The rest of the FAANG has invested very heavily in cloud while Apple seems to be a laggard. GCP, AWS and Azure are all publicly available products, and cloud at Netflix, Meta seems very mature for a private offering.
This is not a huge disadvantage in my opinion. Let the rest of big tech fight each other to death over cloud, while controlling a very profitable differentiated offering (devices+services). Apple keeps the M series HW out of data centers, even though it presents some very attractive performance/w and per-core numbers.
I think you're correct on it not being a disadvantage. Apple's competitors are the Android OEMs, Microsoft, and Dell. Apple Intelligence is a failure only in the sense that we hold Apple to a higher standard. Can anyone argue that Apple's AI implementation is more flawed than Microsoft? I don't think so.
Market for DDR4 is crazy, but not as crazy as DDR5.
Also a symptom of how inelastic hardware demands are. You would expect the purest k8s people to just shove workers on older machines and completely dodge this crisis, but we don't see that at all. Despite being an almost-commodity, many of the hw vendors still have a decent hand.
As they say, the rules are written in blood. I don't think we should be disqualifying projects because they are not Mponeng-scale or complexity.
I am not a civil engineer, but I did spend a bunch of time looking into building an underground range. Way more relaxed life safety reqs, smaller bore, etc. However, when you start reading, it is clear that much of the work is empirical, heavily localized and based on a great deal on the experience of the builder. I found very little in the way of solid theoretical modeling, but lots of measure, adjust, etc.
I think Grady does a reasonable job highlighting the dangers and risks.
The only time theoretical modelling happens in most civil engineers careers is at university. Especially so with geotech/soil mechanics - it is the most inexact and variable/unpredictable area especially when water gets involved.
Structural steel would be the most predictable. Concrete and timber are in the middle somewhere.
Basically nobody ever died from leaky pipes or substandard weatherproofing. The code is as much about a) homogenizing the industry so big business can statistically reason about it at scale b) turning the subjective into the quantitive so that things can be done, checked, sight off on, etc, etc, without anyone using "judgement" as it is about protecting life and limb. Just about every professional has a laundry list of complaints about their area of code that boil down to it being theoretically useful but at great "not worth it" expense or a similar "not worth it" expense being incurred in lieu of very basic judgement. Arc fault breakers, and engineering requirements for small retaining walls come to mind as oft cited examples. And of course there's the myriad of wrangling that goes on wherein things get looser/stiffer requirements depending on whether their use is deemed worth incentivizing (this stuff usually lives in local addendums to the code).
I'm not saying there isn't value in there, but this habit people have of acting like it's all relevant to safety and screeching about "written in blood" is exactly what creates room for unrelated stuff to exist in the code.
>However, when you start reading, it is clear that much of the work is empirical, heavily localized and based on a great deal on the experience of the builder. I found very little in the way of solid theoretical modeling, but lots of measure, adjust, etc.
Which is a point very much in favor of the amateur.
I know you're probably intending to only remark on leaky water pipes, but:
The New London School explosion was caused by a leaky pipe. It killed 295 students and teachers, and led to the inclusion of smelly thiol in natural gas, as well as the Texas Engineering Practice Act.
A drip or stream from a leaky pipe isn't gonna do it. You need to get the bacteria into the respiratory tract to get legionaires disease. And even then a specific temperature water is necessary for it to grow.
A dehumidifier (or an HVAC system, which is where the name of the disease came from) is more likely to give you legionnaires disease than even the most substandard plumbing.
> Basically nobody ever died from leaky pipes or substandard weatherproofing
Famously, moist wet areas only grow molds that are safe for humans to live amongst, and absolutely never rotted away wooden structural components of a building.
Sounds like a tacit admission that those bits are written in rotten wood rather than blood.
The part I take issue with isn't the building code in principal or that non lethal things can be regulated. It's that the people being to lie though their teeth and pretend it's all written in blood when a whole bunch of it isn't are essentially stealing credibility from those parts that are. A few bad apples (handouts to industry) spoils the bunch (very clearly important stuff, like floor and roof loading).
A lot of regulations are also for consumer or user protection - not just for the first owner, but later owners too. Substandard waterproofing may not be dangerous, but it does bankrupt people and ruin lives.
> I'm not saying there isn't value in there, but this habit people have of acting like it's all relevant to safety and screeching about "written in blood" is exactly what creates room for unrelated stuff to exist in the code.
meh, I understand the point, but it is about your risk tolerance being different than whoever writes the code. I have a long list of complaints about the NEC, (including AFCI requirements), but IMO, these kinds of requirements do save some amount of lives -- the issues comes down to how much do you value your own life, and/or the lives of others. The tradeoff, as always, is cost -- inspections, permits, impact studies etc push up the cost of new and remodel jobs substantially.
Where I really take issue with different code is when we hammer down on a specific issue of small significance while neglecting a more significant problem. For example, I have never in my life seen an inspector check the torque of a main lugs, polaris connectors, etc. Might just be my inspectors, but I have seen way more failures due to loose or over tightened connections than anything else.
I am all for gradually raising the bar for safety, but it has to rise faster than the increased cost, along with a level raising of the bar across all facets.
This is hackernews, do the math for the love of god.
There are good business and technical reasons to choose a public cloud.
There are good business and technical reasons to choose a private cloud.
There are good business and technical reasons to do something in-between or hybrid.
The endless "public cloud is a ripoff" or "private clouds are impossible" is just a circular discussion past each other. Saying to only use one or another is textbook cargo-culting.
Ever seen a default ubuntu splash screen/wallpaper on a train, coffee machine, airport terminal kiosk, bus, or other big piece of slow moving, appliance-y thing?
That is why Ubuntu Core (and similar) exist. More secure, better update strategy, lower net cost. I don't agree with the licensing or pricing model, but there are perfectly good technical reasons to use it.
I have been arguing this point for several years now -- but wrt to the Democratic party's relationship with guns. The same justification used to limit the second amendment is the same justification that can be used to limit the 1st, 4th, etc.
Both parties seem to be on an authoritarian bent over the last 10-15 years, which sucks.
bah humbug. In the age of LLM edited text, having a distinct style (be it written or spoken) is a breath of fresh air. AI has already ruined the em dash, no need to let it ruin your unique voice, style and cadence.
Hear, hear. I used to dislike low effort replies and comments that were written hastily and contained spelling/grammar mistakes, weird turns of phrase or even feeling incomplete like two different thoughts mashed together and presented as they happened to be. Now I just enjoy them, flawed as they are someone bothered to "hand-craft" them.
The US has lost multiple wars to goat herders in pickup trucks with small arms.
As Ukraine has demonstrated, a shaped charge and consumer drone is highly effective against even heavy mechanized armor. ERA doesn't work well for multiple hits, and drones and HMX/RDX are cheap.
reply