Computing desk
< January 31	<< Jan \| February \| Mar >>	February 2 >

Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

February 1

unwanted clicks

Hello. I have a new laptop (kubuntu) with a touchpad. I like having a tap on the touchpad being equivalent to a primary mouse click. But, when typing, I very often accidentally touch the touchpad, with the result that the typing cursor moves to where the mouse cursor is. This is very disrupting, and happens orders of magnitude more often with this new laptop than other PCs I've used. Does anyone have any advice? Is there a software or configuration solution I might try? Robinh (talk) 00:24, 1 February 2012 (UTC)[reply]

Most touchpads can be configured to reduce sensitivity so that the problem happens less frequently. This is a feature provided by the manufacturer of the touchpad and laptop, not by the operating system, so you need to either search the laptop manual or tell us the manufacturer if you cannot find the touchpad configuration routine. (In Windows it is often on the right of the task bar, or available via "Control Panel". What's the Kubuntu equivalent?) Dbfirs 00:33, 1 February 2012 (UTC)[reply]

thanks for this. My machine is an HP Elitebook 8560p. I don'tsee a suitable option in the Touchpad system settings box. Cheers, Robinh (talk) 00:45, 1 February 2012 (UTC)[reply]

For an extremely low-tech solution, you can place a bottle cap or something similar over it, so the force of pushing on it would be deflected to the sides. Use tape to hold it in place. (Looking at a pic, it looks like the buttons are long and rectangular, so you'd need something shaped like that.) StuRat (talk) 01:14, 1 February 2012 (UTC)[reply]

Hi. The buttons are indeed long and rectangular. But it's accidental touches of the touchpad that are bothering me, not the nipple mouse. I have been expecting some sort of retraining of my hands to occur over the weeks, but that hasn't happened. best wishes, Robinh (talk) 01:24, 1 February 2012 (UTC)[reply]

I think there are disposable food storage container tops that would cover that, like the ones I get at the deli. StuRat (talk) 01:30, 1 February 2012 (UTC)[reply]

OK, but when I want to move the cursor (ie move the mouse), I'll have to shift a sticky-taped boxtop from the touchpad. Robinh (talk) 01:42, 1 February 2012 (UTC)[reply]

Just tape it at one edge, so you can flip it off and on like a hinge. StuRat (talk) 02:36, 1 February 2012 (UTC)[reply]

If I were you, I would turn off tap-to-click and try that for a week, and see how often you have been disrupted by this feature so you can weigh the disruption against the benefits of having tap-to-click on. Personally I now turn tap-to-click off for the very reason you cite. Comet Tuttle (talk) 01:57, 1 February 2012 (UTC)[reply]

That's probably good advice (aargh it happened again). Pity, tho', I find tap-to-click really (aargh again) useful on my other machines. Wonder what it is about the 8560p that makes (aarg) it so (aargh) difficult (aarg aarg) to use. Best (aargh) wishes, Robinh (talk) 02:04, 1 February 2012 (UTC)[reply]

(replying to self). Tried the above, find it intolerable. What I want is to require an actual button click to place the text cursor at the mouse cursor, but to leave the other settings (aarg) untouched. Anyone? Robinh (talk) 02:13, 1 February 2012 (UTC)[reply]

I had the same problem on my Acer Aspire 4315, which came preloaded with Ubuntu. I turned off tap-to-click and never missed it. Usually, I prefer to use an external mouse on a laptop. I miss my old Thinkpad which had a nipple mouse, which is far superior to a trackpad.-gadfium 03:07, 1 February 2012 (UTC)[reply]

I use Ubuntu 11.10, so I don't know how much this applies to Kubuntu, but I can use System Settings > Mouse & Touchpad, and then select "Disable Touchpad when typing", which would seem to be what you want. I assume there's some fancy-schmancy way to do it via the terminal as well, but I'm not really into that. Maybe this will jog someone's memory who knows more about Kubuntu and can help further. - Cucumber Mike (talk) 07:51, 1 February 2012 (UTC)[reply]

That sounds exactly what I am looking for. But I see no such option. Frustrating! Robinh (talk) 08:17, 1 February 2012 (UTC)[reply]

Relevant information is here. Nevard (talk) 08:25, 1 February 2012 (UTC)[reply]

Or maybe not. Apparently the Ubuntu developers decided to make X configuration depend on a bunch of flaky individual files. Have a fiddle with "/usr/share/X11/xorg.conf.d/50-synaptics.conf". There's also a even more obtuse way of doing things referenced here. Nevard (talk) 08:29, 1 February 2012 (UTC)[reply]

The touchpad is by Synaptics, and they provide a configuration setting to reduce sensitivity, so if you want retain tap to click (and I find it very useful), you will find it in Properties --> Device settings --> Settings --> Sensitivity --> Palm Check & Touch Sensitivity (if it has the same driver as mine). I would try this first. (The driver for Windows is called sp52860.exe and is here, but possibly Kubuntu doesn't support it?) Dbfirs 08:33, 1 February 2012 (UTC)[reply]

@dbfirs: thanks for this, but I don't see a chain starting "Properties, Device settings" anywhere. What am I missing?

