MollieTheMare
No bio...
User ID: 875
Edit: Sadly while the post rendered correctly in the preview, it looks like some of the tags do get stripped when they get published in final form. I guess because arbitrary xml has some theoretical vulnerabilities. I guess we'll have to live without math support for a while longer.
The Motte has (limited) math support!
Inspired by the recent discussion of innumeracy, controversies caused by the refusal to specify problems symbolically, as well as past lamentations about the lack of math support I endeavored to research the simplest way to add MathJax to the codebase for limited TeX support. During my journey I discover that W3C has apparently been working on web native math in the form of MathML since 1998. It has varying implementation completeness for its 2½ components. There is semantic MathML, Presentation MathML, and MathML Core. The good news is The Motte appears to freely pass along the raw MathML XML tags, the bad news is implementation appears to then depend on the users browser. Fortunately, it seems like at least Core is implemented in most modern browsers. I have tested the latest Firefox for Windows and Android, as well as Chrome for Windows and Android.
The raw code is a bit of a mess. To render:
x=−b±b2−4ac2a
The LaTeX:
x=\frac{-b\pm\sqrt{b^{2}-4ac}}{2a}
becomes:
<math xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mi>x</mi><mo>=</mo><mfrac><mrow><mo>−</mo><mi>b</mi><mo>±</mo><msqrt><mrow><msup><mi>b</mi><mn>2</mn></msup><mo>−</mo><mn>4</mn><mi>a</mi><mi>c</mi></mrow></msqrt></mrow><mrow><mn>2</mn><mi>a</mi></mrow></mfrac></mrow></math>
Fortunately there is a relatively easy workaround. There are a number of applications that support translation, including online. Allowing you to construct decently complex equations. e.g.:
L=∫0∞Bν(T)cos(θ)dν=2π515k4T4c2h3cos(θ)π=σT4cos(θ)π
The rendering is still not perfect, but much better than plain text.
@ZorbaTHut thanks for all the hard work you do keeping the site running. DM me and I'll send you a patch if you want a note for the 'Formatting help' page, and that's somehow easier than just adding it the next time you're working on the code base.
I think it's more or less coherent, but the more elegant framing is already established by the Groisman paper. It is a bit excessively long for what's its trying to say though. The TLDR is:
There are two different probability spaces being discussed in the canonical problem. People implicitly switch between them mid-argument. Once you keep them separate, there is no paradox.
I'm pretty sure @ControlsFreak is just making the two probability spaces explicit by assigning them to Alice and Bob, rather than both to Beauty.
edit: I guess the culture war angle is twofold. One is epistemic, what does it mean to know or have information. Two, Grokipedia said the solutions to the problem are controversial. This was disputed as incorrect since the truth is obvious, but the subsequent dispute proves by example it is in fact controversial. Thus showing Grokipedia in fact spewing hard truths the woke left don't want you to know (mostly /s).
MollieTheMare
1d ago
UCSD has the crazy system where the gen-ed math requirements are set by the college not the university. So if you are in sixth college in a non-STEM major you can get away with one "Structured Reasoning" course like Intro to Philosophy, and one "Exploring Data" course like Stats for Psychology, and no actual math. It looks like if you are in Revelle college you are required to take Calc I/II.
Interestingly, all the undergrads do have a climate change gen-ed. How they are supposed to understand any climate modeling without knowledge of PDEs is a bit of mystery to me.
IMO if your job does not require the level of competency required to pass Calc I, your job should not require a bachelors degree.
A bachelor's degree is supposed to an academic laurel, and mathematical reasoning is a key pillar of the academy:
Let no one ignorant of geometry enter
MollieTheMare
6d ago
It is decently advanced.
That being said, a 2X body weight squat is a pretty achievable strength standard for most men. Very roughly, about 50% of a 1RM is a weight you can do indefinitely. It is supposed to be a challenge though. It's also about half of the 405x50 mark set by Tom Platz. So if the left tail of the log normal stops at 0, and the right tail is Tom Platz, halfway in-between doesn't seem that far out there for the average person to shoot for as a benchmark.
MollieTheMare
9d ago
That's part of the appeal.
The first time you can't do it you have to face down the question of: "Have I just let my self go, am I just not trying hard enough, or am in that era of appreciable physical decline."
If it's because you've let yourself go it gives you the opportunity to correct with no more than a years lag.
I also assume your ability to grind through progressively more unpleasant tasks grows with experiencing life's shit, so you'll be able to try hard enough?
Having a benchmark to know when you need to acknowledge when you are in decline is the real value though. Not that the day you can't do it is the end, but maybe that day is a sign that you should figure out what to do for when the decline gets serious. How many people do you see who should have retired before their job destroyed their body. Or should have long ago given up driving and are now a hazard on the road. Or had plenty of opportunity to make a living will but instead are rotting in agony leaving their family in an impossible position.
MollieTheMare
9d ago
Uhh, sorry if you've posted more extensively about your training history in the past and I didn't remember it. The only themotte fitness I readily recall is that 5hour is training for a century ride, jdizzler sometimes posts about veganism and general endurance training (mostly running I think), and Skookum did not actually attempt The Hock (though in my head canon he's still out there climbing the glacier).
Squats for (BW + BW load) x age
is decently advanced. Although I did say
Whatever scheme you want.
I apologize if I completely misunderstand and you are 6'7" and yours knees are shot from playing O-line in the NFL. Based only on the implication that you do not lift weights regularly and weigh over 300 lbs, I would for sure suggest some easier scaling. If you are looking for a serious suggestion for a physical birthday activity for a total novice at >300# body weight <40 Years old, I would maybe suggest no external load and breaking it up into 5-10 sets of body weight squats.
I don't think some sort of physical challenge for your birthday is all that bad of an idea or even that unusual. For example @thejdizzler:
running my age in miles on my birthday
It's a way of regularly measuring where you're at. More importantly, it's a physically manifestation of facing the continuing but ultimately futile battle against entropy that makes up life.
@orthoxerox did already summarize the broader point I was gesturing to though:
lift heavy stone make sad head voice quiet
It's not at all unusual for people to find that lifting (and other exercise, jdizzler making the rest of us look bad casually running 28 miles) can have a positive impact on physical and mental health.
MollieTheMare
9d ago
Isn't using physical pain as a distraction from emotional pain the idea behind cutting, and other forms of self-harm?
Yes. The difference is that the scars from cutting leave you externally disfigured, while the scars from lifting leave you swole.
MollieTheMare
9d ago
I'm not sure this will actually help your situation, but I'm still going to suggest birthday squats. Whatever scheme you want. The formulation I think of is (BW + BW load) for one giant set of (age) reps. So a 185 lbs 40 year old would squat 185 lbs of external load for 40 reps. Pause at any point, but don't rack the bar until you've finished the set (or collapse). It seems unlikely you will be worried about future suffering after finishing, with any luck you'll be in so much pain you won't have any spare capacity to worry about anything else.
MollieTheMare
10d ago
Not OP, but depending on what you are doing and your config you could almost certainly run glm-4.5-air or openai/gpt-oss-120b, which are roughly 100B class models.
A name brand box with a AMD Ryzen AI MAX+ 395 and 128GB of unified LPDDR5 RAM would probably be just over that now, but you can probably find a no-name box on sale for around $1.5K from time to time. Performance would obviously be worse than duel RTX 4000 Adas or something, but a lot cheaper.
The use case I'm imagining is like a background task doing a code review or auditing a highly sensitive code base to check for potential vulnerabilities, intentional or accidental. I could also imagine using something like that to slowly scrub through heath, financial, or other sensitive files. Either for auditing purposes or converting to structured data.
It would probably be a bit slow, but for anyone who has to work in an air-gaped environment it seems like it would actually be supper useful. It saves you having to send a query to the public internet the majority of times you have to look things up. Just replacing google searching, or (bleh) having to look something up in a paper book. It doesn't take that many uses from an engineer making 200k a year saving a few minutes to make it ROI positive for a business. Even just the time it takes to transcribe something you looked up from the internet facing machine to your offline machine. I suppose it depends on how many people are in the working group whether it would be more efficient to have some beefier centrally hosted machine on the intranet.
Even if it doesn't have to be air-gaped, I imagine if you have like 100+ employees dropping like $20k might still be cheaper over like 1-year than paying for an API provider. Especially if there are a bunch of compliance problems with sending things off site.
