BillinghamJ

Isn't that... exactly what the Mac Pro was? PCIe support was the primary point

Main distinction I tend to see is just whether you're doing line management or not, which tends to be the EM track

Beyond that, agree it seems like it can just be anything in virtually any title

Systems running core government functions should be set up to be able to efficiently execute their functions at scale, so I'd say it should only restrict extreme load, ie DoS attacks

Then why mince words, and why accept new non-enterprise clients - just be clear it's maintenance-only and existing customer-only.

Seems strange not to just... say nothing and merely remove any mentions of an enterprise offering from the website.

All this blog post can do is make people nervous and lead to customers moving elsewhere. Revenue will drop, and further compound their desire to not invest in the platform. What's the benefit/upside in publishing such an article?

They'd still at least buy reinsurance etc anyway.

All unlimited liability insurance companies (e.g. motor insurers in the UK) have reinsurance to take the hit on claims over a certain level - e.g. 100k, 1m etc.

For extreme black swan risks, this is how you prevent the insurance company just going bankrupt.

Reinsurers themselves then also have their own reinsurance, and so on. The interesting thing is that you then have to keep track of the chain of reinsurers to make sure they don't turn out to be insuring themselves in a big loop. A "retrocession spiral" could take out many of the companies involved at the same time, e.g. the LMX spiral.

Changing/layering architecture adds risk. If you've got a standard way of working you can easily throw in on day one whose fundamentals then don't need to be changed for years, that's way lower risk, easier, faster

It is common for founding engineers to start with a preexisting way of working that they import from their previous more-scaled company, and that approach is refined and compounded over time

It does mean starting with more than is necessary at the start, but that doesn't mean it has to be particularly complex. It means you start with heaps of already-solved problems that you simply never have to deal with, allowing focus on the product goals and deep technical investments that need to be specific to the new company

ChatGPT can easily do Monte Carlo simulation in its "thinking" step, and has done many times for me. e.g. I asked it to compare savings interest between regular banks and median returns from premium bonds. It's not difficult at all for it to do, you can see the code it's generated to do it + the output, easy to inspect

Interestingly, in the UK, comprehensive insurance is now generally cheaper than "third party only" or "third party, fire & theft" cover

The reason is because the insurance companies want you to care about the car as an asset, on the basis that statistically they are driven more carefully (and therefore cause less third party property damage, bodily injury, etc.)

Legit from whose perspective though? Around an authoritarian government maybe widely viewed as reasonable/justified, but doesn't sound legitimate to those governments

So a CDN with... higher latency for people on Earth?

On iPhones, there is an option - "reduce white point" in the vision/display accessibility settings, and also low power mode. You can force low power mode to remain on with the Shortcuts app

How trustworthy really are humans?

Although it has a core section that does run underground through tube-like tunnels, it isn't classified as a tube line :) So isn't part of the "London Underground"

Google Sheets can do that - https://officewheel.com/group-dates-in-pivot-table-google-sh...

Maybe they mean fast in terms of the screen refreshing?

This is actually only an option if you qualify under Apple's definition of "reader" apps - ones that are pure content-consumption, nothing interactive etc.

If you don't qualify as a "reader" app, what you're suggesting actually isn't allowed - you're _required_ to offer IAP if you provide access to the service in-app

3.1.3(b) - "may allow users to access content [...] provided those items are also available as in-app purchases within the app"

https://developer.apple.com/app-store/review/guidelines/#oth...

The app I work on is a content app, but they decided it did not qualify as a "reader" app because we had polls alongside the content, which were interactive and therefore excluded us from the definition.

Password managers can store the passkeys just like they store passwords. 1Password has had strong support for them for quite a while now

I'm not sure why that'd inherently be a problem? Knowledge that some record exists about a medical event at a particular time is not too problematic, compared to _who_ it happened to

It will work - the magnetic interaction happens through the plastic, just will be very slightly weaker than one with additional magnets due to the tiny gap. The video shows it mounted on a MagSafe charger

https://support.apple.com/en-ie/mac/repair

289 EUR for all 15"/16" MBPs

The only reason they'd require a full top case replacement would be if there was additional damage making them unable to guarantee safety of the new battery

I don't know how you'd define "light load", but I have a few services that average about 100rps. Only 1 active container most of the time, they're allocated 128MiB RAM, with a p99 utilization of 55%. They don't have any GC config or special optimization - just operating with regular JSON POST requests, making MongoDB queries

Go just does incredibly well in these kinds of conditions. The only reason we allocate 128MiB is because that's the minimum Google Cloud Run allows - we'd probably go for 64MiB on most systems otherwise

The AI bits seem to be the one feature that is completely useless. I've tried all of the functions and eventually ended up with all of them disabled

One of those board members already did sign!

The year 1 is also completely arbitrary though? Humans existed long before then too. It doesn't really matter when the epoch is, as long as everyone uses the same one, and negative values are supported as appropriate

Isn't the real issue that the control plane isn't decentralized/redundant? Entirely dependent on PDX

I meant in terms of achieving the intended outcome for the end-users. Is it actually desirable to ensure there's a resolution when really they need to decide?

I don't think there's truly any technical solution to "two users edited the same piece of text" tbh. You can try to be smart about it in various ways, but at the end of the day you're still going to hit situations where it's really not clear what the users' intent would be to handle two mutually-unaware edits on the same words

Git has much more explicit conflict resolution though, which covers a lot of the risks involved

Despite that, Notion actually does work okay in unstable/briefly offline connectivity situations. It syncs changes when it reconnects.