“Completely exchangeable”
Obviously people are not fungible, replacing one person with another will never provide the exact same results. The question in each case then is how different would the results be, and would the replacement be better or worse? For a very simple job, perhaps pressing a single button, the results may only be subtly different. But what happens when it’s a complex job with no right and wrong answers, where work input is affected by output (like a chaotic system), spanning multiple areas of influence? The work output of the individual changes drastically, and just like in a chaotic system, the results to the organization vary increasingly over time. Nobody is fungible, but of all people, decision makers like politicians, CEOs etc are the butterfly wings flapping in New York that causes a cyclone in Japan. The only real way to evaluate if their impact is likely to be positive is looking at previous results. Due to rarity of top performers and importance to systems, they have negotiation power.
Dependable leaders really do have that much value to their organizations. This is similar to why in critical areas like medicine, old-and-dependable things are valued over new and shiny. The older things have lower risk, and a strong track record. That added dependability is more important than being the newer “better” but riskier option. Back to this topic, how many CEOs with track records managing 80 billion revenue AI organizations are ready to replace Altman? Because Open AI is well ahead in the field, they don’t need big risky changes, they need to reliably stay the course.
When calculating energy requirements on a flat plane of travel, the weight of the load is only relevant* when accelerating. On a flat plane, once the load is up to speed, it doesn’t take any significant increase of energy to maintain speed, regardless of load weight. The main reason fuel is required to maintain speed on a flat surface is overcoming air resistance. At the speeds that road traffic travels, only the front and back shapes of the vehicle contributes substantially to air resistance, not “air friction” from an elongated middle body. This means that on a flat plane like a highway, doubling the length of a vehicle to accommodate double the load, does not require anywhere near double the energy to maintain speed. It does however provide much more space to capture solar energy.
By the way, you can take advantage of this yourself on long highways by driving close (but safe) to large trucks, driving in their slip stream. It cuts fuel requirements substantially because the truck is doing some of the work of moving the air for you.
One puzzling thing is why trucks aren’t designed to be more aerodynamic, instead of a giant box shape. Anyone got any thoughts on this?
*aside from small things like added friction on bearings, changed tire geometry etc.
>aside from small things like added friction on bearings, changed tire geometry etc
This is rolling resistance, it's generally proportional to weight, and it's usually not negligible for heavy vehicles: it can easily make a large fraction of the energy usage of a truck, though at highway speeds it's rare for it to be the majority of it.
Afaik, it has partially to do with regulations. In Europe the max length of a truck includes the cab, thus it's as short and upright as possible to leave max room for cargo (cab over engine), whereas US trucks have the long "hood". Old trucks were quite angular but the new ones are a lot more aerodynamic.
Second, those "tapered" add-ons in the back can help a lot to approximate a teardrop shape better given that there are a lot more constraints in general in the front.
> The main reason fuel is required to maintain speed on a flat surface is overcoming air resistance.
Yeah, 200 horsepower's worth.
Wind resistance at semi-truck highway speeds is very significant. Multiply the numbers here[0] by 9.5x to convert from BMW to Semi-Truck (accounts for increase in Cd and frontal area). So semi trucks might take on the order of 180hp at the wheel just to maintain speed. With drivetrain losses you'd need to be generating 200hp.
I'm from Melbourne - Forget hotels, you can stay in a great hostel for USD$39 (AU$59) a night here, which is very affordable, and an amazing way to meet people. Don't be afraid of dorms, I've literally had the best times of my life backpacking the world and staying in dorms. Here's a hostel in St Kilda I'd recommend that's $39USD/$59AUD a night for a 10 bed dorm (which isn't too big, avoid huge dorms >18 people) https://www.hostelworld.com/pwa/hosteldetails.php/Nomads-For...
You will get more than accommodation. You will be welcomed in to a group of young travelers from all over the world who are looking to seize the day.
