Raid wasn’t designed for data safety but to minimize downtime. Just swap the drive an continue operating the server seamlessly. Full backups are still required as the chance of complete failure isn’t zero

It’s still in alpha but hoarder is promising

It’s designed to organize bookmarks, but can also support markdown notes with picture (a single picture, not multiple pictures)

Unfortunately at the moment the mobile app is so alpha that doesn’t support creation or editing such notes, only new bookmarks or new photos.

It uses a headless chromium to make screenshots for URLs.

Optionally, can use a bullshit generator like ollama or openai api keys to automatically create a lot of useless tags to each note

Maybe they really believed that this shit was so good that nobody would ever return or even sell it to third parties in the future

What I’ve been doing:

Easy option: because I only have around 40gb of music, I sync it between my PC and my phone using syncthing since 128gb is the minimum nowadays

Hard option: streaming is cooler so I installed nextcloud with an optional plugin called “music” which allows to connect an app called “ultramusic” and it becomes “self hosted Spotify” with android auto support and all the bells and whistles. Disadvantage: Nextcloud is a moving target. For some reason they have to release new incompatible versions every two or three months. So for plugin developers this is a very annoying upgrade threadmill that eventually leads to burnout and that plugin dies. Even officially supported plugins sometimes don’t support the latest version when they launch it. If you choose to use nextcloud with docker, make sure to stay behind 1-2 versions (tag nextcloud:28 when nextcloud:30 is released) or your plugins might suddenly break without any warning. According to fanboys this is the industry standard nowadays and it’s up to the user to manually check the GitHub issues of each of the 30 plugins if it’s compatible before updating. Even if it’s official plugin. They call it “stable” but they mean “beta testing for the paid enterprise version”.

Unfortunately the hardware is so closed that even themselves can’t reset the software for refurbishing

https://www.theverge.com/2024/8/7/24211339/humane-ai-pin-more-daily-returns-than-sales

Once a Humane Pin is returned, the company has no way to refurbish it, sources with knowledge of the return process confirmed. The Pin becomes e-waste, and Humane doesn’t have the opportunity to reclaim the revenue by selling it again.

You phone doesn’t have the low-res laser projector gimmick that is invisible in daylight, runs extremely hot and drains the battery in one hour

The fact that suddenly it went to watch a leisure website during work… did they use stolen screen recordings from human activity for training? Like if some corporation allowed them to record all the activity of their employees for training

I went back to watch my Facebook since years and I scrolled and scrolled. 24 sponsored posts before seeing a post from someone that I knew! And it was a post from someone out of the country who did a 1 week workshop with me ten years ago (=someone that I barely know). Next post is from a schoolmate after 21 sponsored posts. Need to-scroll through another 21 sponsored posts to find another real post. Again from someone who I barely know (the teacher of that workshop where I went ten years ago)

How can people enjoy this shit? They actually enjoy watching ads?

that’s good, didn’t notice that

yes but the old docker requires the old licenses that aren’t on sale anymore to share more than 10 directories or other QOL improvements

It says “unlimited file sync is a $5 in-app unlock” so I’m guessing they can make money. Main problem is the apple developer fees that will eat the profit of the first 25 sales each year

That client seems really nice, is that new? I was thinking that it was just a download link for the browser but now it’s like a self hosted steam 😲

Years ago it was really interesting, could follow all my friends and see what they were up to. Now it’s ads, suggested content and sometimes, accidentally, I can see some uninteresting post from someone I barely know

Does anyone know why it was forked and the fork got all the improvements while the official app is in the exact same state of when it was launched years ago?

It was because all the proposals got rejected?

Because if he rejected all the improvements I don’t really understand why he’s saying “nobody wants to help development”

Well, Gemini is the stupidest of the bots

It’s pure greenwashing.

They made a deal with a company that has zero practical experience with nuclear reactors, literally built nothing.

It’s impossible that in just 6 years they will manage to:

1. Build a test reactor for their new unproven technology that as of now exists only as a PowerPoint presentation to show to investors
2. Have it approved by the government
3. Build the full scale reactor
4. Have it approved by the government
5. Get a license to use enriched uranium in a private setting

Even if they finished yesterday to build the final version of the full scale reactor, 6 years aren’t enough to go all through the regulatory red tape

Now that they promised that will use “green” energy in the future, Google can continue to use energy from coal and in 2030 everyone has forgotten about this vaporware deal

this is assuming that mozilla won’t ruin firefox as usually by copying the bad parts of chromium

seeing the latest 5 years, i don’t have much hope

Yes, prestashop is faster than Woocommerce. I manage two e-commerce with both. But while with Woocommerce all I needed extra was a $15/lifetime stock synchronization plugin, with prestashop I would need $100/month of plugins to have feature parity. So I keep prestashop basic as an simpler store under a different domain that doesn’t need stock synchronization or mass import or blog.

Odoo, I loved it at first sight when I tried the 30 day trial 5 years ago. So much snappier than Woocommerce, and with so many features. Its main problem was price and complexity. Official hosted version required a subscription for every single feature. Invoices? That’s $19/month. List of clients? Another $19/month. Blog? Add $19/month. For the tiniest extra feature, needed a subscription. In the end the full package was completely unaffordable and the bare minimum was unusable. The free self hosted version is the most complex install that I had to do in my life. I installed a third party plugin and I broke it beyond recovery. Because it’s in python it requires a dedicated server and not a normal hosting. Unless you’re a Linux guru you have to pay for their hosting service. Luckily recently they realized that their pricing was unaffordable for everyone except huge corporations, so now the full package is around $20 per month.

I feel that’s even worse than Woocommerce in that regard. There was a reason it was forked as thirty bees: v1.7 broke almost all the plugins and had a very slow adoption rate, and plugin developers continued to target 1.6 as it was more popular. The situation stalled for years

Edit: but as a e-commerce is faster than Woocommerce

Maybe, a good alternative that offers everything while also being fast is odoo. But you need to pay for their service, as the free version is very difficult to install and maintain

sure, but why solve problems in 10 minutes when i can do it way more sophisticated using 10x more time and resources?

(at the moment reverted to the easy html form + php send mail)