Their AAA games are mid so I suspect their ‘quadruple-A’ games are also mid. Price 'em how you want, I’ll pass anyway, thanks Yves.

One of the best games ever made. Not sure if Switch is going to be the best way to experience it, but as long as it works properly and runs well enough it’s a must-play in my books.

Arakane Austin clearly didn’t want to make that game as evidenced by their 70% dev exodus during development, and you really need a dedicated dev team to achieve a “redemption arc”. This is likely another head office decision made with no understanding of their development team(s) or how to actually achieve it. I’m not holding my breath.

I’m interested to see it took this long tbh

man that bald english dude has to be losing his damn mind rn

I don’t think it’s unfair to point out that many of the people who were interested in Starfield leading up to launch thought they were getting more of a space sim than they did, proceeded to look for alternatives, and NMS was there being pretty good at what it does now. The OP article demonstrates this and is not a comparison between the games. In my case, Starfield just reminded me that NMS exists and I decided I’d rather be playing it. Fundamentally comparing the games is ridiculous, but it’s no surprise that NMS ended up in the conversation.

Really? You can’t fathom how someone would consider NMS a better game? Both games are barely comparable other than using space as a backdrop. Judging by the reaction online, it seems like many people were lead to believe that Starfield would be a space sim and came up wanting when it was more of a sci-fi Fallout, with mostly optional engagement with the space elements. For those people, I can see merit in recommending they check out No Man’s Sky, which has a shallow, bit widely-spread space simulation to engage with.

I don’t think it’s useful to try and argue which game is better, but I would much rather play No Man’s Sky any day of the week. Bethesda RPGs have long lost their luster for me since the Oblivion days, and now just stand as a testament of disappointing writing, stagnant technology and under-baked systems. Starfield does not show any meaningful signs of breaking the norm.

Andrew Wilson and my personal definition of ‘very exciting’ likely differ greatly, so I’m not going to pay attention to anything until we see the products. I suspect he is very excited to make a lot of money on a bunch of new Star Wars games.

I respect the writeup, although personally think the use-case described is too specific for general mail hosting. I have had a different experience for a similar amount of time running a couple of mail servers for home and work myself. I didn’t have the luxury of avoiding spam/virus filtration on the work server due to the domain’s history and the nature of 3rd party users with varying degrees of tech literacy. Most issues I have faced with maintaining these servers have been down to the filtration elements the author was able to avoid, specifically the virus scanner growing in memory footprint as hot new virus definitions are included. The overall virtual footprint of my postfix/dovecot/sql/nginx/roundcube/spamass/clamav stack has grown significantly over the years on clam alone, depsite no real change in usage patterns. Ongoing maintenance outside of ClamAV has been minor, but something will pop up now and again when a large 3rd party makes a decision that forces others to follow suit, or a new mail client is picky about protocols, etc.

At the time I needed to deploy these servers, the task was more difficult and required a lot more scrutiny than most other admin work I had done at that point (from a history of web server and backup system maintenance). The mail servers tended to require more active maintenance than most other small/self-hosting roles like web/file/game servers, or deploying a NAS or network gateway with a taylor-made distro/OS. Familiarity was the main roadblock; there was a lot of mail-specific terminology and best practices that differ from other server software. There is also a lot of ‘legacy friction’ related to bolting on separate daemon interaction that SMTP was never meant for while still maintaining backward compatibility with SMTP servers and mail clients. I have seen a lot of parallels with deploying and troubleshooting fediverse and ActivityPub driven software, likely due to the similarly decentralized behavior and reliance on 3rd party uniformity. I think it’s probably fair to call mail hosting ‘hard’, at least comparatively.

No shade on the writer though, and there are plenty of other ways to make mail hosting easy on yourself in 2023 (containerism and automation, or all-in-one solutions like Mail-in-a-box come to mind). Despite the difficulties, I’d rather the option to self-host mail not be yanked from the average user just because Google or Microsoft has the user-share to disengage with the rest of the network without much consequence, as they have done in the past for other things.

For what it’s worth, NVIDIA’s failings on Linux tend to be mostly in the desktop experience. As a compute device driven by cuda and not responsible for the display buffer, they work plenty good. Enterprise will not be running hardware GUI or DEs on the machines that do the AI work, if at all.