I really really hate hostels to be honest. I actually lived in Melbs for a couple of years in the past and have an Aussie ex wife too. I used to live in Middle Park (of all places) and so going and staying in a dorm in St Kilda just isn't my idea of a fun time right now. Can appreciate I'm in a bit of a strange place but I'm in some kind of middle ground between tourist and resident, and the days of entertaining ideas of dorms are well behind me.
Edit: but that said, thanks for the tip. I realise I'm being diva-ish so I'll give it some thought. Would likely have a good time there. Just still sad at being stiched up for someone's kicks at Christmas.
That's fair. Sounds like your friend really hung you out to dry. Sorry to read about your experience these holidays, and hope you find a way to turn it around. Merry Christmas!
I would second this - I've had some incredible times in such places. I've put up a few other people in hostels tonight and expect a few more might take me up on the offer before end of day tomorrow. I understand not wanting to be a drain on resources, but it is very cheap and I'd feel much better knowing you've got a roof over your head.
In the macro economic sense, fiat money isn't 'used up' or 'locked away' when you buy something like crypto, it's transferred from your account to someone else's bank account. Worse, it goes through the process of fractional reserve banking and multiplies about ~10x after changing hands repeatedly.
There's no such thing as fractional reserve banking. It's an urban myth that has been debunked by QE for over a decade. Lord only knows why people still believe it.
Banks create money on demand by discounting collateral. Government creates money on demand by discounting the power to tax.
Fiat money disappears by the drain to taxation, to repaying loans and to 'rainy day funds'.
QE is so thinly related I can hardly imagine how you could contort it to have "disproved" something which is codified in law and taught in basic finance and economics courses.
1) It appears to me that the document you provided actually refutes what you are saying. It supports fractional reserve banking. Here is a quote from the conclusion: "Most of the money in circulation is created, not by
the printing presses of the Bank of England, but by the
commercial banks themselves: banks create money whenever
they lend to someone in the economy or buy an asset from
consumers. "
2) "infinite money" without a legal limit to reserve ratios would only occur if every single bank actually had exactly 0% reserves, and it would take infinite time and infinite transactions for that to occur.
FYI using the observed absence of 'infinity' as a proof is generally poor logic as there is lots of mechanisms blocking infinity from occurring in reality.
That's correct. However, investing in Crypto using dollars will inflate the dollar. Thus, the arguments that the government will probably leave crypto alone, still holds.
But if someone with money to spend transfers it to a crypto scammers account rather than buying say a car, that avoids inflationary pressure on car prices.
Although if you have X in circulation as money, and Y in stock market valuation, you could say Y/(X+Y) of total value is in the stock market.
If the market valuation goes up to Y+Z, you could say money has "entered" the stock market, pushing its share of value to (Y+Z)/(X+Y+Z) even though the money in circulation, X, could be unchanged.
Not sure about that. Money velocity went down a lot. If you pay Apple money, Apple – the company – keeps that money as cash reserves in some form or another. This might be reinvested and circulates a bit more, but is it really spend in the real economy so that average Joe benefits from this?
Yes, when you buy newly issued shares from Apple (rare), cash flows through the stock market into Apple's accounts, but again, no money went 'into' the stock market.
If you're talking about Apple selling devices, then it's another concept entirely.
Yes, although you could argue that money that was transferred from a checking account to a brokerage account has « flowed into » a market, at least for the time it takes to settle any trades and for the counter party to withdraw theirs (since it will not be used for consumption)
Try and break the problem down with systems thinking. Here's an example of this:
━━1=> The output effect you're trying to eliminate/reduce is pain and swelling in respiratory tract.
┗━━2=> Hypothesis (needs testing): This is caused by physical irritation of the tube
┗━━3=> Hypothesis (needs testing): That irritation is caused by A) friction and B) physical pressure
┣━━4a=> How can we reduce friction?
┃ ┗━━5a=> Would a lubricant on the tube meet engineering constraints and reduce friction?
┃ ┗━━6a=> Would the use of lubricant in practice create added risks or difficulties in the operating room?
