Hey, I store recipes on my home computer! Having a portable handheld terminal that can view the recipes makes it much more practical than it would have been in the 80s.
I keep my bread recipes in Google Keep on my phone. It's extremely useful, since the phone takes up much less room on the counter than the laptop does.
I have a fanny pack. I usually put my phone, a notebook, my wallet, some band-aids, and a couple diapers. Sometimes I add a charger if I think I'll need it. It's quite convenient, and I basically don't put anything in my pockets. Phone sits on its charger or in the bag, usually.
This misses the fundamental information problem. Your recommendation algorithm is centralized—it only knows what its signals can measure. Ads create a decentralized market mechanism where businesses themselves can signal “your algorithm is underweighting me.”
Consider the failure modes of pure algorithmic ranking:
Cold start problem: A phenomenal new restaurant opens. It has no ratings, no historical visit data, no repeat customer signals. Your algorithm buries it. How does it escape this trap? Organic discovery is glacial—it might take months to accumulate enough signals while the business burns cash.
Structural bias: Your algorithm might systematically underweight certain business types. Maybe sit-down restaurants generate longer “time spent” signals than excellent quick-service spots. Maybe your visit detection misses certain building types. The algorithm doesn’t know it’s biased.
Local knowledge asymmetry: The business owner knows their value proposition intimately—they know their recent quality improvements, their new chef, their differentiation. The algorithm is looking backwards at historical data.
Network effects lock-in: Once a place is highly ranked, it gets more visits, more ratings, reinforcing its position. Even if quality declines, the algorithm is slow to react.
Quality-weighted ads let businesses with superior local information challenge the algorithmic ranking. If you’re genuinely better than your algorithmic position suggests, you can bid to prove it. The quality weighting means you only profit if you’re right about your own quality—it’s costly signaling backed by conversion economics.
This is “outside-in” because you’re not trying to perfect a centralized algorithm. You’re creating a market mechanism where distributed information surfaces through economic incentives. The businesses that are most undervalued by the algorithm have the strongest incentive to correct it.
Pure algorithmic ranking is central planning. Quality-weighted ads are a market.
I think these are the salient concerns I've faced at work using pgvector. Especially getting bit by the query planning when filtering -- it's hard to predict when postgres will decide to use pre- vs post-filtering.
As for inserts being difficult, we basically don't see that because we only update the vector store weekly. We're not trying to index rapidly-changing user data, so that's not a big deal for our use case.
A gas station? No, they'd be selling prepackaged sodas in the 20-oz size.
You might be able to do it at a 7-11, since they sell empty cups that you're meant to fill with a slurpee. I don't know if they also have soda fountains to fill those cups.
It's a reason why you and your organization should depend 0% on open source software because any open source software could be ruined by a bunch of people who don't get along.
In my mind DHH and the people who are still fighting with him years later are all bad people and if I had the power I would cut them off from the net completely, not even let them have a credit card.
It seems plausible that the value accrued by any given extreme response to binary [discrete-classed] emotions is upper-bounded by zero
One good thing about opensource: there's realtime leakage of how things are run so wiseguys have time to fork and/or pivot.
Some people factor that attention premium into the routine cost of a sw business. Because the other scenario-- the downside of depending on proptech-- is not recoverable without a team of lawyers
I find it heartening that lack of backprop from reality to character is (albeit not always career ending) still generally relationship-ending..
Maybe similar mechanisms can be developed wrt intra-/inter-org relations.
My pediatrician always charges us for an office visit + preventative care when we go in for a preventative care visit. It's obviously to get more $$ from insurance. I feel like this goes both ways...
Yeah enough gets talked about insurers acting in bad faith, but let’s not forget hospitals also acting in bad faith for their end. Some personal examples:
1. Sitting in a Urgent care. They get you in the exam room. You sit there for 15 mins, doctor comes and sees you for 5 mins (mostly rushes the exam), do a blood draw, ask me to sit around while they run the test, doctor leaves, as soon as 45 mins are over the nurse comes over to let me know it’s taking longer to run the test so I can go home and they’ll call when the results are out. A month later charge thousands of dollars to insurance for a 45 min Urgent Care visit that doesn’t cover the lab work.
