125 comments

[ 3.1 ms ] story [ 177 ms ] thread
GNU Radio isn't bad. As for the rest of them...
This is a mixture of WordArt inspired logos, with some more "traditional GNU" ones and some that are just WTF

DDD, EMMS, look nice

What is GTick supposed to convey? Groff? Recutils?

I think Emacs and The GIMP look nice (though GIMP is somewhat unfortunate...).

DDD and EMMS are detailed but not nice looking.

Groff has a letterpress machine in the background, so it's appropriate, but I guess not very logoish.
Indeed most of them are pretty unprofessional looking, but then again GNU isn't like companies who will spend tons of effort on finding a good logo.
Or tons of effort into producing streamlined and technically efficient software.
Care to cite examples?
- One Linux Window Manager I used understood the concept of icons, but used the same icon for every piece of software.

- Most apps didn't install .desktop files for years after they were standardised.

- OpenOffice/LibreOffice is still a shotgun blast of icons, years after MS abandoned the concept.

TBH, good examples of OSS UI are harder to come by than bad ones. The only one I can think of is Firefox.

Edit: as the poster below points out, those aren't official GNU apps, they just run on Linux. That said, I don't think GNU is much different from other OSS.

OpenOffice is not a GNU project, and the GNU window manager (GNOME Shell) doesn't suffer from lack of variety in icons.
I don't like everything the GNU project produces, but I do find that they tend to produce working software, at least among the stuff I commonly use. One of the first things I do on non-Linux boxes is install GNU coreutils, because the alternative coreutil options (Solaris, FreeBSD, etc.) tend to be some mixture of slow and buggy. One of the things GNU got right in its early days, imo, was to declare "no hard-coded arbitrary constants". So GNU awk and sed don't crap out with messages like "awk: record too long" because your line was longer than some hard-coded buffer size. Also, everyone else's grep seems slow.
That's simply not true. A lot of these projects provide software with excellent capabilities. The problem is, after someone writes a good algorithm the project is not done. Producing good end user documentation, a user interface that makes sense, or choosing UI paradigms that enable casual users to be productive - those are tasks which can easily take up the majority of a developer's time. A good case could be made that many open source developers don't even recognize these areas as goals or see how their project could possibly be structured any differently. This results in a lot of wasted time for the users, use case failures, a lot of duplicated effort across inaccessible projects, and the impression that these programs are not efficient.

However, by far not all GNU projects look like this - a ton of them could be considered "good as is" even by commercial standards. For example, Octave just works as it should, is well documented, and doesn't fight its users all the time. Even Gimp recognized usability was important, they just got it completely and mind-bogglingly wrong.

EDIT:

Wow. An innocent question turns into a downvote madness here. Oh boy. So much hate...

What is up with the bull logo? what's the significance of it?

- c-graph, the guy must love trumpet

- classpathx, doesn't even look like a logo

- clisp, this looks like some icon from the 1990s website that I used to see a lot as a kid :)

- dejagnu, oh i know this. this was the lovely MS Office Word's Word Art! They generated from that. I used to make powerpoint and documents with word art when I was a kid.

- emms, this looks quite normal to me

- gcc, I never knew gcc has a logo

- gdb, fish as logo, what's the reason behind it?

- guix, wow, that looks like Chrome logo

- httptunnel, that looks like a tunnel to hell

- gnowsys, This one is little cute

- parallel, with a massive logo but I don't see anything resembles parallel. I am definitely not mathematician or geometry expert

- pspp, looks awesome. dat pie chart though

- a few don't actually have a logo, but a text logo

and finally

- wb, this is the most impressive one; looks like someone just crop something off a photo using photoshop.

But then again, neither I nor most people are designer and expectations of how a logo supposed to look might be quite different back then.

It's not a bull, it's a gnu.

http://en.wikipedia.org/wiki/Gnu

Thanks. New vocabulary. Didn't know what gnu is/was. So are they actually in the same family as bull/cow?
To find out, you could, for instance, read the second sentence of the wikipedia link the other poster posted for you and for which you thanked him.
lines are parallel in parallel logo. it is a trick
I don't get it. Mind to explain further?
All grey horizontal lines are parallel while they don't look like it. Shifting black/white rectangles makes the lines look angled

Well, actually when image is huge, it is not obvious but if you scale page zoom down, you can notice the trick.

