Slashdot Mirror
Microsoft Patents OS Shutdown
An anonymous reader writes "You would think that shutting down software could be fairly simple from an end user's view. If I ask you to shut it down, would you mind shutting it actually down, please? Well, it's a bit more complicated than that, because you need to ask the user if they really want to shut down and if unsaved documents should be saved. And that warrants a patent that also covers Mac OS X. Next time you shut down Windows, remember how complicated it is for Windows to shut down. Perhaps that is the reason why this procedure can take minutes in some cases."
They only cite documents going back to 1998. There has to be prior examples of this happening is there? Also, software pattents are getting ridiculous now a days. What is next, are they going to patent moving a cursor on-screen to select the option to shutdown?
GENERATION 25: The first time you see this, copy it into your sig on any forum and add 1 to the generation. Social exper
When can I, um, patent shutting down the USPTO?
One has to wonder if they are also trying to patent the inadvertent "BSOD" shutdowns. They seem much more complex. ;-)
I can see someone patenting "Are you sure?" prompt.
Actually, when I think of it, alot of dumber patents have been accepted.
If you post as an AC, don't expect me to spend a mod point on you.
Are they going to patent the process by which you tell Windows to shut down, and it freezes?
A standard Windows crash is much quicker.
Slashdot: Failed Car Analogies. Amateur Lawyering. Anecdote Battles.
Slashdot user shutdown -h now, i've got some bad news for you.
Is all the fault of lawyers? Now it all makes sense.
Here's the USPTO link. The abstract:
A user interface and scheme is provided for facilitating shutting down an operating system. Aspects include the operating system receiving a command to initiate shut down, and automatically terminating graphical user interface (GUI) applications that delay shut down which do not have top level windows. Also, aspects provide a user, through a graphical user interface, the ability to automatically terminate all running applications in response to determining that a running GUI application has a top level window.
Read the EFF's Fair Use FAQ
What is ironic is that Microsoft patents this, but my Mac running OS X 10.6 shuts down and off in literally 2 or 3 seconds, whereas Windows 7 on the same machine (and without virtualization) takes 15-20 seconds to shut down and off.
It's called pulling the plug.
Seriously though, wouldn't virtually any version of unix or other multi-user OS prior to MS releasing Windows 3.x qualify as prior art?
That's all well and good, but having a software option to shut down the computer was a Macintosh thing (introduced in Finder 4.x in 1985) about 10 years before it became a Windows thing. One need not worry about this patent.
It is now safe to turn off your computer.
The story of windows shutdown development:
http://moishelettvin.blogspot.com/2006/11/windows-shutdown-crapfest.html
All that work surely deserves a patent.
Shutting down Windows, evacuating ones bowels.
Potayto potahto.
No folly is more costly than the folly of intolerant idealism. - Winston Churchill
So, in effect, they are patenting the ability to NOT use something. I typically just flip the switch to 'off', or leave it running and turn off my monitor. What's next, patenting user logon?
Someone flopped a steamer in the gene pool.
How about those instant "clicks" where windows just turns off. Those never even show up in event logs. Can somebody write a widget for that?
Hard shutdowns = win. Who wants to walk away with the knowledge that some notepad used as a clipboard augmentation prevents the shutdown for 8+ hours because Microsoft's code is retarded. Let them patent a broken shutdown system -- in reality it's not that hard to do. Just make a signal for indicating to a program the OS wants to shutdo... oh wait.. nvm.
I guess I should patent MY shutdown technique - goes back way before then. Make sure nothing important is going on (like a write operation), and just cut the power.
It still works great on modern OSes with a journaling file system - and the best part is that your whole desktop, including open apps and files, is restored next time you log in, and you only lose 2-5 seconds on reboot (which is less than the time you lose doing a clean shutdown), and you don't have to answer 3-4 dialogs asking if you want to save your session, etc.
Do that every time, and over the course of the year, you've saved 30 seconds x 250 days, oe 125 minutes - that's 2 HOURS of electricity. Be green - pull the plug :-)
Seriously, most of the time I shut down properly, but if I hear thunder close by, I just cut the power unless it's a laptop. Lightning doesn't have to be close enough to hear to induce surges in power lines, so I figure if I can hear it, it's already too close. I haven't lost any data doing this, but I *have* had to replace one cpu because of a power surge (and that was in the bad old days when you had to hand-solder them to the board).
Pull the plug. A *real* OS can handle it.
