Most python code that was written for Python 3.7 will run just fine in 3.12, a few things that were already marked as deprecated will not work though.

We work mostly with Go and Python, and the Go release cycle is more aggressive with releases getting security back ports for about 12 months. We have to upgrade to a new major release twice a year, and we do have some compatibility issues almost on every Go release.

Kubernetes has a new release 3 times a year with deprecated APIs on regular basis.

With python we upgrade once a year and most issues we have are related to third party SDKs that are too opinionated about their own dependencies. We do have breaking changes, but mostly we find pre-existing bugs that get uncovered thanks to better type annotation, which is vital in larger python projects.

Not fixing bugs is still better than pulling the plug completely on $5k hardware. My Kindle Paperwhite from 2012 still works just fine even if the OS upgrades ended years ago.

Former Googler here, from before it became Alphabet.

When I worked there, Google was proud to be a company run by engineers and not by sales, and sales was pretty much non-existent.

One of Google problems is that their ad revenues from their main projects is so profitable that they have a huge amount of disposable money that they need to re-invest on R&D, any R&D. Next issue is that the internal politics and promotions are all based around doing new things and there is very little professional growth to be expected for making a product actually work well and be usable by the general public.
The devs that built the product will switch to a newer project within 2 years of product launch to go on an other project that is promotion worthy while the remaining staff gets no hope of recognition.

This is why you see all these product announcements, that are just experiments that are thrown at the wall, some will stick (GMail, GCal, Maps, Chrome, ...), but if they do not get massively profitable they get no TLC from the better staff and they get canned.

For ransomware concerns there are good solutions such as Clumio: https://clumio.com/

Disclaimer, I work for Clumio.

Have you look at the Deep Archive pricing? It is $0.00099 per GB/month, or about 1cent per GB per month, or around $12.5k/y for 1PB (slightly more depending on your average file size). The recovery time is under 12hs. Overall this is quite competitive compared to offsite tapes and a lot more flexible/faster. They even have a Virtual Tape interface to work with your existing backup software. If your data is already in AWS there is no extra networking cost, within the same region, but if from on-prem then you have to deal with e-gress cost and speed of course.

https://aws.amazon.com/blogs/a...
https://aws.amazon.com/s3/pric...

https://aws.amazon.com/storage...
https://aws.amazon.com/storage...

Personally I don't think I will ever work for a company that has a on-prem presence ever again. I never have to deal with bad, or aging hardware, addressing rewiring of racks, or deal with lying hardware vendors that bill you more.

Granted the main problem with companies using AWS is that they they get sticker shock over time because they forget to design their stack to autoscale. Most kind of do scale up, but scale down is rarer, but that's where cost savings really exist over on prem. An other thing they forget is to garbage collect stale data, tune log verbosity, and that disk is cheap but not free. All these things add up and are easy to overlook with on-prem designs.

When I was working in the self driving cars industry, this is how our cars could be 'driverless': someone would make key decisions for the care several times per mile from a desk located a few miles from the car ... and this was on a relatively short loop.

When I see such a car on the road I still stay away from it, if not to avoid potential crashes but to avoid potential eye damage:
https://cleantechnica.com/2021...

I worked at Google a long time ago and the employees count went from 10k to 20k. I was quite impressed by the quality of most engineers I worked with, only a few were awful engineers but good at office politics. That got worse at 20k employees and I left for a better, smaller, company. I know several average former coworkers that do fairly well at Google.

I'm much happier at small companies.

We actually use some of their OSS products, we did consider using some of their 'SaaS' offering and were quite surprised at how it was managed internally. It was more of a somewhat hosted service where they run and maintain the hardware for you, but not much more. Since we run our infrastructure in AWS, it was actually going to make things harder, while being a lot more expensive for us.

There are some technical decisions that they have made over time that I have been scratching my head over. For example the http provider v2 to v3 which is now failing silently if you get a 400 or 500 error back. You have to update your code with a condition to check the http code explicitly. They could have handled this with a new rule name very easily but they decided to force everyone to rewrite their code, or just ignore failures from now on.

https://registry.terraform.io/...

Overall I do like their products but there are definitely things that make little sense technically sometimes.

Considering her liver and lung lesions started to heal after taking the medication it is safe to assume that they made their way to these organs. My guess is that the patient has permanent brain damage but things stopped getting worse.

Granted in this case the patient eat the worm, but they had a compromised immune system and that plant grows close to the ground where it is more likely to pick droppings.

In most cases it generally safe to eat wild food without a sterile environment to wash them in. I do eat berries in the woods when available, they are usually much tastier than store bought. Of course I do not pick the ones I'm not familiar with and I avoid the ones close to the ground where wildlife could pee on it.

+1 for zstd, much better compression and speed.

My wife got me a nice pair of such glasses. I got laser surgery years ago so I would not need to wear glasses so I was not looking forward to wear glasses again. The results are that I got headaches from wearing them. What happen is that the inside of the glasses caused some glare from the surrounding light sources. I stopped using them after a few days, never again.

I second your feelings. I installed the Steam app on my personal MacBookPro years ago and it would launch itself and try to download upgrades every single time I would open my macOS user session. If it did that in the background with little cpu/memory I would not have been bothered as much but since I'm only a casual gamer and mostly use a playstation when I do, it was not worth it at all, and never tried it again.

Soda cans, while mostly metal, actually have a plastic liner in them which is probably leaching forever chemicals in the soda. Only glass bottle are plastic free (if you ignore the plastic liner of the metal cap, but store your bottles upright).

True, it was definitely a smaller startup.

I actually used to work at Google, over 10y ago, and my manager asked me to write a bot that would scrape data from websites in a specific field ... I had to remind him it was strongly against company policy. That was not a good career move for me, but that's an other story.