2) I think it’s important to recognize that one of the things models are good for is astroturfing, and any given conversation you see may be direct or secondary effects of that (among other marketing).

An argument can be as old as the search engine and hold real value. There are ways in which unreflective search engine use has misled and mistrained people.

There’s always been argument to be had about how we manage and offload attention, what we gain and what we lose when resistance is reduced. It’s part of reflection that’s been necessary in order to make progress solid ground, and is more necessary with non-deterministic tech.

The phrase “Tactical tornados” may be older than web search and describes people who also got a lot done.

Models can be incredibly helpful boosters and situationally effective subordinates… and also patchy as a real engineering IC or org.

(I’m team CSS standards although my knowledge is a bit tilted towards late 2010s, currently revisiting)

L2 issues are already involved in some way often revealing some kind of system failure, requiring context and exploration to understand, and judgement (and perhaps even system overrides) to fix.

I could see “automated L2 is the new L1” improvements, but without a big capability jump and/or a resource bonfire, I don’t think even frontier models would effectively replace good L2 staff.

They might magnify good L2 staff so fewer are needed (and maybe even help L1 staff become L2).

But not one about extra lines of code when it came to table layout.

And claiming non-table alternatives always needed polyfills and more code doesn’t sound like an accurate reflection of the time either. It sounds more like resentment of people who actually did invest in understanding of the domain because they might not just let you use the small toolset you knew without thinking about anything else.

And I say that as a person who did a lot of table-layout markup too.

A careful re-reading of my comment will reveal that I did mention rent as a factor in at least two places: one as an opportunity cost to be reckoned with for people following the PFI plan (with which my example still comes out looking good), one as a cost of living substantial enough for many working people that they do not have significant disposable income, which makes leveraging their largest living cost appealing.

> Interest rates are closer to 6.7% which means your $2500/mo doesn't even cover your principal and interest of $2600 which is to say nothing of PMI (which will be required since you didn't put 20% down), homeowner's insurance, HOA fees, or property taxes.

Using a 5% interest rate was one of several simplifying assumptions that I chose to be generous to the Buy Housing on a Loan plan.

You are correct that interest rates are presently and historically higher than that, and that mortgage insurance, homeowners insurance, property taxes, and some maintenance costs that under the BHL plan can add up to significant housing costs that aren't going to equity and therefore aren't well-leveraged. In other words, the BHL plan actually comes off worse than I made it look.

(If there's a counter side of that, it's that landlords can and will pass on those costs so they're reflected in rents... but sticklers will notice that landlords who are done with amortized costs or who financed at lower rates can choose not to do that and may have incentives to depending on the market.)

That's in absolute terms. There's a relative point too: the higher the interest rates, the more the field tilts towards the PFI. It magnifies debt/leverage, making that path more expensive, and it magnifies return from invested income, making that path more rewarding if you can swing it.

> And don't forget that other than insurance and taxes, your mortgage payment is capped for 30 years.

I did leave out the bounding effect that a mortgage can have, and that's arguably an important missing point.

Wy would someone do that? My observation is that incomes also tend to grow in rough parity to rents in many markets -- in fact, local income growth is probably the primary variable local rents are dependent on (at least in a functioning market). This means during prime earning years decades from retirement, rent changes might be an acceptable simplification. But you're probably right that the closer you get to retirement, the more important bounding costs is. And there might even be other situations where the tradeoff starts to make sense even for earners with significant disposable incomes.

I considered putting it up in the bullet points. Apparently deciding against that lost my expressions of this point to some readers, including yourself.

But yes, this is why the analysis after the bullet point mentions the profile of people who don't have $2500 disposable income. The leverage matters more to people in this situation.

Having seen this conversation play out more than a few times and even turn a tad fighty, I think this is the fault line:

* people who do this kind of analysis frequently and generally have high disposable income often see that they can leverage compound interest rather than pay it, so the Pure Financial Investment plan seems like a slam dunk to them, and for their profile they're probably right.

* people who generally don't have high disposable income see that they can use leverage to make their rent payment do double duty, which seems like a huge win for them, and for their profile they're probably right.

