I’d imagine somewhere there’s an engineer-artist putting such a print in a brightly lit locked box, which is programmed not to open until the interior light has shined long enough to eradicate the image on the print. (No idea if we have circuitry and LEDs that last long enough to make such a thing feasible).
Yeah, I felt like the “you have to be wealthy” hand-waving in the quoted section wasn’t very explanatory. Are lenders giving the ultra-rich great interest rates here as a loss-leader to try to attract other business from them?
> First, this type of planning is generally not economically feasible unless the taxpayer has a net worth exceeding around $300M. If you’re worth less than that, you’re not going to be able to command attractive loan/line of credit terms from investment banks. You’re going to have to get a plain vanilla product from a retail lender which is going to have relatively high interest rates (typically the Secured Overnight Financing Rate plus some amount of spread) and other terms that make implementing “buy, borrow, die” expensive enough that you aren’t much better off (or you’re much worse off) than you would have been had you sold the asset and taken the after-tax proceeds. (Caveat: even loans/lines of credit at retail interest rates can still be very useful for short-term borrowing needs.) Clients with a net worth exceeding around $300M, however, can obtain bespoke products from the handful of lenders that specialize in this market, and the terms and conditions of these products make “buy, borrow, die” a no-brainer for virtually everyone who has this level of wealth.
It's an investment for a bank which is middle way between regular loan/bond (where you get fixed interest and 100% of the loan at the end, but no appreciation) and stock share (where you get no fixed interest and all the appreciation when you sell it). The hybrid product would be you get some interest and some of the appreciation, but not as much interest as for a loan, and not all the appreciation at the end. How much would obviously be negotiated depending on interest rates, projected appreciation, and other factors. The point here would be to defer paying the interest (to make the asset owner's life easier while they are alive) while leave enough enticement for the bank to agree to the whole scheme (banks usually don't just buy shares in people's 401k's). I do not know which combination specifically works but it doesn't seem implausible for me for such combination to exist.
On the other hand if taking off work means forgoing income, taking the slower rail journey is even more expensive. Instead of taking 6 days off work to fly and take a cruise, maybe now I need to take 8. For retirees or folks who can work on a train, the calculation might be very different.
I get that, that’s one way of looking at it. But other folks look at train vs plane for more utilitarian “getting there”. Depending on your situation and what you consider the train ride to be, the value offered by the train can be considered a reasonable deal or outlandishly expensive.
Economy class airplanes are all utility and zero experience. The business is viable because utility and torment can cancel out because the trip itself has need to have a perceived value - all the value is in moving from A to B.
A drive, a cruise, or a train trip have intrinsic value beyond moving from A to B. You experience the movement in a different way.
We need to start thinking a bit differently about the economics of transportation when travelling for leisure.
As a native Marylander I always find myself forgetting about Cumberland, which is a shame. As someone who has mostly lived in and around Baltimore, you head west to Frederick (aka Fredneck) for the small city in the middle of rural farmland. If you keep heading west you get to Hagerstown which feels way out there in farm country. And if you keep heading west you eventually move from farms to mountains and you hit Cumberland, which looks like a city that time forgot.
As other folks have commented, there’s some beautiful architecture and the old part of the city seems like it could be a bustling place. There’s a train station and easy access to the great outdoors. But the jobs have long gone and drug addiction has taken root for so many there. I don’t know the best way to revive a place like that but I hope something eventually works.
And if you travel farther west on 70, you'll eventually reach Wheeling, WVA. At one time its position on the Ohio river and near railroads made it a transportation hub, it made money in iron, textile, and logging - they used to float logs down the river. The vestige of wealth is still visible in its architecture, beautiful brick homes, ornate porches, windows and roofs. It's this glimmer into this past, not so far in the distance, that is so sad to witness. A lot of the town has fallen into disrepair, not slum exactly, but heading there. There is a central market building with some kitschy arts and crafts, and food stalls that supply tour buses. The buses come for Wheeling Island Casino, which has one of the last two remaining greyhound racetracks in the US. There's some attempt at preserving the historic buildings and downtown. People keep leaving, and the tourist attractions are more of a detour stop than a destination point. There used to be a pie stall - best pies in the US, handmade, fresh ingredients, $15, baked to order by a retired teacher. He sold the shop, his kids didn't want it, it was too much work and they made more money doing other things.