Yes, sorry, I thought that Kubuntu might display the Synaptics driver icon like Windows does, but evidently not. There ought to be a config command to reduce sensitivity, but I don't know it. Sorry. Dbfirs 16:30, 1 February 2012 (UTC)[reply]

@Nevard: thanks, it's all a bit intimidating (I worry about breaking something). Disable tap-to-click sounds good, I've been looking at 50-synaptics.conf, does this have the same syntax as xorg.conf (so the tips will work)? Robinh (talk) 08:55, 1 February 2012 (UTC)[reply]

and it looks as though ubuntu and kubuntu differ in this respect too (I didn't realize they were separate things until today) Robinh (talk) 08:57, 1 February 2012 (UTC)[reply]

Yup, the separate xorg.conf.d files are just a really 'clever' way of making configuration 'easier' by spreading stuff out all over the place. Ubuntu and Kubuntu differ only in the desktop environment you get- which happens to include configuration utilities. I gave up on KDE a while back, but it's still better than a vanilla version of the latest Gnome (and way better than the horrid Ubuntu version of Gnome.) I suggest you install 'gpointing-device-settings'- it should do everything you want, and doesn't have any major dependancy issues. Whether you're using some KDE program, or a Gnome configuration utility, or g-d-s, it's just configuring the same basic X-server settings behind the scences. Nevard (talk) 09:05, 1 February 2012 (UTC)[reply]

heh, gpointing-device-settings worked fine. I have increased the "tapping time" from its default very short, to a longer, time. Actually (sorry for the mission creep), I now realize what I want is precisely this: "leave tap-to-click enabled except for when this would change the position of a typing cursor". But basically I am pretty awesomed out by gpointing-device-settings. Kia Ora! Robinh (talk) 09:18, 1 February 2012 (UTC)[reply]

Good as. If the settings don't stick for you, all that's being altered are 'XInput' settings. The command-line utility for these, 'xinput', is relatively easy to fiddle around with, and the folks at AskUbuntu are likely to be able to help you with any issues. Tot siens. Nevard (talk) 09:39, 1 February 2012 (UTC)[reply]

Understood (and I even learned some Afrikaans for free!). E noho ra, Robinh (talk) 09:47, 1 February 2012 (UTC)[reply]

If one needs a very low-tech solution - this is exactly the reason why I use a trackball with my netbook, with the touchpad disabled altogether. --Ouro (blah blah) 10:29, 1 February 2012 (UTC)[reply]

Scheduling with multiple deadlines

I have a list of exam dates like this:

schedule

16.3.
4 Sahkomagnetismi
4 Fotoniikka
4 Suhteellisuusteoriaa
4 Materiaalifysiikka I
5 Varioppi
4 Kvanttifysiikka II

13.4.
5 Lineaarialgebra A
5 Analyysi III A
4 Differentiaaliyhtalot B
8 Diskreetti matematiikka
8 Numeerinen analyysi

20.4.
4 Kvantti- ja atomifysiikka
4 Kvanttifysiikka I
4 Optiset kojeet

11.5.
5 Fysiikan peruskurssi I
5 Fysiikan peruskurssi IV
4 Fysiikan matemaattiset menetelmat
4 Elektroniikan perusteet
5 Tahtitieteen perusteet

16.5.
4 Analyysi III B
4 Differentiaaliyhtalot A
8 Sovellettu analyysi

25.5.
4 Johdatus topologiaan
4 Todennakoisyyslaskenta B
8 Analyysi IV

15.6.
4 Sahkomagnetismi
4 Fotoniikka
4 Suhteellisuusteoriaa
4 Kvanttifysiikka I
4 Materiaalifysiikka I
5 Varioppi
4 Kvanttifysiikka II

20.6.
5 Lineaarialgebra A
5 Analyysi III A
4 Differentiaaliyhtalot B
8 Diskreetti matematiikka
8 Numeerinen analyysi

24.8
5 Fysiikan peruskurssi I
5 Fysiikan peruskurssi IV
4 Fysiikan matemaattiset menetelmat
4 Kvantti- ja atomifysiikka
4 Elektroniikan perusteet
5 Tahtitieteen perusteet
4 Optiset kojeet

31.8.
4 Analyysi III B
4 Differentiaaliyhtalot A
4 Johdatus topologiaan
8 Sovellettu analyysi
4 Todennakoisyyslaskenta B
8 Analyysi IV

The numbers before each course name denote how many ECTS points each course is worth (in principle 1 ECTS = 15-30 hours of study). If you look carefully you will see exams for the same course are offered on multiple dates. On any exam date I can sit as many exams as I want. I only have five hours total to complete all of them so three exams per date is preferred.

I can't attend the actual courses so I'll have to study on my free time. On weekdays I'm reserved from 9am to 6pm but weekends are free, so I can spend perhaps 30-40 hours per week studying. Does there exist a free program that would help me build a schedule and update it to match my (lack of) progress? I'm on linux. I doubt I will have time to pass all the courses. --145.94.77.43 (talk) 00:27, 1 February 2012 (UTC)[reply]

That's an exam schedule ? Is "31.8." = August 31 ? Is it in a foreign language ? There seem to be a huge number of classes (50), you can't be taking all of those in 5 and a half months, right ? StuRat (talk) 01:19, 1 February 2012 (UTC)[reply]

Yeah it's August 31.8. I'd like to do as much as I can. --145.94.77.43 (talk) 01:25, 1 February 2012 (UTC)[reply]

So is the problem deciding which classes to take, or tracking your progress in each ? StuRat (talk) 01:27, 1 February 2012 (UTC)[reply]

All classes are about as good, but If I don't progress I'll have to drop classes. --145.94.77.43 (talk) 01:29, 1 February 2012 (UTC)[reply]

One issue that comes up is whether you think you can actually learn all the material permanently, or just cram it into your brain long enough for the test, and then forget it. If the latter is the case, it would be far easier to cram for just one test at a time. (I've always thought it was stupid that they expect so much memorization for each test that you can't possibly remember it permanently, but they do just that.) Also, aren't some classes pre-reqs for others ? StuRat (talk) 01:33, 1 February 2012 (UTC)[reply]

It's hard to make a trade-off between learning and "learning" before I get a feel of what's doable. A problem with cramming is the spacing of the exam dates. There's anywhere from two months to less than a week between them. I'd rather try to cram for an earlier exam date than try to sit too many exams at once later.

Pre-reqs are only recommendations and there aren't that many. In my experience the up-front effort required to pass courses in the wrong order often is a lot less than the sum of the courses. Mastery of the previous courses is not tested for. --145.94.77.43 (talk) 02:46, 1 February 2012 (UTC)[reply]

Let's clarify those dates, a couple seem wrong:

16.3.
 4 weeks
13.4.
 1 week
20.4.
 3 weeks
11.5.
 1 week ?
16.5. (should it be 18.5. ?)
 1 week
25.5.
 3 weeks
15.6.
 5 weeks ?
20.6. (should it be 20.7 ?)
 5 weeks ?
24.8
 1 week
31.8.

Did I guess correctly ? StuRat (talk) 03:02, 1 February 2012 (UTC)[reply]

It's all correct, except for the guesses. --145.94.77.43 (talk) 03:17, 1 February 2012 (UTC)[reply]

Do you mean my corrected version is right, or your original version ? (If your original version is right, then some of the tests are on different days of the week.) StuRat (talk) 03:59, 1 February 2012 (UTC)[reply]

My original version is right. --145.94.77.43 (talk) 04:05, 1 February 2012 (UTC)[reply]

OK, then, here's the schedule with the time intervals between each:

16.3.
 28 days
13.4.
 7 days
20.4.
 21 days
11.5.
 4 days
16.5.
 9 days
25.5.
 21 days
15.6.
 5 days
20.6.
 65 days
24.8
 7 days
31.8.

For those tests which are 9 days or less apart, you might want to skip one. Here's one possible way to do it:

16.3.
 28 days
13.4.
 28 days
11.5.
 13 days
25.5.
 26 days
20.6.
 72 days
31.8.

That would give you 6 test days and at least 13 days to prepare for each. If 13 days isn't enough, you could revise the plan to have 5 test dates, as follows:

16.3.
 28 days
13.4.
 34 days
16.5.
 34 days
20.6.
 72 days
31.8.

Personally, I wouldn't want to do more than one test each day, but that's up to you. If you do want to do 2 tests in one day, I'd suggest the last day, since you will have 72 days to prepare. So, this gives you 5-7 tests, depending on which of the two schedules you choose and whether you double up on the second test day. I wouldn't recommend anything more ambitious than that, while working full time. StuRat (talk) 05:38, 1 February 2012 (UTC)[reply]

Do all modern browsers support 8-bit PNGs?

I noticed a setting in Photoshop's Save For Web feature. Beside the normal PNG-24 setting, there's also a setting for PNG-8. It has a restricted color palette like GIFs do. And the files are amazingly tiny. Generally speaking, would any current browser/email client/app that supports normal PNGs also support these type of 8-bit PNG? --76.79.70.18 (talk) 02:06, 1 February 2012 (UTC)[reply]

It's part of the standard PNG definition so any browser that can truly support PNG can support PNG-8. For a list of those see Comparison_of_web_browsers#Image_format_support and in particular Portable_Network_Graphics#Web_browser_support_for_PNG for more granularity. --Mr.98 (talk) 03:26, 1 February 2012 (UTC)[reply]

I use 8-bit PNGs I've not had any trouble. Windows, Firefox, Opera works with them too. Don't know about IE 82.45.62.107 (talk) 18:36, 1 February 2012 (UTC)[reply]

Intel Turbo Boost

I have a computer with an Intel Sandy Bridge i7. How can I tell if it is using Turbo Boost? Bubba73 ^{You talkin' to me?} 03:07, 1 February 2012 (UTC)[reply]

Read this. I recommended downloading Intel® Turbo Boost Technology Monitor. Is it a good idea to add information about this software to the article Intel Turbo Boost? Von Restorff (talk) 03:46, 1 February 2012 (UTC)[reply]

Thank you, that is very interesting and the program works nicely. (It also explains some behavior I've seen.) And yes, it would be a good idea to include it in that article, as an external link at the very least. Bubba73 ^{You talkin' to me?} 04:01, 1 February 2012 (UTC)[reply]

Resolved

I found another way: start Task Manager, select Performance tab, click Resource Monitor button, and near the top it shows the percentage of CPU speed. So for mine, 103% there is boosting from 3.4 to 3.5 GHz. Bubba73 ^{You talkin' to me?} 04:17, 1 February 2012 (UTC)[reply]

YVW. Would you be so kind to describe your alternative method and the software in the article? I am sorry, I am not a native speaker, otherwise I would've done it myself. Von Restorff (talk) 05:09, 1 February 2012 (UTC)[reply]

OK. Bubba73 ^{You talkin' to me?} 16:15, 1 February 2012 (UTC)[reply]

Thanks! Looks good. Von Restorff (talk) 21:19, 1 February 2012 (UTC)[reply]

That Intel program is over 22 megabytes in size! It should be small. Bubba73 ^{You talkin' to me?} 04:01, 2 February 2012 (UTC)[reply]

I agree that could probably be smaller but the releasenotes say: "Size optimized for web download" in the section "Features Added in Intel® Turbo Boost Technology Monitor 2.0 v2.1". Von Restorff (talk) 11:30, 2 February 2012 (UTC)[reply]

The maximum frequency percentage in Resource Monitor (RM) is not consistent with the Intel program. I have a quad-core 3.4 Ghz i7, which should turbo boost to 3.5, 3.6, 3.7, and 3.8, depending on the load. When I had all four CPUs running, the Intel program was showing 3.5GHz and the RM was showing 103%. Since 3.5/3.4 = 1.029, that seemed right. With four programs running on low priority, TB was 3.5, 3.6 with three, 3.7 with two, and 3.8 with one. So decreasing the number of processes by one increased the TB by one step (0.1 GHz).