I can't figure out which patent or application the article is referring to. This patent issued to Microsoft last year and covers OS shutdown methods, so I think it's the right one. The first claim is this:
Basically it covers delaying shutdown while an application wraps something up and informing the user that this is happening via a GUI. The more detailed claims cover the circumstances under which this might occur (e.g., a negative response from the application, no response from the application, etc).
This patent does not cover what Windows XP or OS X do in this circumstance. In fact, the behaviors of XP and OS X are explicitly mentioned in the specification, and the patent is meant to cover an improved method for handling the situation.
There are lots of ridicilous patents
I can without hesutation say that there are no ridicilous patents.
I'm starting to think some of the ridiculous patents and patent trolls lately are trying to push the buttons of copyright law in the US just enough so that someone in the government gets a hint and enacts some kind of reform, or better yet, gets rid of software patents. Patenting OS shutdown, I would like to see some of the patents that get shot down in the IT industry. Someone should patent a whole bunch of useless stuff but just not sue. Get the patents just to see what kind of stuff they let through, and then point at them for the kind of ridiculousness they're allowing.
I know *that* is a bunch of crap too and it's probably just an arms race of stupid patents before the other guy gets it and sues you.
Wednesday, November 11 2009, "Microsoft Patents Sudo?!!"
http://www.groklaw.net/article.php?story=20091111094923390
Instead of 'asking if they really want to do it' make it harder to do accidentally in the first place.
For example, require them to specifically select it, and check a box in the same dialog to confirm it.
Instead of every program getting to ask the user if they really want to exit... demand every program save a 'crash copy' of every open file in a special temporary directory, and exit with no user interaction.
Instead of giving every program an opportunity to cancel a shutdown, send a WM_YOUARESHUTTINGDOWNIN5SECONDS
They have two options... either save everything they need to save in 5 seconds, or have the OS "hibernate" them by dumping their user memory area to crashfile, so they can be re-launched at boot
And no program or system driver has the ability to cancel or delay shutdown for any reason.
They only have the ability to advertise ahead of time that they're doing something critical and don't want to be shutdown. The 'advertisements' appear in the dialog when the user clicks 'shutdown'.
If the user is running a defrag, special programs running as administrator might have the ability to gray out the shutdown 'Ok' box, or rather cause windows to prompt the user "XXX program is executing a critical task", requiring the user to 'OK' a manual override --- in this case, the software tells the system in advance that shutdown should not be done right now, but letting software delay the decision to block shutdown until a shutdown's actually attempted, is braindead.
A shutdown should be either "OK" or not. The system should now at all times when a shutdown would not be OK, so it can be performed immediately, and software running on a machine should not be allowed to change its mind and block a shutdown once it has been initiated.
After shutdown's started, some "Begin Critical Task" API should be disabled/blocked.
Claim 1: A computer readable storage medium storing computer-executable instructions for performing a method for shutting down an operating system, the method comprising the steps of: receiving a command to initiate operating system shut down; sending a shut down request to a graphical user interface application without a top level window; receiving no response to the shut down request with a predetermined period of time; determining that the graphical user interface application without the top level window is not hung; automatically terminating the graphical user interface application without the top level window; determining whether any graphical user interface applications with a top level window delay shut down; prompting a user for a user command to selectively shut down the graphical user interface applications with the top level window that delay shut down after determining that the graphical user interface applications with the top level window delay shut down; and then after the determining step, automatically terminating all running applications responsive to the user command received from the user that has been prompted.
It's obviously not very innovative, but it's also probably patentable under US patent law. I don't personally know of any prior art, and in fact I think I wouldn't want my OS to shut down like that.
Qxe4
Because the proposed shutdown process is so blatantly wrong that any inhibition that blocks misguided developers from copying it is a good thing.
What should happen is that the apps all shut down and go away without any IF/AND/ORs/BUTs about it. If they need to save some state or or whatever then they should get a chance to do so. They should NOT EVER get to communicate with the user or otherwise delay the shutdown process. The OS layer should chose, prior to sending the message, if there is any opportunity for GUI interactions (and should not normally do so).
In an ideal world this would force application developers to maintain proper state and properly deal with auto-save and other features to make the user's life easier. The current model, as detailed in this patent, just pushes the problem off to the end user (which is wrong).
"because you need to ask the user if they really want to shut down and if unsaved documents should be saved"
This is one of the most annoying things about computers. If I want to shut it down, shut it down!
It is to late for questions, I probably already left after I issued the shutdown command.
Any question about unsaved documents can be asked the next time I start the program, just save them in a temporary location in the meantime.
Standby and hibernate have somewhat mitigated this problem, but for multi-user systems there is still no practical solution.
