Products for Dumb Suckers


  • Garbage Person

    @Groaner said:

    Clearly, someone needs to make a motherboard with at least 100 DIMM slots so that the entire contents of the hard disk can be loaded into memory.
    PowerEdge R920.



  • @blakeyrat said:

    handling the "out of virtual memory" case, where Linux's method is criminally idiotic

    Yeah, the OOM killer is kind of stinky, but it works better than Windows's method: endless thrashing until the machine is force-powered-down because it won't respond to anything else.

    @blakeyrat said:

    aggressive, intelligent and proactive disk caching. Which practically guarantees (after watching your behavior for a few days) that what you need is in memory when you need it

    That's the design intent. I'm completely unconvinced that it works better in practice than just suspending/hibernating instead of shutting down when the box is not in use, but if anybody has measurements that say otherwise I'd be interested in seeing them.



  • @flabdablet said:

    That's the design intent. I'm completely unconvinced that it works better in practice than just suspending/hibernating instead of shutting down when the box is not in use, but if anybody has measurements that say otherwise I'd be interested in seeing them.

    Well for one thing, it doesn't require leaving the machine on 24/7 for no fucking reason.

    For another, in the "leave the machine on 24/7" scenario, the newly-run programs would tend to push the long-ago-run programs out of the file cache. Even if you launch your email client every day at 4, it'll be out of the file cache because Photoshop ran more recently and filled it up.



  • @accalia said:

    even visual studio has trouble using it all up

    Visual Studio is a 32 bit app. It would be very unlikely for it to ever put any memory pressure on your system.


  • FoxDev

    @Jaime said:

    Visual Studio is a 32 bit app

    yes, but it is large address aware and so can use as much ram as you want to throw at it.

    well maybe not that much but it can address 4GB in a single process. add multiple instances (plus the multiple processes (debugger is separate last i checked)) to run/debug complex projects that eats away memory fast.


  • ♿ (Parody)

    @blakeyrat said:

    What makes Windows superior is its aggressive, intelligent and proactive disk caching. Which practically guarantees (after watching your behavior for a few days) that what you need is in memory when you need it.

    Linux has nothing like that.

    Mine does that all the time. Most of my memory most of the time is cached disk stuff.

    @blakeyrat said:

    So I'm guessing

    Actually, this is painfully obvious.



  • @boomzilla said:

    Mine does that all the time. Most of my memory most of the time is cached disk stuff.

    Well duh. But it's cached disk stuff you used in the past, not cached disk stuff you're likely to need in the future.


  • ♿ (Parody)

    @blakeyrat said:

    But it's cached disk stuff you used in the past, not cached disk stuff you're likely to need in the future.

    Seems to be pretty much the same stuff for me. Though that's on a system with 24GB of RAM. Anyways, this seems to be the Linux equivalent:



  • Couldn't possibly be as good as Superfetch. Just look at the names. What kind of wimpy bullshit name is "preload"? Fucking hopeless open sores garbage. Also that project was first registered on SoresForge in 2005, which is before Vista came out, so it's obviously just a time pod ripoff of Vista's idea. Also shoulder aliens.



  • Linux always has every Windows feature "first", if you define "first" as "a experimental buggy non-working alpha which doesn't ship in any production Linux distribution and whose development was entirely stalled mere minutes before Microsoft announced the feature."

    You have to know your Linux-speke. It's all about redefining words to mean the opposite of what people think they mean. Words like "free".


  • ♿ (Parody)

    If only we could harness the power of blakey's goal posts, we could stop talking about wind and solar boondoggles.



  • I'm the wind, baby.



  • What's a windbaby?



  • It's a baby you wind up.


  • Discourse touched me in a no-no place

    @tar said:

    What's a windbaby?

    The "bad translator is bad" thread is over that ➡ ↗ ➿ way.


