GLITCH COP!3R

I recently got a very nice package in the post all the way from Australia! Drew Taylor (aka bowsneak) sent me a package of art and zines, one of which was GLITCH COP!3R

GLITCH COP!3R

GLITCH COP!3R

[GLITCH COP!3R] was created as part of Sticky Institutes Target 168 2012 zinemaking challenge, in which a zine had to be created within 168 hours (1 week). The ‘secret’ theme for the challenge was ‘C3100 photo copier’.

All of the images in this zine were ‘stolen’ from the internet, and glitched to produce ‘a new work’. Images were selected from Google image search using the phrase ‘c3100’. It was maintained that the file name of each image must include the letter ‘c’ and the number ‘3100’. The masthead of the zine – titled ‘GLITCH COP!3R’ – also includes the alphanumeric ‘C3100’ (the ‘O’ is made up of an ‘O’ and a zero).

GLITCH COP!3R

GLITCH COP!3R

GLITCH COP!3R

GLITCH COP!3R

If you take another look at the front page you’ll notice that the font used in the masthead is Dataface. I think this is the first time that someone – other than me – has used the font for a project!

Zines!

I was also sent lots of other zines including many issues of Death-Beam Dinosaurs, which you can download and print for yourself

A grid of Death-Beam Dinosaur zines

Thanks again for the zines!

Speaking of zines, Birmingham Zine Festival is coming back again later this year.

Create your own glitch typeface

Making Dataface was really quite an exciting journey. What started off as an attempt to make a typeface inspired by glitch art turned out to be a story of collaboration, exploration and hours of research. Here, I will go through my process.

As you may have seen from my previous experiments in vector databending it’s totally possible to manipulate vector files. My original method for creating Dataface was to save each glyph in the Liberation font to an SVG file and then go through the process of glitching it for each file. Obviously this would’ve taken me a long time, hence why there was very little activity between my original announcement in January and when I started work on it again a few weeks ago.

At this time I thought about writing a script to do this for me. sed is a great command-line utility for Linux that essentially does the same as using find/replace on a character. As it’s command-line it means I can do a lot of automation with it. So, I wrote this simple script that attempted to solve the problem

[sourcecode language=”bash”]#!/bin/bash
rand=$(($RANDOM % 9))
sed -i s/[0-9]/$rand/g fontfile.svg[/sourcecode]

The only problem was that it would replace all numbers in the file with whatever random value was chosen by $rand as the script was executed. Not only is this bad because it would result in a lot of strangely similar glyphs but also because it would modify the header data of the font file, thus rendering it unreadable. I soon remembered that recently the SVG Font specification was finished, which aided my cause by putting all of the glyphs in one big file, but I still couldn’t find a way to efficiently randomise values in the file.

Thankfully fizzPOP came to my rescue. I’m glad that hackerspaces have people with a range of abilities in hardware and software, as I was soon presented with a solution to my problem by GB. After a few revisions he created a script that would replace only specific values in the file and wold even let you specify how much it should be randomised. You can download the finished script and source files and have a go for yourself.

Click to download

 

Simplified instructions on compiling the script:

  • Unzip the file in a clean folder. This will give you three files:Font_Sample_-_Liberation_Sans.svg, glitch.l and makefile
  • Type “make” into the command line (without the quote)
  • If you haven’t got make, type:
    [sourcecode language=”bash”]flex -t glitch.l >glitch.c[/sourcecode]

    [sourcecode language=”bash”]gcc -o glitch glitch.c[/sourcecode]

in either case, you will get a program called “glitch”.

Please note this has only been tested on Linux, requires Flex (available in the Ubuntu repository) and it is designed to work on SVG font files. I only know FontForge that is able to create these fonts files. To run the script do the following

[sourcecode language=”bash”]./glitch 0.50 outputfile.svg[/sourcecode]

That tells the script to glitch the file by 50%. I have noticed that sometimes you get errors if you put in 1.00 or more.

Once you have generated the file you can import it back into FontForge to save as a .ttf, .otf or whatever font type you choose!

(I still hate Comic Sans)

Here’s everyone’s favourite Comic Sans glitched at 50%

Dataface

After months of hard work I’m happy to finally release Dataface

Click to download

Special thanks go to various hackers at fizzPOP for all of their help, GB for programming help and Scribbleboy for general guidance. You can view of preview of the font here on Flickr.

There will soon be a lovely writeup of how it was done and the tools to use to create your own. Until then, I hope you enjoy Dataface!

Dataface update

Dataface is coming along rather nicely!

Dataface (by hellocatfood)

It looks different from my previous updates as I’m using a different method for glitching the typeface. This may change yet again if a script that I’m working on decides to work.

Dataface update

You may remember from my earlier blog post that I’ve been working on a databent typeface. It was mentioned a fair while back now, but I have been doing bits of work on it every now and then. Here’s a bit of my progress so far:

Ass you can see some of the characters are more recognisable than others. In fact, looking at it again I can’t really remember what some of them were. As I’m planning on having most characters mapped out, in upper and lower case, progress will be a bit slow, so I’ll aim for April for a completed font.

Beginnings of a glitch typeface

I was speaking with Jon earlier about my work and he’s noted that a lot of it has been text based and then asked if I was working towards making a typeface in the same style. I must admit, my recent text based work has mostly been an excuse to use the awesome Kawoszeh typeface, but I feel he’s onto something.

Whilst I’m quite far from a complete typeface I’ve been doing a few experiments:

Believe it or not that is the letter A glitched in the same way (replaced 9 with 15), but under different conditions. The reason for the above experiment is wanting to find the best environment in which to make the typeface. For example, the more nodes you have on a shape the more variance you get. The type of nodes that you have also has a major effect.

I’m also thinking about what typeface to use as a base. Being mostly Brummie I’m drawn towards hacking Open Baskerville although using just Arial Black provided some good results, as can be seen in my short glitch animation

I’ll have something produced next year and, should I finally do some coding, actually have a script to databend for me and make the whole process a bit more random!