┃ ┗━━7a=> Hypothesis: Yes, due to the application of lubricant necessitating a glove-change afterwards.
┃ ┗━━8a=> How can we overcome problems with lubricant application to the intubation tube?
┃ ┗━━9a=> Would passing the intubation tube through a no-mess, easy-to-use "self lubricating ring" mitigate added risks or difficulties in the operating room?
┗━━4b=> How can we reduce physical pressure?
┗━━5b=> Would reducing the diameter meet engineering constraints and reduce pressure?
┗━━6b=> If there is a minimum diameter requirement needed to deliver oxygen, would dynamically changing the diameter of the tube (e.g. inflating it) after insertion create a reduction in pressure?
So that's a very simplified example of how to solve problems with systems thinking (although in reality it would be WAY more detailed and actually test each hypothesis).
The problem is very rarely "money" in medical fields, the problem is someone actually identifying the root problem/s, i.e. the "problem behind the problem behind the problem", and applying engineering to solve that, which solves the problem one level above it, etc.
If you're interested in this, read up on things like "the five whys", a systems approach to problem solving invented at Toyota.
I don't want to be overly critical, but what exactly do people see in this book that eludes me?
I've read it a few months ago with high expectations, pen in hand, but… I didn't really make many notes.
Sure, if you've never heard about feedback loops (and that there is positive and negative feedback), but, is the book just way too basic for me or did I miss the point?
Is Weinberg's book maybe more practical or more advanced?
I'm still thinking that systems thinking has much to offer (and skimming the International Encyclopedia of Systems and Cybernetics confirms that hunch), but where do I continue?
I don't think inflating would work, since it's a tube. Maybe you could achieve the same thing with a sort of skeleton inside the tube that can push it to different widths.
I wouldn’t know whether that would help, though. I think it’s at least equally likely the problems are caused by having to make fairly tight turns in a tunnel that changes shape all the time.
I appreciate your method of breaking down the problem. I'm still stuck on the idea of using different materials but after applying your thought process I think a gelatin coating may solve the lubrication issue and possibly allow a topical numming agent like Ambesol or anti inflammatory to be infused in the gelatin layer.
It is possible that because my procedure ran longer than expected due to complications, they either had to reinsert a different tube or the tube they used was not designed to be used for that long.
I had my gallbladder removed and unfortunately some of the gallstones got stuck in the bile ducts and they had to chase them down. This turned a 45 minute procedure into a 3 1/2 hour operation and once finished it took another 2 hours to wake up.
The issue with coatings and lubricants is that you’d need to use a nontrivial amount for them to be effective, and that means the patient is at risk of aspirating the substance. That happens, and boom, you’ve got aspiration pneumonia. Even with some kind of encapsulated or solid gelatin coating you’d have to be very careful to make sure it doesn’t leave any kind of residue or break into pieces that could be aspirated.
This is a dumb idea that would presumably cause other problems but ... what if they put the patient upside down? It's certainly make using some sort of lubricant a lot less dangerous. Now presumably it would introduce a host of other problems that would offset this but who knows.
I don't know what kind of anesthesia you were on, but a quick google suggests some kinds of anesthesia cause difficulty swallowing. I also can easily pull up a link suggesting that dehydration is a common cause of sore throat following any major surgery.
So this may not be entirely about the physical contact of the tube.
I am not a doctor, just someone who has seen too many of them over the years for my own medical dramas.
Edit: Just to be crystal clear, I am absolutely not trying to suggest anything for you, medically speaking. The above observations are food for thought concerning the problem space under discussion. That's it. That's all it is.
TL;DR
1. Consider Ishikawa (“fish bone”) diagrams before applying the 5 whys or as a way to help organize the 5 why process.
2. To improve a system, first determine what is constraining the performance of the system. This is 80/20 thinking and attempting to optimize globally instead of locally. Local improvement may not result in system improvement.
Longer explanation and examples follow.
