A big one for me is Microsoft office (desktop), Libreoffice and other FOSS alternatives just simply don’t come close, and feature wise are 20 years behind. Especially since I basically mastered MS office 2007+'s drawing features, which the FOSS alternatives don’t replicate very well.

And of course Microsoft loves to push Office 365. I don’t pay for that and just use desktop office, but Microsoft prefers you don’t know that you can do this.

And I’m going to get shit on by Lemmy big time for this but while Linux is great and has made vast improvements in recent years, I still use Windows, not only because of MS office, but because a lot of games tend to only support Windows. I know that wine and proton exist but they’re not perfect and don’t feel quite the same as running native.

I wish an operating system existed with a hybridized Linux and clone NT kernel (using code from FOSS Wine and ReactOS of course) so that the numerous back catalog of NT software can run similar to as intended while also interacting with Linux programs better and using a shared environment. Since it would probably become vulnerable to viruses for windows as well, maybe? (my programming knowledge is extremely rusty) an antivirus similar to Windows defender is bundled with the operating system. Hopefully if someone makes such an operating system it can be a Windows killer and would switch immediately

  • Rose@slrpnk.net
    1·
    21 hours ago

    One of the most frustrating programs for me is digiKam. On paper, it’s the perfect DAM/photo manager. But it’s kinda slow for day-to-day use. The user interface is janky in a lot of ways. It doesn’t see constant refinement either. It doesn’t even speak to me as a metadata nerd because I don’t want to turn my metadata into a janky mess. Yeah, you have a powerful metadata editor. It’s like a welding torch without any eye protection.

    I’m using ACDSee on Windows, because it’s operating on pretty much the same principle (image file metadata is canonical, app database is just for indexing), but it’s faster and smoother to use. Not perfect, it has its mild limitations (like why the hell doesn’t it support OpenStreetMap - Google Maps kinda sucks for nature trails, you’d think photographers would have pointed this out), but it’s just so much more efficient. If digiKam ever gets a huge UI overhaul, switching over will probably be fairly easy though.

    Also about a decade ago, I would have said that as far as novel writing software/large structured document word processors go, nothing beats Scrivener. Scrivener is still probably the best software in its niche, but it looks like a bunch of open source word processors in this niche have come a long way. Currently looking at novelWriter, which seems really rad.

    • slax@sh.itjust.worksEnglish
      1·
      6 hours ago

      I have to ask you about metadata nerd status…

      I have a bunch of exported Google Photos and icloud Photos… photos… what’s the best way to fix the metadata as the “date taken” keeps using export date.

      • Rose@slrpnk.net
        1·
        1 hour ago

        My immediate thought was that there’s some inconsistency with various types of metadata. For example most software will pull the date from the Exif DateTimeOriginal field. But there’s also XMP tags that have the same purpose. Or similar purpose. These standards have plenty of date fields for various uses, and while they serve a noble purpose, the software just craps all over them. (Don’t ask which software. All of them.)

        My guess is that at some point of time, one of those tags got updated, but not the other tags of similar purpose. So the program you’re using could be pulling the date from one field, and when you update it, you’re actually changing some other field.

        Of course all of this is wild because usually no one needs to touch the datestamp anyway (unless you, like, have to correct daylight saving time or clock drift or something). Software changing this to a batch import time? That’s weird and silly.