  • BINNED

    @flabdablet said:

    It's a baby you wind up

    Or a baby that farts constantly? 🚼 💨


  • Discourse touched me in a no-no place

    @boomzilla said:

    If only we could harness the power of blakey's goal posts, we could stop talking about wind and solar boondoggles.

    He's a veritable army of badgers, moving goalposts like anything.



  • As long as you're not the mediocre song that plays at the end of Castlevania: Symphony of the Night. Luckily, if you buy a version for a newer platform, that song was yanked.



  • @flabdablet said:

    You often toss in this little aside when you're in full Windows Vista+ Memory Management Is Glorious spate, but I have never had any clue what you could possibly be on about. If you can provide actual numbers - or even a vaguely coherent argument - in support of the claim that Windows MM performs better than Linux MM for any given use case, I'll be amazed.

    I've heard from coworkers that our software (both Linux and Windows compatible) has a few issues on Linux specifically because it was originally written for Windows, and supposedly Windows does a very good job of cleaning up after you when the process terminates while Linux does not, causing the process to retain some memory after it dies.


  • Discourse touched me in a no-no place

    @mott555 said:

    I've heard from coworkers that our software (both Linux and Windows compatible) has a few issues on Linux specifically because it was originally written for Windows, and supposedly Windows does a very good job of cleaning up after you when the process terminates while Linux does not, causing the process to retain some memory after it dies.

    About the only way for that to be possible would be if they were using shared memory. Which would be insane for virtually any application; there are a very few places where (deliberately) shared memory makes sense, but most people don't need to touch that stuff ever. If they're doing the sorts of things to make this be actually true, you've got the makings of a full TDWTF story.

    More likely that they're ignorant nitwits. The odds are ever in that option's favour.


  • kills Dumbledore

    Or it's terribad on Linux because of other things that make Linux awful



  • @boomzilla said:

    Seems to be pretty much the same stuff for me. Though that's on a system with 24GB of RAM. Anyways, this seems to be the Linux equivalent:

      en.wikipedia.org
    

    preload (software)

    Sorry, if it's not installed be default then it doesn't count, and if it isn't in the dist's repository, it doesn't even exist.

    Isn't that how it works? Lintards (as opposed to ordinary Linux users) are always telling people how Windows sucks because it doesn't do anything out-of-the-box, and you have to go to third parties to get any software to do anything.


  • ♿ (Parody)

    @SirTwist said:

    if it's not installed be default then it doesn't count

    Really? I'd be a little surprised if no one ever installed it by default.

    @SirTwist said:

    and if it isn't in the dist's repository

    $ aptitude search preload
    p   preload                       - adaptive readahead daemon
    p   preload:i386                  - adaptive readahead daemon
    

    @SirTwist said:

    Lintards (as opposed to ordinary Linux users) are always telling people how Windows sucks because it doesn't do anything out-of-the-box, and you have to go to third parties to get any software to do anything.

    I think Windows' refusal to play nice with 3rd party updaters is majorly sucky. It's definitely less convenient to get to a useful state from start with Windows than a typical Linux install these days.


  • kills Dumbledore

    Don't forget when Windows does include something new "OMG they're muscling in on third party developers to put them out of business. MONOPOLY!"


  • Discourse touched me in a no-no place

    @boomzilla said:

    I think Windows' refusal to play nice with 3rd party updaters is majorly sucky.

    You know who else is to blame? 3rd party updater writers. I see Adobe's finally gotten on the Task Scheduler bandwagon instead of a resident program. But then WHY does it want to run every hour? Hell, it probably can get away with being run weekly. Ditto Chrome, which has not one, but 2 tasks, both of which are "run every hour when triggered" and one of which has a bonus trigger. And ditto GoToMeeting, which is another "run every hour when triggered" bandwagoneer.



  • @FrostCat said:

    WHY does it want to run every hour?

    Windows doesn't have anacron.


  • Grade A Premium Asshole

    @FrostCat said:

    And ditto GoToMeeting, which is another "run every hour when triggered" bandwagoneer.

    Fuck GoToMeeting. But mostly fuck webinars and conference calls.