It's an optical illusion. Lines are actually parallel, but seem not so.
I would say the websites as well, compared with other startup websites, those website design still stays in 90s.
You think those are the worst logos? I guess you haven't seen the Perl 6 logo: http://perl6.org/camelia-logo.png
That's not a logo. It's Camelia, the spokesbug for Perl 6!

(Perl reaching out to pre-schoolers, it seems)

UPDATE: It appears that the logo was designed by Larry Wall. I now see its brilliance!

I read a talk by Larry on how Perl was a postmodern programming language. I, too, see the brilliance of this logo.
Ehh, it's cute. Which is more than can be said about most GNU logos.
The Smalltalk and Speex logos could be easily turned in something good with some work by a professional designer. Some others are hopeless since there is no idea at all.
Goes hand in hand with GNU software being some of the clunkiest and most awkardly engineered UNIX software out there.
Programmer art at it's finest. I think the idea is functionality over presentation though, just look at the sites most GNU projects are hosted on.
It's not like it's a balancing act between the two, though. Spending more effort on presentation doesn't necessarily mean that the functionality will suffer for it. People that are more aesthetically inclined can contribute better presentation for the software. Since it's open source, those people could be anybody, not necessarily software developers. That someone might want to design logos seems like a pretty different desire than making functioning software, so it's not like you would be diverting effort away from the actual software. (OSS programmers could also want to design logos, but that desire doesn't seem like something that would contradict their desire towards contributing to OSS.)

If you want more design-inclined people to contribute you'd have to market towards those kinds of people. So it seems like a catch-22.

Some of them (clisp) are not functional. You can't read clisp in the logo because of the color choices.

I looked at the logos first from a branding and functional perspective and my thought was that like all things there is tremendous variety in the set. Some logos were really good at doing the things you want to do with a brand. Others were metaphorically off-color. Others were really bad.

Again, though, that's normal. Logos are, well, a balancing act between functionality, uniqueness, character, and polish.

I see it as part of their charm - speaks to the homebrew, grassroots, piecemeal nature of them.

Not too slick and not too polished.

For instance, I feel that the artistic execution of the old NetBSD and FreeBSD logos really felt inline with my perception of the projects themselves. When they were replaced with thought-out carefully designed ones, in my mind, this abstract connection was lost.

git did this recently too. I preferred the tree-eating monster and vertical logo - it had so much more character and soul than the current look does.

(comment deleted)
> I see it as part of their charm - speaks to the homebrew, grassroots, piecemeal nature of them.

Which also incidentally are words many people would not want their projects to be associated with.

> git did this recently too. I preferred the tree-eating monster and vertical logo - it had so much more character and soul than the current look does.

The old one might have had "character" but the new one looks professional and the whole new image that git has nowadays makes it a ton easier to pitch to non developers. Why does the whole idea of appearing "professional" have such a bad attitude among certain developers?

> Which also incidentally are words many people would not want their projects to be associated with.

Then those people can use different logos for their projects?

> Then those people can use different logos for their projects?

They are. FreeBSD and git for instance changed their logos. Everybody is free to use a logo of their choosing, but they have to accept that others will make comments about them.

Sure, and I can make comments about how git's new logo looks unprofessional, like a logo of a freemium Android game. But is this a good use of our time?
I don't think it's about professionalism. It's about aligning the presentation with the reality.

I believe such designs are appropriate and well-suited for all-volunteer, community-run projects and there should be no generalized push to change them.

git may not have been the best example - but I did really like the monster.

  Why does the whole idea of appearing "professional" 
  have such a bad attitude among certain developers?
Professionals get paid, so professional projects need to make money somehow.

If someone plans to make money, and I can't see how they plan to make money, and they offer me something for free, I gotta wonder if they're going to start charging me, discontinue or pivot the project.

hmm, the logo's don't look like some giant-and-evil inc. is trying to sell you something so this could really appeal to the target community. makes sense.