When I use windows and I ask it to shutdown, he doesn't always do so... is this covered in the patent?
I worked at Microsoft for the Windows 95 launch, where I provided Tier-1 support for BOOP (Bill and the Office of the President, i.e. CEO tradeshow tech support). I do recall that Bill specifically called out the 'shutdown' function on Windows 95 as an error. He didn't like it, he hated the idea of waiting for the OS to shutdown, and wanted simply to be able to push the power button to immediately turn the system off, like a DOS PC.
He may or may not have understood the concept of in-memory caches and unsaved user work, but it didn't much matter to him.
Not mad really, I just I wish I had the nerve, and lawyers, to patent something so totally ridiculous. One could get the same output if you put 1000 monkeys into a room with some Win dev machines; 1000 monkeys all throwing chairs and their feces onto the keyboards will eventually come up with a patent to shut down a system. Good work!
I would probably rather wish to be a televangelist; lots of money and no sense of playing fair or trying to using your own ideas, or using real skills, to make you money. Yeah, that's the life for me; fucking over people, taking their money, and having no remorse. Good work! Be proud, you assholes!
This is the NSA, we're gonna geet U h@x0r5! Also, what is a h@x0r5?
...I thought the BSOD was already patented---it's a feature, not a bug, right??? :-)
Twelve-and-three-quarter inches. Unyielding. This wand belonged to Bellatrix Lestrange.
GNOME: Official release: 3 March 1999
As the first poster in this thread notes, "They only cite documents going back to 1998."
Not saying that it should be a valid patent, just that GNOME doesn't qualify as prior art.
99.9% of lawyers gave all of them a bad name.
Even a new version of "Hello World."
That, by itself, doesn't make the effort patentable. It also has to be non-obvious to other practitioners of the art, namely other programmers in the operating systems domain.
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
This is not a very powerful Idea but still very confusing incident in software processing proocess. Cheap Cars
Probably the fact that it doesn't seem to be getting any better. It's a waste of resources for pretty much everybody involved and it makes software more expensive/less available for everybody as a result.
Not tied to a specific machine and therefore fails test In Re Bilski?
Donte Alistair Anderson Roberts - hi son!
Karma: Chameleon
I know it is a slashdot pass time to make fun of strange patents like this but it is worth noting that there are government incentives (beyond the patent itself) to do something like this. In Canada, at least, there are programs like SR & ED
Applying for these tax incentives can often lead to outrageous descriptions of basic things. Each employee is asked to say what they worked on this year and to fill in a lengthy form describing what is new about it. A programmer that designs an options menu might actually be working on "A graphical interface that allows a user to list, navigate and manipulate system configuration parameters." I think the idea is that if you make the description dense enough the bureaucrat reading it will just approve it without thinking about it too much.
Asking about saving is the wrong question. Saving shouldn't be a question. A document's current state should be persisted at the drop of a hat and that means undo info as well.
A small faction at MS gets this.
Android, as a platform and as recommended dev practice, gets this. Many great IPhone apps get this.
An app should expect to be terminated rudely and abruptly at any time. You'll impress the hell out of your users if you follow this rule.
meh.
I'M taking out a patent on the letter 'M'. I can see quite a few people here are going to owe Me Money.
As far as I can see, the patent is not about the OS shutdown in general, but the very specific way in which it is done in Vista and Win 7. Particularly, this involves the full-screen notification that some applications are preventing the system from shutting down that you eventually get after clicking the shutdown button, and the surrounding OS behaviour towards such applications (recognizing which running programs warrant putting up the prompt, signaling/terminating them...). This is not the traditional Unix shutdown stuff in any way, so unless you give the user the choice on shutdown to kill processes that don't want to quit by themselves yet, you should be rather safe. Still, this is yet another software patent :-(.
Personally, I find this full screen thing rather annoying anyway, mainly because Firefox interacts badly with it, showing its own exit prompt on shutdown and being overlaid with the that fullscreen prompt, preventing me from clicking the save and exit option in Firefox.
http://www.moonlight3d.eu/
What is with all the smarminess about shutting down being easy? Says who? Sure, you could just halt everything, simply send a command to the board to cut power. That would be simple and quick. It also would be a good way to lose data or have problems. To gracefully shut things down as quickly as possible is more complex. You want a way to signal all apps to quit quickly and get them the CPU time they need to do so, but not stall out the system. You also probably want a way to back off for a bit if an app needs more time, but not for too long. You need to check and make sure everything is consistent, etc, etc.