1:
The 5 whys are a great problem solving tool. Often there are many potential root causes which need to be investigated. Sometimes there is a single root cause that is responsible for the problem being investigated other times there are multiple factors that interact to cause the problem.
An Ishikawa or “fishbone” diagram is a useful tool to help organize many potential root causes.
In the example of pain and swelling caused by intubation we could start a little higher before diving into detailed 5 whys.
We can start with top level groups, such as:
- material
- method
- person
- patient
For each group, we determine potential root causes. This can be done through asking the experts performing the process, reviewing SOPs, observing the process, looking at process parameters, literature reviews.
Some examples, material could be the material of the tube, the type of coating being used, anesthesia, etc.
Method could be how the tube is inserted.
Person can be if the person is new to the process or experienced, is the person following procedures...
Patient could be if the patient has any allergies to materials or adverse reactions to medication or anesthetic...
The above are illustrative. You can decompose the groups in different ways. You can define the groups differently. Of course the above isn’t comprehensive or unique. Try to be MECE or mutually exclusive and collectively exhaustive when decomposing the problem space.
We often try to investigate the potential root causes by performing statistical experiments to try and determine cause and effect and which variables significantly impact the outcome.
2:
I want to provide an alternative point of systems thinking. The medical procedure is a process. There is a system at work to deliver that medical procedure. There are measurements to determine how the procedure is working. Efficiency (are you using the resources planned) and effectiveness (are you achieving the outcome the procedure was designed to provide) are two high-level ways to look at measuring system performance. Many measurements can be derived from these two.
From that perspective, you want to focus improvement on the areas/steps in the process that will most improve the outcome measures. With that in mind, it may have been determined that the combination of intubation materials and procedures that are used provide for the best outcomes as determined by the procedure provider. It may be to reduce cost while keeping recovery, discomfort, risk of infection, etc. at some baseline acceptable level. If you change those variables to reduce discomfort, it’s possible you may negatively impact cost, patient recover, or something else.
Note: I’m not a health care professional, so I have no knowledge about medical procedures or the various trade-offs and optimizations that take place. I am speaking from the point of view of someone who works in quality and process improvement.
“I love working here, and because of that I need to let you know that other opportunities are becoming financially better alternatives. Would you reconsider the maximum salary that you think I’m worth here, so that we can keep working together?”
I think this is the best answer. It shows you _want_ to work there still which your boss will hopefully appreciate. Pretending to already have an offer could come back in your face.
This is going to be more or less my strategy when I ask (rather soon). I do love working where I am. The only problem is that other offers are starting to come in, and I know that they'll be offering a lot more money. :(
Are you guys coming to Swift Summit in SF in a couple weeks? (see: https://swiftsummit.com). I'm one of the organizers, and previously ran a Swift meetup in SF. It would be great to have a chat about this with you there if you're coming!
My .io has been up and down about 25 times in the last 15 hours. The downtime has been about 1.0 - 2.5 minutes each occasion, with a down period of 1 hour 48 minutes at the start.
The latter downtimes look planned, as they are for exact periods (1 minute 0 seconds, 1 minute 30 seconds, 2 minutes 30 seconds, etc)
What are you using for monitoring / what generated these logs?
Unless your monitor is attempting a connection every, e.g., one second, you're going to end up with the "exact periods" you describe (i.e., presumably your monitor is only attempting to connect once every 30s).
Also, this would be more indicative of a problem with your web host, not the .io root name servers, assuming the TTLs on your RRs are set to > 30s.
Dependable leaders really do have that much value to their organizations. This is similar to why in critical areas like medicine, old-and-dependable things are valued over new and shiny. The older things have lower risk, and a strong track record. That added dependability is more important than being the newer “better” but riskier option. Back to this topic, how many CEOs with track records managing 80 billion revenue AI organizations are ready to replace Altman? Because Open AI is well ahead in the field, they don’t need big risky changes, they need to reliably stay the course.