I drove by Wheeling on my way to Texas from Maine. Literally looked like a city that was one great but is now dying. Very sad stuff. Apparently they are revitalizing downtown though.
Sure, but problem Cumberland has same problem as rest of Appalachia, it's geography isn't very good. Mountain areas make everything 10x times harder to build.
Let's say some big software company wanted to build second HQ. Even if Cumberland was attractive in workforce, education options and so forth, the architects would say "Building your HQ2 is going to be rough. There isn't enough flat land, flooding could be problematic, fiber companies are screaming about the trenching" Not to mention, where are you going to put all your workers since housing will run into same problem. So if you wanted to stay in MD, somewhere like Hagerstown or Salisbury would be a better choice since usable land is plentiful.
If there was an economy worth building for, the geography wouldn't be a blocker. Look at Seattle, Oakland, or the Hollywood hills. They're all built on rugged, mountainous terrain just as difficult as the Appalachians, but they don't suffer the same issues attracting wealth because their economic situation is so different. In fact, each of them has the opposite problem of demand to build vastly outstripping permission to build.
>Look at Seattle, Oakland, or the Hollywood hills. They're all built on rugged, mountainous terrain just as difficult as the Appalachians, but they don't suffer the same issues attracting wealth because their economic situation is so different.
Wrong. Seattle, Oakland and Los Angeles are mostly built on much flatter parts of those areas.
California entire geography is about "Hey, check out these massive valleys or coastal land we can build in." Same thing with Washington State, Seattle is in between Cascades and Olympics where there is all this flat land to build on. Yes, they running out of land and building into mountains now. That problem is like having FAANG scaling problems. It sucks but it's good/manageable problem to have and you have massive checkbooks to help solve it.
Have you been to Appalachia? It's not on the coast and does not have these benefits. If you want to compare it to West Coast areas, it's more like Sierra Nevada. Inland Mountains with only small valleys to build infrastructure in.
I've been to Appalachia. It didn't strike me as notably more rugged than any other reasonably hilly area, and quite a bit less rugged than many places known for it (e.g. the Alpine cities, or the Himalayas). The west coast cities were a reasonable comparison, because they're (1) in the same country and (2) Fairly comparable in elevation and grade, if not a bit worse. The oakland hills (and other bay area communities [1]) rise to comparable heights despite starting at sea level with 25% grades, for example. Queen Anne in Seattle [2] has almost exactly the same elevation gain, but the last 200 feet are basically a cliff. I'm not picking distant suburbs here, but rather historic parts of these areas that have been developed for almost a century. They only maintain the illusion of flatness now because the landscape has been intensively modified over that time to appear less severe.
Instead of just a few pictures showing rich area around a mountain that can afford it because companies like Apple have plenty of flat area, let's use available data provided by government agencies: https://topochange.cr.usgs.gov/topochange_viewer/viewer.htm
Cumberland, MD vs San Francisco. Wow, when I overlay urban areas of Bay Area, it's heavily built up in much flatter areas and gets much less dense as elevation change gets steeper. Not to mention, sea access gives you massive advantage since you are not having to move as much over really tall mountains.
Cumberland, MD has none of these. It's in a valley between two larger ridges of mountains but is heavily constrained. Also, not having sea access means transporting goods there is much more difficult and requires more infrastructure. And since we are one nation, you could just move west to much flatter Ohio and Great Lakes or East to flat parts of Maryland and Chesapeake Bay.