  • @flabdablet said:

    Windows doesn't have anacron.

    It doesn't need it; that functionality is built-in.


  • ♿ (Parody)

    @FrostCat said:

    You know who else is to blame? 3rd party updater writers.

    You have to assume most software will suck. Still, because Windows makes them all do it themselves, either they suck or they don't exist at all. Mostly, windows is the problem here.

    @FrostCat said:

    Ditto Chrome, which has not one, but 2 tasks, both of which are "run every hour when triggered" and one of which has a bonus trigger.

    See, on Linux, it just gets updates as they're available, like most of the rest of the stuff I have installed.


  • Discourse touched me in a no-no place

    @flabdablet said:

    Windows doesn't have anacron.

    Irrelevant and also wrong (not that it has anacron, but not that that's why it's doing that.) There's no good reason for it to do that. All these triggers are set to run at a certain time, and then to run every hour after being triggered, for one day. In practice that means hourly whenever the computer's on. Again, there's no good reason for any of those programs to check for updates hourly.


  • Discourse touched me in a no-no place

    @Polygeekery said:

    Fuck GoToMeeting.

    Can't be helped--I have to use it for work.


  • Discourse touched me in a no-no place

    @boomzilla said:

    Mostly, windows is the problem here.

    This (in the context in which we were typing) does not follow from

    @boomzilla said:

    You have to assume most software will suck.


  • Discourse touched me in a no-no place

    @boomzilla said:

    See, on Linux, it just gets updates as they're available

    How is "as updates are available" determined, and by what? In the end, it probably comes down to polling by something.



  • @boomzilla said:

    I think Windows' refusal to play nice with 3rd party updaters is majorly sucky.

    I think the necessity for third party updaters is majorly sucky. Third party updates and some way to connect them to a single update mechanism is the right way to do it.

    If Windows had something like ninite built in as standard, with a standard mechanism provided to extend it to include installation/update sources the user chose, then the Windows ecosystem would be far less unpleasant.


  • ♿ (Parody)

    @FrostCat said:

    This (in the context in which we were typing) does not follow from

    If Windows didn't make them write software (which is bound to suck) there would be less sucky software.


  • ♿ (Parody)

    @flabdablet said:

    I think the necessity for third party updaters is majorly sucky. Third party updates and some way to connect them to a single update mechanism is the right way to do it.

    Exactly. The obvious thing is that when software installs, it tells Windows Update what URL to check for an updated MSI. When Windows Updates does its check against MS' repository of updates, it also looks at the third party updates that have registered. When there's a new MSI, it gets downloaded and run. Scale this out to the enterprise stuff.

    @flabdablet said:

    ...then the Windows ecosystem would be far less unpleasant.

    There is a massive blub effect going on with Windows users and software updates.


  • Discourse touched me in a no-no place

    @boomzilla said:

    If Windows didn't make them write software

    Of course, if MIcrosoft DID put in a hookable updater, people would scream that Microsoft is trying to put everyone else out of business.


  • ♿ (Parody)

    @FrostCat said:

    Of course, if MIcrosoft DID put in a hookable updater, people would scream that Microsoft is trying to put everyone else out of business.

    I hear this a lot, but it sounds about as dumb as making 3rd parties write their own updaters. What's the theory on how this puts anyone out of business? I've never seen more than a simple assertion that they would.


  • Discourse touched me in a no-no place

    @boomzilla said:

    The obvious thing is that when software installs, it tells Windows Update what URL to check for an updated MSI.

    Since video driver manufacturers are on Windows Update, others should be able to do so as well. I don't know how much effort that is, but the obvious answer is "too much for people to bother to do", however much that is. I tried a quick search, but all the results were about how to run Windows Update.


  • Discourse touched me in a no-no place

    @boomzilla said:

    I hear this a lot, but it sounds about as dumb as making 3rd parties write their own updaters. What's the theory on how this puts anyone out of business?