MollieTheMare
11d ago
When William Shockley invented the first transistor, he probably didn’t have the modern computer in mind.
As an aside it's a bit inaccurate, or at least incomplete to say Shockley invented the first transistor. Probably more accurate to say "contributed to the invention of" or "developed the bipolar junction transistor."
From the 1956 Nobel citation:
In 1947 John Bardeen and Walter Brattain produced a semiconductor amplifier, which was further developed by William Shockley. The component was named a “transistor”.
Shockley's main contribution to the first transistor was suggesting using field-effect to control a junction, but this had already been proposed by Julius Lilienfeld. He probably does deserve much of the credit for the bipolar junction transistor.
This does emphasis the point that a given invention is confluence of a variety of circumstances such that, as you say:
some fail and some succeed
It is quite a testament to Bell Labs that they not only were able to recruit such a large stable of geniuses, but were able to harness that power in a synthesis of cooperation and competition. It can't have been easy to manage so many (justifiably) huge egos.
MollieTheMare
12d ago
I remember when this paper came out.
There are important linguistic aspects of learning a programing language, duh it's called a language.
It's also completely stupid they they observed fluid reasoning to be the most important factor, then concluded the emphasis on advanced mathematics in introductory computer science is unjustified. Static analysis, like the f(x)=O(g(x)) kind, is related to fluid reasoning much more than arithmetic skills. In sufficiently advanced mathematics you see essentially no numerals or arithmetic. You even see extensive prose in addition to notation for sufficiently advanced mathematics:
Take advantage of the English language - Terence Tao
In classical programing you do the analysis the computer does the arithmetic.
MollieTheMare
15d ago
Re cycling clothing vs athleisure clothes:
more or less indefensible distinction
There are three major factors at play that make areo cycling wear look goofy while competitive exercising clothing makes you look hawt.
- The cut of areo clothing is for when you are in an areo tuck. When you are just standing around you end up with random folds, creases, and bridged areas that look frumpy. While the UCI has slightly cracked down on the flying squirrel sleeve to torso transition, areo clothes are cut so the panels are in the most areo shape first, for displaying sponsorship second, and comfort third. Modern leggings for women and ranger panties for men are cut and patterned to first to make your butt look good and for freedom of movement second.
- The chamois makes it look like you are wearing a diaper. It's 100% worth it from a comfort perspective, but the chamois disrupts the drape of the fabric front and back.
- Your average pro-level cyclist has tiny stick arms. If you have a jacked enough upper body to make a shirt look good you'll have to size up in order to fit your arms through an off the rack areo jersey. This will make it not-areo and leave loose fabric flapping around.
Re. calculator:
I think I was confusing that calculator, this calculator, and plugging in various Cd⋅A from random wind tunnel reports I'd seen. But yeah, I think it's generally in the right direction. Anything that makes you more aero without sacrificing comfort, or reduces rolling resistance, is free speed, and it's not even necessarily negligible speed.
Re. cadence:
I'll have to take a look at the paper when I can get access, since it's paywalled. Based on the abstract I think we're, more or less, on the same page. It looks like their average OPR and FCPR for trained cyclist was well within my 70-110 RPM range I quoted. @FiveHourMarathon was talking about 60 RPM, which seems a bit on the low side, but I assume it will creep up into the 70 with more experience. From there I suspect it depends mostly on individual physiology and ride type. Since the paper was from 2006 I assume that short cranks hadn't been "discovered" yet. My intuition is that higher RPM is more efficient the shorter the cranks, but this adaptation would occur without any explicit cuing.
Re. Single leg:
I see they are talking about much higher intensity than I was. My logic for warming up with single leg drills was:
- Your cardiovascular system isn't primed yet, but you're limited by using a single leg for total power output so there is some time efficiency in using that time to do the drills.
- Part of the point of warming up is for more efficient motor unit recruitment. If (the sensation of) pedaling circles is more efficient there should be some synergy between priming the motor unit recruitment and training the specific pastern.
