Makes me wonder what problems they faced with a Client/Server architecture.

Scope creep maybe? Supporting lots of platforms can be very time consuming.

I mean. I like compiling stuff myself but when I have to update 100 tools and all are source code to be compiled, then I can’t use my pc for a whole day or so.

Where I’m from there’s certificates a company can get, that confirm a certain level of process and IT security. Also a company existing for at least 5-10 years without incidents is a “vetted” company in my books. At least anything that managed to produce a working IT system before 2021 when AI came around.

I also believe there’s a bit of bad wording going on with the original comment. Take it up with that guy, lol.

This situation would have been easily preventable with basic understanding of what they’re doing is what OP is saying. This leak is not something highly complex, it is painfully stupid on the side of the developers.

There’s a difference between a hack, where data is exposed, compared to data exposure due to negligence or ignorance on the development side.

Or it gets them into a negative feedback loop since AI hardly ever tries to contradict you.

But yeah. At least they’re opening up to someone/something.

If you’re behind a conventional router they still do NAT afaik.

Per default your IPv6 address should be an internal one if it’s enabled.

Yeah, we managed to recreate that in a lab. Those old OS’s are super vulnerable.

EOL means no more security updates, which means attack vectors don’t get patched.

If you keep using a Windows installation (or any OS for that matter) that isn’t patched regularly you are very likely to be victim to some malicious actor eventually. It’s not manual hacking anymore, it’s bots scraping the whole internet exploiting known vulnerabilities completely automated.

The risk is much lower if you’re in a home network with NAT, where your PCs IP is not publicly reachable, but if you communicate with any webservices you’re still vulnerable.

As example. If you nowadays put a Windows XP machine live on the internet with a public IP, it will be compromised within minutes.

So yeah. Good call switching to Mint, but please don’t use unpatched Windows.

Isn’t the Wikipedia article usually already the summary of the topic?

If there’s an article with more than 20 references to papers it’s usually already abridged enough.

Just auto-generate videos with AI images and voiceover and add subway surfers gameplay on the side for those who think this slop is needed.

Seems true.

I use Hugo for static site generation and it makes the RSS stuff for me.

Glad it worked out for you!

But it’s definitely one of the deadly sins of selfhosting.

Everyone that thinks self hosting E-Mail is easy, I urge you to run your own mailserver and see how many mails actually reach their targets.

Your mailserver won’t be trusted by anyone, which makes your email always be delivered as spam, if they don’t get blocked outright.

Otherwise this scoring system seems to be quite alright. Even though it could use some more detail and citing some sources for the numbers would be great.

Since no one mentioned it yet.

You can also add the executable to Steam as a non-steam game and in its settings you can configure Proton as compatibility tool.

That works quite well for some stuff.

As I’ve said. Nextcloud is a great example of FOSS working out for a business, haha.

I guess we just have to agree to disagree then. Which is fine.

Your points are valid and thank you for detailing them for me. If I was in their shoes making others able to steal my IP, even if they’re not allowed due to licensing and having to deal with constant scrutiny of the source code are k.o.-criteria, which hinder the project and lead to potential revenue loss.

Well said.

Then that’s a moot point I guess, haha.

Still a great way to pay for Obsidian to support the development though.

It’s not just about syncing files. It’s also the fact I can edit stuff on my tablet and see the changes in almost real time on my laptop with Obsidian Sync. I believe most other solutions wouldn’t play nice with such a workflow.

That’s a bit naive imho. Remaining closed source is a form of IP protection and that’s really ok for what Obsidian is (a markdown editor). There’s just not any benefit for them other than appreciation from FOSS enthusiasts. Also maintaining an open source repository causes a higher workload and they lose a lot of freedom.

If privacy is your concern you don’t need source code anyway. It’s quite easy to sandbox an application like that and analyse network traffic and such. Also Obsidian is built using Electron. That means with enough motivation one could quite easily reverse engineer most of the app. Most of the applications behaviour can also be observed via the integrated dev console, which lets you view source code.

In short I don’t really see the need, unless I want to build or maintain it myself. And I think the negatives far outweigh the positives from the perspective of Obsidians team.