What I did leave out is how a mortgage can bound your living costs. Another commenter correctly pointed out rents can expand dramatically. Where incomes track rents, I don't think this makes a dramatic difference, and that's why I didn't include it, but it's true this isn't guaranteed, and mortgage can function pretty well as a hedge.

It's probably worth making a closer comparison though:

* Buying a House on Loan: commit to paying off a $450k loan over 30 years at 5% interest, with an immediate $50k down payment and the home itself as collateral. So ~$2500/mo payments, another 400k in interest by the time you're done. Your home probably appreciates by that much in most markets, which gives you a million dollar asset at the end. In some good markets, it may appreciate by 3-4 times, which would mean you have a 1.5-2 million dollar asset.

* Pure Financial Investment: put $50k into a fund, add sustained regular $2500/mo contributions. Let's imagine that the fund averages a conservative 5% annual return and we do this for 30 years. The outcome should be... a bit above 2 million dollars.

All investment involves risk and variable outcomes, but the BHL plan probably has a more varied outcome. Parity may be as common as substantial profit.

The PFI plan, on the other hand, performs really well even considering conservative 5% returns: over 2 million dollars (minus 400k you would have probably paid in rent). Bump it to 8% returns and we're looking at 3 million, a performance even many good real estate markets couldn't match.

Its major problem is that you need to be disciplined about putting the chunky contributions in, which means you need to consistently have rent-payment-level disposable income to make this work. Many working people don't.

Leverage lets housing costs go to equity and interest payments, which is key leverage for people who don't have disposable investment income. But less key for people who do.

Things have reached the point where I probably could use open sync+storage options to achieve what I do with Dropbox (and perhaps eventually I will do that as a hedge against the risks of Dropbox enshitification).

But I'd love to see Dropbox continue to provide worthy convenient service.

The lynchings? Part of the problem was that enough of the societies where they happened did accept those as part of the social order. And they were undisciplined in a way that allowed them to be esentially lotteries -- no requirement for any kind of hearing, no right to defense or appeal, no right for others to know how well any accusation stands up, no right for anyone to even know who the lynchers are. Supposed offenses could be trivial if not entirely fabricated, to cover nothing more than ugly bullying.

The contrast between the two situations could scarcely be starker.

And that's before we get to the ridiculous racial fearmongering that you're selling. You want to talk about 2018 murders? It doesn't look like you want to do that carefully and honesty. If anything the 2018 data shows white people as the disproportionate threat to white people.

Because of the 3,315 white murder victims where an alleged/established perp has been identified closely enough to talk about their race, it appears that 2,677 were killed by white people.

That's 80% of white murder victims killed by white people. This is above proportion of the population that's white at a level that's beyond noise (about 60% of the US is "non-hispanic white", maybe we move up to 75% if we're lumping in white-alone-identifying hispanics and assume that the criminal justice system also errs on the side of assigning perps "white").

514 murders where the perp is black would mean about 15% of white murder victims were killed by black people. Estimates of the portion of the US that's black run about 12-15%, so this tracks proportion.

2018 statistics make it look like, if anything, whites are in more danger from their fellow white people than from a racially integrated society (and this sure tracks the lived experience of most white guys I know).

And if someone says "well, of course white people are killed by more of their fellow white people because that's where they voluntarily associate" ... that wouldn't change the statistically demonstrated nature of the threat, it would only highlight how limited racial integration actually is (and how much more ridiculous it is to complain it "kills whites").

https://ucr.fbi.gov/crime-in-the-u.s/2018/crime-in-the-u.s.-...

Like "they didn't prepare correctly" or "they didn't do the right things" or "mondays amirite" there may even be cases where it's true, and a robust analysis of lottery situations sometimes reveals local maxima or tradeoffs that are tough to shake.

But they can also be spoken with a post-hoc resignation that discourages the very analysis that might confirm them... because such an analysis might also disaffirm them.

One question to ask is whether a way of addressing a "lottery" encourages you to stop analysis and reflection, or to work your way through analysis and reflection.

Giving examples of specific "lotteries" I see is just as likely to activate those psychological mechanisms I talked about (or a partisan frame) as it is to open anyone's eyes.

If you want hints, though, watch for where you see the psychological mechanisms in yourself or others. "Those people just didn't do the right things. They should have been more careful, more prepared, more like the people who didn't get stoned. They should have done it the right way. They should have known their place. And if it's their time, well, what are you gonna do, mondays amirite?"

