The Official Status Thread



  • @Jaloopa said:

    You're right, it does look a bit like Windows

    90% of Linux desktops look like Windows.


  • kills Dumbledore

    Status: Just saw that Motorola have an offer on the Moto X for £160 off, extended until the end of today. Except that when you click through to Moto Maker it shows full price


  • Discourse touched me in a no-no place

    @cartman82 said:

    90% of Linux desktops look like Windows.

    It's got rectangular things with buttons in it and you can move a mouse around and type stuff! </ordinary_user>



  • What part of that is impressing you?

    At least the mixer seems sensibly-designed (via ripping-off Microsoft's design). But goddamned is that one useless taskbar or what! It hurts my eyes just looking at the screenshot. In the second screenshot I can't even tell the name of the window open behind Mahjong (oh, excuse me, KMahjongg) due to the lovely grey-on-grey color scheme.

    @cartman82 said:

    90% of Linux desktops look like Windows.

    It's less blatant a rip-off than it used to be.



  • STATUS:

    "Hmm, I can't seem to get the resize cursor to show on this dialog's borders. There's got to be something I'm missing..."

    *google, google*

    And the dialog crashed while I was googling.

    Now that's the linux I know and tolerate.

    I have a feeling this might end up in the sidebar after all...



  • @blakeyrat said:

    What part of that is impressing you?

    At least the mixer seems sensibly-designed (via ripping-off Microsoft's design). But goddamned is that one useless taskbar or what! It hurts my eyes just looking at the screenshot. In the second screenshot I can't even tell the name of the window open behind Mahjong (oh, excuse me, KMahjongg) due to the lovely grey-on-grey color scheme.

    It just looks and feels nice and polished. Has all the gizmos and widgets. Full audio mixer. Auto-updater. Session management. A bunch of GUI config options. Tasteful design choices (IMO). All this is a given when using Windows or Mac, but is a rare sight on Linux.

    Of course, things start bugging out once you dig deeper (as usual), but it's a nice effort IMO.



  • @cartman82 said:

    It just looks and feels nice and polished.

    Except when it crashes.

    @cartman82 said:

    Has all the gizmos and widgets.

    Like if you click the date in the taskbar it says "Monday!"

    @cartman82 said:

    Full audio mixer.

    Welcome to 2007.

    @cartman82 said:

    Auto-updater.

    Welcome to 2001.

    @cartman82 said:

    Session management.

    I don't even know what "session management" is. What exactly is a "session" and why does one need managing?

    Do you... do you mean like... sessions with your dominatrix?

    @cartman82 said:

    A bunch of GUI config options.

    Welcome to 1984.

    @cartman82 said:

    Tasteful design choices (IMO).

    If you're superhuman and immune to eye-strain. If you put that desktop in a library as a kiosk, you'd be basically responsible for crimes against humanity. "Here's your computer, grandma, good luck reading ANY FUCKING THING!"

    @cartman82 said:

    All this is a given when using Windows or Mac, but is a rare sight on Linux.

    So you use an OS that sucks ass, you admit it sucks ass, and your solution to this is to keep using it.



  • @blakeyrat said:

    Except when it crashes.

    It does show you the memdump in a nicely designed dialog, though.

    @blakeyrat said:

    Like if you click the date in the taskbar it says "Monday!"

    That's a hover.

    @blakeyrat said:

    Welcome to 2007.

    YES WE CAN!

    @blakeyrat said:

    Welcome to 2001.

    For the hundredth time, the planes will be abducted on September- Look, it doesn't matter how I know- Voices? No, I- No! NOT THE JACKET! NOOOO!

    @blakeyrat said:

    I don't even know what "session management" is. What exactly is a "session" and why does one need managing?

    Do you... do you mean like... sessions with your dominatrix?

    That's actually one of the things where Windows lags behind. Nix-like systems can detect what apps are started (and their arguments, windows positions etc) and bring them back after restart. This saved state is called a "session". Mac does this flawlessly, Linux a bit crummy as usual, but in this Distro it seems to work OK.

    @blakeyrat said:

    Welcome to 1984.

    Sigh. No music for me for 6 years.

    @blakeyrat said:

    If you're superhuman and immune to eye-strain. If you put that desktop in a library as a kiosk, you'd be basically responsible for crimes against humanity. "Here's your computer, grandma, good luck reading ANY FUCKING THING!"

