sure, instead of 50 kids with OLPC, you could give 1 mac. Also, what you do with a mac without internet and little included apps? no brew install, no new apps (and fewer free apps, most that exist are paid apps)

>>i don't care if 1/4 of the internet goes down, i care about my site.
>To the 1/4th of the Internet that couldn't reach your site, as far as they care, your site is down

i actually mean that my site is down 100%, that is what i care, i don't care other sites being down (unless they are a requirement for my site)

>>do the CF downtime was bigger or smaller than a downtime on my side?
>Bigger. One is inclusive of the other.

again, you didn't understood what i mean, i'm saying that a downtime in my site due to CF is probably shorter than a downtime on my infra if i tried to replicate cloudflare on my infra

>Which would be better if 1/4th of the Internet could use that bypass config to reach your site, but they can't, their traffic routes through CF first and can't get anywhere else.

ignoring the 1/4 traffic, as i pointed above, but related how to bypass cloudflare.
right now my www.example.com site is a CNAME to www.example.com.cdn.cloudflare.net and them i have in cloudflare that www.example.com points to my backend ip. To bypass cloudflare, i can point my original www.example.com to the backend directly, bypassing cloudflare. yes, i will expose my backend IPs, but we can either change it or firewall block any traffic that isn't cloudflare after restoring the original config

exactly!! :D

can i see your code to see what assumptions did you make on early stages, that you never bother to go back and fix?! :D

who said that that flow (dev->qa->staging->prod) didn't happen?

how many people had bugs in prod that didn't show up in the previous steps? how many people had problems only hours after deploying to prod?

you can't not always test all conditions, dev and qa may not have the amount of users/access/info to really replicate a problem, staging may read the prod DB, but not trigger the conditions, maybe they are intermittent, require a special corner case, etc

in this case, a prod DB change is ALWAYS harder to test, and even if you replicate the DB, make the changes and still get the prod updates in real time, you may not trigger the issue. As they said, sometimes the file reach the 200 entries, other time it was lower. Everything can be fine until certain problem show up and you end with a cascading, runaway problem. staging may not have real life load for that runaway problem to show up!!

even adding canary deploy or traffic shifting deploy could help, but we already know that cloudflare already do that... but again, DB is harder, most DBs can't have a cluster of multiple masters with different configs, so it is hard to have a master DB in the old config and a master DB in the new config so both app versions can use their own DBs... much worse if those DB need to have performance!

so talk is easy, but there is not perfect solution.

Several years ago a datacenter went dark because the extremely redundant multi-ups, generator setup, different power sources, very complex setup that was build to never fail... failed due to a small problem that a simple setup would not care, but the extremely complex setup cause a small cascading issue and in the end totally failed. Yes, it may be possible to fix that, but maybe we are adding just another layer that can act strange and cause itself another cascading problem.

Similar the Iberian power failure a few months ago, all systems were distributed, redundant and like... a small issue and cascaded out of control

you can not predict all possible problems and the more complex the system is, the harder is to predict and control them

i don't care if 1/4 of the internet goes down, i care about my site.

do the CF downtime was bigger or smaller than a downtime on my side?
can i replicate their features even, to get a similar service?

So while this downtime is always bad, all sites have some downtime... maybe i was luck , but was little affected by this
either way, no, i can not really replicate the cloudflare solution locally, the costs would be huge, more people, more servers, more knowledge and still would not reach the same level. Just the bot detection and CDN is more than worth the downtime... that again, i would also face sooner or later on my side, with my own solution

sure downtime is bad, but we can also mitigate that, we can have a fallback config that bypass cloudflare (domain must not be hosted in cloudflare, of course, so you can change it) if something goes very wrong. people with more money can have even multiple cdn

as for the rest of the world... go outside, touch grass! it is not the end of the world you can't talk to chatGPT or view some tiktok videos, accident happen, people should adapt, relax and move on

well, what the conference was simply telling is that people that brag they own a fortune in crypto are putting themself at risk... and we all know some people that do that, right?! :D

toyota hybrids are perfect (yes, complex beasts, that i agree), but they work fine and zero problems ( i know, i own one with 13 years)
most other brands just rushed to hybrids and fall in to mistakes and problems... but even those improved a lot

cash, gold and jewels are physical and can have size limits to stole in bit quantities, bank transfers take time and may trigger alarms or blocked, crypto can be stolen quickly, transferred to multiple accounts and get some cash out before even someone suspect you were kidnapped. that is why they are better targets

while some people were banned for using wine/proton, most were by accident, all cases that i know were reverted after some noise... some were really banned for using cheats and trying to use linux/wine/proton to hide it

getting kicked from not having the anti-cheat working in wine/proton, sure... ban usually require much more (unless someone adds a stupid check or use AI ... but like i said, accidental bans)

Also, you can know the status of the anti-cheat in linux for each game before even trying: https://areweanticheatyet.com/
so don't buy games that do not work, complain to the game owners to report that you can't play because of the anti-cheat and go play something else, there are many great games out there that work fine and are probably better!

True, car manufacturers were sleeping on the wheel, they kept copy&paste their (and other) models, without little innovation.
Hybrid cars were a detail that most ignored, Tesla was a alarm ringing, time for them to improve...yet they look only to top class and quickly it was saturated. Where is demand, lower and middle classes, there were too few alternatives. Now Chinese cars, "same" but cheaper, is a big slap in the face.

Big car brands are too slow to innovate and that creating a market for new brands that can have simple processes.

Is like Nokia phone development, full of process, analysis, tests, market segments... them apple launched the iphone and any valid response required too many development years just to start testing.

but the reverse is that all those plan and test increase quality and quality is good. Toyota even created what is considered now a standard in quality control
new brands skip many of those and them you have tesla headlights glued, things breaking, etc

Chinese for sure also have quality problems, worsen by the fact that many workers (and factories) do cut corners to increase production numbers

Cars build 30 years ago were much simpler and cheaper to build, but doing the same today is not valid due to security, quality and expectation. Tada Motors have very cheap cars... are they safe? do they work after 2 years? not really

that is what they are selling... but is it true? (not the price, while with government funding, the price is also fake, we will pay that later, when there is no alternative)
almost all Chinese hardware of several kinds i get, the announced battery life is totally fake with real world usage. Some the battery lost almost half capacity in one year.

in some cars reviews, i also read that real live range is much lower... but all cars like to lie in their real range. we need more real work metrics and those only show up with time

why are you even here?! go make your own news!!

news for nerds. stuff that matters!
while in the USA apple rules, the rest of the work prefer not wasting their money and wants good android table for the money... problem, that is a rare hardware, the race to bottom means you have many cheap and bad tables and few good ones. This is a valid question. Samsung and Lenovo are taken usually as the last good enough tablets, but with some downsides... maybe other people know other alternatives, specially ones that can be upgraded

people buy laptops, many times not for the mobility but as a good setup that is pre-build, works and doesn't take much space
people buy consoles, to plug in the living room and not have to worry about updated, virus, etc, just turn on and play

can you trust that Chinese car to drive from Romania to France? how many stops you will need to recharge?! are there enough chargers in the path?! too new to know, to many unknowns. They may even be good, Chinese quality is increasing, but is a too high value for most people to bet. My city car is 25 years old (not that much used to be replace by a EV), the family main car is 10 years (hybrid), both still good, low cost maintenance... will a Chinese car last that longer? at what maintenance price?!

for full families and long distances, you really want a different kind of car, again, hybrid or hybrid-plugin, not EV