I’m not sure exactly how to describe this behavior and I don’t really know what does it, but I think it’s something with the trackpad (not sure about this)?
I find when I’m typing that the cursor can just suddenly jump wildly around on the screen. It often moves way up and then suddenly I’m typing in the middle of some other sentence.
It also often selects everything between where I was and where it jumps, and then as soon as I type another key it erases everything (because it often happens so quickly I don’t notice it has happened before making the next key stroke)
I have looked in the trackpad settings and nothing in there seems to relevant. It seems like maybe this is some kind of macro or shortcut I don’t know about. I’d like to turn it off but I honestly don’t even know what I’m triggering.
It happens kind of a lot tho. It might be some kind of keyboard shortcut I’m triggering as well? Unsure, but it seems to have something to do with the heel of my palm grazing the trackpad.
It sounds like your hand is hitting the trackpad while you’re typing, as you say. I don’t have pureos, but there should be a setting in the trackpad settings that says “disable while typing” or something similar. I also believe there’s a function key that disables the trackpad, its icon will look like a small trackpad with an x on it (from what I recall).
Sorry these are guesses. I don’t have a L14 in front of me.
thermochemical, I second the motion (pun intended).
It seems to be slightly better under Gnome, which I don’t use. If you ever find a solution please post it here.
I use Xorg, and I’ve tried using synaptics Xorg driver to configure the touchpad as well as the ‘synclient’ utility provided by the synaptics driver. In the end the tweeking and more tweeking never accomplished anything of value for me. Maybe, if you’re using Xorg, you’d like to give it a try. Here’s a place to start: http://unix.stackexchange.com/questions/28306/ddg#139262
Jumpy mouse pointers go back to the 1980s. The original PS/2 serial port mouse interface sent 3-byte packets to provide information on mouse movements and button states. Every once in a while, noise on the line or perhaps a clock synchronization issue would cause the packets to lose alignment with the host, so mouse coordinates would be misinterpreted, resulting in erratic behavior. Smart mouse drivers have since corrected the problem by adding some crude statistical logic to realign packets in the case of too many or too few bytes arriving.
When USB mice emerged on the scene, the industry usually just installed a USB bus interface on top of the existing PS/2 logic on the mouse side, so the problem wasn’t actually fixed. Trackpads are more or less just mice in a different mechanical manifestation, so they suffer from the same problem.
The right answer, absent a redesign of mouse logic that isn’t going to happen, is to apply the same sort of statistical resync to the mouse driver. My guess is that this has never been done for certain Linux drivers, simply because the programmers were unaware of the problem (and the cause is totally nonobvious from bug reports).
The practical answer is that your trackpad might be suffering from the same sort of mechanical integrity problems that I’ve observed with my Librem 15v4 keyboard. I get repeated keys sometimes, although the problem is much rarer now that I’ve entirely removed the warped right bracket from the spine, which was compressing the electronics, resulting in electrical noise. My trackpad hasn’t worked at all in a long time, which tells me that yours[[ may be connected to the same problem.
It would be interesting if the problem comes and goes depending on the angle at which you open the laptop.
Anyways, I gave up long ago. I use a USB mouse. It might help to have one handy to unplug/replug (thereby smacking the mouse driver stack) when the problem occurs. (Perhaps one could also suspend/resume. Granted, suspend has been a dumpster of failed hardware state restoration since the day it was invented.)
Hopefully Purism will deploy more robust mechanicals and/or noise proofing in the future.
I also had issues in the past with a jumping cursor. I had fun with it by loading a couple dozen operating systems on one laptop. I noticed some O/S gave me the problem and some did not. So I eventually leaned toward Mint Cinammon since it rarely gave me that problem.
I say rarely, a few times when using Zoom and Thunderbird at the same time, it came up. So I closed T-Bird and it also went away. Must have been an early version of Zoom since it doesn’t happen any more.
Long story short, it may be a difference in libraries utilized? (KDE vs Gnome?)
That does make some sense. Depending on interrupt service latency, among other considerations, one OS or one particular driver might do better than others at mitigating this problem. But in the end, it probably can’t be eliminated. It’s an artifact of bad hardware design that we’re probably stuck with.
That said, while your experiment is illuminating, I obviously wouldn’t recommend choosing a less secure OS because it has better mouse behavior. So we’re back to stuck.
A secure o/s was never my main modus operandi here; for me it is just a side issue. Considering constant monitoring is a just part and parcel of my retirement, I’m just an old guy saying to other parties,“get off my lawn.”.
(I’ve used that last line before, this is repeat.)