    Hah. KDE is clearly aimed at power users. You want a grandma linux, look into ElemetaryOS.

    @blakeyrat said:

    So you use an OS that sucks ass, you admit it sucks ass, and your solution to this is to keep using it.

    I sacrifice GUI to get great CLI environment that matches my servers.

    So yeah, everything is a tradeoff. Welcome to reality.



  • @cartman82 said:

    That's actually one of the things where Windows lags behind. Nix-like systems can detect what apps are started (and their arguments, windows positions etc) and bring them back after restart.

    So can Windows. It's called "hibernate". Welcome to... uh. Well pretty sure Windows 98 had it. Let's say 2000, to be safe.

    @cartman82 said:

    I sacrifice GUI to get great CLI environment that matches my servers.

    Oh really? What's it called?

    Oh... you meant... you meant Linux is "great"? You're using the word "great" to describe Linux' CLI?

    Hm.



  • @blakeyrat said:

    So can Windows. It's called "hibernate". Welcome to... uh. Well pretty sure Windows 98 had it. Let's say 2000, to be safe.

    That's like saved memory state. With linux, it's more like they remember app startap args and positions and then bring them back. I'm pretty sure that's what Mac does too. Much better to start the apps from clear state than keep their memory garbage around forever.

    But I could be wrong about technical details. I just know I can restart Linux and get my environment back and I can't do so with Windows.



  • @blakeyrat said:

    Oh really? What's it called?

    Oh... you meant... you meant Linux is "great"? You're using the word "great" to describe Linux' CLI?

    Hm.

    Compared to alternatives, yes.


  • kills Dumbledore

    My shit sandwich is better than your shit sandwich



  • @Jaloopa said:

    My shit sandwich is better than your shit sandwich

    No it's not, mine has undigested watermelon seeds from yesterday.



  • @cartman82 said:

    That's like saved memory state.

    Yeah, that's what you said.

    @cartman82 said:

    With linux, it's more like they remember app startap args and positions and then bring them back.

    Ok? And that differs from Hibernate which, apparently in your little fantasy fairy world, doesn't bring back "positions" (positions of what?), how?

    @cartman82 said:

    I'm pretty sure that's what Mac does too.

    Ok, first of all, apps don't start with "startup args", WTF.

    Secondly, are you just talking about applications saving their own window position? Why the fuck would that be the OS' job? Windows applications that aren't broken garbage already do that.

    @cartman82 said:

    Much better to start the apps from clear state than keep their memory garbage around forever.

    If you have memory garbage, stop using applications written in C and C++ and other memory leak-y programming languages.

    @cartman82 said:

    But I could be wrong about technical details.

    More "vague" than "wrong".

    @cartman82 said:

    I just know I can restart Linux and get my environment back and I can't do so with Windows.

    Yes you can. It's called fucking HIBERNATE.


  • Discourse touched me in a no-no place

    @blakeyrat said:

    It's called fucking HIBERNATE.

    Does it work if you restart to apply an OS update? Because that's the sort of thing that we're talking about here. Not just sending the machine to sleep.



  • @dkf said:

    Does it work if you restart to apply an OS update?

    I don't believe so, but Windows 8 has that thing where it'll like hibernate driver and OS state, so I honestly don't know if my info is out-of-date or not.

    @dkf said:

    Because that's the sort of thing that we're talking about here.

    After suffering through the vagueness, I think he's literally just saying, "when I move my Steam window into the corner, then restart Steam, the window's back in the corner." Which is:

    1. something every OS (that isn't shitty Linux) has had in its UI guidelines for decades and

    2. isn't the window manager's business anyway and

    3. is such a fucking basic piece of functionality is it even worth mentioning? (I guess Linux is usually so broken that it is.)


  • kills Dumbledore

    I think it's more "do a full restart of the PC and all open apps will reopen in the same place as they were, with any command line arguments used to open them". Not sure why Linux would need that, since it seems to be a point of pride among its users that they never restart anyway...



  • @blakeyrat said:

    After suffering through the vagueness, I think he's literally just saying, "when I move my Steam window into the corner, then restart Steam, the window's back in the corner." Which is:

    1. something every OS (that isn't shitty Linux) has had in its UI guidelines for decades and

    2. isn't the window manager's business anyway and

    3. is such a fucking basic piece of functionality is it even worth mentioning? (I guess Linux is usually so broken that it is.)

