• 1 Post
  • 33 Comments
Joined 1 year ago
cake
Cake day: July 28th, 2023

help-circle






  • If you remove the app-platform role from Nextcloud by separately hosting the individual apps, what benefit do you get from having both Nextcloud and File Browser?

    Nothing really. For almost any Nextcloud feature out there, you can find a server app that does the same.

    But that’s the point in my opinion. I don’t want to waste time managing tons of apps if I can manage one Nextcloud instance. Nextcloud basically decides for me what’s the best way to get those features running, so I don’t need to figure out myself.

    Now if you’re into self hosting one container for each feature, go for it, no reason to not do so.





  • This hate comes mostly from Linux communities like here and on Reddit. When you see actual numbers, both are widely used for production use. They have lots of active users as reported in their respective blogs and websites.

    That said, it is aware that both had problems. Most hate towards Flatpaks that I can see is from purists that prefer their distro shipping their packages with dynamic dependencies and uprated by their package manager. Also there is complains with outdated runtimes and stuff like how sandboxing works.

    Snaps has all problems than before with some extras. When they were released, because of compression, they were painfully slowly to open and they affected boot time. Nowadays this is mostly gone, but they still keep a proprietary store, inability to have multiple repositories (stores) and they don’t respect your home directory structure by placing a “snap” folder in your home.

    Personally I use both and I’m happy with them. The proprietary store stuff does not bother me because I’m already trusting canonical binaries by using Ubuntu and they are easy to use and be productive with them.



  • As an addition to other responses, think that most apps (specially smaller ones) are developed using some framework or set of libraries that might or might not support those protocols.

    So let’s pretend that I have an app buit using Electron and that framework does not support Wayland. There’s nothing I can do on the app side until Electron supports Wayland in this fake example.

    So it actually takes time for the libraries to support the new protocol and then app developers to update their apps to support it aswell.

    That’s why you see that the Wayland migration is incremental and not all at once.






  • Sorry, I’m not following you. We’re discussing if deregulation is always bad or not, and my point is that it can be bad in excess.

    The article points to some corruption that happened at the time, which is correct, but it does not discuss anything about the benefits (or lack of it) for the end user after the deregulation.

    If we could prove that if we kept excess regulation in the market as it was before, people still could have access to telecommunication as we have today, my point should be wrong. But its not the article discussion or what you’re saying now.

    Notice that I’m not defending 100% deregulation, even in the Brazilian example the market is still regulated today, just way less than it was 20 years ago. If we have no data caps for residential internet around here, its because of a regulation, not companies good faith for example.

    Out of curiosity I asked chatgpt to list examples similar to the Brazilian one and it listed a few:

    United Kingdom - Energy Deregulation Japan - Rail Deregulation New Zealand - Telecommunications Deregulation Australia - Banking Deregulation Sweden - Postal Service Deregulation

    Truth is, its really hard to prove that something “is always” good or bad, as it will need to go over all cases and prove the point one by one. Normally economic discussions are applied to a society or sector, so we could say “in the US deregulation is often bad”, which is much more fair and easy to prove.




  • In Brazil, during 1990 - 2000, telecommunications were a highly regulated market. Due to difficulties to enter the market, there was not that many companies around, and outside big cities, phones are mostly inexistent or too expensive for the average customer.

    I remember that in my city (100k people) there was only one option, and in my mother’s city (5k people) there was no option.

    After ~2000, the government did a big deregulation, making easier to other companies to enter in the market and do investments. A few years later, we got coverage across the entire country. For example, in 2005 my city had 5 options and my mom’s city had its first option. At this time my family finally could afford paying for a phone plan.

    So yeah, excess regulation can harm markets, and deregulation can be good. Finding balance is key here.