I remember reading an article about discount supermarket chains where the designs of the product packaging and the store itself were intentionally made cheap-looking to communicate the message "you pay only for the product, not for the fancy packaging, advertisement or store".

So the logos of GNU maybe communicates just the right message instead of looking fancy.

git has a logo? Ditto for most of the other projects. I have recognized the Gimp horned-dog-or-whatsit (and that only because it's on the splash screen); for the others I use, it's approximately the first time I have noticed it.

(Cue wild speculation whether that signifies a need for a better logo, or whether each and every piece of infrastructure needs to have its own corporate identity manual)

As a graphic designer I agree, and wrote on this before:

http://i.liketightpants.net/and/my-favourite-logo-ever-is-fo...

> I heart the complete absence of symbolism. And the departure from traditional corporate marketing is all the more exciting because it rhymes with the product itself. GNU Pascal is open source software: it is changeable and redistributable at will. As such it is built more on the values of the academic world than on the traditional corporate ethos. And while visual identities for open source software tend to be pale imitations of corporate aesthetics, GNU Pascal is one clue of how such software could find a style of its own.

I think that's right for some of them. However, a few of the logos I looked at and thought "wtf?"

This isn't unique to GNU projects or open source, and I don't think these are worse than other ones I have seen in other contexts.

Haha... I knew most of this logos but in honesty most distros either don't use them or use their own. Seeing them together like that it really shows how terrible they are.

And if you think that's ghetto look at their respective homepages. I do approve of minimalism for websites though.

Some logos are like "Kill Me Now, Please"
Welllll... Gimp one is fine.

I agree with the rest though.

I think the emacs one is great. Guix and and c-graph could possibly be salvaged by simplifying and stylizing.
I have skipped emacs thanks to the monstrosities of emms and electric around it. Thanks. Yes, that's all right too.
"Do logos even matter?", I was asked recently by a talented and young web designer. While I think they do, I'm also open enough to the idea that maybe my view might be dated.

What's Facebook's logo? What's Asana's?

Colour and typography are, imho, 80-90% of the elements one needs to come up with an good "logo".

When a logo is just color and typography, they're called "wordmarks".
I don't really enjoy these submissions making fun of other people for their free, open source work.

Yesterday there was this thread about the "three = 1" in the linux kernel source code, now this. If you don't like them I'm sure most of those project would welcome a new design. Not to mention that many of those projects are tremendously useful and complex pieces of code. So what if gcc or gmp don't have cool logos? It's not as if they needed to market them.

This is what happens when we believe that open-source needs marketing to succeed: everyone who doesn't put in the time to make a Bootstrap-based landing page, buy a domain, and create branding is then viewed as lesser-quality. Why would I want to create any of that crap? I already write docs, comment, and keep up with it on Github.

Never mind that some of the linked projects are tackling problems orders of magnitude more difficult and hairy than some commenters here will even fathom undertaking in their 'careers.'

This has nothing to do with marketing, it's more a question of taste in my opinion. It's hard to argue that these logos don't all look extremely dated.
(comment deleted)
> This is what happens when we believe that open-source needs marketing to succeed

It does, if you want it or not. A project that is not being marketed will not be successful. If you are suggesting that GNU has no marketing you're wrong. GNU might have marketing not everybody agrees with, but it has a lot of it.

> GNU might have marketing not everybody agrees with, but it has a lot of it.

Exactly. It's just a different type of marketing from the Bootstrap-powered, coffee-house-backdrop intro video with indie guitar: "Hi! My name's Rob! Sometimes, I need to cross-compile to an embedded..."

> It does, if you want it or not. A project that is not being marketed will not be successful.

The homepage of vim, the editor every hipster who wants to look like a Unix greybeard tries to learn as if it were the holy grail of programming knowledge, has a very 1990s look. It also has more users than a lot of startup-published tools that claim to revolutionize how you do X.

There's no bootstrap landing page shit, no ads and no pitches. It finds users by word of mouth but there's very little actual advocating and no intentional conversion effort so as to deserve the word "marketing". No one tries to sell you vim, or make you open an account with it or anything else.

Same with emacs, autogen and emms and whatever.

