• 0 Posts
  • 26 Comments
Joined 2 months ago
cake
Cake day: February 5th, 2025

help-circle


  • Linux was getting pretty nice by 2004. In 1996 it was a LOT rougher.

    I basically left Windows in 2006 and never looked back. I did some cross platform work in Qt where I’d develop in either iOS or Linux and then hand the product over to the test team to compile in Windows - worked beautifully. Sure, there were things that worked in one OS that wouldn’t work in one or both of the alternatives, but when I figured out the problem it was 90%+ me “getting away with” bad practice on my development machine that once cleaned up ran everywhere just fine.

    These days the Browser is 99% of the OS that means anything to anybody.



  • I started programming in DOS professionally in January 1991. It was pretty clear how bad Microsoft sucked by February of 1991, and blindingly obvious when they “updated” DOS more than annually with “95% backward compatibility” which translated to: "we just broke all your programs and you’re only going to have to figure out which 5% of your code you’re going to have to update to make it work in this version - aaaaand, by the time you do that we will be releasing a newer version! ;-P "

    Something called DrDOS came along and we used it just because it wasn’t updating and breaking backward compatibility so often. Since 640k wasn’t enough for us even then, we ended up putting the kludge “Phar-Lap 32 bit extender” libraries on our product so we could access all the cheap RAM that systems were being shipped with (2MB was pretty much standard by 1992).

    Then there was the day that McAffee decided that our product’s main .exe was a virus. It wasn’t. It wasn’t infected with anything. It didn’t do anything vaguely resembling malware. McAffee just had a false positive pattern match with our software.

    The Microsoft treadmill was a very real thing all through the 1990s - much like Android and iOS are today. Sure, you’ve got a cool idea for an app, but we’re going to keep shifting the OS underneath you so that you’re spending 90%+ of your time just recoding your same old app for the latest OS release. That way you don’t have any time to innovate and maybe threaten our business model.




  • How many times did I read the condescending “HOW-TO sound drivers” with it’s condescending opening of “If you must hear biff bark” and it’s virtually always unhelpful content. Maybe one install in 10 would benefit from the pearls of Linux Audio wisdom contained therein, and it was the best available source for YEARS.


  • I tried to use Slackware in the mid 90s. After two solid days trying to get my ppp dialup connection to come up after a reboot - it would come up in the first boot after install, run for days like that, but after any reboot it was dead and gone and nobody on the internet seemed to know how to deal with it. “Real men” didn’t use dialup, and people on dialup (self included) had no clue.

    I declared it “not ready for prime time” due to that, and issues with sound drivers, and ignored it until 2003. In 2003 I tried some Cygwin and was impressed with its performance, so very close to “bare metal” Ubuntu. In 2004-5 the 64 bit AMD chips were coming out and I used Gentoo to build a true 64 bit system addressing 8GB of RAM - there wasn’t really any other option.

    I got tired of compiling every little part of the system from source for days on end and migrated to various flavors of Ubuntu / Debian, which by 2006-7 was becoming a viable desktop alternative. Before that you ALWAYS had to have a Windows machine for something, usually several somethings. At this point I only use my company issued Windows laptop when I need to connect to the company VPN, which can be months between needs depending on what I’m doing. My wife and I use Ubuntu full time now.


  • I started using Linux more or less full-time in 2014. I find it to be just as “stable” as Windows or OS-X, which is to say: it’s stable until you do something that makes it not stable.

    If you’re staying in the mainstream, using a “stable release” from a big distro (Ubuntu, Debian, there are others…) and waiting at least 6 months after the release of that stable release before using it, I have found Ubuntu to be just as stable as Windows or OS-X. You might want to use an unstable app, that can be a problem in any OS, but granted: there aren’t as many “stable” apps to choose from in Linux as Windows.

    OS-X and their apps have burned me hard, repeatedly, for things that Windows and Linux had under control 10 years earlier.

    The major difference in my WIndows vs Linux experiences has been: when you want something to work and it just doesn’t, in Windows you have to shrug your shoulders and explain to your customers: It just doesn’t work, there’s nothing we can do. In Linux, you have the option to do the heavy lifting and make it work. It will frequently not be worth the effort, but if you’re really determined you can fix just about anything in Linux.





  • MangoCats@feddit.ittoLinux@lemmy.mlWhich browser do you use and why?
    link
    fedilink
    English
    arrow-up
    3
    arrow-down
    2
    ·
    2 months ago

    I read the new language to mean: they are going to record your input streams and feed them to AI/LLM - thereby recording your previously private info that they used to discard and protect. Up to you, I use Chrome because it integrates well with the gmail account I’ve used for 25+ years and I appreciate the “login anywhere and get your same setup” functionality, as well as the ability to nuke remote login sessions.




  • I had a remote relay box: 8 channels of power control, so I could at least power cycle machines from remote when all else failed.

    I actually ended up not using it much at all, it was a nice security blanket, but the last time I decided that I wanted to power cycle something was about 6 years ago, and at that time I realized it had been over 3 years since I had previously used it, and that usage was more of a “let’s make sure this thing is working like I think it should” test.




  • Back in the 1990s I developed an app over the course of 6 years, first 3 in C/DOS then we ported that to C++/Borland/Win95 and continued developing it for another 3 years. I was the only coder, we had a dedicated tester / documentation specialist and the algorithms lead who was more of an idea guy than any hands-on code work.

    We got bought out. Buyers “needed it in native Win32 because of the depth of the talent pool.” Whatever, I’m here to help if they want it during porting. Buyers estimated 2 developers could port it in about 2-3 months. Yeah, o.k. Never asked for help, but at 6 months in they had expanded the dev team to 6 guys and were still struggling and looking to hire more. Ultimately they reduced scope a little and called it “ready to use” in Win32 after about 15 months. Glad they got it “maintainable” by switching to that Win32 dev environment with such a deep talent pool to hire from, they easily spent more man hours on the port than we spent developing it in the first place.