The terminology is all a bit loose. Proteins are peptides, but we don't usually call them that. But when a hormone is a protein, rather than a small molecule, then we call it a "peptide hormone". I suppose because we're making a chemical point, that the hormone is in the family of peptides rather than steroids or eicosanoids or whatever?
As you say, insulin is a fully fledged protein, albeit a small one, and i wouldn't call it a peptide myself. But then, insulin is 51 residues, whereas tirzepatide (Mounjaro) is 39; is that such a big difference? FWIW i think a lot of other "peptides" are even smaller, eg CJC-1295 is about 30 residues.
peptides (2-50 amino acids) are smaller than proteins (50+ amino acids)... it's impossible to call a complex protein a peptide, it fails to meet the definition.
Logs and metrics are both derived from events. A log takes the whole event and records it somewhere. A metric takes some numeric value from the event, aggregates it over time, and records it periodically. You can reconstruct a metric from logs for the underlying events.
A trace is a period of execution between two events. You could record a trace as a pair of log entries, or one log entry at the end. You can then reconstruct a trace from those log entries. If you want to associate multiple spans, and separate log entries, within a trace, you use a shared ID, which is just the same as a context entry for logging.
All three of these pillars are just ways of looking at events. They are not fundamentally different at all. This is a mistaken idea in "Observability 1.0" whose correction is the basis of "Observability 2.0".
The pillars still have their uses, but the choice between them is really a non-functional one - storing a log entry for every event might be too expensive, so just store metrics instead, and index every log entry so it can be correlated with nearby ones might be too expensive, so just store specific traces instead.
This is the literally the "everything is a graph" argument from database architecture. The conceptual abstraction fails badly because it has to be implemented on real silicon that imposes constraints not considered in the abstraction.
Logs, metrics, and traces are all derived from raw events but none of them are intrinsically discrete events in a systems engineering sense. They are all different data models with different patterns of traversal over raw events. As data model, you need to build secondary indexes over the raw metrics to reflect the orthogonal data access patterns depending on if you are evaluating them as logs, metrics, or traces. This famously has poor scalability and performance.
In analytical processing we largely manage the inherent performance and scalability issues using denormalization, which allows processing pipelines with very different requirements to be optimized independently. Or in this context, treating logs, metrics, and traces as unrelated things with independent infrastructure.
"Observability 2.0" deeply embeds an architectural assumption that all systems are small. It is not a tractable architecture in high-scale or high-performance systems.
Real silicon has a long history of destroying beautiful conceptual abstractions in software engineering.
You are conflating the challenges of ingesting and querying at large scale with the what the original comment is about, which is emitting them more easily.
I don't see them as separate issues. Emitting them directly runs into the inherently poor memory locality (and potentially concurrency) of trying to produce logs, metrics, and traces from the same underlying event data representation.
It is only "easy" if performance and scalability don't matter.
> Logs and metrics are both derived from events. A log takes the whole event and records it somewhere. A metric takes some numeric value from the event, aggregates it over time, and records it periodically. You can reconstruct a metric from logs for the underlying events.
No, metric is just value. Some are derived from events (like histogram/rate of given event duration) but others are wholly independent (like returning app's CPU/memory usage)
The app's memory usage is an aggregation of the alloc/free events. I think the original point was that all of the metrics, traces and logs are conceptually the same but for efficiency, we store less data in each place, not the full history. Personally, for the systems I work on, having an easy way to turn logs into metrics and vice versa, without deciding up front, would be a slight benefit.
Pretty sure you've been able to interrupt a thread blocked on IO in a normal InputStream since 1.0.2 in 1996.
There is a significant caveat to this, which is that if the underlying blocking IO operation isn't natively cancellable (and the JDK doesn't use the trick Zig uses here), then the interrupt is allowed to close the IO resource as a way of ending the operation early. In practice this is usually fine, as you're interrupting the thread precisely because you want it to give up on whatever it's doing, but it precludes some particularly subtle possible IO designs.
Not sure if this changes under virtual threads, where the thread is much more loosely coupled to the syscall.
I think that specific issue was a huge mistake, because it means the interruption, which was already messy, is now unusable.
First, you can get deadlocks as interrupt now calls arbitrary code to close channels. Second the interruptee thread cannot recover from an interrupt, if it chooses to to. Lastly it means that a server that shares a file descriptor with multiple threads breaks unrelated operations on interruption.
Mild counterpoint: a friend of mine, a professional woman, started on GLP-1 drugs some time ago. She has lost weight, but she also says she's had significant cognitive improvements, more focus, better sleep, etc. It would be great to have research which explicitly teases out these various effects.
This research already exists though. Obesity and being overweight increases the risk of sleep apnea, strokes, heart disease, diabetes, mental health problems, etc. etc
Causality is an issue there, I would suggest that genetic auto immune issues that are rather wide spread can cause all of the above. I think people are finding that GLP-1s help with auto immune issues.