    I dunno, but ask the EU about putting browsers in the OS--maybe they can tell you.


  • ♿ (Parody)

    @FrostCat said:

    I dunno, but ask the EU about putting browsers in the OS--maybe they can tell you.

    Yes, exactly. This argument is total bullshit. There were actual organizations that could be put out of business.

    @FrostCat said:

    Since video driver manufacturers are on Windows Update, others should be able to do so as well.

    How do those things get there? Do the manufacturers submit to MS? That's my assumption. Like, part of their certification process or whatever (they still do that, right)?



  • Or, shock and horror, MUSIC PLAYERS!!!!! (Guess how many copies of Windows -N sold? Hint: zero. Even the governments that imposed that dumb rule and made Microsoft create and support a new version for it didn't end up buying it.)

    Seriously, though, I think Microsoft puts a lot more pressure on video card drivers because bad drivers can really make Windows look bad. Whereas a botched Adobe Reader update only makes Adobe look bad.


  • Grade A Premium Asshole

    @FrostCat said:

    Since video driver manufacturers are on Windows Update, others should be able to do so as well. I don't know how much effort that is, but the obvious answer is "too much for people to bother to do", however much that is.

    It is a lot of work, and quite a bit of expense. It would have to be easier and/or cheaper to just tie in to Windows Update than it would be to just write your own (crappy) updater.

    We did it a lot more simply for one of our products. When we update our server-side stuff, we restart the services and it compares versions and updates clients if they differ. The client-side service handles the updating by itself. No need for a separate service. It won't work for everyone though.



  • @FrostCat said:

    video driver manufacturers are on Windows Update

    drivers am special


  • ♿ (Parody)

    @FrostCat said:

    I dunno, but ask the EU about putting browsers in the OS--maybe they can tell you.

    If anything, the fact that Office updates via Windows updates says that if your theory had any merit, they would have either forced MS to open it up or make Office run its own updaters like everyone else.

    @blakeyrat said:

    Whereas a botched Adobe Reader update only makes Adobe look bad.

    Given how many security problems result from unpatched programs sitting around, I'm really surprised that MS never did something to help 3rd parties keep their stuff updated. Because virus ridden machines do make Windows look bad. That was a big OSX marketing point, wasn't it?


  • Discourse touched me in a no-no place

    @boomzilla said:

    How do those things get there? Do the manufacturers submit to MS? That's my assumption. Like, part of their certification process or whatever (they still do that, right)?

    Apparently. I found a page talking about driver submissions, and one that talks about "drivers and software" which implies that non-driver software can be on WU.

    I have vague memories during the Windows XP beta of people screaming about the fact that the starting hurdle was a $100 fee to join a program that let you have the ability to put your software on WU. That seems less of a concern in the days of app stores that require similar fees. My thought is that barring other hurdles, like the kind Apple puts up to get an app in its app store, a small fee isn't a big deal for someone like Adobe or Google, but may be more of one for GPF Software.


  • Discourse touched me in a no-no place

    @flabdablet said:

    drivers am special

    Well, yes. But like I said, I did find a page that mentioned "drivers and software" as if they were two different things, but of course now I can't find it again to see if that distinction was meaningful in context. I thought I've seen things on WU that weren't drivers.



  • @boomzilla said:

    You have to assume most software will suck.

    QFT



  • @FrostCat said:

    a $100 fee to join a program that let you have the ability to put your software on WU

    This is complete bullshit from MS. Adding the ability to register third-party sources for application package updates as a normal MSI installation step ought to be a trivial change to WU client-side. And that's all they'd need to do. They wouldn't need to certify anything or host anything and it would make their OS better for their users.


  • Discourse touched me in a no-no place

    @boomzilla said:

    your theory

    It's not my theory. I am just saying that if MS announced third-party software could be on WU, someone would scream, and you know it's true.

    But MS would probably also put some kind of restrictions, like how drivers have to be WHQL certified, and people would scream about THAT, too.


Log in to reply