The first half of the article says "namespaces, cgroups, and seccomp aren't 'security boundaries' because if the kernel had a bug it could be used to escape from a sandbox". Then in the second half it says "use gvisor and do all this other stuff to avoid these problems." This presentation feels kind of dishonest to me because the article avoids acknowledging the obvious question: "well what if gvisor has a bug then?" I mean, sure, another layer of sandboxing that is simpler than the other layers probably increases security, but let's not pretend like these are fundamentally different approaches.
If the successful people you meet in real life are nice and you see lots of meanness on the internet, it probably just means that anonymity causes meanness.
Or that the internet is full of mean losers. Your premise assumes anyone with an internet connection is successful and I very much doubt that assumption holds water.
Not really - I've seen some pretty mean people back in high school, because high school is one of the places where people are forced to be together, no matter what their preferences are.
But once one becomes an adult, there is a much greater leeway in choosing whom to interact with, so it is often possible to not interact with mean people at all.
Wait. There are people who aren't individually educated by private tutors while they accompany their parents on travel, and at their summer, and winter homes throughout the globe? /s
> We already know many useful things to do; there are already 10,000 startups (9789 out of YC alone, 4423 of which are coding-related) doing various ostensibly useful things.
Val.town seems to be serverless, where as this is explicitly a server. One is really a subset of the other though, so I suppose if you're deploying ts functions to a service/server, and your execution costs match up with the tiers here, exe.dev could be cheaper.
Learning homological algebra is a long way away from learning about varieties and ideals in any sensible order in which to learn material. You have to do a lot of work in between in order for homological algebra to seem motivated. Sounds like you ought to try an introduction to algebraic topology.
> Well, dhh considers torrents outdated (I’m not kidding, check the tweet), so it’s only officially being offered as a single download from a Cloudflare server. Which sounds cool until you’re on a weak-ass connection in constant danger of dropping halfway through the download.
That has nothing to do with bittorrent vs http; use a download manager instead of a browser.
Don't you think that the vast majority of dangerous human drivers would be perfectly capable of changing their behavior during a driving test? Even without any malicious intent most people would be more careful during a test.
I want a camera on every traffic light and stop light - or better, cameras on a random 20% subset of intersections. The system would automatically flag infractions for human review. Combined with docking points off people's licenses and/or fines based on income/wealth percentage, this would be a decent deterrent.
> targeted specifically high ticket items (jewelry, luxury)
My immediate instinct upon reading this is "of course those businesses wouldn't be interested". They are selling a brand and image first, and if it looked like they wanted people to download some new app and put in their bank details to make a purchase instead of swiping a credit card it would cheapen the brand. Luxury stores are all about a streamlined experience and the appearance that money is no issue.
Yeah. High-margin, high-end businesses can afford the fees. If there's a space to compete it's low-margin businesses where 3% really stings, even if it's 3% of a smaller gross. I used to work with a company that served mainly immigrants and blue-collar workers, and something lower-fee than credit/debit was a constant conversation (we were partway through building an ACH backend by the time I left, but I don't know if that ever made it into production given concerns we had).
We tried with SMBs (it was our original focus, low net margin businesses). But 1. Deployment times are way worse in person, 2. Merchants still actually didn’t care (they said they did! But there was no strong need). 3. conversion at checkout is lower for small tickets than high tickets.
SMB business model just didn’t work, with tons of white glove customer support needed for a customer we’d make $20-$50/month in fees
With high ticket sellers the purchases are high intentionality enough that you can convert customers.
Not only only low margin but even more low price products. If you sell ice cream for $1 there is not only 3% fee but also minimal transaction fee like 30cent.
I recently made a purchase in an Ontario used book store that only took debit cards and cash (and the clerk was behind a COVID shield). Today I went to a used book store in Buffalo, NY that took credit cards (and the clerk was roaming around the store). Somehow that openness and lack of friction got me to buy more books for much more money. And I would gladly return...
I think when the guy mentioned he studied past failures it's prime example of crunching the numbers and applying theories without actual common sense. "Ah yes, I wouldn't buy a Gucci bag for $999 but for $995 it's a deal done" you don't need PhD in economics to realize this is stupid.
The fact that the second part of his comment is full of acronyms "we ASDFed the GHIJ over KLMN" suggest it's yet another case of a manager completely detached from reality of average consumer.
Haha i wish i were just a bean counting manager. A number of venture backed startups have tried exactly this idea, and it was a tarpit.
Confused about your hate of acronyms. The bank rails are a practical limitation on how effectively a startup can compete in this space. Do you feel the same hate when people use technical terms about coding eg “we deployed with Kubernetes on AWS”?
This is ecom, not in person, and we found these places are brand agnostic about payments— they don’t control the look/feel of PayPal, Klarna, etc. And for every Hermes, there are a dozen luxury merchants less anal about the brand.
White labeling the payments infra as “their pay” did intrigue them, but not enough