Additionally, it's easiest to do the drills when using a stationary bike or turbo trainer, since you don't have to worry about traffic or balance. Single leg stuff introduces some novelty which might slightly combat the extreme boredom from training indoors. Also, depending on configuration, the inability to freewheel will highlight dead spots in the pedal stroke. The gain is probably too marginal to be worth the expense or hassle of counterweighted or split cranks for us mere mortals. On the other hand, there isn't a whole lot of cost to warming up on the trainer with single leg drills. I think it's unlikely it's harmful, and there might be some small marginal benefit.
MollieTheMare
16d ago
There is some natural variation, and it's sometimes argued that it's not clear if it's a selection effect that efficient cyclist ride with a higher cadence or if a higher cadence is just more efficient.
Controlling for constant power, the subjective feeling and from experience, <70 RPM will toast your legs in no time. Presumably from greater muscle tearing from greater muscular tension, lower lactate clearance, less oxygen and glucose uptake since the venous system is passive, etc. But just like when someone tells you the sky is blue because of Rayleigh scattering, the question is then why isn't the sky ultraviolet. Why is >110 RPM inefficient if turning over the pedals faster is so much better? For cadence, I suspect this is because you become less smooth if you exceed the speed at which you can maintain neuromuscular control.
In fact, your body more or less just does this regulation for you. As with many things: slow is smooth, smooth is fast. All of this to say cadence will naturally improve as you gain neurological coordination from practice. I haven't kept up with the latest meta, but circa mid 2010s, most sports scientist thought there was limited or no evidence for the effectiveness of pedaling drills while most professional coaches and elite level cyclits thought there was value in doing them. Two that come to mined:
- If you are doing all out intervals, indoors or otherwise, try to fully spin out each gear before shifting up. This is bad practice in a race because someone could get a jump on you while spun out, but might help with learning to turn the pedals over quickly while producing decent force.
- Single leg pedaling, possibly while warming up or cooling down, obviously while using clip-less peddles. I think the sports science people are firmly in the camp that most of the work (the F⋅s kind) is done on the down stroke, most concluding that practicing pedaling circles is thus pointless. I know at least one Olympic medalist who prescribes single leg drills though, and most experienced cyclists describe losing coordination and thus power as feeling like pedaling squares. My take is single leg drills train the coordination of pedaling circles. This feels smoother, you can thus turn over the pedals faster, so at a constant force level you are making more power.
Assuming the drive train is in at all decent shape, don't under estimate areo and rolling resistance. Even at modest speeds it can really add up. I can't readily find it, but there used to be a calculator where you could put in various different bike configurations. My rough recollection is that even at 100 W, going from arms extended, relaxed cut jersey, & cheap clinchers to sphynx like tuck, fitted jersey, and tubulars was like 11 → 15 mph for an average sized cyclist. Fortunately, with modern tiers tubeless and high quality clinchers are almost as efficient as tubulars were back in the day. There's no way not to look goofy in form fitting cycling clothing though.
MollieTheMare
17d ago
The point of Huel is it's complete nutrition that's convenient, in the sense you don't have to think about the micros and phytochemicals, there's various options for macros, it's easy to prepare, and has minimal cleanup.
Less easy to prepare and more cleanup for the powdered version vs RTD but has the advantage of very long shelf life and compact storage. The essentials line is (I think) £26.70 per 22x400 kcal servings in the UK, or 330 kcal/£. It's not meant to be the cheapest per-calorie though. Assuming you are in pure survival mode, you can probably get away with three scoop (600 kcal) servings twice a day, supplemented with one meal of rice and beans or rice and lentils to make up the calories. In the US they have retort package rice and retort Dal, not nearly as cheap as making from scratch, but something even a student could afford.
Tamago kake gohan with furikake on top was an old standby of cheap, taste, pretty shelf stable ingredients of mine. Even made with microwave rice, it was pretty cheap and still decent tasting. But it's not clear to me if eggs can be found cheaply in the UK right now, or if they were if they would be safe to be eaten very lightly cooked.
I assumed that the "private" free tier of most commercial models do not actually allow you keep the full thread in the context, and that the performance for summarizing a long and meandering thread would be degraded if it was not keeping the whole thing in context. I don't recall exactly, but I think the thread was like 25k tokens. It's possible the free online models have expanded to that allowed context size by now, I haven't done a comprehensive survey. If not the best case is they reject the prompt because it's too large. The more tricky case is the model "forgets" important parts of the thread because it falls out of context.