Sure, Bay area built up with creating really excellent schools that created really high paying industries but thinking it was "We lifted ourselves up by our bootstraps" and instead "Our geography and World War with Navy on our coast desperate to win really did help us."
I'm saying nothing about bootstrapping. What I'm saying is that difficult geography is not a blocker if the economics are sufficient. People built and regraded the hills around Oakland, Seattle, and LA because the economy was there first. The geography is an expensive inconvenience no one wants to deal with, not an unmanageable fact of life.
Seattle was MADE flat by literally using fire hoses to flatten hills and mountains [0].
That said, I disagree with the role geography has with developing a tech industry - most of it can be directly related to investment put during WW2 and the 1950s into innovation clusters.
For example, Seattle and aerospace (Boeing), Bay Area and computers+electronics+nukes (HP, IBM Almaden, LLNL, LLBL, Los Alamos managed by UCB), San Diego and Biotech+Defense Tech (Salk Lab, Navy), Portland and electronics (INL, PNNL, Tektronics, Intel), etc
I saw an interactive exhibit at Museum of History & Industry (MOHAI) in Seattle. It showed how different sections were regraded and you could push buttons where that section of hills and lakes would get lifted or dropped down. This wasn't a computer display, this was a real physical model of the city.
The Bay Area really benefits from Stanford and Berkeley being there. You need a steady stream of educated new grads to grow from.
Love that museum. Paul Allen really made Seattle such a great city - amazing museums, good sports, amazing biotech research, strong entrepreneurship scene. It's like Boston but better.
> Bay Area really benefits from Stanford and Berkeley
Also UCSF (major biotech hub) and SJSU (major electronics hub - imo EE@SJSU makes EE@Stanford look like child's play).
> Because Elasticsearch is not a reliable data store, organizations that use Postgres typically extract, transform, and load (ETL) data from Postgres to Elasticsearch
I’ll admit haven’t kept up with this but is it still the case that Elasticsearch is “not a reliable data store”?
I remember there used to be a line in the Elasticsearch docs saying that Elasticseach shouldn’t be your primary data store or something to that effect. At some point they removed that verbiage, seemingly indicating more confidence in their reliability but I still hear people sticking with the previous guidance.
It got a lot better in the ~7 series IIRC when they added checksums to the on-disk files. I don't know if you still have to recover corruptions by hand, or whether the correct file gets copied in from a replica.
The replication protocols and leader election were IMO not battle-hardened or likely to pass Aphyr-style testing. It was pretty easy to get into a state where the source of truth was unclear.
Source: Ran an Elasticsearch hosting company in the 2010's. A little out of the loop, but not sure much has changed.
I'm not sure what the author was referring to, but in our stack ES is the only non-serverless tech we have to work with. I know there's a lot of hate in HN around serverless for many reasons but for us for several years, we've been able to scale without any worry of our systems being affected performance-wise (I know this won't last forever).
ES is not this way, we have to manage our nodes ourselves and figure out "that one node is failing, why?" type questions. I hear they're working on a serverless version, but honestly, I think we will be leaving ES before that happens.
We have a similar experience except we’re using AWS’ version, so any inquiry as to what happened ends with support saying “maybe if you upgrade your nodes this won’t happen again? idk”
Just read the article: „Elasticsearch’s lack of ACID transactions and MVCC can lead to data inconsistencies and loss, while its lack of relational properties and real-time consistency makes many database queries challenging.“
A problem here is that ES (and Solr too) are pathological with respect to garbage collection.
To make generational GC efficient, you want to have very short lived objects, or objects that live forever. Lots of moderately long-lived objects is the worst case scenario, as it causes permanent fragmentation of the old GC generation.
Lucene generally churns through a lot of strings while indexing, but it also builds a lot of caches that live on-heap for a few minutes. Because the minor GCs come fast and furious due to indexing, that means you have caches that last just long enough to get evicted into the old generation, only to become useless shortly thereafter.
The end result looks like a slow burning memory leak. I've seen the worst cases take down servers every hour or two, but this can accumulate over time on a slower fuse as well.
