Skip navigation.
home | news feeds | archives | search | forums | submit news
Home

Suspend and hibernate names

Submitted by hughsie on Sun, 2006-05-07 12:22. GNOME
GNOME

Microsoft Windows and OSX have decided on names for the sleep actions and it's about time Linux did the same. We are hurting the user experience with ad-hoc and confusing policies about naming.

Developing gnome-power-manager (and a chunk of the power management in HAL) gives me first-hand experience of the ways users, distros, packagers and developers can and do get this wrong.

From an Ubuntu bugzilla entry:

here's from NOTES from 'hotkey-setup'
When the machine should be put to sleep in some fashion:
KEY_SLEEP signals Suspend to RAM (Suspend, technically called "standby")
KEY_SUSPEND signals Suspend to Disk (Hibernate, technically called "suspend"...)

I hope this page can convert some of the wacky/zany/confusing/crazy software out there to stick to common names.

I'm now trying to get project, vendor and distro buy-in so that we can all be using the same names. Comments appreciated.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Is this a parrody?

Submitted by Scarlet on Mon, 2006-06-05 22:29.

I SWEAR this must be a joke or something, at least I hope it is.

"Microsoft Windows and OSX have decided on names for the sleep actions and it's about time Linux did the same." sounds like is must be coming from someone lampooning the sort of soapbox rhetoric on slashdot.

If this is serious, then that is saddening for the following reasons:

  • Trivial issue: I've never suspended (or whatever the heck it is called) a computer under any operating system, let alone multiple OSs. I'm also sure that if I did, I wouldn't let the names confuse me and even if I did select the wrong option, I wouldn't care because the end result would be similar. With GTK, TCL/TK, XAW, QT, AWT etc. all functioning just differently enough to make life hell who really would honestly say that fixing the words for partially deactivating a computer is the best place to start.
  • Unilateral abetary declarations: For all the many different possibilites of names, the developer of one piece of software states his preference. We all assumed he had a preference, what difference will this make.
  • Publically grandstanding: if this is an issue that needs to be resolved amongst people involved in power management then what is it doing on gd.o? Simply inviting a lot of people to give their remarks as to the colour, size and foundations of a bikeshead never helped anyone. This needs to be discussed with the parties involved, not us.
  • Citing of Windows and OSX as reason to change something. That has never gone down well, especially within gnome. If people wanted to use Windows they'd use it, it's not all that expensive. Rarely has windows done something right and there is no point emulating them on the off chance that they have. It is a particually bad opening point and is likely to bring up the same stupid debate as to whether emulating windows has any merit blah blah blah.
  • Insulting of other software: I'm sure calling software written by people who name things differently "wacky/zany/confusing/crazy" will really make Klaus von KDEwriter want to switch his strings for you. That's some top grade leadership there.

If I could go one day without seeing another computer geek thinking he can establish unity by stating what he wants on a glorified blog I would be happy. Hopefully this is all a joke and today is that day for me.

AAAAAAAAAAAAAAAARRRRRRRGGGGGGGGGGGGH! by Anonymous George (not verified)

Don't you think?

Submitted by Anonymous George (not verified) on Thu, 2006-05-11 05:08.

Don't you think that a lot of the people thinking about moving from Windows are doing so because they don't like the user experience?

If you have a heartbeat you've been trained to use windows in 95% of cases.

Emphatically not. Most computer users I know were never trained to use Windows, they stumble their way through it. So many people don't know the most basic Ctr-X Ctr-C Ctr-V. They use the mouse because they don't know any other way. If they are brave enough to use Help, it just confuses them anyway.

I'm not a hacker, I do hold two design degrees with more than a little knowledge and experience in ergonomics and user interaction. However much M$ spent on developing the Windows interface, a lot of it is terminally broken.

People dislike Windows because they don't understand what's going on half the time and don't feel in control of their own data. Windows interface is not consistent, is full of ambiguity, leads you round in circles.

