thanatos_dem

I am having a hell of a time trying to parse that sentence. Can you rephrase?

Can confirm, grandfather died from a heart attack when shoveling snow.

From the article:

> The Apple Watch also alerted Gabe Burdett that his dad had fallen.

What a terribly written and overly editorialized article.

A bit concerning to see infrastructure companies getting involved with politics tho.

Seems like the poster meant that going after private prisons was the easy PR win, and that going after the unions would be more difficult, but effective at addressing the underlying perverse incentive structure.

Apparently Apple has given up on the whole “common sense thing” that the touted during the iPhone 5 days - https://youtu.be/O99m7lebirE

I don’t see an option I don’t need to wear, unless I’m looking at the wrong thing.

I have the same issue as the original poster in that I find wearing something on my wrist while I sleep uncomfortable. The Beddit is a strip that goes under my sheets. So I guess that’s why not.

Wait, it isn't 0118 999 881 999 119 725 3 any more?

I use a Beddit for this very reason - https://www.beddit.com

Maybe remove “Apple” too in that case?

Finally a new and refreshing take on the MacBook keyboards.

> Human configuration

Your engineer is showing

Yeah, killing inbox moved me away from google for mail entirely, and after that the dominos started falling. The only google product I use regularly at this point is YouTube, which I pay for to avoid ads.

How about java projects? How can I add a dependency to a project I’m building without it being published to a maven repo, for instance?

Somewhat related - when I was working at Union Pacific in 2011, there was a hugely expensive effort to replace the old 1024x768 monitors in the locomotives with 1080p equivalents, including redesigning a lot of the UIs to work in a wide screen form factor.

After the rollout completed, all the feedback was negative, saying that the fonts were too small to read on the new displays, especially when the locomotive began vibrating at speed.

In the end, the whole fleet had a software update pushed out that set the resolution back to 1024x760, and all the UIs looked stretched. The engineers (train, not software) were ecstatic.

Lesson to take away is that with user interaction, it’s important to look at how humans will actually be using the system. That seems self evident, but given stories like these clearly isn’t.

Yeah, unfortunately the strongest plausible interpretation of that comment is not that great

Except at this point humans are unquestionably the apex predator. What animals are we going to need nukes for? This is humans fighting humans.

Unless your argument is that there’s a distinction between classes of humans comparable to that between species, which is a... problematic position, so I hope that’s not what you’re claiming.

How often do you find yourself writing or using `interface{}`? It’s littered all over the go code I’ve seen, including the standard library.

`interface{}` is the equivalent to using `Object` in java land. It completely punts on any sort of type safety, relying on the developer to handle messy edge cases correctly. I think one of the only reasons that it hasn’t been as severe a problem in go as java is thanks to that explicit error checking that go makes developers do (and that gets so much hate online), forcing developers to consider the case that the type isn’t what they assume.

But genetics provide a useful way to safely make those assumptions, and cut down on the boilerplate around them.

Not just frameworks, but the JVM as well. Having go compile to a reasonably small statically linked binary (which can be shrunk even more by things like UPX if size really matters) is great.

Needing to bring a very large, very complex virtual machine to execute your program adds a lot of “fatness” IMO.

I really wish go code didn’t have so much reliance on interface{}. I know it’s not the fault of this library, but even once generics make it in, it’s going to be a nightmare to unwind all the piles of interface{} scattered around.

As far as I’m aware there is no explicit holdout period on reads, however, on read you do need quorum of the range holders for consistency sake. The exception here is if you’re talking to the “leaseholder” of that data, in which case it can respond to you directly.

If you have a workflow that consistently exercises the same ranges, leaseholders will be moved to the nodes that you’re connected to, in a process called “follow the workload”.

If you run a load generator against a cluster you’ll see that - for the first 30-90 seconds or so it’ll have some latency, but after some period of time, the leases should be moved closer to your generator and you should see latencies drop.

When you’re looking at data access patterns and partitioning you should keep this in mind, as it can be a big performance improvement if properly considered.

NTP is plenty accurate for Cockroach to work correctly.

> What happens if I'm monitoring and see an elevated clock-skew?

If you have monitoring that can detect that, NTP should also be able to run and correct it, apart from some very weird failure modes of the network itself (like traffic dropping in one direction).

> I think CockroachDB will be amazing when cloud providers start offering APIs for High-Resolution Clocks as a service

With the hybrid logical clocks that cockroach uses (basically wall time + vector clock), they don’t need to be all that precise. The trade off it makes is giving up linearizability (aka external serializability).

Spanner gets that by making every transaction wait for the maximum clock skew (7ms, with their atomic and GPS clock systems). But in order to run on commodity hardware Cockroach forewent it, in favor of allowing more acceptable skew.

> What was different about Huawei? Except that it was a Chinese company.

Well that and they blatantly steal from other companies, disregard copyrights and patents, and have the backing of the Chinese government protecting them from any real repercussions.

They don’t even try to hide it with their product naming. Take the “MateBook X Pro” for example - https://consumer.huawei.com/en/laptops/matebook-x-pro/

Ah, yes. Blindly dismissing ideas you disagree with. That’s how we improve political discord.

Yeah, pretty similar overall. Tho Tekton is also the underlying technology for Jenkins X. With Jenkins being such a common system in the CI/CD space, I think we'll see Tekton preferred going forward.

https://jenkins-x.io/

> Not even the battery thing.

That was an attempt at graceful degradation that wasn’t communicated well.

The general belief was that having a phone that runs a bit more slowly, but consistently used it’s battery as the battery life naturally degrades is better than a phone suddenly dying with 40% battery remaining if the peak power draw exceeds what a several year old lithium ion battery can produce.

I much prefer the former personally, but I get where the misunderstanding and public outcry came from.

Beyond that, what made your phone unusable? The last 4 or 5 iOS versions have improved performance across the board, even for phones several years old.

If it’s not branded as a google product that sure reduces the perceived risk of it being cancelled out of the blue one day.

Eek. The content is interesting, but the channel’s style and sound effects are just too much for me. Same problem I have with “Mad Money”.

I believe it uses iCloud/CloudKit to sync data from all devices associated with your account.

It’s a bit more messy than that in California even. Non competes are not forbidden, just generally unenforceable due to some other labor laws. So a lot of companies will still to this day have employees sign non-competes, and threaten them with legal action if they violate them.

They’d almost certainly lose, depending on circumstances, but the threat is enough to have a pretty strong chilling effect.