I'm not saying this is some Herculean feat but acting like shutting down quickly and orderly is trivial is silly.
IDK what the summary was complaining about,
Windows 7, 32 bit on an old athalon x2 2800.
seriously 6 seconds.
How much is your data worth? Back it up now.
The current X windowing protocol, version 11, was created in 1987. X itself was first created before I was born. I think Unix is pretty old.
Support my political activism on Patreon.
You need to be able to recover from a crash anyway, so why write both: infrequently used but critical crash recovery and frequently used, but not always run clean shutdown? Just crash. Crash-only software: More than meets the eye
Now imagine somebody else asking you these questions... you see how complicated this can become? Luckily, it doesn't have to be that way. All you have to do is license Slashdot's patented OS Startup Procedure and you'll no longer have to lie awake, agonizing about how to deal with all this complexity!
I forgot to mention she has everything on an x-25m.
That might have something to do with it.
I tell her "she has the faster" computer but mine has WAY more horsepower.
Since 99.9 % of her utilization is zunepass, outlook, and IE. I really can't fault microsoft. She for some reason really likes them.
How much is your data worth? Back it up now.
If designing a shutdown process is this complicated maybe it does deserve a patent.
It doesn't shut down until you click on some dialog button. Which is not "would you mind shutting it actually down, please?"
Maybe Computers will never be as intelligent as Humans.
For sure they won't ever become so stupid. [VR-1988]
It'll shut down real fast.
Why is Microsoft applying for a patent when the Macintosh has had this since OS 1.0?
What? I'm supposed to send Microsoft some money every time I want to write an application (or an app) that declares an interest (wants to be notified,) when a system shut down request event occurs.
What kind of idiocy is this crap?
Has Microsoft finally lost its marbles? (The only thing I can think of is that Balmer is an idiot. Somebody has to hurry up and kill all the damn lawyers before they ruin everything.)
MSBPodcast.com The opinions expressed here are my own. If you don't like 'em... Think up your own stuff.
... smashing the computer with a 9 iron out of frustration?
Check your premises.
A: The USPTO looked in the mirror and found prior art!
Hope is the currency of fools
(it was on slashdot, even, but I'm not going to try and use slashdots shitty search for it)
http://yro.slashdot.org/story/09/11/11/2055226/Microsoft-Patents-Sudos-Behavior . A one word search ( sudo ) and it was the FIRST LINK.
I know bashing slashdot is the cool thing to do, but try not being a total assclown about it, mmkay? At least bash the stuff that doesn't work.
They complained you had to click Start to shut down.
They said it took too long.
Why do MS want it? They just want everything. They probably want everyone to hate them, too.
If they want everything and want everyone else to have nothing, they want to force everyone to use their system. Seems a bit anti intuitive to me. I don't like companies that piss about like that, that's why I'm fed up with them, and won't buy from them ever. They should innovate, not shut everyone else down. They think they'll get money, but the real way to make money is to make something better, something more competitive. something more likely to make people actually want it, not to force them.
Anyway, how simple is... terminate tasks, sync disks, unmount, send shutdown signal? I can do it in 5 seconds.
________________
Sorry, wrong convo.
Indeed shutting down an OS is very complicated because you have to clean it up for the next startup. Think e.g. of running apps keeping open handles on network files. Think of tunnels of tunnels over everything. Components have to be shut down in order and the order is not always trivial to find. Then you give some time to apps to clean up their stuff. The whole thing can be a real mess...
Correct me if I am wrong here but is that not totally up to the specific applications?
The OS would ask all the running programs to close, but it would be up to the programs themselves to save the content if applicable.
Troll is not a replacement for I disagree.
Infringement is not decided on the abstract, they all revolve around the claims, which are usually conveniently buried in the middle. What are they?
I don't support the patent, but I also don't support the submitter's complete troll of a description.
Windows 7 shuts down very fast, and it's actually really nice that it asks if you want to close unsaved documents - I'd rather have to click one more time on "Force Shutdown" than have it force Photoshop or Word closed when I forget one of my tens of documents isn't saved yet.
WHO NEEDS SHIFT WHEN YOU HAVE CAPSLOCK/ DAMN1
...I miss my original Amiga and my old Palm device. The former, which could be instantly shut off (off, not "down") by flicking the switch, and the later, which instantly went idle with a push of it's power button, and instantly was back up to it's previous state with another push.
I can't wait until modern computers catch up to the 1990s and 1980s respectively.
