I got a steal of a deal on a Samsung ML-2150 laser printer from Buy.com (plus $10 off for signing up with Google Checkout). I was going to hold out for a good duplexing one, but I figured if this printer lasts six months, I got my money's worth. Before I received it, I read somewhere that that Foomatic hpijs driver would work. Upon hooking up the printer, I found this not to be the case. I figured I could post the process of getting it connected here.

First, make sure you have cups installed. I print through a NAS device that doubles as a print server, so I also needed to get Samba installed as well. If you don't have cups installed, a quick apt-get install cupsys will do the trick.

Conveniently, Samsung includes drivers on their disk. However, the way the installer runs is bad. It's set up to run all print jobs as root, and well, that's just not a good idea. We work around this by just using the vendor supplied Linux drivers without the installer. I went digging around and found more current drivers than were on my install cd here.

Unpack the tarball. The unpacking will create a directory called cdroot. Execute the following commands, as root:

# cp cdroot/Linux/noarch/at_opt/share/ppd/ML-2510spl2.ppd /usr/share/cups/drivers
# cp Linux/x86_64/at_root/usr/lib64/cups/filter/rastertosamsungspl /usr/lib/cups/filter/

You've just installed the needed drivers. Isn't that MUCH less painless than say, running an install disk? Now, all that's left is to navigate to http://localhost:631, which is a convenient web interface for setting up printers. Select "Add Printer" and go through the steps, naming your printer (without spaces), and giving it some metadata (location, description). When it comes time to select a .ppd file, click the browse button of the file field at the lower part of the screen, and navigate to /usr/share/cups/drivers. You should see ML-2510spl2.ppd there. Select it. Move on, making sure to enter the path to your printer (in my case, it was smb://nas1/lp, but your path will most likely be different).

Once you've completed the wizard, print a test page and pat yourself on the back. While you're at it, pat Samsung on the back for providing Linux drivers that make this process so freakin' easy.

Disclaimer : This post may come off as a rant. I'll do my best to prevent that. However, I do not hate Microsoft. They have a lot of great products. The purpose of this post is merely to point out the quite cultural differences between the open source world and the Microsoft world.

Today I attended an MSDN event at a local movie theatre. I must say that it was indeed a different culture than my local LUG, and the Hacking Society branch here. I haven't used Windows exclusively since I was 17, so I've definitely found myself out of my element, and in an "community" that I had never experienced. I did not stay for the whole presentation (all four hours of it, with no wifi available in the theatre), mostly because I was bored, but for a few other reasons as well.

If It's Not Broke...

Ian Bicking recently wrote a post about why he doesn't buy into Microsoft's Silverlight. The basic idea is that Microsoft's (as well as Adobe's) current attitude toward the web is that they are trying to "fix" the web. I was confused by this post, until I attended the MSDN conference. While I'll admit that Microsoft had quite the number of tools that they had built for their developers (including a cross .Net Framework compiling infrastructure in Visual Studio 2008), but their attitude was that they were fixing something that was broken.

There's a large difference between improving something, and fixing something that is broken. When something is broken, it means that nothing works, and it's worthless. If the internet is broken, why are we using it on a normal basis? Sure, we could use improvements. That's what the whole "Web 2.0" thing was about: improving the internet.

Integration != Assimilation

My initial interest in attending the MSDN event was their talk of integration. Recently, the boys in Redmond have been making an albeit mixed effort to become "not evil." I commend them for. They've been trying to find open source "rockstars" who are willing to use their tools and promote their tools. They've published the source to .Net (even though their license is a freakin' joke). So I was naturally interested in how they want to "integrate."

Instead of hearing about their efforts to help out the Mono project or something like that, I hear them talking about making Javascript more like a .Net language. WTF!? Then it hit me. Instead of integrating, conforming to standards, Microsoft's idea of integration is assimilation. That means that everything that is worth anything in this world should be molded to conform to Microsoft's standard. Having worked almost entirely in web development for the last few years, I understand how well Redmond's finest adhere to standards. I suppose this isn't news, but when the presenter said "We wanted to make JavaScript more like a .Net language" I turned off mentally. Provide me with a library, give me an API for interfacing JavaScript with your services, but don't change Javascript, and don't change the Javascript language because YOU don't like it.

If You Can Compile It...

