Once I had decided on the colours, I was able to refine the cover some more. I got rid of the edges of beige, so that the colours visually flowed down the edge of the box better and also doubled the strokes of beige that crossed on the surface. I was advised to keep it the thickness of the rim, but it looked better this way, especially when I decided to put a floral motif in the centre of the intersection. I was inspired by motifs used in henna designs and rangolis.
After finalising my design and visual system, I moved on to double checking the chunks of text that google translate had given me. I had to reach out to everyone I could to get help with hindi and malayalam (hindi because I can’t read or write the script and malayalam because I know nothing about it).
Malayalam gave me a hard time because of the script. One word would take up an entire line, giving me either really long character lengths, or really crazy rags or just problems in general. Making the type fit and connect with the other languages was way harder than I had imagined. A line that would be 1 and a half lines in english and hindi would take up 3–4 lines in malayalam.
This unevenness was picked up when I went for further feedback. There were widows and orphans that I had missed and was sacrificing the way the type looked to make it align with the shapes I had created. I had to go back and alter the grid I had created for myself to lessen the line lengths and get a more normal rag. The problem was, I had to treat each language differently. I was using the same lengths for each language, but it wasn’t working, especially for malayalam and arabic. I had to adjust everything and re-check to see if it looked balanced or not visually, without the gridlines.
The other problem was the way I was aligning my type around the edge. It looked disconnected because of the up-down composition I had done, and because of the change in size. The thought here was that if I centred everything and it all looked the same, then it would seem dull or boring. I was advised to consider stability as well. Because I had so much going on in the rest of the box, so many colours and languages, it was already far from dull. I needed some kind of consistency to give the readers eyes a rest and some kind of balance to follow.
I considered a few ways to incorporate hierarchy, like varying changes in size and colour. I was advised by friends that enlarging keywords didn’t work with arabic and hindi and just by looking at malayalam it was clear it didn’t work in that language either. The only place it worked in was in english. Which is why I decided to keep all the type sizes constant, and just subtly adjust some of the words in english. It made sense, since english is widely spoken, and I also didn’t want to emphasise it too much, otherwise it would create a hierarchy of languages.
Printing presented it’s own nightmares (in the epson printer). Initially, the colours were way off, even though I had exported it as CMYK, it printed in RGB (far right in the image). It seemed to be something to do with exporting it as a pdf. I tried to reexport it into photoshop and then export it again, but alas, no success. I don’t know why I made it sound so dramatic, but it was really frustrating. Even when I did get it to work (by exporting the files as a high quality jpeg, and then saving as a high quality pdf at 300 res), the colours were SO deep. It made the pink look very close to red (far left in the image). I wanted the bright happy colours that I had chosen (centre in the image), so I decided to use the inkjet printed to print. I could have experimented with the tints to adjust how much ink was used, but it would have wasted a lot of paper and time to test and get the exact colours.
I could have experimented with tints to adjust how much ink was printed by the epson, but it would have wasted a lot of paper and time to test and get the exact colours.
^ I also made a last minute decision to round the corners. It clashed less with the rounded motif in the centre.