I love Richard Stallman's often recounted printer story, and as a long time user of a Free software desktop (exclusively for nearly 15 years now) I empathize with the pain of getting printers working well with Linux based systems. Well, CUPS has become so good these days and printer manufacturers seem to care enough about Linux support that things are actually pretty good, if not perfect, these days.
Still, when we decided to get a new printer for the house I approached the task with a certain amount of personal surrender and the expectation of annoyance. I spent some time one afternoon looking at the various options for a simple black-and-white laser printer, preferably with duplex printing support (aka "print on both sides of a page") and settled on an entry-level Samsung printer, the ML-2955DW. It seemed a good product on paper in terms of the price/performance ratio. Even though it's ~1 cent more per page over the life the printer (something that doesn't matter too much to us as we don't print that much), it got very good reviews online and has features like Wifi and the sought-after duplex.
Going back to the Linux support, I was a little nervous when I searched for reports on using this printer with Free software operating systems: I only found a couple of hits, the primary one being one fellow's somewhat manual and only semi-successful effort to get it working under Fedora. So with the knowledge that it does work (if with some effort) and figuring "Hey, I'm a geek with Linux know-how so I should be able to get it to work, right? Samsung makes all kinds of Linux devices these days so they must have reasonable Linux support, right?", we placed an order.
When it arrived we set it up in the home-office and were very satisfied with the weight (it's light) and look (it's elegant from the blue-LED light buttons on the control panel to the fold-up lid on the paper output to the paper size markings on the adjustable paper tray) and happily surprised with some of the nicer features like the WPS button for wifi setup. I pressed the WPS button on our wifi AP and then on the printer and waited a few seconds until the WPS button stopped blinking on the printer. In theory, it was set up and ready to go. Now to set it up under Linux.
We fired up S.'s laptop running openSUSE 12.1 and went into the printer config. After not finding the exactly right driver, we went to Samsung's website where the "Universal Driver" package was available for download. After unpacking it to disk, expecting to start a bunch of cp and vim as root, I saw that it had a GUI installer. Even more cool: it uses Qt! There is also a text mode installer, but the Qt GUI worked like a champ and was amazingly simple and dreadfully uneventful to use. It set up CUPS perfectly, detected the printer on the network quickly and when it came time to print a test page .. up popped the Qt printer dialog! The only thing that could have been better is if they had included the Linux software on the CD that shipped with the device.
Still, my Linux hardware pessimism kicked in: it was all going a little too smoothly. Would it print properly, or would we get some strangely bad output? Would the features like duplex printing work without any tweaking? We too it for a test drive and ...
... everything. just. worked. Yes, including duplex printing. LibreOffice seems to have some problems with that feature (though perhaps I did something wrong there: it's print support is so baroque and difficult to use that I would not be surprised if that was a PEBCAK error ;), but all the Qt and KDE software I tried worked flawlessly.
The moral of the story, at least for me, was that the Free software desktop has come leaps and bounds in the last decade. My dinosaur-era expectations of trouble were not rewarded, and instead I was treated to a great experience. Thank you to everyone who has worked on making this possible: those who have slaved over printing support, those who have made awesome toolkits like Qt to tap into all the features, those who have guided companies like Samsung from the relative dark ages into the light when it comes to Linux support ... to all of you: my deepest gratitude.
Amazing how one little printer can tell such a big story to the right person. :)
About Me
Labels
- akademy (3)
- amarok (1)
- appeal (1)
- build system (1)
- community (7)
- crash (2)
- d-bus (1)
- dataengine (1)
- dolphin (2)
- events (1)
- family (1)
- free software (1)
- gnome (2)
- graphics (1)
- guademy (1)
- kde (52)
- kde e.v. (3)
- kde; kdegraphics (1)
- konqueror (1)
- krunner (4)
- libs (5)
- lug radio (1)
- media center (1)
- misc (1)
- openid (1)
- oxygen (2)
- performance (1)
- photos (1)
- plasma (26)
- podcast (1)
- pr (1)
- prose (1)
- release (1)
- social (1)
- solaris (1)
- techbase (6)
- travel (6)
- trolltech (3)
- web2.0 (1)
10 comments:
With Samsung printers I never had trouble installing them. Usually I plugged them in, Printer Manager showed up, and everything was set.