At the MSDN event, it was stressed that "If you can compile it, we [i.e. Microsoft] care about it." It was also very apparent that they didn't much care for an language that wasn't compiled. Scripting was unneccessary in their eyes. Granted, there are benefits to compiled languages. Bugs that are usually hard to track down can be caught at compile time. However, discounting scripting languages jsut because it doesn't have anything to do with the Common Language Runtime doesn't help your "integration" cause. Scripting languages have their place in the world just as compiled languages do. That "compiled only" fanboyism just makes me roll my eyes.

Now, I'm not gonna say that these attitudes are wrong. In fact, like I said, it was more of a change of culture to me. I'm used to the LUGs, and reading blog posts about Richard Stallman and the crazy things Eric Raymond has to say. The differing opinions, as much as it annoys me, is what I'm used to. What I'm not used to is people cheering when a new ORM (Orcas) is introduced to the .Net framework. So I'll stay in my open source world, with all the turmoil of GPL vs BSD, Vim vs. Emacs, and Gnome vs. KDE, because the "hive mind" culture is just not for me.

This is big news. Apparently, AMD is promised to publish specs for cards >=R500 which is great news. I've been a fan of ATI for a long time, and I've even become a bit of an expert when it comes to the fglrx driver. I don't have any problem getting hardware acceleration to work on any of my boxes, but I've had to fight with it on SuSe and RedHat.

While I suspect that Dell had something to do with this, with their fancy shmancy Ubuntu Linux line of desktop computers, this is great news. If I thought for one second that the radeon driver would compare with fglrx driver (which it honestly doesn't), then I would have taken the free driver over the proprietary one. However, I'm not gonna cut off my nose despite my face in the name of "Free as in freedom" I'm a GPL kinda guy, but there's a limit to my fanaticism.

I am excited about this. However, a point was brought up on an IRC channel I frequent that while the GPU specs may be published, there are other decoders on the card for HDMI, DVI, and other encryption and DRM enforcements that most likely will not be published. Why? ATI is probably under a pretty strict NDA from their vendors. Although I'm not super familiar with HDCP, it also has engineering requirements which restrict the firmware on those cards, and can't be controlled via software, so no go for HDCP with any new drivers. Of course, I'm sure you'll be able to find something similar to the non-free repositories that we all go to for libdvdcss2 (which is illegal in the U.S. - come and get me...)

This is great news for the linux community though.

The sysadmins over at Tummy.com are selling a lot of their old gear. I'll be picking up a few servers, one to use as a new network firewall, and one to serve as my new svn server. Their prices are pretty reasonable for such old hardware, but if you're planning on running Linux, they're pretty cheap boxes. They even have equipment for sale to build your own Asterisk server.

All the stuff for sale can be found at http://forsale.tummy.com/. They also update pretty regularly right now, so keep checking back!

My wife and I have been reading a lot of graphic novels recently. One the graphic novels that we've been reading is Marvel's "Civil War" series. It's all about the U.S. passing a "Superhero Registration Act" which requires all superheroes to register their "secret identities" with the government. The superheros then naturally either choose to register or not, which results in a superhero civil war, with Iron Man fighting against Captain America. Hopefully you get the idea

Today I had a rather interesting conversation with a few "friends" of mine in the #django channel of irc.freenode.net about choosing between GPL and BSD. While this is a personal preference, my preference was the GPL, mostly because I have this feeling deep inside that it's almost morally wrong to modify an existing piece of free software and then sell it as a new product. Initially, it felt a lot like those graphic novels I've been reading, where each side is insisting they are "right" even though both sides were asserting that is was merely a preference. Eventually, we all came to an agreement that neither was necessarily superior to the other, but more of an answer to the question "What do I want to do with this?"

I've had conversations like this many times with other developers, sysadmins, managers, and the like. None of them has persuaded me as did those few other developers in #django tonight. I'm actually very seriously considering releasing all the source code to this site (which I will do in a few weeks anyway) under the BSD license instead of the GPL. Why? Well, I figure that the only people who would probably want to see my source code would be developers, not likely end-users (because nothing I've done is incredibly useful). Since it's more in the name of education, I don't really care what someone does with it. If they want to modify it and make money, be my guest.

While I'll continue to use Linux (GPLv2) instead of switching to *BSD, I feel like I had someone give me valid arguments for the BSD license, and like anyone who is willing to learn and grow, I'll accept and understand it.