I patent OS boot up!
here is claim 1 " 1. A computer readable storage medium storing computer-executable instructions for performing a method for shutting down an operating system, the method comprising the steps of: receiving a command to initiate operating system shut down; sending a shut down request to a graphical user interface application without a top level window; receiving no response to the shut down request with a predetermined period of time; determining that the graphical user interface application without the top level window is not hung; automatically terminating the graphical user interface application without the top level window; determining whether any graphical user interface applications with a top level window delay shut down; prompting a user for a user command to selectively shut down the graphical user interface applications with the top level window that delay shut down after determining that the graphical user interface applications with the top level window delay shut down; and then after the determining step, automatically terminating all running applications responsive to the user command received from the user that has been prompted. " Does anyone skilled in the art see any obvious holes ?
Then they'll really be able to get rid of the competition. They'll have them coming and going
We need a "+1 -- nice sig" moderation.
I patent the process of locking up my house when I leave? Or maybe the process of how I get out of my car when I go somewhere? Is there really a need/ability to patent this type of thing? And more so why does the patent office still allow it?
I will shred my adversaries. Pull their eyes out just enough to turn them towards their mewing, mutilated faces. Illyria
Didn't we learn yesterday that the need to shutdown is obviated by ksplice --- wait that would be Linux wouldn't it.
If you shut down, it gets aborted, then shut down again without changing anything, it's not legitimate.
More likely the application accpets the shutdown, but returns the wrong value to the OS. The app shuts itself down but says 'abort'. Try again and the app is already gone, and shutdown proceeds normally. This is an application bug.
Or something else like the app gets sent the shutdown message and takes too long to process. The OS thinks it's hung, so it starts hung app processing. But the app finishes working and goes away in the meantime. The OS thinks there's a hung app, but there isn't. This is a Windows bug, assuming nothing has changed since you went through the previous list.
AFAICR, my Amiga 1000 had an Exit function in the top RMB menu wich would kill all apps and leave you with a blank screen. They you could turn off the Amiga. (Not that it took any damage from just turning it off!)
Can I light a sig ?
If a very simple-sounding process is complex enough to be patentable, then it's probably too complex.
Unix shutdown is a simple matter: shut down all programs that have a known shutdown procedure, tell all remaining applications "we're shutting down now, so you'd better save your shit or something", then tell all remaining applications after that "we're shutting down now, I already told you to quit nicely, so blow up already", then unmount filesystems, poof, done. Application developers know that SIGTERM/SIGHUP is a good hint that the application should terminate as gracefully as possible, and the end of the world may be imminent so it'd better hurry up. Application developers should know that failure to terminate as a result of that signal may mean the user and/or the system may bring up the big guns next. There's no need for complex heuristics to see whether applications are hung or whatever: "Does this app shut down gracefully? No? Who cares, then - the user wanted a reboot, the user shall get a reboot. Off ye go." There's already a good mechanism for shutting down applications gracefully. What else would anyone need to implement shutdown procedures?
Now if an app is holding things up, after a time Windows will pop up a screen saying "This application is holding up shutdown, you want to wait on it or just kill it?" If the app finishes its business before the user responds, shutdown continues. I used to see this with Steam. It shuts down, it doesn't zombie, but it took longer than Windows wanted to wait sometimes so Windows would let me know, but then right after that came up it would go away and proceed.
Seems like a reasonable compromise between just waiting on apps forever (which could literally be forever if the app is zombied) and just killing things without saving.
They should patent illogical naming of processes.
Come on, you need to click "START" in order to STOP something???
I think of it as "start shutting down"... once I accepted that double-think I got on with my life.
XML is a known as a key material required to create SMD: Software of Mass Destruction
#> shut down
?
Who does that anyways?
I only ever do a
#> reboot
after
#/usr/src/linux> make && make install
maybe
#> hibernate
on my laptop from time to time.
"More likely the application accpets the shutdown, but returns the wrong value to the OS. The app shuts itself down but says 'abort'. Try again and the app is already gone, and shutdown proceeds normally. This is an application bug."
Sorry, but if the user tells the OS to shut down and an app has the power to override that indefinitely, that's an OS bug, not an application bug. An application shouldn't have that power.
All the Mac and Linux machines run forever, so they don't need any shutdown patents anyway.
I only ask since you typed it as:
shutdown -s -f -t 0
and I would think the correct syntax should be:
shutdown -s -t -f -u ;-)
Is it just my observation, or are there way too many stupid people in the world?
They forgot the step where you check the hanging process isn't the shutdown process itself. Someone patent that step quickly.
Is that when the screen turns blue? Hell, they can have that patent.
Yes.
Believe it or not, the world of GUIs did not start with windows 95.
