| Summary: | new guide: using a mouse within a console (gpm) | ||
|---|---|---|---|
| Product: | [OLD] Docs-user | Reporter: | nm (RETIRED) <nightmorph> |
| Component: | Submit New | Assignee: | Jan Kundrát (RETIRED) <jkt> |
| Status: | RESOLVED FIXED | ||
| Severity: | enhancement | CC: | docs-team |
| Priority: | High | ||
| Version: | unspecified | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
| Attachments: |
guide: gpm.xml
gpm.txt guide: gpm.xml guide: gpm.xml guide: gpm.xml guide: gpm.xml guide: gpm.xml guide: gpm.xml guide: gpm.xml |
||
|
Description
nm (RETIRED)
2005-07-19 00:16:41 UTC
Created attachment 63752 [details]
guide: gpm.xml
Created attachment 63753 [details]
gpm.txt
Comment on attachment 63753 [details]
gpm.txt
Same as previous attachment
Created attachment 63768 [details]
guide: gpm.xml
Created attachment 63769 [details]
guide: gpm.xml
i think the config section is wrong ... /dev/input/mice is a cumulative device for all mice on the system, not just one chances are your USB mouse shows up as /dev/input/mouse1 and your touchpad shows up as /dev/input/mouse0 ... the usage of /dev/psaux is deprecated (and can be disabled in the latest 2.6 kernels) the guide should tell people to try /dev/input/mice before anything else and if that fail, fall back to other devices also, i'd add some more notes about different protocols to try if the default imps2 and ps2 fail to work ... Created attachment 63828 [details]
guide: gpm.xml
Fixed with vapier's suggestions. I searched for quite some time, but
unfortunately, the only protocols seem to be ps2 or imps2. At least, this
applies to the stable gpm available in Portage. Kernel =< 2.6.2 could be
patched with experimental support for using evdev with outdated versions of
gpm, but this has been deprecated for well over a year. The problem has since
been solved.
The only other place to set protocols, such as ms3 and ExplorerPS/2, is
actually in an xorg.conf file. Xorg has many more options, but they are
independent of what is available to gpm. As it is, gpm is extremely flexible,
and IME handles serial, USB, and PS/2 mice with only ps2 or imps2; somehow it
still gets the job done.
looking at mice.c in the gpm tarball suggests the following exist (i guess you could look at the gpm info page too but i hate info) reveals about 35 protocols ... however, i dont really recognize any of them except for 'imps2' and 'ps2' :) maybe add a note that says 'if imps2 and ps2 both fail to work for you, please refer to the gpm info page for other protocols to try' Created attachment 63829 [details]
guide: gpm.xml
Still has the changes suggested by vapier, but this should be the *right* file;
sorry, I uploaded the wrong version. This one no longer has "FAQ" at the end.
It should have been "guide" all along. Darned Bluefish...
Created attachment 63831 [details]
guide: gpm.xml
Includes the info page suggestion.
i'm not sure if we have any guides that cover lynx / links, but if we do we should link to them in your conclusion otherwise, the only other change i'd make is: -This guide shows you how to set up and use gpm, the general -purpose mouse server, from within a command line interface. +This guide shows you how to set up and use gpm (the general +purpose mouse server) from within a command line interface. looks good, nice job Created attachment 63838 [details] guide: gpm.xml Fixed with your suggestion. Oh, and I looked around for more documentation on links/links2/lynx, but there's nothing in the Gentoo docs except what's referenced in every chapter of the various handbooks, the bit about how to browse documents online or offline. However, the handbooks are arch-specific, so I'd feel kinda strange about including a pointer to http://www.gentoo.org/doc/en/handbook/handbook-x86.xml?part=1&chap=2#doc_chap4 , even though it references links. AFAIK there's an identical chapter for every other arch, but not one _non_arch-specific reference to links. Seems very good. However, I have some comments: a) I'd write "General Purpose Mouse server" instead of "g... p.. m..." (capital first letters). b) "In my case, I have a USB mouse on /dev/input/mouse0. So, I have uncommented /dev/input/mice" deserves explanation, IMHO. c) I'd add something like "(<c>info gpm</c>)" after "look at info page". d) First paragraph of "Working with GPM" calls both mouse and real cursor "cursor" in one sentence which could cause troubles for readers. Created attachment 64022 [details]
guide: gpm.xml
Now with 75% more jkt changes! :) Reasonable enough suggestions, so I
incorporated them. Well, except (d). I'm not sure what you meant exactly; I
certainly didn't call a mouse a cursor and a cursor a mouse. That paragraph
reads well to me. Clarification?
Well, I was inaccurate :-), here it goes: "Simply highlight the text with the left mouse button (it will stay highlighted when you release the button), switch to a different terminal if you wish, position the *cursor*, and press the middle mouse button to paste the text where you placed the *cursor*." What I meant to say is that average user would think that you are talking about mouse cursor, not the cursor controlled by arrow keys on keyboard. Some applications support this while others (IMHO most of console apps) don't. I'd recoommend adding a note saying that the rectangular cursor controlled by mouse is independent from your "real" cursor and that pasting will take place where you have your real cursor. (In reply to comment #16) Seems okay to me. Still not entirely sure about what changes you would make to it. That paragraph seems straightforward as it is. It will be pretty obvious to a user--even a new one--that if clicking on a line with the mouse doesn't bring the cursor there, then that user will just have to use the keyboard to bring the cursor to the right spot, just as if gpm wasn't installed at all. Doesn't seem like that big a deal. Users will find out pretty fast if the mouse will work in every circumstance. I will commit this. In CVS with minor fixes, thanks. |