2. Go to PCP with cold symptoms that haven’t cleared in 10 days. I insist it’s a sinus infection, they send me back with no antibiotics and ask to schedule and online appointment in 2 days. I insist I come in in person, but they schedule an online appointment anyway. Nothing gets better and I see the doctor online after 2 days, they say I’ll have to come in so that they can evaluate me in person and prescribe antibiotics. I go in person, get antibiotics and get cured. Insurance gets charged for 3 separate hour long visits ($750 each and none of them lasted more than 10 mins).
Or it was viral after all and you cleared it on your own.
Doctors who specialize in this have a hard time accurately distinguishing bacterial infections from viral. There’s no reason to trust your own opinion on the matter. It’s too easy to fool yourself.
If doctors prescribed antibiotics to every person who came in insisting they have a bacterial infection, we’d all be in for a bad time.
> Go to PCP with cold symptoms that haven’t cleared in 10 days
It was a bacterial infection. That was the correct diagnosis. Flu (viral) doesn’t get progressively worse after 10 days and then get better immediately after a couple doses of antibiotics. My symptoms were in line with a sinus infection (I’ve had them before just like I’ve had flu before) and even if they are not able to diagnose correctly after 10 days, there are other tests that can be prescribed that weren’t and there was absolutely no reason to schedule an online appointment when they clearly knew that they’d need an in person check anyway.
There are viruses that can last 2 weeks and mimic bacterial infections.
Most cases of bacterial infection will also clear on their own after 2 weeks.
There are no good noninvasive diagnostic tests to distinguish bacterial sinusitis from viral because is the presence of normal
nasal flora.
The standard of care is to consider antibiotic treatment after 2 weeks of symptoms for adults and 3 weeks for children.
There’s a reason for these standards:
As of a few years ago physicians were prescribing antibiotics for 80% of cases of sinusitis. Despite the fact that only about 1%-2% of cases actually needed it.
20% of antibiotic prescriptions in the US are for sinus infections.
This is a massive contributor to antibiotic overprescription, which is why the current criteria is 2 weeks of symptoms.
I don’t know what happened in your online visit, but scheduling an online visit and then if your symptoms persisted past 2 weeks prescribing antibiotics during the the online visit would have been entirely appropriate.
I also confirmed with a MD friend of mine, 10 days or more of worsening symptoms could be indicative of bacterial sinusitis. For adults you can begin the treatment on that diagnosis. You can also start antibiotic treatments before 10 days in certain conditions but it is generally not recommended. Just because a patient shows up on 11th day with worsening symptoms doesn’t mean you have to wait 2 weeks.
Sure. It’s a guideline not a hard rule. Doctors have wide latitude. A doctor could give you antibiotics whenever they want.
The doctor didn’t do anything wrong by asking you to wait a few days.
In addition to residency, my wife is fellowship trained, she’s who PCPs end up sending patients to. She’s basically level 2 tech support (an ENT would be level 3 for this particular problem). She sees the results of this stuff all the time.
And the majority of PCPs vastly overprescribe antibiotics for sinus infections. The vast majority of patients who come in saying they have a sinus infection and asking for antibiotics are wrong.
Doctors are tasked with antibiotic stewardship, but people complain on review sites when they don’t get what they want, so many of them just do it.
An obligation to pay is always good for the billing side. Think about the sociopathic prices of US pharmaceuticals.
Afaik any other country with mandatory health care also puts a ceiling on prices. In germany, there is a price catalog for any service, with only few exceptions, and doctors/hospitals cannot legally charge anything else for these covered services. Now guess what the US does not have, even thought obama had foreign consultants explicitly advising for it.
Health ensureance companies are certainly not the most altruistic but any profit oriented company trying to cut cost where ever possible is hardly a supprise.
> In germany, there is a price catalog for any service, with only few exceptions, and doctors/hospitals cannot legally charge anything else for these covered services. Now guess what the US does not have
Well, we sort of do: we have Medicare's reimbursement rates, which are indeed a price catalog for every service... but only if you're covered by Medicare, of course.
I've heard that price negotiations between private payers and providers are often done with reference to the Medicare rate: "I'll pay you 20% over Medicare for this."
reply