I didn't want to spend money (okay I spent like $0.03 of electricity) or have real name accounts associated with something quite so depraved as researching "Are traps are gay?" I suppose I could make a burner somewhere, but seemed like more work. If it was an important thing I simultaneously really didn't want to pipe to an API endpoint I have no control over I would have used a better model. I might have managed qwen3-30b-a3b-thinking-2507 or openai/gpt-oss-120b (probably good enough?), but that combination is so rare I'm too cheap to buy good enough hardware to be able to run that kind of thing in a decent time span. I also thought that excepting Grok, the main commercial models might be too nerfed to tell you if traps are gay. In retrospect I should have chosen a spicier local model for the goal of maximum luz.
So yes, the paid commercial models are likely to produce a better summary than I provided. I don't have 100% confidence that they are not going to associate even private chat queries with my real identity when they decide who to mark for liquifaction though.
MollieTheMare
20d ago
I suppose just dumping some AI slop in response would be frowned upon.
I'm actually more interested in how people would handle AI summarizing a thread this size. What models do people prefer for that kind of task? What hosts do people use to run those models. For general RAG what types of techniques have people had good success with?
My attempt was to run locally, since I don't want to waste tokens on such a ... singular task. In order to summarize the whole thread I used 'inject-full-content' rather than a chunking RAG to try to get a complete summary of the thread. The content I injected was a PDF of the fully expanded thread I printed. Due to too weak hardware for such a large context simultaneous with a decent model and finite time I ran qwen3-4b-thinking-2507 via llama.cpp v1.53.1. The only prompt was:
Summarize the attached thread.
A notable excerpt from the "thinking" (emphasis mine)
Some discussion about transgender identity and related topics (which seems to be a side conversation that got somewhat out of hand)
The acctual summary was not as exciting as you would hope. As an alternative I did a chunked RAG and asked:
Are traps gay? Answer with "yes" or "no."
Based on the thread, the AI answer is apparently "no." There you have it definitive proof.
Actual response to the summary request below:
Summary of the Quantumfreakonomics Thread on Political Money
This thread discusses a post by Quantumfreakonomics about the evolution of political money in the United States since the 2010 Citizens United Supreme Court decision. The key discussion points include:
Main Argument
Quantumfreakonomics argues that despite the widespread expectation that Citizens United would create a "wild west of corruption" in politics, the predicted corruption didn't materialize. Instead, political money has evolved more slowly and strategically.
Key Discussion Points
-
Why the predicted corruption didn't happen: The thread explores the concept of "cultural inertia and shame" as barriers to political money. Big corporations like Exxon didn't immediately engage in large-scale political spending because they operated with constraints against appearing "corrupt" or "cartoonishly evil."
-
AIPAC as a successful model: A significant portion of the discussion focuses on why AIPAC (the American Israel Public Affairs Committee) has been so effective at political influence without spending massive amounts of money. Commenters debate why AIPAC's strategy works when other industries (tobacco, pharma, oil) haven't replicated this success.
-
Different political money strategies: The thread examines various approaches:
- Tech billionaires' strategies (like Musk's Twitter acquisition)
- The Koch brothers' libertarian think tanks
- Targeted political pressure on specific constituencies
- The role of "astroturfing" (creating fake grassroots movements)
-
Economic analysis: Some commenters discuss the ROI (return on investment) of political influence, debating whether the market for political bribes is efficient or not.
Notable Insight
The thread concludes with the observation that "money in politics" is far too broad a term - what matters is how effectively money can be converted into political capital through specific, targeted strategies rather than simply buying ads or making large donations.
The discussion also includes a significant side conversation about transgender identity that became quite extensive, though this appears to be a separate thread that developed from a comment about gender roles.
This thread illustrates how political influence has evolved in a more strategic, culturally constrained way than many predicted after Citizens United, with AIPAC serving as a particularly effective model of political influence that other groups have struggled to replicate.
MollieTheMare
21d ago
This is part of the tragedy of the situation. It's easy to have such an impression, but knowing at least some of the details, it just seems like lives were needlessly ruined.