"linux is just DOS with a UNIX like syntax" -- Galactic Dominator (944134)
Push and hold the power button. Works every time.
(I am not responsible for any data loss that results from the above advice).
-ted
First of all - shouldn't the patent, to be valid, cover all possible exiting scenarios and possible events, including unexpected crashes? ie. close app A, dont save changes; close app B with changes saved; close C.. oops C crashed, 3 hrs of work gone, tough luck; close "system" ... wait, wait, get some tea, wait, wait... cold reboot.
Therefore, if crashes are included in the patent, doesn't that prove that shutdown procedure is unpredictable, hence not patentable? Separate "crash scenario" patent perhaps?
On another note, isn't this equivalent to patenting something like "exiting the subway" (gather your belongings, make sure you have you wallet, cell phone, etc., get up, wait for the train to stop, step out on the platform?
Weird days we live in.
I had the painful experience of helping a friend figure out why he could not get onto his banks web page using Microsoft Windows Vista. Turns out it was Vista screwing up the system time but on the way to figure that out, I had to reboot the computer a half dozen times due to software updates and many of them application level updates.
So no wonder someone at Microsoft thought it was important that they map out and get a patent on how to shutdown a computer. They've probably spend more man-hours working on it than any company in the history of computing and still fail at it. IMO
LoB
"Anyone who stands out in the middle of a road looks like roadkill to me." --Linus
"Well, it's a bit more complicated than that, because you need to ask the user if they really want to shut down and if unsaved documents should be saved." ..................
No, the OS notifies the applications that they're about to be killed, and they take care of prompting and user interaction. The only difference is whether the OS decides to force KILL or not if the app doesn't close.
Yes, I did. You haven't looked at any desktop environment prior to windows 95 though have you?
"linux is just DOS with a UNIX like syntax" -- Galactic Dominator (944134)
Now all we have to do is enforce this patent, and get that BS out of the other systems!
If someone is passing you on the right, you are an asshole for driving in the wrong lane.
I've basically been running my systems continuously. Glad to see someone making progress on this no shutdown bug.
All of those companies have an OS that uses a similar shutdown.
This is so simple it does not merit a patent. 15 minutes design time will cover it, 30 minutes at the most.
I think Unix is pretty old
It's 41 years old. Next year it will be the answer to the question of life, the universe, and everything.
X itself was first created before I was born
In 1984, Bob Scheifler and Jim Gettys set out the early principles of X
That's the year my oldest daughter was conceived. Is your name "Leila"?
Free Martian Whores!
Maybe it shouldn't, but I can remember before that feature was introduced shutting down my computer and going "oh shit" and having to hurry up and save a file before the app was force quit. Most of the time I made it.
Heck a number of times the "save, don't save, cancel" dialog has come up and I have relaized that actually I forgot to do something (needed to print a documented, etc) and I hit cancel.
From a pure engineering perspective it is correct? Probably not. But it sure is useful.
Need help treating your acne? Come here!
As a naive user, why should I have to ask my computer for permission to shut down?
Because if you agree to let yourself be inconvenienced slightly around the edges, we (the systems designers) can make the big part in the middle much more convenient.
Ever encountered thrashing (excessive swap file reads/writes)? If you want to be able to turn the system off on moment's notice, you're asking for all data to be written to disk at all times. That is, instead of having RAM between CPU and disk, the CPU should just write straight to disk. That is, it should write to disk all the time.
You're asking for thrashing to be the way computers operate by default. You don't want that. We are in fact so certain you don't want it that we are arrogant enough to make the edge-inconvenient way the default without asking you.
Or rather, given what most people do with their computers, that's the best way for them to work. If you're really insistent, you're welcome to run on a diskless workstation or off a Linux LiveCD, or mount all your file systems read only.
Let's see, your TV doesn't store much data and can afford to sync every time anything changes; neither does your car. Your VCR, I would assume, can sync rather rapidly. Also, you don't install new applications on any of those, and you don't complain when your VCR player can't play the new "DVD" format. I don't know about your cell phone, but my 5 year old dumbphone has a cute shutdown animation to cover up the fact that it's a computer with all its inherent complexity. And my N900 which runs Linux; well, go figure...
In short: computer behave differently because they have to meet different demands. If you want something other than what computers give you, well, all the more power to you I guess. It might be expensive to build if it's only you who wants it, though.
I live in South Florida, otherwise known as the lightning storm capital of the world.
