@bagder I shudder to imagine AI with full access to backend systems; the risks are real, and deeply unsettling.
Views expressed are mine alone and not that of my employers. Accuracy, completeness, currentness, suitability, or validity is not guaranteed.
@bagder I shudder to imagine AI with full access to backend systems; the risks are real, and deeply unsettling.
@danie10 Looks like a nice service!
My instinct tells me that the next concern may be privacy. As a third-party service, OpenRSS could gain too much insight into users’ reading habits. Though, probably what happened with Google’s RSS reader and similar platforms as well. I guess there's a fine balance between privacy and convenience, where users must weigh the trade-offs.
Follow updates across the web in a feed that only you control, even if the site does not offer an RSS feed
Open RSS offers feeds that are a much healthier alternative to the intrusive, algorithmic feeds on websites that harm and manipulate us. But several sites, including Tumblr and Craigslist, have removed their RSS feeds, so that you are forced into th ...continues
@uastronomer it bothers me when popular sites I enjoy reading, like MyBroadband, focus too much on numbers for the sake of numbers without explaining how they arrive at them. This takes away from the real value of the content and the bigger picture. It also sets a misleading standard, making others believe they should prioritize metrics over meaningful insights, but maybe that’s just me.
I can’t help but wonder how they arrived at that number, 193,000 Linux users. What kind of analytical wizardry or data sorcery could have generated such precise numbers with such certainty? And don’t even get me started on the claim of 27 million South African readers. I’d genuinely love to know more about how these figures were calculated.
https://mybroadband.co.za/news/technology/575064-mybroadband-in-2024.html
@bagder nice topics. Wish I could be there and just listen!
The HTTP workshop, day one: https://daniel.haxx.se/blog/2024/11/13/the-2024-http-workshop/
very relevant
This comic is about the difficulty of dealing psychologically with 50/50 odds [...]
https://www.explainxkcd.com/wiki/index.php/3007:_Probabilistic_Uncertainty
@inetpro Not minor upgrade, no - just the bigger ones like 4.1 to 4.2, or 4.2 to 4.3.
Minor upgrades tend not to mess with dependencies, so there's usually no change to the system outside ~mastodon, so there's not much point in a rebuild. But when dependencies *do* change, I see it as an opportunity to see if I can improve things, make my install process more efficient, try include things that I'd previously avoided because I wasn't yet comfortable with them, and so on.
@uastronomer I like your approach, question is, do you do this with every upgrade?
As is my habit, I won't be upgrading my Mastodon servers from 4.2 to 4.3
Instead, I'll build all-new servers, from scratch, so that I can troubleshoot and update my install notes, and then I'll migrate my instances across.
This is a long way around, but it helps clear out cruft from previous efforts, helps make sure my understanding of how this stuff works is as current as possible, and finally serves as both practice and verification of my ability to help new customers to migrate their instances to my service :)
@bagder noted. Good to be aware of this, thanks!
@bagder in the case of http://example.com/one/two/?id=42
, what happens when the file 'two' already exists and then you change it to id=43?
@jpmens you nearly confused me for a moment. I thought you were referring to the pv command for monitoring data progress through a pipe.
@inetpro since this is how it has worked on Windows for decades already, I consider it unlikely that they can or even want to change this.
But it would be useful to have control over it on a per-connect basis.
@bagder that’s fascinating! Thanks for the interesting explanation. Let’s hope someone at Microsoft sees this and addresses the issue soon.
slow TCP connect on Windows
Presumably you just have to be more patient on Windows.
https://daniel.haxx.se/blog/2024/08/14/slow-tcp-connect-on-windows/
@inetpro because it would change behavior and break scripts for N% of users
@bagder I guess you're right, of course. ;-)