Was kind of funny (and embarrasing for the others) when they had trouble printing on their Windows machine, I came over with the notebook, took the USB cable, and printed. :)
Glad to hear it. The last three printers I set up have been Samsungs; the first one at work back when I was running a book café; the second at home for me and Hanna, and the third one at my parents place.
At the last two places it was in a mixed Linux/Windows environment.
Never had any trouble! Samsung printers are definitely my first choice.
While I find this fascinating to read, I also sadly notice our expectations towards Linux devices.. That somehow autoconfig doesn't happen. I hope that will change, because that's a story on it's own!
Better check whether they still do this:
http://digg.com/news/story/Samsung_Linux_printer_driver_modifies_the_permissions_of_many_executables
:) Had a similar experience when I recently installed Ubuntu on a laptop. The installation etc. had taken place earlier in the afternoon, and I was in bed later at night when I first tried 12.04. The thing that I found really, really spooky was that Ubuntu just told me the printer model I had (2 rooms away) and all I had to do was to confirm it. Works. Thanks. Next.
This may not only be against some pessimism about printers and GNU/Linux. I'm one of those people who grew up with Windows 3, parallel printer cables and dot matrix printers. Had someone told me back in those days that I'd have a computer in bed that would instantly find, recognize and configure a printer without cables… No way I would have believed that. :)
Oh: There's Windows Vista on the same laptop. According to the printer's documentation it should be possible to connect to the printer under Windows as well. It didn't connect automatically, though, so I decided it wasn't worth the effort. :)
You are aware that those drivers are proprietary? I think it is a very bad idea to use or recommend those drivers, and to praise Samsung for providing them is entirely counterproductive! If anything, those easily available proprietary drivers will make it much less likely that somebody will develop a Free driver, so it would be better if those drivers didn't exist at all.
Your praise of those proprietary drivers (without even a mention of them being proprietary!) also makes me very suspicious about the proclaimed openness of your Vivaldi tablet.
PS: The installation process with a Free Software driver is either:
1. plug printer in
2. The printer works!
or:
1. plug printer in
2. say yes when PackageKit asks whether you want it to search for a driver
3. enter your root password
4. The printer works!
The fact that you need to find, download and run an installer at all is very user-unfriendly, and a direct consequence of the driver being proprietary.
PPS: And it seems you completely missed the point of Stallman's printer story, where the entire problem was that the driver was proprietary!
@Kevin Koffler: the point of this blog entry was that things "just work" these days far more often than one might expect when using a Free software desktop.
i agree that it would be infinitely better if the drivers were open source. at the same time, i'd prefer the printer to work rather than not work.
that vendors are providing support for their hardware "out of the box" is a good improvement. eventually we can expect more and more of these available drivers to also be released as f/oss.
getting pissy at me doesn't make that happen faster.
please note that i'm not recommending any particular printer or non-free driver .. simply that a piece of hardware worked because the Free software desktop has gotten to the point where we can begin to simply assume drivers that work will be available.
oh, and this:
"also makes me very suspicious about the proclaimed openness of your Vivaldi tablet."
is so utterly out of line and random. seriously.. i ate jam on bread yesterday for breakfast, i suppose that must also say something about the clothes i wear while hacking? it makes as much sense as your stupid jab above.
the reason it bugs me so much is because it is an example of the "crabs in a bucket" BS that too many people involved with Free software subject others to. so instead of pushing forward together, we get to create imaginary points of conflict out of thin air with each other and waste time and energy on that instead.
very, very disappointing.
Just for the record, I'd like to point out that all the Samsungs that I mentioned above have been PostScript printers, and hence they simply worked out of the box with free drivers. I did not use any proprietary driver (and wasn't even aware Samsung provided them).
I'm not going to stop recommending Samsung since they offer proprietary drivers for some of their models.
Post a Comment