The problem with Kramnik's cheating allegations is that they are based entirely on his own subjective evaluation, and Danya is not the only one he has accused. His knowledge of statistics is farcically bad. His internal evaluation seems to be: at my peak was I better than them at classical, if yes then I should be destroying them in online blitz. The problem is that even though classical and online blitz are highly correlated, they do not have a correlation of 1. Blitz and classical are slightly different skill sets, and this is magnified online vs OTB. Imagine your boomer uncle trying to navigate a web-app with his worn out 20 year old Circuit City mouse and complaining the app is broken every time he miss-clicks. That's what it's like watching Kramnik playing online. For example, chess.com allows you to pre-move to minimize the time spent on your turn. Kramnik for some reason refuses to use this feature, then also complains that his opponents are moving with super-human speed. No your opponents are not cheating, you just don't understand how the platform works. Unfortunately, his accusations no matter how baseless come from the voice of a former World Champion. He could have been known as a great champion that helped train the next generation, instead he just looks like an angry old man who can't accept that he has declined and can't keep up anymore.
On the Danya side, the various rumors make it seem like an overdose was more likely than suicide. Either way, I don't think the incentives necessarily indicate cheating. While very good Danya didn't really make his living playing chess he made his living as a chess influencer. In particular, an influencer known for his wholesomeness. As that type of influencer your profession and self-identity revolve around how your are perceived. It seem from his last stream that despite the fact that many of the top chess names privately supported his innocence, and his fans remained loyal, the trolls were weighing heavily on him.
MollieTheMare
22d ago
I typically use leather teardrop pattern straps for anything that's grip limited. Hook grip for competition lifts.
The trick is to just lift more than the people mocking you.
I would typically silently judge someone using gloves. That being said, if you're lifting enough (800 solid-ass pounds), chalk, gloves, straps, and mixed grip all combined somehow does not look goofy at all.
MollieTheMare
23d ago
Generally just leave them be.
If they are so thick they are painful, have a distinct ridge, or in danger of tearing remove them. Typically using a 1-3 blade safety razor and very carefully shaving them down. It's easier than it sounds. Don't go too deep. Going over the residual with your wife's pumice stone can help the transition, just be prepared to answer some questions if she sees you using it.
MollieTheMare
26d ago
As I said down-thread I'm not a YIMBY.
The closest I've come is not objecting when a neighbor wanted to build a garage that required a zoning waiver because it was 10% bigger than allowed. I didn't object because their property is otherwise adequately sized, I have to live next to him, it doesn't obstruct my view, and it was so he could store his boat. I think it's more attractive than keeping the boat on the drive, and I wouldn't mind being invited out on the boat. If he had wanted to up-zone his lot from R-1 to R-5 you bet I would have broken out all the NIMBY tropes.
Whether a particular policy proposal from a self-identified YIMBY makes sense probably depends on the YIMBY and the proposal.
For the laundromat example, I do think there exists some system where the zoning and permitting requirements do not have to be so onerous. This would still be a small or marginal change to the total cost, but not zero. For example, if the opportunity cost for the zoning of that building was reduced by $3MM, there were 40 residential units built, each with a value of $1MM. That is a 7.5% cost savings. $1MM is not exactly affordable, but 7.5% is 7.5%.
I think it's also clear that the North America, specifically the US, specifically high cost of living US also has other problems. General cost disease being a big one. I don't think just deciding to build will fully solve this. The Golden Gate Bridge cost $630MM in 2024 dollars to build in 1933. The Golden Gate Bridge suicide net cost $400MM in 2023. Currently San Francisco has density of something like Copenhagen. Clearly they can "afford" to do more density as Singapore has much higher density and much lower GDP Per capita. Do I think they can just build their way to Singaporean quality of life? No. Do I think there is some fundamental limit that would cause a public services collapse above current density? Also no. Do I think public services could collapse if people don't get their act together? Yes.
MollieTheMare
26d ago
Sorry, I wasn't maximally clear in what I was even responding to.
You asked the question:
But what kind of housing?
My answer is the kind of housing where there is:
- Strong latent demand
- By purchasers who will pay market rates
- Where market rates will support direct constriction costs and marginal infrastructure requirements
With current aggregate zoning requirements, new market rate construction is on average purchased by families well above the median income family. Therefore you can make marginal changes to relax aggregate zoning requirements without reaching the point where
you build housing that the median income family can afford
Further, because of effects from initiatives like Prop. 13, public services are not necessarily diminished by even a median income family making a housing purchase.