I can speak only for myself. But when I was overweight, the heels of my feet would crack into such deep fissures that I could remove skin like thick pieces of jerky.
I lost 90 pounds and now the heels of my feet are as soft and supple as a baby's bottom.
Being overweight causes all sorts of weird side effects.
The incidence rate of genetic issues generally doesn't change much over time due to how slow genes changes, so an increase in obesity rates from 5% to over 40% in just a few generations suggest the cause is very unlikely to be due to genetics.
In the US, I'm convinced that it was the food pyramid.
As a kid in the 80s, I wondered how it made sense that my diabetic father was told to follow the same model as non-diabetics, which emphasized carbohydrates.
Anectata: I sleep much better after losing about 25 kg, and don't snore anymore. I didn't really think before about how your airways can get fat, but now that they're thinner, I had to relearn how to swallow, because food/drinks tried to go up my nose. I guess my point is that resting better has all sorts of secondary improvements.
Interesting. Part of the reason I was eager to use retatrutide was the reversibility. Fantastic drug and entirely reversible. I self-tested and the effects I had on it were reversed entirely in its absence. Similar to caffeine or amphetamine at the doses I have tried (which are not high), removing the intervention resulted in a return to a prior state over a fairly short period. I'd say that, in general, some of the good properties essential to an intervention for a non-debilitating condition are:
* uniform dose-response: easy to increase effect by changing dose
* reversibility: automatic reversion to pre-intervention in a short-term is gold standard - allows for regret escape
* low-cost
GLP-1 drugs have these properties which I like. I'm surprised to hear that someone would prefer the alternative, but I must imagine that cost and the dosing structure (sub-q injection is annoying) must dominate decision making here.
The side effect of drastic muscle loss isn't quickly reversible though.
Because muscle is about 3x as expensive to maintain as fat, unless you are on some major strength training to maintain, muscle is the very first thing that your body gets rid of when losing weight.
So now you are off the GLP-1. But you also have to eat fewer calories to maintain a calorie deficit because you no longer have all that calorie burning muscle.
* And depending on when you start your GLP-1 journey, men when they hit 40 naturally lose between 5% and 8% of their total muscle mass ++a year++ a decade (women, probably earlier). In addition to the muscle loss due to weight loss.
* And it takes approximately 3 times longer in your 50s to create a pound of muscle that it does in your 20s.
* And you can't train like you did in your 20s or you will hurt yourself.
* And as you age, your body can't absorb the protein it needs to build muscle as efficiently as it did when you are young. So you need to eat even more food to compensate.
Ok. This is now more of a rant about getting older than it is about GLP-1s.
-- Edited to reflect that muscle loss % is decade, per the corrections in replies.
It doesn’t have to be a foregone conclusion that GLP-1 causes muscle loss though. The primary reason for the muscle loss is the reduction in calories and protein intake. If you want to keep most of your muscle mass you can just make sure you keep up with your daily caloric and protein needs.
Muscle requires protein, yes. Unfortunately, we don't get to keep unnecessary muscle just by eating protein. The body requires a lot less muscle to lug around 200 pounds than it did when lugging around 300 pounds.
We have to do the strength training. "Use it, or lose it".
GLP-1s aren’t “forever drugs.” They’re eliminated so fast from your system you need to keep taking them to keep the effect. That’s the opposite of a forever drug. If you go off them, you just return to the way you were.
It certainly can be a good thing! But it can also be considered a bad thing that you have to keep taking them, and have to keep paying for them forever. A one-time fix (a cure) can be more desirable than a lifelong treatment regimen.
how do you permanently cure obesity? you can try rewiring metabolism with crispr or some epigenetic changes but thats insanely risky, it would be very likely to mess up the immune system and give people cancer or nerve disorders.
maybe i could see something based on modified gut bacteria that produce GLP-1 drugs inside your body. that would run into problems with dose regulation but at least you can flush them out with antibiotics if you get a bad response.
Which means if you want to maintain their effect, you need to keep taking them forever. So maybe GP used the wrong term but I suspect they meant exactly this.
In any case, we just don't know what kind of long-term effects these drugs have. No one knows what happens if you come out of them 30 years after you started.
Water and food have a milder risk profile. The FDA (https://glp-1.health/blog/glp1-obesity-eligibility-bmi-chart...) recommends "GLP-1 weight loss medications (Wegovy, Zepbound) if your BMI is 30 or higher, or if your BMI is 27 or higher with at least one weight-related health condition such as high blood pressure, type 2 diabetes, or sleep apnea". This is because there are side effects such that GLP-1 meds have not been shown to produce a net positive benefit-to-risk ratio for people who are only mildly overweight.