> The homepage of vim, the editor every hipster who wants to look like a Unix greybeard tries to learn as if it were the holy grail of programming knowledge, has a very 1990s look.

But that's often not how people perceive vim these days because other channels have started filling the void. Vim's marketing these days is people referring the editor to others and letting them know how cool it can be.

Just because the original developer does not do marketing does not mean that other people won't fill the void. I for instance have been using vim for years and years and I'm very commonly showing off my vim setup to others at conferences and sharing my configuration online, tweeting screenshots etc.

Did it occur to you that people use vim because it's useful, and not just because they are hipsters who want to look like Unix greybeards?
That this was posted to make fun of the logos is one possibility, but I'd prefer to think that the poster might have wanted to simply bring attention to it. Knowing there's a problem is the first step toward resolving the problem, right?

I completely agree that the GNU projects are very useful. That said, no matter how useful and valuable a product is, all that value is simply lost if no one uses it. Making these projects more attractive to people not already familiar with them will help them to flourish the way that they should.

I think most would consider that at best a distraction.

More often then not overly flashy websites seem to look bad on my book. To me it seems more like they are trying to slip something under the rug while simple designs come across as more focused. Like http://prgmr.com http://tarsnap.com and http://nearlyfreespeech.net .

I agree, I'd rather have a website like http://www.slackware.com/ than any of that bootstrap stuff.
Why? The "bootstrap" stuff utilises modern, standards based semantic markup. The linked site is poor both aesthetically and technically.
Calling something "World's Worst" is, IMHO, not a very nice way to start a discussion about improving it.
Maybe open-source does actually need marketing and an appearance of quality and modernity to 'succeed', whether you like it or not. Code quality or complexity is often invisible and largely irrelevant to end-users (as long as the tool does the job), and it's a known fact of life that people are (at least initially) attracted to things that look good.

If I just look at myself as someone who's been using OSS for years, and absolutely depends on tools like gcc for my day-to-day work, even despite all that I still sometimes find myself discarding certain OSS software solutions I have no prior experience with, for nothing else but a crappy amateurish logo or an ugly website. I know it doesn't say anything about how good the software is, but somehow it's an instinctive reaction, which I think is pretty universal among humans.

In a broader sense, if you look at some of the reactions this article seems to trigger, I think you can safely see that tunnel-vision and a complete disregard of the differences between OSS developers and potential OSS users is a huge missed opportunity. You can be confident in the quality, complexity and utility of a piece of software whatever you like, but that doesn't 'sell' it to end-users.

If OSS wants broader acceptence and ever become a real-alternative to commercial software it needs superficial polish and marketing. This may not be relevant for compilers or debuggers, but it is relevant for tools facing non-developer end-users.

Nothing's preventing you or anyone else with good graphic sense to jump in the team and propose better logos. That's what OSS is about: sharing and improving each other's work.
Well said. If you don't like it, improve & contribute.

I'm indebted to the amount of amazing tools available by GNU and can't imagine how much they've contributed to the growth of numerous lean startups. Not all open source contributors are great designers and they mostly do this in their spare time.

making fun of other people

not sure the intent really is to make fun of people. I read it as (light) criticism, which isn't the same thing, and which should be allowed by all means preferrably without people getting all politically correct about it. Imo.

Apart from that: I'm no designer but some of these logos look hideous to me. Though I also think some of them are just meant to be like that, as a f*k you statement: "it's about the content, not about the logo". Which is prefectly understandable: ever had to come up with a decent original logo? That can take days which could as well be spent programing for instance.

> ever had to come up with a decent original logo?

Yes, don't under-estimate the time-cost of this for non-designers. I tried it once. The design sucked, I got advice and got totally lost in a 3 day tour of typography and colour theory.

In the end I think it looked ok but it was a lot of effort for apparently 'not much' output.

However as others have pointed out this kind-of presentation IS important and totally worth it.

Redesigning them is actually a very good opportunity for someone in need of a good portfolio.
I don't think that open source should be above making us laugh every once in a while.

You can look at this way instead: The focus is on good code, not nice logos which fits the Linux image well.