But now I have only one process running. The Intel program is showing 3.7/3.8, but RM is showing 101% (when it should be 109-112%). I don't know why this is so. The RM figure may not reflect the TB. Bubba73 ^{You talkin' to me?} 04:27, 2 February 2012 (UTC)[reply]

I am afraid it is not that simple. To be honest I have never really understood how this works; all I know is that its pretty complicated nowadays. Maybe someone else comes along who can answer your question. My guess would be that it has to do with dividing threads, not processes, over the cores. Von Restorff (talk) 11:30, 2 February 2012 (UTC)[reply]

I asked on Microsoft Forums, and that percentage is not the Turbo Boost factor (so I removed it from the article). It is how fast the base speed is compared to the rated speed. In Control Panel/Power Options, you can set the maximum CPU usage to less than 100% (to save electricity). You would mainly want to do this on a laptop. So my machine is running a little faster than the 3.4GHz base, not counting Turbo Boost. Bubba73 ^{You talkin' to me?} 19:38, 2 February 2012 (UTC)[reply]

search.iminent.com

That's the one, it's back again so I can check now. this website seems to have taken over my internet, every few days it suddenly sets itself as my homepage, and when I change it back, it waits and then suddenly reappears. How do I fix this?

148.197.81.179 (talk) 03:35, 1 February 2012 (UTC)[reply]

Did you try using any anti-malware programs ? StuRat (talk) 03:41, 1 February 2012 (UTC)[reply]

Try scanning your computer with malwarebytes antimalware, it is free and it works really well. Von Restorff (talk) 03:44, 1 February 2012 (UTC)[reply]

Seconded, Spybot can also be a good program to run occasionally. Just don't let it 'immunize' the computer- older computers can take almost as big a performance hit as if they were infested with spyware, although without all the credit-card stealing. Make sure you reboot into Safe mode before running the scans. Nevard (talk) 08:17, 1 February 2012 (UTC)[reply]

Try to clean your register, and deleting all those binaries --190.60.93.218 (talk) 14:59, 2 February 2012 (UTC)[reply]

TalkTalk / SKY in the UK

"TalkTalk" and "SKY" are trying to persuade me to take a package of 40 GB Broadband, and 24/7 free telephone calls. Has anyone had good or bad experience with these packages please. i would welcome any advice.--85.211.142.228 (talk) 11:49, 1 February 2012 (UTC)[reply]

Relatives of mine have the Sky package (in combination with a Sky TV subscription) and they've had no issue. Another has a TalkTalk package, and again has had no issue. Neither is a very high user of internet. -- Finlay McWalterჷTalk 13:10, 1 February 2012 (UTC)[reply]

Not really an issue but in some cases the cost of 24/7 free phone-calls can be as much as a lot of calls. Check your bill, if you are paying less than the extra for free calls then it probably isn't worth it. -- Q Chris (talk) 13:50, 1 February 2012 (UTC)[reply]

I've heard someone complaining that one of Sky's free calls packages has some unusual restrictions on which numbers you can call for free (they claimed that they are charged a very high rate for 0800 numbers, for example). Googling also turns up complaints that the '40GB per month' limit actually means '1.3GB per day'. 81.98.43.107 (talk) 14:26, 1 February 2012 (UTC)[reply]

Yes, it's wise to check your existing bill and look at the composition of calls (for stuff like 0800, 0845, and mobile calls). In general (that is, I've not looked at the Sky or TalkTalk plans in detail), call plans that are cheaper for regular land-line calls can be unexpectedly expensive for other numbers; for example, many SIP plans charge more for calls to mobiles than BT does, even though calls to landlines are cheaper. -- Finlay McWalterჷTalk 14:42, 1 February 2012 (UTC)[reply]

... and remember, of course, that the speed of downloads will be limited by any copper wire in your phone connection, whatever the provider. This may not be relevant to your situation if you have fibre to your premises. Dbfirs 16:14, 1 February 2012 (UTC)[reply]

Thanks all. --85.211.142.228 (talk) 08:13, 2 February 2012 (UTC)[reply]

Network diagram for a wiki

I'm working on a small Wiki (a few hundred articles so far) and wonder if there's any way of displaying graphically the interconnectedness of the various articles so we can see what links to what, and thus highlight orphaned articles or isolated groups. --rossb (talk) 20:03, 1 February 2012 (UTC)[reply]

Not sure a graphic approach is best for this type of thing. Even with only a few hundred articles, you may have thousands of cross-links so far, which would make for a very confusing graphic requiring you to zoom in and out and pan around. Once you get into thousands of articles it will be even worse. Perhaps just a tool that lists all orphaned articles (and all articles with fewer than N articles in it's ring) would be a better choice. StuRat (talk) 21:12, 1 February 2012 (UTC)[reply]

Maybe you knew this already but if the wiki is small you can easily see the orphaned pages by choosing "Special pages" in the menu and clicking on "Orphaned pages". On this wiki the page is disabled but the default setting is enabled, and it should look something like this. Von Restorff (talk) 21:48, 1 February 2012 (UTC)[reply]