If you hear someone saying something like that, if you find yourself saying it... interrogate that. There may sometimes be real truth to it. But ask yourself: is that really all there is to it? Does the world have to be that way? If it were your child who "drew the lot", would you be satisfied?

What are the lotteries you don't see, because you're used to them, and they're just part of how the world works, like this one is to the people in the story?

If you're looking, you can find them. But it's also as uncomfortable to find them in real life as it is to read the story. So, most of us are happy to keep some other ideas between us and these lotteries. Those people just didn't do the right things. They should have been more careful, more prepared, more like the people who didn't get stoned. They should have done it the right way. They should have known their place. And if it's their time, well, what are you gonna do, mondays amirite?

And if that's true, then you can be safe because you will do the right things. And nobody has to go to the bother of persuading a society with any changes at the margins on which it sacrifices random people.

In vector function space, no one can hear your eigen-scream.

It's hard because you continue to love pretending that people are calling you names even when they're not. It's hard because you find it convenient to demand people defend arguments they're not making.

My position is that your arguments are bad. I've made my arguments why. I don't have to argue about the policy, because I haven't staked a position on it.

I haven't called you names, so it's disingenuous to respond as if "Don't call people names" is on topic. For sake of contrast, saying something like "You're a dishonest piece of troll shit" might be calling you names; it's pretty clear most of my discussion has instead been focused on pointing out the problems with your arguments, positions, and rhetoric, which is part of why you're finding things so hard.

Though at this point your dishonesty is manifest to anyone following along.

Then something happened I probably should have died from (80-90% of people don't make it). High blood pressure turned out to be a contributing factor. High salt consumption much of it from my favorite cheeses turned out to contribute to the high blood pressure.

I learned to like swiss. And be modest in my consumption of tastier saltier cheeses. I no longer glibly tell people someday I'll be found dead of dairy poisoning with a smile.

I would probably request cheese in a known last meal situation though.

We should all enjoy things, and many of us can still stand to be more restrained in how we enjoy. And what.

I've spent a lot of words on it under the charitable assumption that you were truthful in your desire for better discussion. At this point, you've persuaded me that was overly charitable.

Instead it appears you're determined to maintain a fictional posture implying bad behavior in others whether or not it exists so you can claim whatever fruits of grievance you're here to harvest.

If the time comes when I care about defending or attacking the memo, I'll do so effectively. I haven't taken a position on the memo. I've only asked you to sustain positions you've taken or honestly retreat from them. And not pretend other people are taking positions that they aren't, which is apparently a big ask.

You could have done that in response to the person who asked you what reasons they had not considered. That would have been one reasonable way to engage, it would have mildly trespassed the bounding attempt in their statement that they had considered the memo, but it would have introduced substance backing up your claim and let you interrogate their claim that they had in fact evaluated the reasons given in the memo.

You didn't do that, though. For some reason you instead chose "I don't support the policy" which, as stated, is a non-sequitur in response to the question ""what reasons have I failed to consider?"

If you'd like to talk about the reasons in the memo or other reasons not in the memo, no one has been stopping you.

> > So saying "I don't support the policy" is a non-sequitur.

> No it isn't. It's not an argument. It's just a statement of fact.

A non-sequitur can be entirely factual. This means affirming something as a statement of fact is not an adequate defense against the charge of non-sequitur, it is actually a further non-sequitur.

> my argument here which is "don't just call people liars."

This may be the argument you intended to make. But because you also asserted that legitimate reasons for the policy existed and then refused to defend that assertion with a reasonable response to the question "what reasons have I failed to consider?" (including "the reasons in the memo" up until this level of the discussion) and also appear determined to avoid that quality of engagement, you've ended up engaging in a way that works against a general ethos of better discussion and the micro-dynamics which support it.

There's also the fact that it's reductive to assert anyone has "just" called the administration liars. Given that the federal judiciary has retreated from traditional presumption of regularity when fulfilling their judicial responsibility[0] (ie, that the executive is acting in good faith), that's compelling reason to believe the judiciary has found a pattern of admin dishonesty in the social/institutional setting where honesty is most critical. Retreating from the presumption of honesty in lay discussion is a pretty reasonable step. This in addition to my previous argument that dishonesty certainly exists in general and reckoning with that should not simply be reduced to "screaming liar."

