georgebarnett

Agree this is where the law is now.

It’s an interesting thought experiment to wonder if we would have got here if Google News wasn’t a giant pile of summaries.

If Reuters lost a click because somebody wasn’t interested enough to follow the link, sucks to be them.

If they lost the click because the article was summarised and you didn’t need to click through to read it then that’s on HN.

This is less obvious for Google news, but substantially more obvious for other search snippets (like say looking up synonyms).

Everybody appears to be holding a foot gun here. I’m very much looking forward to seeing which one goes off first.

The outcome is uncertain, but it should be fascinating to watch.

HN doesn’t provide snippet summaries of linked articles so you don’t have to click through.

I continue to be unsurprised at the number of people who are surprised to learn giving their contact book to a random internet app is a poor idea.

Perhaps in the early days, I’d understand however by now there’s a LOT of evidence to suggest that it doesn’t end well.

As an aside, Garret Heaton is one of the co-founders of HipChat. Congrats on another exit mate!

Every company has incidents of customer support nightmares, But only Google ends up at the top of HN on a monthly basis.

Amazon, Microsoft and Apple have similar numbers of users and there are no issues getting in touch with them.

Google chooses this path, it’s not forced on them.

I can sign up to AWS and get world class service on day one with just a credit card.

That’s the bar.

Many other companies of similar size manage to provide customer service just fine.

This is a solved problem - you just have to be willing to realise that magic AI sprinkles aren’t the answer.

As for cost - this continual stream of screwups is costing them a ridiculous amount of goodwill and future business. It’s probably the best ad for AWS there is.

Why would I use an unlicensed app with no description whatsoever about the company other than a generic website with up and to the right charts?

I don’t know what your intent is but this doesn’t fill me with any confidence whatsoever.

If you took this seriously, you would wait until after you’ve met regulatory requirements to launch.

Weird. My 2016 12” MacBook is still almost my favourite machine. I say almost because the M1’s speed is _incredible_, but I do miss the smaller size and flatter keyboard.

With your actions, you build the world around you.

Saying you would take socialised medicine and yet prioritising actions which oppose this goal mean it’s not what you actually want and so you’ll never get it.

The whole reason the cost is high is because there is scarcity.

You’re not “getting the rich to pay” - you’re preventing the poor from accessing the drug, placing vulnerable less wealthy peoples life’s in danger because you prioritised doses for people who can pay (who would _get it anyway_ if they had a risk factor).

You’re literally trading a risk on one persons life for somebody elses money.

Not to mention the moral issue of “how many doses are set aside?”, because why not sell all of them for $25k if that’s the price you can get?

I find it ironic, that in a nation that complains so bitterly about its healthcare system and the associated costs of care, there are so many willing to exponentially increase the costs of an item just because people need it and some can afford it.

Your approach baffles me.

Governments have already paid for the production. Rich people paying hospitals would funds hospital coffers and not go further.

If you think the states should get the money, then taxes are a better approach.

Giving this to rich people does nothing to further immunity in vulnerable sectors of the population - it weakens that goal.

Why is the life of a person with money more important that somebody without?

Why is this an important signal, given healthcare teams have already decided on a priority order?

Whether or not a protocol can be achieved using a certain method is a valuable outcome. Write it up so your peers know what didn’t work.

If you have time to do the work to get a pilot project off the ground then you absolutely have time to do a short write up on how it failed.

That assumes that you can move off the system and that’s enough notice. Not to mention that it blows up your roadmap.

What’s the point of outsourcing heavy lifting if the vendor is just going to surprise drop it back on your at an in opportune time?

Why should private things stay private, only for politicians?

Big tech are installing datacenters to service users.

Servers have to go somewhere - would you prefer it to be on dirty energy far away from you so it both pollutes and provides poor service?

Datacenters are rediculously efficient (bigger ones even more so) compared to alternatives.

I'm in the process of switching a lot of my work onto iPad.

I don't see a conflict here - I'm still using a "compiler" api, its just somewhere else (i.e. not on my local device), its just in somebodies cloud. I can't see why that would go away in the future and I don't really care where the compiler actually is (if I did, I could self host).

The idea that governments should have a ready response for every circumstance is the height of human hubris.

One would think that given there were readily available no pre-canned global responses, coupled with the fact NOBODY can predict the future, faking it/making it up is the only available option.

Not to mention, one could also think of the haphazard response as “exploring the problem space”.

Under those conditions I'd agree theres likely a high overlap of similarity. I have a use case on my home playpen that would likely be served well by that. Thanks for describing your approach.

Why are you comparing HBase with Postgres? They are very different technologies with completely different architectural constraints?

There are metal bars.

Why do you think the team at the tunnel would go to such lengths and expense if there was a simple solution?

Yes, I completely agree with your assessment. For this reason, I prefer separating workloads. Cloud providers make this really easy with managed sql and key value datastores.

In datacenter there are good options using high availability storage arrays and vm systems. I don’t see that kube was ever meant to replace that for this workload type.

I know some have this overly simplistic idea of “one platform only”, which I don’t agree with. I see no reason not to use the correct tools as appropriate for the task at hand.

There’s some interesting ideas in this blog, but also a bunch of weird logic assertions which sound like the type of statements you’d get from somebody trying to sell their thing.

    In case that wasn’t clear,
    you can’t containerize an 
    appliance. That means you
    cannot orchestrate an 
    appliance. That means you
    cannot adopt Kubernetes if
    you keep buying appliances.

Why do you need to containerise an appliance? Containerisation is useful for abstraction of hardware, but that presumes the goal of a homogenous fleet. In the cloud you don’t care and in a datacenter you separate the workload. So long as there’s an API you can poke to manage the thing, you’re fine.

Likewise, you don’t need to containerise something to have kube orchestrate it. You simply need an api object and a controller which can work with that.

Also, the idea that you somehow can’t use kubernetes if you have storage appliances is just plain wrong. Perhaps it’s true that can’t do _some_ things, but the idea that you throw the whole thing out because one workload doesn’t fit seems very extreme to me (although not surprising from a company who sells a Kubernetes based solution).

Why does _everything_ need to be on Kubernetes?

I will not use chrome but I’d love to try your extension. Could you add a way to get a reminder from your team when Firefox or safari support is available?