The words used to communicate what a computer is doing ARE important. Yes, people learn new terms to a certain extent but they need a sensible starting point. Not a familiar but never understood one.

The "What people are used to" argument is tired. In many cases, what people are used to is NOT UNDERSTANDING. Do you want to continue that?

Windows is far too low a target to aim for in UI design.

Snooze and Hibernate

Submitted by Anonymous George (not verified) on Tue, 2006-05-09 10:41.

Snooze or Nap implies light sleep, Hibernate implies deep sleep. Wake is applicable to both states.

It's been said that Hibernate would not translate worldwide and I agree but it can probably be translated as Deep Sleep in most languages.

But what do the two do and what does the user achieve by using them?

One saves energy and the other ALSO secures data
Maybe more appropriate descriptive terms are not a bad idea.

"Save Energy" and "Save More Energy and Protect Data" are clumsy terms but they might be good descriptives for use as tool tips.

already screwed

Submitted by Anonymous George (not verified) on Tue, 2006-05-09 07:55.

the fact we're talking about using different terms than those already used by windows/osx means we're already hurting the user experience, surely?

Definately Not by Anonymous George (not verified)

re: already screwed

Submitted by hughsie on Tue, 2006-05-09 12:26.

The names chosen are very similar to Microsoft Windows, although even MS get muddled when they refer to "sleep" in some documenation.
The OSX SureSleep stuff might have legal restrictions, and so was avoided.

I'm really tired of people saying "I think this is silly, xxxx is CLEARLY a better name" as it's all personal opinions that other people will argue to the death against.

The real point is it doesn't *really* matter what the names are, as long as they are consistent amoungst developers and users.

Richard.

There is no perfect nomenclature

Submitted by Anonymous George (not verified) on Tue, 2006-05-09 06:14.

Sadly the ACPI specification doesn't name the types of sleep states, and while Windows and Mac OS X have their names they don't agree with each other. And yes, lay people are clueless, i. e. many users (even inteligent ones!) can't tell the difference between memory and disk, and that's even worse because disks are sometimes called "secondary memory".

What I'm trying to say is that no matter wich names we pick, it will be necessary to define them for the user, at least with tooltips and documentation, and maybe with text in the preferences dialogs themselves. And, of course, there must be uniformity in the nomenclature inside GNU/Linux and other POSIX OSes!

suspend-to-ram and suspend-to-disk

Submitted by Anonymous George (not verified) on Mon, 2006-05-08 05:40.

I disagree. I don't think suspend-to-ram and suspend-to-disk are bad names for these concepts. They convey the differences very well, since one expects RAM to be volatile but quicker and disk implies more permanence but slower. The only argument? "But the average user doesn't understand these terms." Bollocks. Just about any user knows the difference between RAM and disk, or at least memory and disk. It's a fundamental aspect of using computers, since non-technical users are going to be editing documents and need to understand the concept of saving.

However, with the terms suspend, sleep, hibernate, etc. it is confusing to understand the difference because they are all very similar concepts. Even I, having used computers for years, can't always remember the difference.

To a new user RAM and disk by Anonymous George (not verified)
hibernate by Anonymous George (not verified)
"Pause" is perfect! by Anonymous George (not verified)

Sleep?

Submitted by Anonymous George (not verified) on Tue, 2006-05-09 10:18.

Sleep is probably more widely recognised. You could even have "snooze" - suspend to ram and "sleep" suspend to disk. I suspect those translate worldwide

Don't Apple use sleep for suspend to disk when the battery is low?

The problem is, whatever you call them, your average user is not going to differentiate between the two or if asked to choose, know which to go for. At least sleep can be interpreted as a deeper state than snooze. And most people's computers wake up so why not continue the analogy?

Disagree

Submitted by Anonymous George (not verified) on Mon, 2006-05-08 09:57.

I know a lot of people who are pretty intellegent but can't remember the difference between RAM and disk, it's all "memory" to them. It depends how much into computers you are.

They sure as hell aren't banjo-players they just don't know or want to know how their computer works.

