aard

One exception nowadays: Business notebooks - and that’s only because the rest of the notebook market went to shit. If you want a somewhat compact notebook with more than 64GB of RAM, decent CPU performance and good battery life Apple currently is the only one offering something.

The space used by the smallest solar charger I’ve seen on Amazon seems to be similar to 6 or more batteries in the format the N900 was taking - so if you look at space, slow charging from solar charger, and reliance on sun conditions taking individual batteries seems to be the better option for a few days hike. It’s also easier to stow individual batteries to wherever you still have space left.

With my N900 I used to travel with 6 to 10 charged batteries to have a few days of runtime. Things got better now with powerbanks - but for something like hiking just carrying a few spares would still be smaller and lighter.

About 20 years ago I made a script that converts pictures to HTML tables. Back then RAM was a severe problem for this, and even for more powerful hardware browsers tended to just crash on larger pictures.

I checked it again a few years later, and things looked way better. I guess using CSS it’d be rather trivial nowadays to do the same with a short video by just cycling through showing/hiding tables of each frame.

I did not sign with them after I had some issues with the contract provided, and the resulting interactions with my future manager. I’d say at least for someone from Europe the company culture is less than ideal from that encounter.

AMD keeps some older generations in production as their budget options - and as they had excellent CPUs for multiple generations now you also get pretty good computers out of that. Even better - with some planning you’ll be able to upgrade to another CPU later when checking chipset lifecycle.

AMD has established by now that they deliver what they promise - and intel couldn’t compete with them for a few generations over pretty much the complete product line - so they can afford now to have the bleeding edge hardware at higher prices. It’s still far away from what intel was charging when they were dominant 10 years ago - and if you need that performance for work well worth the money. For most private systems I’d always recommend getting last gen, though.

This doesn’t have anything to do with user control - modern windows versions need drivers to be WHQL signed to get that kind of access. Alternatively you’ll need to enable developer mode on your system, and install your own developer certificate into its keyring for running own code, which has its own drawbacks.

Crowdstrike is implemented as a device driver - but as there is no device Microsoft could’ve argued that this is abusing the APIs, and refused the WHQL certification. Microsofts own security solution (Defender) also is implemented as a device driver, though, and that’s what the EU ruling is about: Microsoft needs to provide the same access they’re using in their own products to competitors. Which is a good thing - but if Microsoft didn’t have Defender, or they’d have done it without that type of access it’d have been fully legal for them to deny the certification for Crowdstrike.

Both MacOS and Linux have the ability to run the type of thing that requires those privileges on Windows in an unprivileged process - and on newer Linux versions Crowdstrike is using that (older versions got broken by them the same way they now broke Windows). So Microsoft now trying to blame the EU can be seen as an attempt to keep people from questioning why Microsoft didn’t implement a low privilege API as well, which would’ve prevented this whole mess.

I have a 96 core one. While it’ll be fine as a desktop for compiling I’d stick with an AMD system.

The devkit has 6 memory channels, and you’ll want to fill them all - there’s a surprisingly high performance penalty if you don’t. Even then, compiling a code base which could be spread over hundreds of cores is still significantly slower on the ampere compared to my old 3970x.

At the time of sending the mail I need the metadata - so offering a SMTP server implementation which keeps this in memory while forwarding is not hard. You’d lose a persistent spool in case of delivery errors - but we’ve been doing relays that keep the client connection open while trying to deliver the mail to relay errors directly to the client already 30 years ago, so that also isn’t an excuse.

For IMAP - if you don’t do serverside searching or similar it’ll work with fully encrypted mails.

They will have access to metadata - otherwise they wouldn’t be able to work as email service. That’s sufficient to implement those protocols.

The client then would have to bring their own crypto, and you’d probably want the SMTP server to reject mails if delivered unencrypted (though their FAQ says you can send unencrypted mails).

The reason they claim they can’t is probably trying to keep full control over what users are doing, in which case I agree - fuck them, don’t use services like that.

Don’t have links anymore, but few months ago I came across some startup trying to sell AI that watches your production environment and automatically optimizes queries for you.

It is just a matter of time until we see first AI induced large data loss.

And as soon as I learned about that I stopped using it. Turns out it was the right choice - since then more then one company had breaches where authenticator seeds extracted from a google account were used to bypass 2fa.

Did pretty much the same with a new server recently - spent ages debugging why it didn’t find the SAS disks. Turns out, disks like to have power connected, and no amount of debugging on software level will help you.

You still had a 4GB memory limit for processes, as well as a total memory limit of 64GB. Especially the first one was a problem for Java apps before AMD introduced 64bit extensions and a reason to use Sun servers for that.

I was referring to work setups with the overengineering - if I had a cent for every time I had to argue with somebody at work to not make things more complex than we actually need I’d have retired a long time ago.

Unless you are gunning for a job in infrastructure you don’t need to go into kubernetes or terraform or anything like that,

Even then knowing when not to use k8s or similar things is often more valuable than having deep knowledge of those - a lot of stuff where I see k8s or similar stuff used doesn’t have the uptime requirements to warrant the complexity. If I have something that just should be up during working hours, and have reliable monitoring plus the ability to re-deploy it via ansible within 10 minutes if it goes poof maybe putting a few additional layers that can blow up in between isn’t the best idea.

Everything is deployed via ansible - including nameservices. So I already have the description of my infra in ansible, and rest is just a matter of writing scripts to pull it in a more readable form, and maybe add a few comment labels that also get extracted for easily forgettable admin URLs.

Shitty companies did it like that back then - and shitty companies still don’t properly utilize what easy tools they have available for controlled deployment nowayads. So nothing really changed, just that the amount of people (and with that, amount of morons) skyrocketed.

I had automated builds out of CVS with deployment to staging, and option to deploy to production after tests over 15 years ago.

I personally prefer bzip2 - but it needs to be packed with pbzip, not the regular bzip to generate archives that can be extracted on multiple cores. Not a good option if you have to think about Windows users, though.

Nowadays it matters if you use a compression algorithm that can utilize multiple cores for packing/unpacking larger data. For a multiple GB archive that can be the difference between “I’ll grab a coffee until this is ready” or “I’ll go for lunch and hope it is done when I come back”