I can also see how someone may nevertheless feel that calling the memo transparent pretext is not adequate. The productive response to such a failure where you feel it has occurred would be to bring the official reasons from the memo into the discussion, then ask people what they specifically think is wrong with those reasons.

I didn't engage your comments to defend or attack the policy -- I probably could do that, but it certainly didn't seem to be where you've focused. Instead you've focused on the quality of the discussion, and seem to be confused about why people have been critical and even hostile towards your engagement. It seemed like if you wanted better discourse, explaining how some of your engagement draws that criticism would help. It's strange if you're not interested in that, given that your stated position is about policing the quality of engagement in general, but no one can make you focus on what you don't want to, only point out the contradiction in that as well as the problems of your engagement.

[0] https://www.justsecurity.org/120547/presumption-regularity-t...

You kicked off this subthread by saying (among other things):

> the other side actually has reasons that you haven’t bothered to consider?

In response, it's entirely reasonable for someone to ask you to speculate what those specific reasons might be whether or not you agree with those reasons. You suggested those reasons exist, asking someone who expresses confidence they exist is a reasonable place to start.

Here's one of the ways that you're evading answering that question:

> I have repeatedly said that I don't support the policy.

No one here required you to publicly marry yourself to the policy. We're all aware of situations where we can speculate on possible reasons for a position without agreeing with it in the end. So saying "I don't support the policy" is a non-sequitur.

If you don't want to answer the question "what reasons have I failed to consider?" one way of resolving the tension left by your assertion that such reasons exist would be to say something like "I don't want to speculate on specifics, I acknowledge that this weakens any assertion that such reasons exist, but I still think we shouldn't just call the reasons given dishonest." Perhaps there are other ways of resolving that tension.

But saying "I don't support the policy" is not in any way adequate. You were not asked whether you support the policy, you were asked to back up your assertion that "the other side actually has reasons that you haven’t bothered to consider."

> I want you to tell me why you believe that

I don't think I've personally staked a position in this discussion regarding the administration's policy, much less whether or not they're lying, so it's not clear why I'd have any obligation to defend a position before we hear what your reasons are for asserting the administration "actually has reasons that you haven’t bothered to consider."

It's also not clear why someone who has staked a position that the administration's reasons are pretextual and dishonest would be obligated tell you why they think that before asking you to back up your assertion that other reasons exist, though of course you are also free to ask people why they believe something (and some people have at least mildly elaborated on specific reasons they believe the administration is not honest).

And I understand perfectly well what Motte and Bailey means. I specified exactly what I identified as your Motte (“maybe they have honest reasons that aren’t anti-immigrant, ever think of that even though I won’t speculate on what those would be”) and your Bailey (disavowing any investment in defending the new policy). Repeating your Bailey does not defend your Motte. Insisting that people are misrepresenting your Motte as a defense of the new policy does not defend your Motte, it is simply repeating your Bailey.

Your original position was "the other side actually has reasons that you haven’t bothered to consider?" and you have retreated to "I don't defend this policy so I have no obligation to defend my assertion that the other side has actual reasons I just want better discussion."

If you want better discussion, an explicitly acknowledged retreat from "the other side actually has reasons that you haven’t bothered to consider" will look more honest and less evasive. You could also speculate on what those reasons might be, and that would also strengthen your asserted position that "the other side actually has reasons" to the extent those reasons look credible.

Some people may also be considering the possibility that your claim that you disagree with the administration's position is not honest (as well as approaching the admin as dishonest actors). That's always discouraging of course -- we certainly want to be perceived as honest when we believe we are, and it's also convenient to be perceived as honest even when we are not. I haven't staked my criticisms of your engagement on whether or not you are dishonest: I allow room for the possibility that you're honestly wrong, and even some for the possibility that you may eventually make a substantial counterargument as yet unconsidered. Still, dishonesty is a real possibility to be reckoned with, and bringing it under consideration is reasonable enough. In that light, reducing reckoning with that possibility to "just screaming 'liar!'" also looks like rhetorical evasion.