Slashdot Mirror
Design A Standard For the Linux Standards Base
The widely reknowned HeUnique points to this LinuxWorld article, writing: "LSB wants to ask the Linux community people -- well, the artists among them -- to create an LSB logo." Rather cool to see a contest one of the rules of which is "All submissions must be created using Linux and native Linux tools. Frankly, most of us don't have a clue about how to check for violations. Just do it. We trust you." You've got until March 1st to submit two copies of your award-worthy artwork. See that LinuxWorld site for the full schmear, though. The LSB has been quiet for a little while, hopefully this contest hints (like the article does) at some action in the near future.
This is great fun watching slashdot folks talk about graphics. About as entertaining as listening to a bunch of graphic designers debate cobol versus Java, and just as accurate (see, the null pointer variable is creating an overflow into your read-only memory, dude!).
:)
First: Its absolutely idiotic to demand the logo be created in Linux. I mean, I know there are a lot of k3w1 linux kids out there who just love to get down with the gimp, but lets face reality here: if your goal is to have a good, professional logo that reflects well on the companies and community, dont set artificial limitations on the toolset. If you want some half-assed gimp graphic ("Look, your logo's on fire!") then by all means eliminate from consideration anyone who knows what they are doing.
Second: Please, dear god, everyone stop talking about specs.
Having Petreley say "Four colors or less often works quite well" is as annoying to a designer as saying "Try to keep the buffer overflows to a minimum" would be to a programmer. If anyone is planning on using Hexachrome for their logo design (or more than 4 spots), I'd love to know about it so I can be sure to never, ever work on a project with you ("what do you mean my business card will cost $10,000 to print?")
For those of you who can't figure out the issue between eps and TIFF, don't sweat it but please stop suggesting that you can just resample a TIFF to make it fit at any size. You can't. Yes, you can put a TIFF in an eps (hell, you could put the entire encyclopedia brittanica in an eps -- its just a wrapper format, like quicktime), no it won't make it scale any better. We'll just assume that when they ask for an eps that they are really asking for some vector graphic file in an eps format.
For future Consideration:
The kinds of things that might actually get you a good logo are never mentioned, presumably because they don't know the questions to ask (consider someone telling you to write a program for them, telling how many lines the source should be, what compiler to use, and never telling you what its for!).
What is wrong/disliked about the current logo? It does look very MS-office-ish (similar to the intelocking puzzle pieces) but is that the only thing they didn't like? Do they want something that conveys cooperation, cutting edge technology, stability, or what? These are all very different concepts, with different ways of representing them. If you focus on making an identity that highlights cooperation, you're making a trade-off against a feeling of speed and cuttin-edge tech. So which do they prefer? Damned if I know, they just want it at 640x480.
We seem to go through these same messages every time an article on GUI, logos, etc comes up here -- its fine if programmers don't want to know about how that stuff works, but its more than just a "pretty picture". Fedex did not spend millions of dollars redesigning their logo in the 90s because they wanted it to be "prettier". Having a window manager with "more colors" does not enhace the GUI.
Focus on what you are actually trying to do, and what you want to communicate -- not what color it should be (here's a design hint: if it matters what color your logo is, your logo IS BROKEN)
In a typical identity project, this would be an iterative process -- you'd bring one design for "cooperaton", one for "high-tech", etc -- and discuss with the client which they prefer, why, and then go back and incorporate the feedback. This is a one-shot deal, equivelent to programming an application without ever talking to the user, without having a beta test, and just dropping off an executable, never to be seen again. Sounds like a project I'm sure most programmers would LOVE to work on, right?
---------------------------------------------
Recursive: Adj. See Recursive.
"All submissions must be created using Linux and native Linux tools." - That's the point where I stopped reading. It's just too silly for me.