    No, I mean: "When I have a chrome cluster with 5 windows / 50 tabs, and 8 consoles and 2 IDE-s and a File Manager arranged how I like, and then I restart, I'd like everything to be brought back as I left it."

    The only OS that does this reliably in my experience is MacOS. Linux can do it sometimes, but there's always that one special snowflake that's not brought back, or that other app that's brought back duplicated.

    Windows can do it with hibernate, but it's a dirty hack really. I had problems with apps being saved in a dirty state and then crashing after restart. So I turned it off. I dunno, maybe thing have gone better. But seeing the technical background of this thing, I'm doubtful.


  • FoxDev

    windows hibernate in 7 was awy better than vista, and it improved again in 8.

    i rarely get apps crashing on resume from hibernate in 8.

    what i do get is BSOD from the craptastic Broadcom based wifi card if i hibernate and then resume somwhere that the wifi i was connected to on hibernate is not instantly available.......

    i keep meaning to order a $30 Intel wifi card to fix that.

    actually now i have because i looked the card up on amazon to linkify that for you.


  • BINNED

    @cartman82 said:

    I had problems with apps being saved in a dirty state and then crashing after restart

    So? That is a shitty app, not MS fault unless it's a MS app. Hibernation and sleep have been around for 10 years and still devs get this wrong again and again.



  • @accalia said:

    windows hibernate in 7 was awy better than vista, and it improved again in 8.

    Think about how this technically work.

    Application has all sorts of sockets and handles open to various devices, R/W files, web sockets, cached knowledge about system resources etc. Then its state is suddenly frozen. Then brought back an indeterminate time later. Now the app is supposed to resume operating in an environment that perhaps has different drivers or even devices than it had moments ago. This handle might be for a device that has been removed during the downtime. That cached local IP address could have been invalidated by the DHCP. What now?

    I don't see how this de-hibernation can ever be made bulletproof in a generic way. There will always be crashes and instability. Therefore, I'd rather do a clean restart, than carry some 2 months old dirty memory along, causing who know what corruption in my system.

    @Luhmann said:

    So? That is a shitty app, not MS fault unless it's a MS app. Hibernation and sleep have been around for 10 years and still devs get this wrong again and again.

    Good indication that they will never get it right. Meaning, it's a badly conceived system to begin with.

    Trust in systems, not people.


  • FoxDev

    @cartman82 said:

    I don't see how this de-hibernation can ever be made bulletproof in a generic way.

    As i recall, it's implemented by a window message from windows saying "oh hey, i'm about to hibernate. you'd better release any remote resources you've got a hold of so i can resume cleanly. i'll let you know when we're back"

    local resources can be handled by Windows so they're not an issue, it's only remote resources that are the issue for resume.

    it's not bulletproof but it's at least somewhat bullet resistant.


  • ♿ (Parody)

    @blakeyrat said:

    Ok? And that differs from Hibernate which, apparently in your little fantasy fairy world, doesn't bring back "positions" (positions of what?), how?

    How does hibernating work with rebooting from system updates?

    Sheesh...are you really this dumb today?



  • @accalia said:

    As i recall, it's implemented by a window message from windows saying "oh hey, i'm about to hibernate. you'd better release any remote resources you've got a hold of so i can resume cleanly. i'll let you know when we're back"

    local resources can be handled by Windows so they're not an issue, it's only remote resources that are the issue for resume.

    it's not bulletproof but it's at least somewhat bullet resistant.

    It's kind of like relying on app developers to save their shit in AppData instead of Documents. Or to create consistent Start Menu entries.

    My Documents are still full of crap, never mind what MS wants people to do.


  • FoxDev

    @cartman82 said:

    It's kind of like relying on app developers to save their shit in AppData instead of Documents. Or to create consistent Start Menu entries.

    well yes, like i said. bullet resistant.

    it'll never be perfect, but we can make it better than shit.


  • BINNED

    @cartman82 said:

    relying on app developers

    We all know relaying on developers to get things done is a bad idea.



  • @cartman82 said:

    Good indication that they will never get it right. Meaning, it's a badly conceived system to begin with.
    Bull. Crap. Hibernate is my usual way of achieving the "I want the computer off now" goal. In fact, I almost never straight shut down my computer. 99.9% of programs work fine with it and aren't affected; the only counterexamples I can think of are programs actively downloading something, and there's no way to make that work; they'll just get a timeout when the system comes back up.

    Just because something can't behave absolutely perfectly doesn't mean its not worthwhile, and in this case it doesn't mean that hibernate still does a far, far, far better job at restoring your state than any session restore feature out there.



  • @EvanED said:

    Bull. Crap. Hibernate is my usual way of achieving the "I want the computer off now" goal. In fact, I almost never straight shut down my computer.

    Well I never shut down computer unless I have to (something is corrupted, there's an update, etc.), so hibernate is useless to me.



  • I shut down my Windows computer when Windows update says "by the way, in 10 minutes I'm going to screw over whatever game you're playing", but my Linux computer...


  • ♿ (Parody)

    @EvanED said:

    Hibernate is my usual way of achieving the "I want the computer off now" goal. I

    I tend to work in a locally running VM. I start it up at the beginning of the day and shut it down when I'm finished.

    @EvanED said:

    Just because something can't behave absolutely perfectly doesn't mean its not worthwhile, and in this case it doesn't mean that hibernate still does a far, far, far better job at restoring your state than any session restore feature out there.

    Too many negatives. Parser overload.


  • Grade A Premium Asshole

    @ben_lubar said:

    but my Linux computer...

    But it is still restarting services, and we all know that is basically the same as a complete and total reboot.

    Let's put it in blakeymath: Fractions of a second === several minutes


  • kills Dumbledore

    @Polygeekery said:

    several minutes

    Liar!


  • Grade A Premium Asshole

    @Jaloopa said:

    Liar!

    Show me a server that will reboot in under 5 minutes. One that is not a VM.



  • I shut down my dad's laptop and it took about 10 minutes from the time I clicked the shut down button to the time the "Windows 7 is shutting down" screen disappeared.


  • Grade A Premium Asshole

    Yeah, most machines without a SSD will take several minutes to reboot on Windows updates. My laptop, prior to SSD install, would take 10+ minutes. More than once it happened as I was about to join a video conference call, which was really annoying.



  • It wasn't for Windows Update, though. It was just a normal shutdown because I had turned it on when I was trying to check if it was in standby mode.

    I've rebooted my dhromebook for updates a few times in front of people who have never seen a dhromebook before. The usual response is "when are you going to start the reboot" followed by me saying "it's already done".


  • Grade A Premium Asshole

    Witchcraft! Burn the heretic!



  • @Polygeekery said:

    But it is still restarting services

    Most services can be bounced "on the fly" -- about the only updates you must effectively reboot for are kernel patches, updates to very core system libraries (GLibC itself, for instance, although these are 'rolled in' as new processes are launched), or X11 patches (because taking down the X server takes down all the X clients with it).



  • @cartman82 said:

    No, I mean: "When I have a chrome cluster with 5 windows / 50 tabs, and 8 consoles and 2 IDE-s and a File Manager arranged how I like, and then I restart, I'd like everything to be brought back as I left it."

    ... right, but you can do that with Hibernate.

    You keep explaining this, but what you don't explain is how this differs from Hibernate.

    @cartman82 said:

    Windows can do it with hibernate, but it's a dirty hack really.

    Oh, so now you admit all the time Hibernate was the feature you were looking for.

    And no, it's not a dirty hack. Not by any definition of the words "dirty" or "hack".

    @cartman82 said:

    I had problems with apps being saved in a dirty state and then crashing after restart.

    Sorry you had shitty broken apps?

    @cartman82 said:

    But seeing the technical background of this thing, I'm doubtful.

    Right; the technical background of Linux is SO MUCH MORE RELIABLE. Oops there's another crash.

    @accalia said:

    i rarely get apps crashing on resume from hibernate in 8.

    I've NEVER had that happen. I've been using Hibernate since XP and I've literally not once seen an app crash after resuming from Hibernate.

    @Luhmann said:

    So? That is a shitty app, not MS fault unless it's a MS app. Hibernation and sleep have been around for 10 years and still devs get this wrong again and again.

    Correct; there's a lot of apps that crash with "unknown audio device" errors when you unplug a USB headset. That's not the OS' fault, that's the shitty application developer making wrong assumptions (that sound cards aren't hot-swappable) and not reading the documentation or testing his shit.

    I've never once doubted that Windows has a lot of shitty software written for it. Most of it is open source, but there's also a lot of shitty broken closed-source software too. Part of having the most software, you also have the most broken software.

    @cartman82 said:

    Application has all sorts of sockets and handles open to various devices, R/W files, web sockets, cached knowledge about system resources etc.

    Hardware in Windows is virtualized. I'd be shocked if it wasn't in Linux also, but who knows.

    @cartman82 said:

    Now the app is supposed to resume operating in an environment that perhaps has different drivers or even devices than it had moments ago. This handle might be for a device that has been removed during the downtime.

    Since the hardware is virtualized, the OS can provide proper error resolution.

    @cartman82 said:

    That cached local IP address could have been invalidated by the DHCP. What now?

    That's not even a guarantee offered by TCP/IP; that could happen at any instant, regardless of Hibernation.

    @cartman82 said:

    I don't see how this de-hibernation can ever be made bulletproof in a generic way.

    Yeah well it's happened.

    @cartman82 said:

    Therefore, I'd rather do a clean restart, than carry some 2 months old dirty memory along, causing who know what corruption in my system.

    Why do you think memory gets "dirty", and what possible "corruption" could this cause?

    Memory leaks can slow applications down, I suppose. But how could a memory leak corrupt something? Any other memory problem generally just results in an instant crash of the program in question.

    @cartman82 said:

    Good indication that they will never get it right. Meaning, it's a badly conceived system to begin with.

    Trust in systems, not people.

    So Hibernation is a badly conceived system because some idiot programmer in some third-world country didn't test his software? Is that what you're saying here?

    @boomzilla said:

    How does hibernating work with rebooting from system updates?

    Like I said above, I'm not sure it does.

    @cartman82 said:

    It's kind of like relying on app developers to save their shit in AppData instead of Documents. Or to create consistent Start Menu entries.

    My Documents are still full of crap, never mind what MS wants people to do.

    Right; but how is that Microsoft's fault? What could they be doing differently to clean-up the situation?

    And you use Linux which is far, far, far worse at this.

    @ben_lubar said:

    I shut down my Windows computer when Windows update says "by the way, in 10 minutes I'm going to screw over whatever game you're playing", but my Linux computer...

    First of all, Windows Update gives 72 hours' notice unless you fucked with it.

    Secondly, if your Linux box has been running for 234 days, you're probably riddled with unpatched security holes. Windows doesn't ask for reboots just for shits and giggles; there is a purpose behind it.

    @ben_lubar said:

    I shut down my dad's laptop and it took about 10 minutes from the time I clicked the shut down button to the time the "Windows 7 is shutting down" screen disappeared.

    It's broken. Fix it.

    @tarunik said:

    Most services can be bounced "on the fly" -- about the only updates you must effectively reboot for are kernel patches, updates to very core system libraries (GLibC itself, for instance, although these are 'rolled in' as new processes are launched), or X11 patches (because taking down the X server takes down all the X clients with it).

    With the way Windows file caching works, all the services must be taken offline simultaneously. If you patch CommonlyUsedLib, and services A and B both use it, if you bounce A without simultaneously bouncing B, A will get the CommonlyUsedLib that B's using in memory, and not a fresh copy read from disk. Again, they don't just ask you to reboot for shits and giggles.

    Linux file caching with its "ghost files" (or whatever you call it) is better at this particular use-case. Both Linux and NT were designed before online software patching was a thing, Linux just accidentally lucked into a better way of handling it.



  • @blakeyrat said:

    Secondly, if your Linux box has been running for 234 days, you're probably riddled with unpatched security holes. Windows doesn't ask for reboots just for shits and giggles; there is a purpose behind it.

    0 packages can be updated.
    0 updates are security updates.
    


  • That means the file on disk have no (known) exploits.

    That doesn't mean the files in memory don't.


  • Grade A Premium Asshole

    @tarunik said:

    Most services can be bounced "on the fly" -- about the only updates you must effectively reboot for are kernel patches, updates to very core system libraries (GLibC itself, for instance, although these are 'rolled in' as new processes are launched), or X11 patches (because taking down the X server takes down all the X clients with it).

    :thats_the_joke.pdf:



  • Status: Waiting for things to run, and meanwhile thinking about Dragonball. Anyone else catch the first episode of Dragonball Super? Looks like it's going to go through the recent movies, and it starts right after they beat Buu. It's pretty good so far.



  • Not that I'm a big Dragonball fan, but I was confused about that. The press release said the new series is a direct sequel, but didn't they have like ... 5 movies that are a direct sequel? So is the new series just... the movies? Cut into 28-minute pieces? Or what's going on?



  • Basically, the author has always hated GT, and has been working to expunge it from reality. So this is now the true sequel to the original. Since it's starting with Beerus, I assume that it will cover all the SSG stuff, including Frieza, which could possibly be done in 12 episodes, but I'm rather hoping they do other things as well and make it longer.





  • @blakeyrat said:

    Why do you think memory gets "dirty", and what possible "corruption" could this cause?

    Memory leaks can slow applications down, I suppose. But how could a memory leak corrupt something? Any other memory problem generally just results in an instant crash of the program in question.

    I just can't trust it.

    For example, let's say I'm using a crappy GUI app to inspect a remote database. I have a bunch of tabs with open SQL queries. I turn the machine off (into hibernation), then resume work tomorrow. But now, is my connection with remote database still active? The poorly written app think it is, but it's really not. I click something, the app tries to load stuff from database and hangs. All my SQL queries are lost.

    Is this the app's fault? Hell yeah. But what can I do about it, if there's no alternative?

    And if you think there's no app this incompetent, let's just say I had an exact app in mind when I wrote this.

    @blakeyrat said:

    So Hibernation is a badly conceived system because some idiot programmer in some third-world country didn't test his software? Is that what you're saying here?

    It relies on software makers investing time and effort into something that won't absolutely prevent them from shipping product and can't be put on the "box" as a feature.

    So... hell yeah I'm saying it's a badly conceived system.

    @blakeyrat said:

    Right; but how is that Microsoft's fault? What could they be doing differently to clean-up the situation?

    And you use Linux which is far, far, far worse at this.

    They could have added an easy unambiguous way to access application-specific files. They could have stapled it all over heavily promoted "best practices" tutorials. They could have had Visual Studio show a little warning if you try to get Documents location directly. Instead, you need to dig 17 layers through their crappy MDN site to get partial outdated information about this.

    And yes Linux is just as bad. Doesn't mean I can't whine about Windows, which I also use.



  • Man, that's practically Zoo Race!



  • @cartman82 said:

    For example, let's say I'm using a crappy GUI app to inspect a remote database. I have a bunch of tabs with open SQL queries. I turn the machine off (into hibernation), then resume work tomorrow. But now, is my connection with remote database still active? The poorly written app think it is, but it's really not. I click something, the app tries to load stuff from database and hangs. All my SQL queries are lost.
    How does an OS X-style session restore thing help this one iota?



  • @cartman82 said:

    Is this the app's fault? Hell yeah. But what can I do about it, if there's no alternative?

    That argument applies to every buggy piece of software for every purpose which fails in every hardware configuration. So you'll forgive me if I don't find it very compelling.

    @cartman82 said:

    And if you think there's no app this incompetent, let's just say I had an exact app in mind when I wrote this.

    Sorry you use shitty buggy software.

    @cartman82 said:

    So... hell yeah I'm saying it's a badly conceived system.

    Ok; so how could it be improved?

    @cartman82 said:

    They could have added an easy unambiguous way to access application-specific files.

    There is one.

    @cartman82 said:

    They could have stapled it all over heavily promoted "best practices" tutorials.

    I hope they have. The kind of developers who produce these shitty apps aren't the kind interested in learning "best practices".

    @cartman82 said:

    They could have had Visual Studio show a little warning if you try to get Documents location directly.

    What does "directly" mean in this context? The direct way to get it is to ask Windows where it us, using the SHGetFolderPath function or whatever wrapper your particular runtime puts around it. The danger is people getting the path indirectly, by either hard-coding a guess, or digging around in the Registry.

    I'd love to see your suggestion as to how to implement such a thing.


Log in to reply