Several years ago, lightning hit the pole behind the house. We know it hit that pole because the top of the pole caught fire, in addition to damaging some of the power-company equipment. In my neighbors house, every single electrical device got completely fried. In my house, the only damage was one of the heating elements in the hot-water heater. Not even a light bulb.
I have always has surge protectors on literally everything that could be protected. One under each desk. One under each TV and stereo. One behind the microwave in the kitchen. One behind the refrigerator. The washing machine. Literally, every 110 volt plug in the house. For the computers, the UPS units were even plugged into surge protectors.
After the first lighting strike, I decided I didn't want to buy a new hot water heater again so soon. And, I wanted to protect a few other 220-volt things. So, I found a "whole house surge protector" and got it installed. The thing attaches to the breaker panel at the service entrance. It has varisters on all 6 pairs from ( +, -, neutral, ground ), and it weighs about ten pounds. And, I upgraded some of the little surge protectors to UPSs, for the non-computer devices that we don't like to reboot very often, like the TVs.
Two years later, we had another lightning strike. This one hit the same pole, which caught fire again. My house had no damage at all. Nothing. Three other houses on the street got everything fried, again.
After that, I read up on surge protectors. Turns out, all of the varisters in a circuit work in parallel to protect the whole circuit. If you have a hundred varisters on the (+, ground) path in your hours, even though they are divided into a couple dozen different boxes, the current from a surge on the + conductor flows through all of them in parallel to reach ground.
Seems I may be guilty of overkill on the surge protection. All total, I probably spent close to $1000 on surge protectors. But then, I haven't had to buy any new appliances after lightning strikes, either.
How else would we figure Microsoft is still kind of a company who tries to patent shutdown enhancements instead of adapting to a future which nobody will shutdown? This sounds like not enhancing NTFS driver (note: driver, not the format itself) but "enhancing" defrag instead, making it automatic. E.g. instead of making the OS fragment LESS, lets enhance defragment process.
In such stories, I also want to hear from a developer team/gang who trusts this particular companies "word" or "promise".
If Apple and MS didn't have cross patenting/licensing deals, OS X developers, especially after launchd could have some stuff to tell them.
Well that's shy I shoved in the word "indefinitely." "save, don't save, cancel" dialogs are fine. What's not fine is an app, especially a piece of malware (and yeah this does happen), aborting the shutdown procedure entirely.
Ever encountered thrashing (excessive swap file reads/writes)? If you want to be able to turn the system off on moment's notice, you're asking for all data to be written to disk at all times. That is, instead of having RAM between CPU and disk, the CPU should just write straight to disk. That is, it should write to disk all the time.
The Mac doesn't seem to have this problem. Your argument is invalid.
My current desktop environment is FROM 1998 - Enlightenment 0.16 with the Ganymede theme. Since then Enlightenment 0.16 has only really had bug fixes. I still keep it because it is fast and does those handy things like having iconified apps shown as thumbnails of the actual running window - something that is now in Windows7.
I do that too; on many other things, the car beeps to make it obvious that there's *some* little thing along those lines to take care of.
I listen to both RIAA and non-RIAA stuff if I like the music, tangential business/politics nonwithstanding.
Prior art - that line was used at the end of a computing variation of the Abbot & Costello baseball sketch
I listen to both RIAA and non-RIAA stuff if I like the music, tangential business/politics nonwithstanding.
BillG was right! (Oh shit, I didn't say that did I !!!??)
The application the user was running just took minutes to save the file it has stated that the user data is safely on the disk. From the user's point of view THERE IS NOTHING TO SAVE. So from the user's point of view it's safe to turn off, hard, immediately.
It's even pretty easy to do at the technical level. Databases have been doing this sort of thing for decades. Of course for a filesystem it should be slightly different. Ie: any file that gets an fsync() should be considered important enough to commit to disk early. But everything else gets committed strictly on a snapshot basis because there are no "commit transaction" semantics.
So, every few seconds (or minutes) the system is "checkpointed" such that on a hard reset it will come back to the last, completely consistent, checkpoint (especially including user data). If someone uses fsync() on a file just that file is saved away to disk with instructions to 'fix' the filesystem after a crash. You would do it this way so you could reduce power consumption and improve speed on crappy drives compared to committing every single tiny disk operation.
In technical terms I suppose that could be a copy on write filesystem with a rollforward log for the fsync() and related APIs. I guess doing an fsync() on a file would mean that it isn't copied on write any more, that way a DB could stop external fragmentation and LILO and friends could still work.