> This is because there are side effects such that GLP-1 meds have not been shown to produce a net positive benefit-to-risk ratio for people who are only mildly overweight.
Importantly, this is different from "GLP-1 meds have been shown not to produce a net positive benefit-to-risk ratio"
Anyway nearly all of the side effects for these drugs are just discomfort. Not even really obvious they should be called "risks." Meanwhile, the vast majority of Americans who are not already diabetic or have HTN will soon, and it would be quite a surprise to learn these drugs don't have protective effect.
The health problems that GLPs produce help people avoid are way more costly than the drugs will ever be. Especially now that the market is so competitive and will only get moreso.
For me, historically drugs like this for ADHD or depression or gastric bypass or whatever tend to lose effectiveness over time as the body adjusts. Not for everyone but for enough where you've got a short term gain that ultimately compounds the original problem.
Medication can be a slippery slope into side effects and rotating dosages/medicines. Especially for behavioral medicines.
> ADHD medications do not lose effectiveness over time.
I mean, they're mostly stimulants, which definitely lose effectiveness over time. It would be really weird if they didn't as that's basically how our bodies work in general.
Note: I have been prescribed said medication, but have not started taking them yet so I might be entirely wrong in the practicalities (and how would one disentangle the psychological impact of medication from the physiological impact?).
I had the same intuition as you but I looked it up in light of GP's comment and it actually seems not so clear that's how it works. Apparently the physiological effects of prescription ADHD meds will tend to still be ~same intensity for years and years. Patients often report weakened cognitive/psychological effectiveness, but obviously that's very hard to compare...
> Apparently the physiological effects of prescription ADHD meds will tend to still be ~same intensity for years and years.
I'm not sure how one would even measure this. Like, back in the halcyon days of my youth I took a bunch of stimulants, and experentially there was definitely less impact over time (habituation is a thing).
And given that ADHD is predominantly a psychological issue, then the only measure that matters is that of the patients themselves.
Do you have some of the sources you mentioned available, I'd be interested to understand the details?
> Like, back in the halcyon days of my youth I took a bunch of stimulants, and experentially there was definitely less impact over time (habituation is a thing).
Practical experience is that people cycle through ADHD and depression meds. I agree that it's hard to tell what that means because we don't generally do long term double blind placebo studies.
The issue here is more complicated than just a lack of long term RCTs. It's that we're talking about psychological acclimation within individuals. Given that people's lives change so much, even within an RCT design it's not clear how you get an actual comparison of "effectiveness" from year 0 to year 4.
> t's not clear how you get an actual comparison of "effectiveness" from year 0 to year 4.
You'd probably need a few different measures of attention/executive function and you'd need to counterbalance them across people to prevent habituation.
Given that the regulatory bodies don't require studies like this, it's unlikely to ever happen.
Yeah, money, and just signing up to regularly inject a thing forever.
Anything that’s “starting tomorrow and until the day you die” just seems like a thing I shouldn’t take lightly.
Also I’m not badly, badly overweight. So it’s not like “it’s this or cardiac arrest in a year or two”, for me. I should probably be dealing with my issues the old timey way.
The stakes look very different for other people, obviously.
Bariatric surgery has most of the same upsides as GLP-1s and is a one-time cost. But I agree with the other commenters that "forever drug" is mostly fretting without an actual reason. The Hunger that engineered foods are deliberately designed to induce is a forever thing too, now there's a counterweapon.
If it sounded like I don’t think other people should take them, that was a mistake on my part.
I’m just really not what you’d think of as a model candidate for them, so I’m not quite ready to commit to my first and only, not exactly free, rest of my life pharmaceutical dependence.
That may not seem perfectly rational to some… but ya’know… I’m the one who has to live with my brain.
> A well done monorepo is a huge force multiplier on a large organization
How so?
I work at a company which uses a monorepo, and I haven't seen any upside to it yet. We have a tools team that's invested a vast amount of work in it. Still seems strictly worse than a 'normal' polyrepo setup. I haven't understood why so many people are so enthusiastic about it.
Not op but I do cross repo work, but it's rightfully separate PRs as its separate services, separate contracts and separate deployments. Pretending a monorepo cross service PRs are contiguous is a recipe for deployment race conditions
Is it one or the other? You either need to keep everything in separate PRs or you need to pretend that services are contiguous where deployment order doesn't matter?
1. No submodules. They suck. They don't work with worktrees. They're a pain to work with.
2. Cross-project changes become trivial instead of nightmarish.
3. Testing becomes tractable. Make a change in a submodule? Good luck testing that it doesn't break any of the other repos that depend on it. You essentially turn its API into a fully public API, which introduces a ton of extra work (if you do it right, which nobody does).
Or any of the computer bits in that episoder really. By the way don't watch Golden Boy if there are any people who are not, uh, extremely cultured around.
reply