Hacker Newsnew | past | comments | ask | show | jobs | submit | moepstar's commentslogin

> a perpetual license to Office 2019 runs about $20 and will do that job for me.

Isn’t that only perpetual as long as the activation servers are up?


Probably. I meant "perpetual" as opposed to "subscription" but I agree with your concern.

You just stash a cracked version downloaded off some high seas site just in case.

I recall having "Pinball Dreams" also similar (same?) tech - the soundtrack through the speaker was quite good!

> Perhaps one day some jurisdiction will have the wherewithal to implement legislation to stop this madness.

Oh, and to reply to that point: the EU will have mandated labels on packages that will indicate how repairable something is.

https://joint-research-centre.ec.europa.eu/jrc-news-and-upda...

It seems to be in effect in France already: https://en.wikipedia.org/wiki/Repairability


Sure, having legislation would help tremendously.

What would help just as much: people actually giving a fcuk - as in: researching how durable something is, how hackable, how cloud-dependant or not...

...and not act all surprised when something stops working once the manufacturer calling it quits (or starts charging for a previously-free service).

Today, whenever i talk to others how i evaluate products i still get blank stares and i might as well have talked in a foreign tongue.

Also not happening: learning from $companys previous behaviour - stopped supporting something after a year? No parts, no schematics, no nothing?

Well - welcome to my shitlist of companies that'll never see another $/€ from me, ever again.

Doing this eventually would force companies to change their ways, but as long as they can continue selling whatever dreck they come up with to the masses...


You're blaming the end users. Most end users aren't aware of this stuff, and even if they are, have no practical way to evaluate quality in the way you've described. Even I, as a very technical person, could not evaluate if something is "hackable" without a huge amount of work, and not before I've purchased it.

Like similar cases (is this car roadworthy? are airplanes safe?), this is the classic case for regulation.


> Even I, as a very technical person, could not evaluate if something is "hackable" without a huge amount of work, and not before I've purchased it.

Teardowns, reviews etc… Of course, there is no silver bullet, but researching before buying really goes a long way.

Whatever I buy smart, I always check if I can flash alternative firmware to it or if it can be used locally only - for example.


> researching how durable something is

how I am supposed to know (or research) which fridge or vacuum is more durable?


Mostly by checking teardowns and reviews - that’s the one thing Amazon reviews are still good for - check them, especially the ones with a more negative sentiment and form an opinion based on that…

Also, certain brands more or less got similar issues over all the product ranges, for example power supplies giving up on Hisense TVs, compressors on certain fridge manufacturers etc - there are patterns to look out for - especially when the product has been on the market for some time.


Especially when this changes.

My first Ecovacs robot vacuum lasted 5 years, the replacement died in months and I've replaced that.

Its a complete pain to get open, and the way its wired makes it a complete pain to get around to try and repair.


Don’t tell me they’ve all three been Ecovacs?

The two times I had to open my Roborock S5 (once because our cat decided to vomit when the robo cleaned up and once to replace a broken Lidar motor) I actually was very delighted by its thoughtful design - basically none of what you describe, everything was very modular.

I mean, of course, you’re right - a different model can and probably will be designed differently - but that’s what teardowns are there for - fool me once, check the next time beforehand.


Well, it appears that Neato/Vorwerk robots are not supported - at least they're not listed on https://valetudo.cloud/pages/general/supported-robots.html

> Please note that this list is exhaustive. These are the supported robots. Robots not on this list are not supported by Valetudo. If your robot is not on this list, it is not supported.

From what i gathered so far, Valetudo is actually no custom firmware but modified vendor firmware? So, not sure if anyone related to the project has any interest and capability to reverse that...


When I looked into it, it seemed like a community that loved freeing robots but was absolutely not willing to buy robots that the developers did not have access to. This seems like a fair stance and I think they will start receiving dead vacuums soon. Hopefully Neato's security is as bad as their business side.


> business and domain PC's has Xbox stuff on it and game bar running by default

I know where you're coming from, however Win+G has _one_ use in the corporate world: give users the ability to record their screen without installing (and licensing) additional software to document breakage for support.


Agreed.

So it should be an independant app and not be part of a gaming system that have nothing to do installed by default on a pro computer.

When I needed this functionality and find out it was on the game bar I too was a bit shocked finding out it was installed !


_If_ the device does have a serial by id, you’ll be fine - if not (I.e. it is mapped by /dev/bus/usb/001/002 for example) you’re mostly out of luck. Cases in point recently encountered: USB printers (to be used in container with CUPS), Eaton UPS to be used in container with NUT.

No amount of scripting with systemd path units, pre and post hooks, udev rules could make those work reliably (for me).

Solution, or workaround, after much headdesking: put services into a VM, map USB device by vendor:device ID, done.

If anyone could shed a light into why that is possible with a VM but not a LXC container, I’d be thankful!


Question: did you run the pve8to9 script? Read their extensive documentation [0]on how to upgrade? Fix the stuff aforementioned script comes up with?

My cluster went from 6 to 7, 7 to 8 and recently 8 to 9, along with Ceph - all without a single problem.

Given it’s more or less Debian underneath, not too surprising I’d say? Granted, there’s always a chance for something to go sideways, however, it’s unlikely you’re the first person to encounter this problem and if you check their forums, you should find a solution.

[0] https://pve.proxmox.com/wiki/Upgrade_from_8_to_9


The difference: the free repositories get the updates before the enterprise repo.

So, the software versions that go into the enterprise repo are considered stable by then.

(If we're talking about Proxmox, that is.)


Indeed, leaving Windows behind for the odd application is easy.

With the depreciation of my late 2017 Intel iMac 5k incoming, i however wonder how to ditch macOS for Linux and keep the one odd Mac App I kinda depend on - ideas welcome!

If you’re wondering, the App is MoneyMoney and keeps track of all bank accounts automatically, sorts all spendings into categories etc.

There simply seems to be no equivalent, and running Mac Apps on Linux just doesn’t seem to be a thing yet (at least in a half-viable way I know of, and yes, I at least read about Darling).

Again, if anyone does have a pointer (running macOS virtualized? What’s the status there?) would be much appreciated..

Edit: oh and i fully intend to keep using the iMac, its an i7 with 64GB RAM and the 5k display is still so gorgeous to look at.



Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: