Power management vs. activities (was: Re: [Sugar-devel] stopwatch activity)
On Tue, Apr 27, 2010 at 11:00:15PM -0400, Benjamin M. Schwartz wrote: I'm reluctant to do this, though, because it feels like an ugly hack. The right solution would be for the suspend system to recognize that Stopwatch has a timer set to expire in 100 ms, and postpone suspend. UPower has a DBus API to set latency requirements [1,2]. The UPower backend code uses the kernel PM QoS interface [3] to set the requested CPU (DMA) latency (in µs, up to ~35 minutes) and network throughput (in kbps). cpuidle hooks into the PM QoS framework to provide CPU latency management. We could use the UPower API in activities and provide a helper daemon for powerd to implement this API on the XO until the kernel can do automatic suspends. [1] http://upower.freedesktop.org/docs/QoS.html [2] http://blogs.gnome.org/hughsie/2008/11/06/devicekit-power-latency-control/ [3] Documentation/power/pm_qos_interface.txt in kernel sources [4] http://www.celinux.org/elc08_presentations/elc2008_pm_qos_slides.pdf CU Sascha -- http://sascha.silbe.org/ http://www.infra-silbe.de/ signature.asc Description: Digital signature ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
stopwatch activity
I noticed something interesting with he stopwatch activity on the XO 1.5 C2 with build 120. When the XO goes into suspend, the clock stops display, but upon resume, will show actual time elapsed (clock keep counting). Mark also works correctly, displaying the time when the Mark button is clicked, irrespective of the display. I'm not sure what the behavior should be, though. Should the activity prevent suspend? cheers, Sameer -- Dr. Sameer Verma, Ph.D. Associate Professor, Information Systems Director, Campus Business Solutions San Francisco State University http://verma.sfsu.edu/ http://opensource.sfsu.edu/ http://cbs.sfsu.edu/ http://is.sfsu.edu/ ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: [Sugar-devel] stopwatch activity
On Tue, 27 Apr 2010, Sameer Verma wrote: I noticed something interesting with he stopwatch activity on the XO 1.5 C2 with build 120. When the XO goes into suspend, the clock stops display, but upon resume, will show actual time elapsed (clock keep counting). Mark also works correctly, displaying the time when the Mark button is clicked, irrespective of the display. I'm not sure what the behavior should be, though. I think that's fine behavior. Most stopwatches don't stop running by themselves, so I don't see why ours should. Should the activity prevent suspend? My philosophy is that suspend should be _absolutely transparent_ to the user; i.e. its effects should not be detectable, in the same way that processor voltage scaling is undetectable. This suggests that Stopwatch should inhibit suspend while it is visible onscreen. I'm reluctant to do this, though, because it feels like an ugly hack. The right solution would be for the suspend system to recognize that Stopwatch has a timer set to expire in 100 ms, and postpone suspend. --Ben ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: [Sugar-devel] stopwatch activity
On Tue, Apr 27, 2010 at 8:00 PM, Benjamin M. Schwartz bmsch...@fas.harvard.edu wrote: On Tue, 27 Apr 2010, Sameer Verma wrote: I noticed something interesting with he stopwatch activity on the XO 1.5 C2 with build 120. When the XO goes into suspend, the clock stops display, but upon resume, will show actual time elapsed (clock keep counting). Mark also works correctly, displaying the time when the Mark button is clicked, irrespective of the display. I'm not sure what the behavior should be, though. I think that's fine behavior. Most stopwatches don't stop running by themselves, so I don't see why ours should. Indeed, so when the numbers stop at a certain point, it looks strange. Should the activity prevent suspend? My philosophy is that suspend should be _absolutely transparent_ to the user; i.e. its effects should not be detectable, in the same way that processor voltage scaling is undetectable. This suggests that Stopwatch should inhibit suspend while it is visible onscreen. I'm reluctant to do this, though, because it feels like an ugly hack. The right solution would be for the suspend system to recognize that Stopwatch has a timer set to expire in 100 ms, and postpone suspend. --Ben Sameer -- Dr. Sameer Verma, Ph.D. Associate Professor, Information Systems Director, Campus Business Solutions San Francisco State University http://verma.sfsu.edu/ http://opensource.sfsu.edu/ http://cbs.sfsu.edu/ http://is.sfsu.edu/ ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: StopWatch activity
On Nov 14, 2007, at 2:45 AM, Benjamin M. Schwartz wrote: 1. Project name : StopWatch 3. One-line description : The most ludicrously awesome stopwatch ever Done. Your tree is here: git+ssh://[EMAIL PROTECTED]/git/activities/stopwatch Your usernames are lukego and surendra. Please follow instructions here for importing your project: http://wiki.laptop.org/go/Importing_your_project Let us know if you have any problems with your tree. Happy hacking. Cheers, -- Ivan Krstić [EMAIL PROTECTED] | http://radian.org ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Telling time (was: StopWatch activity)
Nick, At Thu, 15 Nov 2007 17:13:34 -0500, nick knouf wrote: Bert Freudenberg writes: I question the very assumption that continuously telling the time is even remotely important on a learning machine for kids in elementary school age. Dealing with time is a critical life skill that must be learned. Having a clock is thus very important. Whose time? Hours minutes seconds? Days since a recent feast? When the sun is at a certain position in the sky? Since I last saw you on the road? How much do I quantize? Is quantization of time even a concept I am familiar with? Well, it seems that you are responding to a wrong message. The notion of time is _highly_ contingent on situated cultural factors. Just because in the West we measure things using hours, minutes, and seconds, does not mean that the entire world does so. In fact, our conception of time is directly related to churches and clock towers in the middle ages (see Lewis Mumford on this idea) first, and then assembly lines and educational/disciplinary institutions (see Foucault) . The rest of the world has not necessarily adopted our way of dividing days into ever smaller chunks---perhaps there is no quantization at all! A clock application, especially given the areas of deployment, is _not_ something you rush into with the assumption that you can merely write a graphic display of 00:00:00. One must understand the local conditions to know how time is told _on the ground_ and be careful to not impose a Western notion of quantization and temporal division that might be entirely foreign. So, what do you think about the idea of letting kids make their own clocks? -- Yoshiki ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Telling time (was: StopWatch activity)
What I was suggesting though is that there should *not* be a clock in the Sugar frame visible all the time. +1 to including hooks to Sugar for frame-resident mini-apps. +1 to making the frame clock optional (turned on from the clock activity - another reason to keep it an activity) and not the default option. If people want it, they will find it - vice versa is not as true. +1 to a respect that there are vastly differing cultural views of time -1 to the idea that any other culture's view of time is so inherently fragile that it can be shattered by a simple digital clock. The percentage of people who have NOT seen a clock is ever-shrinking - I suspect it's a safe guess that many people reading this message may have grown up when that percentage was several times what it is now. I'll hazard another bet: the xo will NOT have any measurable impact on that trend. And another: many cultural views of time are in fact just as compatible with the clock as yours, even if you (naturally) think that yours is the logical result of the clock. -1 to the idea that we should deliberately leave out features in order to encourage kids to program. O, ye of little faith. My opinions, Jameson ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Telling time (was: StopWatch activity)
On Nov 16, 2007, at 1:57 PM, Yoshiki Ohshima wrote: Well, it seems that you are responding to a wrong message. Not really; if the question is whether or not there is a clock application that is standard on the laptop, implicit there is a decision as to _what kind_ of clock application. It's that question that I wanted to highlight. So, what do you think about the idea of letting kids make their own clocks? I should have made it more explicit in my e-mail that I would certainly be in favor of a variety of different clock applications that reflect local conditions or are based on the logic or whimsy of the user. The only thing I would caution is that the clock construction environment should not privilege one type of representation versus another. Not that I am suggesting that you or anyone else is necessarily doing that; again, I raise the point for the purposes of making the question salient. nick ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Telling time (was: StopWatch activity)
On Nov 16, 2007, at 21:13 , nick knouf wrote: On Nov 16, 2007, at 1:57 PM, Yoshiki Ohshima wrote: Well, it seems that you are responding to a wrong message. Not really; if the question is whether or not there is a clock application that is standard on the laptop, implicit there is a decision as to _what kind_ of clock application. It's that question that I wanted to highlight. Yes, you are answering in the wrong thread. Having an activity for exploring time is obviously valuable. What I was suggesting though is that there should *not* be a clock in the Sugar frame visible all the time. Granted, all office-centric operating systems / GUIs have one. So what? We shouldn't. If a kid wants one, make Sugar easy to modify to allow this and other modifications we might not even think of, yet. - Bert - ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Telling time (was: StopWatch activity)
-1 to the idea that we should deliberately leave out features in order to encourage kids to program. O, ye of little faith. I don't see anybody said this, but yes, that would be bad. The environment should come rich set of tools/widgets etc. that make the environment rich. Several clock examples should be part of it (That is why I just made one). But, these tools should be used by kids to make more stuff, and also these should be openable to see inside. That is what I mean by saying kids should be making clocks. -- Yoshiki ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Telling time (was: StopWatch activity)
This is a Color of the Bikeshed issue. Give it a rest. ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Telling time (was: StopWatch activity)
On Nov 17, 2007, at 0:21 , Mitch Bradley wrote: This is a Color of the Bikeshed issue. Give it a rest. The clock discussion is, you're right. Reminding everyone that we set out to create an environment for kids to explore and construct is not. It's perplexing how few developers seem to share that goal. - Bert - ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Telling time (was: StopWatch activity)
Bert Freudenberg writes: I question the very assumption that continuously telling the time is even remotely important on a learning machine for kids in elementary school age. Dealing with time is a critical life skill that must be learned. Having a clock is thus very important. Whose time? Hours minutes seconds? Days since a recent feast? When the sun is at a certain position in the sky? Since I last saw you on the road? How much do I quantize? Is quantization of time even a concept I am familiar with? The notion of time is _highly_ contingent on situated cultural factors. Just because in the West we measure things using hours, minutes, and seconds, does not mean that the entire world does so. In fact, our conception of time is directly related to churches and clock towers in the middle ages (see Lewis Mumford on this idea) first, and then assembly lines and educational/disciplinary institutions (see Foucault) . The rest of the world has not necessarily adopted our way of dividing days into ever smaller chunks---perhaps there is no quantization at all! A clock application, especially given the areas of deployment, is _not_ something you rush into with the assumption that you can merely write a graphic display of 00:00:00. One must understand the local conditions to know how time is told _on the ground_ and be careful to not impose a Western notion of quantization and temporal division that might be entirely foreign. nick knouf ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: StopWatch activity
On Wed, Nov 14, 2007 at 02:45:31AM -0500, Benjamin M. Schwartz wrote: 2.5. Download: http://dev.laptop.org/~bemasc/StopWatchActivity-1.xo Tested on build 625 on a B4, works okay, problems you probably already know about: 1. the Start/Stop text legend disappears when the cursor is over it and the stopwatch is running, and the keyboard up and down arrows are used, perhaps just white text on white background, 2. the icons for the Start/Stop and Zero buttons are not there, perhaps these are in the later builds, 3. with the activity shared, but one XO active, stopping a stopwatch may result in a reduction of the displayed value, more likely to occur if all stopwatches are running, 30% CPU used on B4 with ten stopwatches running. Switching to text console causes this to stop, as you state. My opinion is that the number of significant digits should be selectable, and a younger kids version could have a single stopwatch with a bar graph, stopwatch clock face, and digital display. -- James Cameronmailto:[EMAIL PROTECTED] http://quozl.netrek.org/ ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: StopWatch activity
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 James Cameron wrote: On Wed, Nov 14, 2007 at 02:45:31AM -0500, Benjamin M. Schwartz wrote: 2.5. Download: http://dev.laptop.org/~bemasc/StopWatchActivity-1.xo Tested on build 625 on a B4, works okay, problems you probably already know about: 1. the Start/Stop text legend disappears when the cursor is over it and the stopwatch is running, and the keyboard up and down arrows are used, perhaps just white text on white background, Yep, white on white. Known bug. 2. the icons for the Start/Stop and Zero buttons are not there, perhaps these are in the later builds, I made these icons myself. I just tested a clean install, using that .xo, and they appear fine on joyride-269. This is probably due to changes in Sugar's path behavior. 3. with the activity shared, but one XO active, stopping a stopwatch may result in a reduction of the displayed value, more likely to occur if all stopwatches are running, This is actually a feature. It takes some time to process your mouse click, and under heavier CPU load, that time may be long enough that the time label continues to redraw before it can be stopped. However, the first thing the code does upon receiving your mouse click is to record the time-of-click, which is what is used to determine the displayed value. Thus, the negative change you can see (if you have very fast eyes) is just a correction for the computer's own reaction time. To understand the motivation for this feature, imagine if the processing delay included several mesh hops, or even a satellite link. In this case, it might take several seconds for the fact that person A has pressed stop to propagate to person B. When that message arrives, person B's clock must be stopped and set back by the propagation delay, in order to make the two clocks agree. StopWatch does this by passing around absolute reference times with every event. My opinion is that the number of significant digits should be selectable, and a younger kids version could have a single stopwatch with a bar graph, stopwatch clock face, and digital display. Every digital stopwatch I have ever seen has precision to hundredths of a second, no more and no less. In this case, more would be infeasible, due to the coarseness of software timing. Less would be pessimistic. To be clear, I am really trying to duplicate the interface and functionality of a standard digital stopwatch, except where I think it can be improved. I don't know what you mean by a bar graph. Younger children can't read analog clocks (I recall being taught how to read them in second grade). Also, drawing clock faces is computationally expensive. - --Ben -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.7 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFHOypMUJT6e6HFtqQRAuxhAKCTcfqnyvfkzXjpmZkF8HSysXf7OACgoK2M ZfQovLMmlalL8r4s8ohRVkE= =QDY8 -END PGP SIGNATURE- ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: StopWatch activity
Is there a reason you haven't made the clock and the stopwatch different functions for a single activity? I second that. I think these could be integrated - Eben Regards, Mako -- Benjamin Mako Hill [EMAIL PROTECTED] http://mako.cc/ -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFHOyxzic1LIWB1WeYRAo9HAKCGVJ81f0coABHNViSJIVU+XENsbACgvyyE wMFJ+UpOpCEtB3Lqcr3oMRk= =wQCB -END PGP SIGNATURE- ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: StopWatch activity
I'm so glad you got around to doing this! Such tool are badly needed on the laptop. Is there a reason you haven't made the clock and the stopwatch different functions for a single activity? Regards, Mako -- Benjamin Mako Hill [EMAIL PROTECTED] http://mako.cc/ signature.asc Description: Digital signature ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: StopWatch activity
Eben Eliason wrote: Is there a reason you haven't made the clock and the stopwatch different functions for a single activity? I second that. I think these could be integrated While you're at it, how about integrating the camera activity with it, so it could be like Dick Tracy's 2-way wrist TV. :-) ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: StopWatch activity
While you're at it, how about integrating the camera activity with it, so it could be like Dick Tracy's 2-way wrist TV. :-) The original message included Obsessive accuracy, so maybe this option would be appropriate: First Atomic Clock Wristwatch http://www.leapsecond.com/pages/atomic-bill/ Back to somewhat serious so this isn't pure clutter... On the suggestion of combining StopWatch and Clock. How about a button on the stopwatch to use its display to show the current time. Or maybe two buttons, one for local and one for UTC so kids learn about time zones. -- These are my opinions, not necessarily my employer's. I hate spam. ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: StopWatch activity
I thought about this a bit more, and think that there may be a valid split between what might be called Clock and Time (currently StopWatch) activities. Clock's primary purpose would be to display a large clock. It would likely have digital and analog modes, but could probably choose one or the other only to maximize the display size. It would be used much like a standard bedside clock, and would ideally have basic alarm functionality built-in once that's possible. It is for simply displaying the current time, and could also be shared and integrate the world-clock idea for an educational purpose. Time, on the other hand, is about the act of actively timing things. This would include the stopwatch functionality it already has, but could also have an egg-timer countdown mode. Both of these are short term, in the moment activities which are more appropriate for experiments and such. It is a much more active activity than Clock. What do people think of this distinction? - Eben On Nov 14, 2007 1:38 PM, Hal Murray [EMAIL PROTECTED] wrote: While you're at it, how about integrating the camera activity with it, so it could be like Dick Tracy's 2-way wrist TV. :-) The original message included Obsessive accuracy, so maybe this option would be appropriate: First Atomic Clock Wristwatch http://www.leapsecond.com/pages/atomic-bill/ Back to somewhat serious so this isn't pure clutter... On the suggestion of combining StopWatch and Clock. How about a button on the stopwatch to use its display to show the current time. Or maybe two buttons, one for local and one for UTC so kids learn about time zones. -- These are my opinions, not necessarily my employer's. I hate spam. ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: StopWatch activity
Benjamin, 1. Clock is non-interactive. It doesn't make sense to share it, or save it to the journal, so I've disabled those features. Human being is good at finding differences, but drawing similarity out of seemingly different things is more fun if you know it. 2. I like small programs that do one thing well. I'm not sure if this is the Constructionist Activity philosophy, exactly, but it seemed like a good idea. If a program has two different main screens, that suggests that it does two completely separate things. Probably the best thing is to provide one of them (or other basic blocks) and to let the children build one from another by themselves. Combining it with the camera makes a lot of sense, BTW. Kids can make an interval timer for camera... (and other things like Pippy.) -- Yoshiki ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: StopWatch activity
What do people think of this distinction? To my prejudice, it sounds like a bad idea. If you have to do some operations on the laptop and wait many seconds just to check the current time, that sounds bad, too. There was an idea of having a little clock in the Sugar frame. How about that? (I think Alan gave a demo of that, and it can be made in 10 seconds, during his talk at Cambridge sometime ago.) -- Yoshiki ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: StopWatch activity
Eben, If you have to do some operations on the laptop and wait many seconds just to check the current time, that sounds bad, too. The clock activity is wholly independent in my perspective from having a clock in Sugar. We still intend to incorporate that - the overhead of launching an activity is silly. Wow, ok. Kids will have plenty of different clocks. That sounds like a rich environment. This is, in fact, why I think we need to clarify the use cases for these activities, and having a computer that is actually impersonating a clock is a reasonable thing to want in some cases, but not what you want while you're actively using the laptop. Exactly. That is one reason why kids should make one. And *ideally* it shouldn't be that hard for say, a 12 years old, like I wrote here: (I think Alan gave a demo of that, and it can be made in 10 seconds, during his talk at Cambridge sometime ago.) -- Yoshiki ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: StopWatch activity
To my prejudice, it sounds like a bad idea. If you have to do some operations on the laptop and wait many seconds just to check the current time, that sounds bad, too. The clock activity is wholly independent in my perspective from having a clock in Sugar. We still intend to incorporate that - the overhead of launching an activity is silly. This is, in fact, why I think we need to clarify the use cases for these activities, and having a computer that is actually impersonating a clock is a reasonable thing to want in some cases, but not what you want while you're actively using the laptop. - Eben There was an idea of having a little clock in the Sugar frame. How about that? (I think Alan gave a demo of that, and it can be made in 10 seconds, during his talk at Cambridge sometime ago.) -- Yoshiki ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: StopWatch activity
quote who=Eben Eliason date=Wed, Nov 14, 2007 at 01:48:44PM -0500 I thought about this a bit more, and think that there may be a valid split between what might be called Clock and Time (currently StopWatch) activities. I agree with your analysis. There are several important ways in which a stopwatch and a clock are different. That said, I'm already finding the number of applications installed by default in recent builds (nearly three full screen-width) to be overwhelming. Until we have a better of way to navigate and find activities in such large collections, I think that reasonable combinations of overlapping applications is a good idea for this and for the normal reasons that we avoid duplicating code. Every time I've used a timer on a phone, computer or watch, it has been a dual clock/stopwatch. They both count and display time and the interface is similar. It may be that all of those systems are combining two things that should be separate. On the other hand, the inclusion of these separate thing in one activity will at least not be surprising to anyone who has used a digital watch. I still think they should be merged. Regards, Mako -- Benjamin Mako Hill [EMAIL PROTECTED] http://mako.cc/ Creativity can be a social contribution, but only in so far as society is free to use the results. --GNU Manifesto ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: StopWatch activity
On Wed, Nov 14, 2007 at 02:36:08PM -0500, Eben Eliason wrote: We still intend to incorporate that - the overhead of launching an activity is silly. More precision would make this particular comment more helpful. How low an overhead (in seconds and MB of RAM IO) are we aiming for? What are we willing to spend to get there? Michael ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: StopWatch activity
On Nov 14, 2007 4:07 PM, Michael Stone [EMAIL PROTECTED] wrote: On Wed, Nov 14, 2007 at 02:36:08PM -0500, Eben Eliason wrote: We still intend to incorporate that - the overhead of launching an activity is silly. More precision would make this particular comment more helpful. How low an overhead (in seconds and MB of RAM IO) are we aiming for? What are we willing to spend to get there? I'm talking, really, about interaction overhead. In order to see the current time I should press a key, or make a gesture with the mouse, or something similar. I shouldn't have to find the clock activity wherever that might be, click to launch it, wait for it do launch (however short that may be), and then close it again just to check the time. I could leave it open all the time for later checking, of course, but I'd still have to perform this exercise every time I rebooted. This kind of things should really be a system device as well. - Eben ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Telling time (was: StopWatch activity)
On Nov 14, 2007, at 22:37 , Eben Eliason wrote: I'm talking, really, about interaction overhead. In order to see the current time I should press a key, or make a gesture with the mouse, or something similar. I shouldn't have to find the clock activity wherever that might be, click to launch it, wait for it do launch (however short that may be), and then close it again just to check the time. I could leave it open all the time for later checking, of course, but I'd still have to perform this exercise every time I rebooted. This kind of things should really be a system device as well. I question the very assumption that continuously telling the time is even remotely important on a learning machine for kids in elementary school age. - Bert - ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: StopWatch activity
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hal Murray wrote: Obsessive accuracy. What's your version of Obsessive? Seconds? Milliseconds? Microseconds? I have no desire to do better than 0.01s. Human reaction times are an order of magnitude slower than that anyway. What I meant is, I have done everything I could think of to maximize accuracy, and this is obvious in the way the code is structured. For example, the first instruction in each user-interface callback records the event time, before any processing is done, to minimize computation delay. Are you assuming that the clocks on various XOs are synchronized? If so, how well? No. Upon joining, a new member asks everyone else what time they think it is. The algorithm assumes that the network delay is the same in each direction. Whoever responds first wins, because this computer experienced the least network+scheduling delay, and so the assumption is most likely to be true. Experimentally, this works very well with two nodes on a mesh; that's about all I can test at the moment. A more sophisticated synchronization algorithm would be appreciated, but I did not know how to make NTP work: 1. From python 2. As a highly restricted non-root user 3. Over Tubes 4. In a way that is resilient to the sudden disappearance of any member of the group. TamTam developers: I would like to know how you do synchronization. I looked through your git repository, but I couldn't find any C source for it. [Long discussion to follow in a separate messsage.] ? -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.7 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFHO3grUJT6e6HFtqQRAvTVAJ9QewEBavAaUz+LSGygTjkljJsb3QCfS8Gk ykQYi9Jefr/CZDT9ESuxEm4= =+KNe -END PGP SIGNATURE- ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: StopWatch activity
Eben Eliason wrote: On Nov 14, 2007 4:07 PM, Michael Stone [EMAIL PROTECTED] wrote: On Wed, Nov 14, 2007 at 02:36:08PM -0500, Eben Eliason wrote: We still intend to incorporate that - the overhead of launching an activity is silly. More precision would make this particular comment more helpful. How low an overhead (in seconds and MB of RAM IO) are we aiming for? What are we willing to spend to get there? I'm talking, really, about interaction overhead. In order to see the current time I should press a key, or make a gesture with the mouse, or something similar. I shouldn't have to find the clock activity wherever that might be, click to launch it, wait for it do launch (however short that may be), and then close it again just to check the time. I could leave it open all the time for later checking, of course, but I'd still have to perform this exercise every time I rebooted. This kind of things should really be a system device as I think the time should be in the sugar frame at a fixed location, so anytime the frame is visible, the time is visible. And perhaps if you hover over it, you can see the date. The Windows System Tray area at the right of the task bar has a simple clock that works like that, and I think it is a good design. Unobtrusive, easy to see when you need it, easy to get a bit more information. Nothing fancy, just text that tells you the time of day. well. - Eben ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: StopWatch activity
On Wed, Nov 14, 2007 at 12:03:08PM -0500, Benjamin M. Schwartz wrote: It takes some time to process your mouse click, and under heavier CPU load, that time may be long enough that the time label continues to redraw before it can be stopped. Good. I suspected as such, based on your original announcement, so when I saw it I didn't dimiss it but tried some more. Hopefully the kids will notice as well, and the teacher can have your notes on how to explain the effect. ;-) Please put your post in the source. Every digital stopwatch I have ever seen has precision to hundredths of a second, no more and no less. Yes, I suspect they mostly use the same internal design. You are lucky, you don't have to. I don't know what you mean by a bar graph. Younger children can't read analog clocks (I recall being taught how to read them in second grade). Also, drawing clock faces is computationally expensive. A bar graph ... okay, perhaps fuel gauge might be a better term, ... creates a linear single axis representation of time, which the human eye is *very* good at estimating and predicting against. It is how a kid catches a ball. A stopwatch clockface, represents the time by rotation of a marker around the circumference of a circle. It does not have the same problems associated with 12-hour analog clocks, since there is normally only one hand, and the full circle doesn't represent half a day. Analog day clocks can be taught to children, but not as easily as numbers and digital clocks, I agree. On the other hand, the lack of momentum perception on a digital clock tends to teach children the art of estimating time without requiring an external beat. Analog clocks remain the cheapest clock per square metre of visual coverage, so our target market will be exposed to them. You cannot dismiss them based on your first-world understanding of digital clocks. -- James Cameronmailto:[EMAIL PROTECTED] http://quozl.netrek.org/ ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: [laptop.org #1581] StopWatch activity
So it doesn't look like there is consensus on this yet - Mako - since you seem to be following this (and I'm at a conference), could you ping me when you think consensus has been reached? Thanks, -- Daniel Clark # Sys Admin, One Laptop per Child # http://laptop.org # http://opensysadmin.com # http://planyp.us/djbclark # http://dclark.us On Nov 14, 2007 4:35 PM, Benjamin M. Schwartz [EMAIL PROTECTED] wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hal Murray wrote: Obsessive accuracy. What's your version of Obsessive? Seconds? Milliseconds? Microseconds? I have no desire to do better than 0.01s. Human reaction times are an order of magnitude slower than that anyway. What I meant is, I have done everything I could think of to maximize accuracy, and this is obvious in the way the code is structured. For example, the first instruction in each user-interface callback records the event time, before any processing is done, to minimize computation delay. Are you assuming that the clocks on various XOs are synchronized? If so, how well? No. Upon joining, a new member asks everyone else what time they think it is. The algorithm assumes that the network delay is the same in each direction. Whoever responds first wins, because this computer experienced the least network+scheduling delay, and so the assumption is most likely to be true. Experimentally, this works very well with two nodes on a mesh; that's about all I can test at the moment. A more sophisticated synchronization algorithm would be appreciated, but I did not know how to make NTP work: 1. From python 2. As a highly restricted non-root user 3. Over Tubes 4. In a way that is resilient to the sudden disappearance of any member of the group. TamTam developers: I would like to know how you do synchronization. I looked through your git repository, but I couldn't find any C source for it. [Long discussion to follow in a separate messsage.] ? -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.7 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFHO3grUJT6e6HFtqQRAvTVAJ9QewEBavAaUz+LSGygTjkljJsb3QCfS8Gk ykQYi9Jefr/CZDT9ESuxEm4= =+KNe -END PGP SIGNATURE- ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: StopWatch activity
Note that X button and keyboard events have timestamps, in milliseconds. (This wraps in some hundreds of days, but I doubt anyone will use the stopwatch that long; you do have to worry in principle about doing modulus arithmetic, though IIRC, X servers generally have been using time since the server reset last for this value. Theory goes that since we are using the kernel's evdev driver, that these timestamps may even get done in the kernel at interrupt level, if the implementation is as has been intended (I haven't looked to make sure), as we did in the original X drivers in the mid '80s This would make the timing pretty independent of user space considerations. - Jim On Thu, 2007-11-15 at 09:22 +1100, James Cameron wrote: On Wed, Nov 14, 2007 at 12:03:08PM -0500, Benjamin M. Schwartz wrote: It takes some time to process your mouse click, and under heavier CPU load, that time may be long enough that the time label continues to redraw before it can be stopped. Good. I suspected as such, based on your original announcement, so when I saw it I didn't dimiss it but tried some more. Hopefully the kids will notice as well, and the teacher can have your notes on how to explain the effect. ;-) Please put your post in the source. Every digital stopwatch I have ever seen has precision to hundredths of a second, no more and no less. Yes, I suspect they mostly use the same internal design. You are lucky, you don't have to. I don't know what you mean by a bar graph. Younger children can't read analog clocks (I recall being taught how to read them in second grade). Also, drawing clock faces is computationally expensive. A bar graph ... okay, perhaps fuel gauge might be a better term, ... creates a linear single axis representation of time, which the human eye is *very* good at estimating and predicting against. It is how a kid catches a ball. A stopwatch clockface, represents the time by rotation of a marker around the circumference of a circle. It does not have the same problems associated with 12-hour analog clocks, since there is normally only one hand, and the full circle doesn't represent half a day. Analog day clocks can be taught to children, but not as easily as numbers and digital clocks, I agree. On the other hand, the lack of momentum perception on a digital clock tends to teach children the art of estimating time without requiring an external beat. Analog clocks remain the cheapest clock per square metre of visual coverage, so our target market will be exposed to them. You cannot dismiss them based on your first-world understanding of digital clocks. -- Jim Gettys One Laptop Per Child ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Telling time (was: StopWatch activity)
Bert Freudenberg writes: I question the very assumption that continuously telling the time is even remotely important on a learning machine for kids in elementary school age. Dealing with time is a critical life skill that must be learned. Having a clock is thus very important. Keeping the activity separate is good. Once the frame clock arrives, the activity clock can either be deleted or mutate into a calendar. The clock activity could also be used to set the time zone. If anything, the stopwatch goes with the ruler and the acoustic ruler. ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
StopWatch activity
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 1. Project name : StopWatch 2. Existing website, if any : http://dev.laptop.org/~bemasc/screenshot.png 2.5. Download: http://dev.laptop.org/~bemasc/StopWatchActivity-1.xo 3. One-line description : The most ludicrously awesome stopwatch ever conceived. 4. Longer description : StopWatch is a multi-user Sugar stopwatch activity. Features include: 10 Stopwatches per instance Named stopwatches Easy to use in ebook mode (with icons indicating the functions of the game keys) Obsessive accuracy. Draws 0% CPU when not visible on-screen. Supports saving and loading from the journal - - If you save a StopWatch instance in which some watches are still running, and then resume it from the journal half an hour later, it will be as if the watches were running the whole time. Crazy sharing: - - Unsynchronized system clocks are handled automatically. - - Sharing is based on a state-machine model, designed so that the effective reaction time of the group is the fastest reaction time of its members, regardless of network delays. - - Completely decentralized, so that any member (even the initiator) can leave the group without messing anything up - - Coherent, so that members' displays never disagree. - - All members can edit names, which are updated in real time. Localization: numbers will be formatted according to the user's locale, even if different users are in different locales. Names may be in any character set 5. URLs of similar projects : http://wiki.laptop.org/go/Stopwatch 6. Committer list just me. 7. Preferred development model [X] Central tree. 8. Set up a project mailing list: [X] No 9. Commit notifications [X] No commit notifications, please 11. Notes/comments: I should thank mncharity for inspiring this activity. I wouldn't have done it otherwise. Philosophically, this activity is complementary to Acoustic Tape Measure (and Ruler). The ability to measure space and time is the starting point of all quantitative learning. Concretely, with a string and a rock, these activities allow you to reproduce Galileo's pendulum experiment, the starting point of physics. -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.7 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFHOqebUJT6e6HFtqQRAkUCAJ0ZxxNiLSqYv6zI1TZuLKmiq/31BgCdGfpQ yOeyWzhjCDAbylMRGDQhw5E= =g6ez -END PGP SIGNATURE- ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