Maybe the intent wasn't to make fun in a malicious way. If hackers can't have a little fun at their expense, who can?

Complaining about `three = 1` was a complete miss though.

(comment deleted)
I view a project as a whole: its face, its code, its features. If people can criticize elegance or efficiency of the code, why can't they criticize elegance or efficiency of its brand? Rails, for instance, is an example of a project where people did not just "do code", but also promoted best practices, designed a nice website, created a blog etc. That inspired Ruby & RubyGems to ultimately have nice websites and logos, contributed for free by professional designers.

Any open source project will help more people if it has a friendly face: well-documented code, nice logo, well-written articles explaining motivation behind technical decisions and so on. All of this contributes to the project's overall utility.

There's a product/market fit thing going on here. Rails is the kind of project that caters to the kind of people who value and can contribute "nice websites and logos" because that is what a lot of them do for a living with Rails.

Add a startup-like website to the GNU binutils project and it's going to look pretty dumb to the no-nonsense crowd that needs assemblers. A "nice website" website is not going to extend the appeal of cross-assembling ARM microcontroller code on a x86-64 one bit.

I agree in principle. But, in fairness, I felt that the "three = 1" post was out of genuine puzzlement. I was certainly puzzled when I saw it, and it was very enlightening to see lifthrasiir's answer.
No need to get pissy - nothing is above mockery and open source is not an exception.

> If you don't like them I'm sure most of those project would welcome a new design.

Funny you should mention that, I literally did this - once. Wanted to improve the landing page for an open source project and improve the logo. The maintainers wouldn't have any of it. In their eyes the project was fine the way it is.

And this isn't just one project. I bet if I were to go to cgraph with a better logo they would reject it. Open source folks are stubborn as a donkey 90% of the time.

Adult package maintainers are having a laugh at their own logos. Teenage package maintainers are getting offended.
A better response to criticism would be to try and learn why people think they are terrible instead of looking for a reason to feel insulted. Whether OP is making fun or not isn't even important. Let him have a go if he wants. Whatever his motivation is, he's got a point. The logos are terrible. These aren't necessarily subjective judgments; there is such a thing as best practices for logo design. Remember what a logo is for. Hint, it isn't to entertain the programmers or designers. It is meant to be a visual cue to help people recognize your group/company/club/whatever. It should be simple and concise among other things.

> It's not as if they needed to market them.

I disagree, IMO a wider userbase is usually a good thing for GNU/Open Source projects, for the community, and for the software/services industry as a whole.

>Yesterday there was this thread about the "three = 1" in the linux kernel source code,

Yes, and that thread generated a reasonable discussion about programming. How is that a bad thing? Do you think someone will stop using ext4 or Linux because of a questionable variable name in a rarely used bit of filesystem code?

I've always puked at how bad GNU logos are (far worse than most other bad open source software logos even), but I think I kind of have a strange soft spot for them now - they are extremely Stallmanesque in their qualities.

Interesting timing with this post, as I just did a redesign of the nginx logo for my own pleasure yesterday[0], and was thinking about doing redesigns of other open source software logos as well. I have started on one for FreeBSD. Any other requests (hoping for software I actually use myself)?

[0] http://dribbble.com/shots/1436258

I like your nginx logo it very much!
Thank you! Maybe I should mention that it is CC-BY-SA licensed, so the project can use it for free if they like. But I am pretty sure the current logo serves the project just fine, and changing it is understandably a low priority for a non-consumer facing project.
I like your nginx logo it very much!
From the linked website:

    Many GNU packages don't have logos—if you are skilled enough and willing to
    contribute logo for a package, please contact the package maintainers and show
    them your work. If approved, let us know at <webmasters@gnu.org> in order to
    add it to this page.
OMG Lol, fornicating turtles?

What is this even trying to say?

Perhaps a campaign to raise $500 to spend on fiverr.com should be organised?
It's actually encouraging to think that those logos are horrible by the standards of most people here. It means that there's progress.

In fact, it's not rare these days to see open-source projects with thoughtful logos and well-designed homepages, especially for web stuff (http://yeoman.io/, http://gruntjs.com/, http://gulpjs.com/, etc.).