"Just about any user knows

Submitted by Anonymous George (not verified) on Mon, 2006-05-08 08:57.

"Just about any user knows the difference between RAM and disk, or at least memory and disk."
Now that's complete ignorance from you. From my experience, a normal computer user just doesn't understand any difference between memory and disk. They just think that when they save a file, it's in the computer's memory. And why would they even think about it any more than that? Just as long as they know how to select the right file and folder, they're happy.
-WereCatf

Average users

Submitted by kelvin on Mon, 2006-05-08 07:33.

> Just about any user knows the difference between RAM and disk,
> or at least memory and disk.

You obviously don't meet a lot of regular users; and no, knowing this distinction is not a fundamental aspect of using a computer. I meet many non-tech-savvy users who plainly state that their computers have "lots of gigabytes".

Also, even in cases where a

Submitted by Anonymous George (not verified) on Mon, 2006-05-08 09:38.

Also, even in cases where a user does know the difference between RAM and disk, they won't necessarily grok suspend-to-RAM and suspend-to-disk. What do either of those mean? In addition to understanding what RAM and disks are, you need to understand why we have both instead of just one, and you need to understand the process of suspending. Try explaining memory allocation to a non-programmer sometime, then try to explain what suspending-to-RAM is and how it isn't the same thing, but it sort of is, well....

No, these sorts of things seem perfectly intuitive to people who have been using computers for a long time. The inner workings of a computer, however basic, are not common knowledge, though.

Animals which hibernate (I

Submitted by Anonymous George (not verified) on Tue, 2006-05-09 09:39.

Animals which hibernate (I assume this is the idea behind using the term) gradually use up reserves, and I believe this is relatively well known in the public.

I think this is really reading too much into the terms. Suspend == a suspension of activity (think freeze tag), implying that the activity will resume again shortly right where it left off. Hibernate == a long period of inactivity. The poster of the article argues that there is ambiguity with Sleep because you can either sleep for a short time (nap) or sleep for a long time (hibernate).

Personally, I don't think the particular names matter that much. If the user doesn't know what each of them do, he/she will figure it out soon enough. The important thing is to gain consistency. You can't have Gnome and KDE calling them different things without confusing the user.

But why does hibernate

Submitted by hughsie on Tue, 2006-05-09 07:19.

But why does hibernate consume resources in your argument any less than sleep does? Sleep is a generic term, and so unsuitable for use in specific cases.

When choosing the names, I had to be very mindful of current uses in the kernel, userspace, common use and the use by hardware and software vendors.

Until you are getting ~20 mails per day about power and suspend (like I do some days), I suggest you talk more carefully about "the average user" as no such user exists.

What confuses people are lots of "opinions" and different points of view, leading to the naming chaos we have now. There's no way a single spec can please everyone, but this middle ground seems to keep most of the key people happy. See the wiki for people agreeing, as the names are increasing every day.

Thanks for your comments tho, but I would be rather more careful in saying that none of the terms are suitable as they certainly were not chosen quickly or without community involvement.

Richard.

There is a wide difference

Submitted by cevans on Fri, 2006-05-26 15:02.

I certainly meant no offense in my comments, but may have been unclear. There is a wide difference between suitability and intuitiveness. None of the ideas are intuitive, but all of them are suitable to different degrees. The most important thing is that whatever is used needs to be well documented in a visible and easy to understand way. Ubuntu does a good job, in my opinion, with the changing text underneath the buttons on the logout/shutdown menu (though the text itself could be better).

Suspend is just as generic as sleep, especially in regards to time scale, and also can have a negative connotation (at least in the US).

Unfortunately, I just don't see the agreement in the community here. The wiki page could certainly be improved in order to better display it, which might convince some of the people who disagree with the page. Various people in the Ubuntu discussion supported Sleep, and Suspend was decided upon because Hal used it. The Red Hat bug has no discussion at all, nor does it appear to have any agreement.

I don't care by Anonymous George (not verified)