Microsoft have tried to do it; NTFS has it's log file, but it only stores the least important data on the disk. So you will get in trouble if you frequently hard power off NTFS. (Only the metadata is logged; Think about it, the user does a backup of the files, they get some metadata but if just the metadata is broken they don't care unless it means they lose the file too.)
Where's my Linux version? The closest is ext3 with full logging but that writes everything twice and checkpoints the entire drive if I tell it to fsync() a single file. I suppose I'm okay with that, for now, but I'm still a bit disappointed... except, I hear ext4 is no better ...
I'm wondering how much of the X Session Manager functionality is prior art?
No but I recall banging a girl named Leila a year my senior... coincidence o_o
Support my political activism on Patreon.
Since we can patent software, I am patenting my new "Hello Universe" program.
Abstract.
Prior art would include the many and various well known "Hello World" programs. This present invention teaches a method and system for a first computer program, when combined with a hardware processor which executes the present invention, that prints out three more Roman characters than inventions taught by the "Hello World" program described as prior art. This present invention claims embodiments in multiple forms and implemented in multiple languages. One embodiment of the present invention would include web server implementations of Hello Universe in multiple computer programming languages. Another embodiment of the invention would be in the form of command line programs written in any variety of programming language. Another set of claims of the present invention would be GUI (Graphical User Interface) embodiments that can be implemented in a large number of user interface frameworks as well as a variety of programming languages.
Additional claims of the present invention include computer programs in any of the forms of embodiment of the previous claims that print any messages of length equal to or larger than the "Hello Universe" message taught by the present invention.
I'll see your senator, and I'll raise you two judges.
Actually, I did have surge protection on the light bulbs, I just didn't realize it. To understand why, it helps to understand how surge protectors actually work.
The electric circuit in a house is a buss with four conductors (+, -, neutral, and ground). The + and - conductors normally supply current from the service entrance breaker/fuse panel to each device. The neutral conductor returns the current from each device back to the panel. The ground is for safety, and does not normally carry any current. Surge protectors work by placing a varister between two conductors of that buss. If the voltage difference between the two conductors is low (say, below a few hundred volts), the varister has high resistance and does not short the circuit. But, if the voltage difference between the two conductors is high, the varister has low resistance, and creates a 'short' between the two conductors. The short provides a low-resistance path for the surge, so more current goes through the varister, and less current goes through other devices on the buss.
All of the varisters in a circuit work in parallel to protect the whole circuit. That is, any surge protector on the "+" side of the main panel, helps to protect everything on that side. And, any protector on the "-" side, helps to protect everything on that side. If you have a hundred varisters on the (+, ground) path in your house, even though they are physically divided into a dozen different boxes, and supplied by different breakers or fuses, the current from a surge on the + conductor flows through all of them in parallel to reach ground.
Of course, the shorter the path to a device, the sooner the surge reaches that device. Electricity flows through copper wire at a speed of a just about 1 foot per nanosecond. So, if the varister is located closer to the source of the surge than some other device, the surge gets to the varister before it gets to the device, and shorts through the varister before it damages the device. Optimal protection is achieved by placing the surge protection varisters at the source of the surge, which is typically at the service-entrance.
'nuff said
Windows assumes you are an idiot...Linux demands proof.
Granted, it doesn't lists *ALL* the blocking applications, only lists all the still opened sessions.
But since KDE3 (launched in 2002) on Linux, when you try to shutdown your computer, you get a dialog listing all the session which are still opened elsewhere (mostly, command line on virtual terminals) giving you a choice to cancel the shutdown or forcibly interrupt those sessions.
So, ok it's not an exhaustive list of application (unlike Windows 7's) but it predates the patent filing (2006) at least by 4 years.
"Sufficiently advanced satire is indistinguishable from reality." - [Tips: 1DrYakQDKCQ6y52z6QbnkxHXAocMZJE61o ]
The Mac doesn't seem to have this problem.
You may not experience this problem, but that doesn't preclude its existence. Unless you have infinite RAM, it could potentially all be used up.
I may be ignorant and OS X might handle memory overuse in a different way; it might start closing your applications "randomly" when they use too much memory instead of degenerating performance, or it might schedule each program for a veeery long time (so as to make the I/O overhead relatively smaller).
But at the end of the day, if your computer doesn't take up the whole universe (and if you think it does and you use a Mac, you're wrong, that's your ego; unearned by the way, it's bought for money), it can run out of resources, and will have to handle that, and there are no good ways of doing that, only ways that suck differently.
Hey, maybe that's a good tagline. "Mac: suck different".
(No trolling intended; Linux sucks differently too, as does BSD, Windows, Haiku, OpenVMS, ReactOS, ...)