It might be better by now with the newer GC options (ZGC, G1, Azul). For a while those had their own problems, but I'm a little out of the loop.
Tuning the older options wasn't really all that beneficial. We tried tuning, then running a custom build of OpenJDK to mess with the survivor space (which isn't that tunable via config), then ultimately settled on more aggressive (i.e. weekly) rolling restarts of servers.
I've been using and supporting ES (and OS lately) for well over a decade. Mostly its fine but a lot of users struggle with sizing their clusters properly (which is costly). Elasticsearch falling over is what happens when you don't do that. It scales fine until it doesn't and then you hit a brick wall and things get ugly.
Additionally, many companies learn the hard way that dynamic mapping is a bad idea because you might end up with hundreds of fields and a lot of memory overhead and garbage collection. I've fixed more than a few situations like this for clients that ended up with hundreds or thousands of fields, many shards and indices. Usually it's because they are just dumping a lot of data in there without thinking about how to optimize that for what they need.
A properly architected setup is not going to fall over randomly. But you need to know what you are doing and there are a lot of clients that I help that clearly don't have the in house expertise to do this properly and are a bit out of their depth.
This seems to be a Lucene/SOLR problem. I used Lucene/SOLR years ago, and it died so often that we just auto-indexed nightly and had a re-index button in the UI.
I've been abusing it as a data store for many years. I don't recommend it but not because of a lack of reliability. It's actually fine but you need to know what you are doing and it's not exactly an optimal solution from a performance point of view.
The main issue is not lack of robustness but the fact that there are no transactions and it doesn't scale very well on the write side unless you use bulk inserts. You can work around some of the limitations with things like optimistic locking to guarantee that you aren't overwriting your own writes. Doing that requires a bit of boiler plate. For applications with low amounts of writes, it's actually not horrible if you do that. Otherwise, if you make sure you do regular backups (with snapshots), you should be fine. Additionally, you need to size your cluster properly. Things get bad when you run low on memory or disk. But that's true for any database.
If you are interested in doing this, my open source kotlin client for Elasticsearch and Opensearch (I support both) has an IndexRepository class which makes all this very easy to use and encapsulates a lot of the trickery and bookkeeping you need to do to get optimistic locking. I also have a very easy to use way to do bulk indexing that takes care of all the bookkeeping, retries, etc. without requiring a lot of boiler plate. And of course because this is Kotlin, it comes with nice to use kotlin DSLs.
You can emulate what it does with other clients for other languages but I'm not really aware of a lot of other projects that make this as easy. Frankly, most clients are a bit bare bones and require a lot of boiler plate. Getting rid of boiler plate was the key motivation for me to create this library.
> a GS-12 getting paid around $68,000 per year (or $99k in very high cost of living areas like DC)
One valid sounding concern that I’ve heard is that the WASHINGTON-BALTIMORE-ARLINGTON, DC-MD-VA-WV-PA GS Locality Area underpays folks in DC by including farflung areas like PA and WV that skew the cost-of-living analysis. Whether that’s an intentional cost-cutting move or bureaucratic incompetence I’m not sure, but in the end the DC-area federal government pay ranges I’ve seen have struck me as quite low.
> most car rental contracts via insurance will only cover up to 30 days, so you end up being on the hook for the car rental
That’s if your insurance is covering it right? My Volvo C30 got rear-ended a few years back and it took about 6 weeks to get all the parts in and get the repair done. The insurer of the at-fault driver covered my rental car the entire time.
I've both had to fight an at fault motorist's insurance on extended rental coverage, and I've also been hand delivered a rental and told to keep it until I'm comfortable with my repaired car by another at fault motorist's insurance.
Ultimately, insurance is in the game of saving money, avoiding court, and reputation damage mitigation. Sometimes you get an insurer that puts their reputation above saving money, but it falls on the claim and the agent to decide how to go about it.
reply