My basic math argument is: the marginal elasticity is sufficiently small, and the existing equilibrium is sufficiently far from public service collapse, that marginal changes should not produce a catastrophic phase transition.
Yeah, I'm not necessarily in favor of any of these things. As an actual property owner IRL, I personally do not want a bunch of shitty development in my back yard. I'm just willing to admit it's down to wanting to protect my own property interests rather than making the argument that it's obviously in the public interest.
You would for sure have to rework the zoning code for those to work.
- For ADUs, they are currently not allowed in my municipality. If it came down to it I would rather my mother-in-law live in a ADU on my property than put her in a home or live in an attached suit. In that situation my MIL would not put significant additional strain on local traffic, because she would not be commuting to work and would probably share grocery runs. If you allowed rental of ADUs in a car-dependent place on minimal lots I do think that would be bad. Is there a way you could allow the former without the latter, IDK.
- For starter homes 400 ft2 would still not allow Connestoga huts. I've guess we've already discussed 800 ft2 homes here. I think that could work for utilizing otherwise unbuildable lots if combined with generous setback requirements.
- The key is eliminate minimum lot-size would be for existing neighborhoods. Again, this would need to be combined with generous setback requirements.
I actually do think the strongtowns people take it too far, but it's a pretty big gap from build 1,000,000 Connestoga huts to allow dense infill where there is latent demand. Might as well the gap to my actual position which is that it should not take 8 years and millions of dollars to get zoning approval to replace a laundromat with medium density mixed use development along an existing transit corridor.
MollieTheMare
26d ago
I'm actually not sure this is correct. The steel-man case for YIMBIsm is for market rate new construction, not incentivizing a bunch of Connestoga huts.
The marginal new construction unit is typically purchased by people above the median income in a given location.
This is obvious at an aggregate national level. The median sales price for new houses sold in the United States is $413.5k. At 20% down, prevailing 6.48% interest rates, and 30% of gross income spent on housing, this is a household income of $98.6k which is well above the ~$83k national median income.
In areas with higher housing prices the marginal income of home purchasers is also higher. For example the media sales price of all homes in California was $833k. While in Santa Monica the median sale price is $1.9MM. Connestoga huts are not competing for developer dollars in an area where the marginal sales price is in the millions. Because of Prop. 13 those purchasers are also paying far more in local property taxes towards local schools than people who were already locked in.
The nice married family with two kids moving into the new built house down the street is well above the average income, and therefore definitionally increases the per capita income. At the national level for example, married family households have a median income of $122.5k, again well above the unconditional median.
A marginal change in zoning will lead to a marginal change in the marginal new home purchaser. You might be able to make a reductio ad absurdum argument for the most extreme straw-man YIMBY, but the "official" YIMBY position is more like revise model zoning codes to allow triplexs where duplexs are now allowed.
MollieTheMare
29d ago
Good luck.
If you don't already own a pair, I highly recommend a good set of cycling bib-shorts at those kinds of distances. Also possibly chamois cream.
Don't underestimate fueling. If you "hit the wall" at hour four of a five hour marathon you only have to survive one more hour. If you "bonk" on hour five of a seven hour century you have to survive for twice as long, and also be lucid enough not to get run over. People have been talking about 60+/-30 g/hour of carbs for such long efforts. The upper end is quite a disgusting amount of sugar to be eating if you are not used to it, so it does require training as well.
What do you consider new Trek?
The Outcast with an "ambiguous" kiss, Rejoined with a "symbiote confusion" lesbian kiss, and The Emperor's New Cloak with a "dark mirror universe" lesbian kiss are all prime time line. More than, about, and just less than 30 years ago respectively at this point.
I haven't seen, and do not consider to be head cannon, anything after Voyager. I suppose the more recent series have explicit homosexual characters, where it's a recurring part of their character rather than incidental to an episode theme?
In TNG era they were relatively slow with their introduction, with only moderate controversy following each showing. It probably doesn't hurt that Terry Farrell as Jadzia Dax is pretty easy on the eyes.
More options
Context Copy link