Saturday, December 02, 2006

printer spiders linux

Uncle Dave was here before Thanksgiving. He wanted to print a Sudoku puzzle that he'd been working on in Excel. I said I had a printer and we could dig it out and hook it up to his computer. We did so, but it refused to print - it had the right drivers and all, but the light wouldn't stop flashing and the head went wild when the thing started up - I mean it would slam home so hard that the printer would jump sideways a few inches.

He finished the puzzle in Excel and we both forgot about the printer - until today. I plugged it into my computer with its fancy new Debian GNU/Linux OS. The computer recognized the printer and configured it in a flash - no 45 Mb printer driver downloads (yeah, that's the actual size of the hefty windows driver for the same printer - I'm sure that about 2 megs of that are driver and the rest HP software crap). But it exhibited the exact same behavior as earlier.

Then I took a look inside the thing. There was crud everywhere in there - it looked like someone used the insides for a place mat for a two year old eating spaghetti. And there was crud all over the encoder strip (the plastic strip with a million little black lines that is used by the print head to keep track of its position). That explains the slamming home issue and probably most of the rest of the funny business too.


So I took the top of the shell off - 4 screws - and proceeded to clean all the crud out of the thing. The the source of the crud became apparent. There were mouse droppings and some mouse collected junk in there. Fortunately the mice did not chew on any of the printer internals. However, while cleaning the plastic strip, some of the encoder lines came off with the crud - that's not good. But it was much better than it was to begin with - not sure what a few missing lines will do, but I decided to fire it up and find out.

Well, it does print now. But those missing lines cause it to loose a bit of ground on every pass - so it creeps closer and closer to one side on each pass. The result is a skewed print - still serviceable in a pinch, but pretty funny looking - I'd post a picture, but I'm too lazy to take one - maybe later.

I have spiders crawling all around my room. Every few minutes I see one dart across the floor from one pile of clothes to another. I like spiders. I think of them as my pets. They keep the mosquitoes and flies down in here.

I really like Linux and open source software in general. When I get bored I like to install some new flavor of Linux on one of my computers. I liked the latest Debian testing release (codenamed "Etch") so much when I installed it on my desktop that I've also installed it on my laptop.

One feature that I really like is smart command line completion. Command line completion has been around for a long time. But in all my experience with it up to now, it used some simple logic to determine if the word it is completing is supposed to be a command or a not-a-command, which is assumed to be a file name. I have not explored the new smart functionality fully as of yet, but it knows about file types, and commands that take other commands as arguments.

the which command takes another command as an argument and tells you the location of the file that would be executed if you were going to execute the second command. The old completion would try to match files in the current directory, the new completion tries to match commands - smart.

some commands only make sense with certain file types. mplayer plays video and audio files - it wouldn't make sense to pass it any other type of file. previously, completion was only semi-usefull. I have ripped some of my dvds to my desktop harddrive which is accessible to my laptop (better screen for movies) through nfs. I have ripped the subtitles with the movies and the subtitle files have the same name as the movie files, but with different extensions (there are 2 files for the subtitles for each movie). Previously, the completion would go up to the extension and stop because it does not know which file is which. If I wanted to watch Return Of The Jedi, a dvd which I own and am therefore entitled to copy/rip for personal reasons1, I would type mplayer je and then hit the tab key for completion and the completion would fill in the filename up to jedi. leaving the extension off because there are 3 files with that name and it would have no idea which one I want. The new smart completion however, would fill in the full filename jedi.avi because it knows that it would make no sense to pass either of the other two as arguments to mplayer. sweet. I'd been trying to figure out how to get the subtitles to work with with filenames different from the movie for this very reason - now I don't have to. sweet.


1. Actually, I don't have this particular movie on my harddrive right now - I did, but had a harddrive crash and lost all the movies on it. And I can't rip it again because the dvd is scratched and won't play or rip correctly. I think I can get it resurfaced, but it reminds me why it should be legal and not unnecessarily difficult to copy/rip movies. The main reason that I rip movies is so that I don't have to load a dvd every time I want to watch a movie - and the dvds themselves can remain protected in their cases. The only reason that Jedi is scratched is that I lent the movie to someone a while ago. Never lend your movies out - they will get scratched. Stupid crash-happy harddrives. Stupid DRM.

0 Comments:

Post a Comment

<< Home