racist regardless of whether you think they should enjoy the same rights.
I spent a lot of time on SSC from 2013 to 2017, and my recollection is that the Venn diagram of commenters promoting HBD and those promoting RDx was nearly a circle. And similarly for related sites like LW. So in that context I would say the two are closely related.
I find it hard to believe you've been exposed to much NRx content if you don't think they consider HBD true and very important. Although don't take that as a criticism -- I would not recommend wading through their beliefs and look back on it as a waste of my time.
I said he was "sympathetic to NRx views" and linked emails from him stating certain views widely held by Moldbug et. al. that he is sympathetic to. Do you disagree that "HBD" is an NRx view, or that the leaked emails express sympathy for it?
In them he says "HBD is probably partially correct" -- do you think that's not a NRx view (I agree others also hold it, but I would argue that it's a core view of NRx -- it doesn't need to be exclusive to them, e.g. Islam considers Jesus a prophet too), or that he publicly held that position previously?
I don't mean to say that he was a closet NRx -- and IMO neither did Metz imply that -- just that he agreed with a much larger subset of their controversial views than he or his allies admitted publicly. I would consider that being "sympathetic" to NRx.
While I don't agree with all of the framing in the NYT story, it's worth noting that Scott Alexander was much friendlier to Moldbug et. al. than the anti-Metz camp argued at the time: https://twitter.com/ArsonAtDennys/status/1362153191102677001
I think Metz's motivation for that framing was his assumption that if SA was not at least sympathetic to NRx views, he would not allow them to be such a big voice in his comments and in his community. You can argue about the reasonableness of this assumption, but he did turn out to be right.
The link you cite says it was worse in the Rust version:
> During the audit, it came to light that the original sudo implementation was also affected by [CLN-001: relative path traversal vulnerability], although with a lower security severity due to their use of the openat function.
> You've adjusted wages for inflation, great, but that's meaningless unless you compare wages to what it costs to live.
That's literally the point of adjusting for inflation. You can of course pick the fastest growing components of the CPI (housing is already a huge component) and complain that we're not outpacing the fastest growing component, but that's arbitrary.
That chart looks at wages rather than income, e.g. excluding healthcare contributions. You can argue that we shouldn't count healthcare because it's gotten way more expensive for no real increase in quality (which is true!). But healthcare is a huge and rapidly growing component of CPI which we are using to deflate wages -- you can't have your cake and eat it too by excluding healthcare payments from compensation but including it in your deflator.
Also that chart just seems to be wrong (they don't actually show the source or data). You can do the comparison yourself easily enough -- take the average hourly wages (https://fred.stlouisfed.org/series/CES0500000003) and use FRED's graph editor to compare to a deflator (e.g. CPI -- https://fred.stlouisfed.org/series/CPIAUCSL). This is up ~10% since 2008. It's not massive but it's not nothing -- and remember the caveat above that this is an underestimate.
And re: switching jobs for low-wage workers -- either they are getting big raises or getting big pay increases from switching jobs, or the BLS is massively wrong. Which is it?
Apples to apples here -- the 2020 GT Premium was $43,315. So we're still not looking at nearly the same size of increase.
You think the average income is being computed incorrectly? Or that there is just a lot of variance so many people are worse off? It is true that wage increases have been uneven, but they are actually much larger at the low end -- this is why you see the most inflation in low-wage service-heavy categories like restaurants.
Unfortunately I could not find data on prices after discounts -- I would love if you could point me to some! I find it hard to believe the average discount in 2019/20 was $8k off on a base trim GT, but if you actually bought one with that discount then fair enough. Note that's still a significantly smaller increase than my parent comment indicated.
My main takeaways have been:
- It mostly just works and simplifies things, but:
- Working with page state (in my case, which sheet is loaded) in individual requests (like updating a cell) is a bit awkward -- you can either supply it via `hx-vals` or via setting a global hidden input which is specified via `hx-include`. I ended up going with the second approach due to https://github.com/bigskysoftware/htmx/issues/1119 but it's not perfect -- in particular you need to duplicate specifying the input if you want to include additional inputs for some node.
- Error handling is also a bit awkward, but it's not too bad once you get used to the idea that you should return HTML elements in 400s. I found https://htmx.org/extensions/response-targets/ helpful as well. - Often HTML structure doesn't allow you to limit your request scope as much as you'd like. For me the big thing is I can't just replace one column since tables are row-major -- I have to replace the whole table if a column is modified.