This is quite possibly the design that almost killed the holiday season for me (thankfully, my love of the holidays pulled me through…)
The project was a collaboration between myself, Brian Pember and ajellebean. We started the newsletter project in November, as a studio original design that we hoped would fill some of the customer requests we’d had over the last year for a design that would allow users to enter a lot of text while also providing an attractive way to include their photos and video.
We had a great late night brainstorming session about what we wanted to do, what we wanted to offer our customers in terms of customization and experience (six pages, 0 – 18 photos, four canned cover options), and who would do what. Our timeline – two weeks.
We had this idea of doing the project using a book model that is presented as a present during the holidays. We formalized through email some of the finer details over the next week before the design went into production.
Brian started the project by modeling the intro and book opening in Maya. When he had a solid working model, ajellebean turned around a great page turning mask in After Effects that I was able to port into Flash and integrate into the book model.
The book model was going to give the users something we’d never done before – while offering six pages for user content, if a user did not enter text on pages subsequent to the first page, those pages would be ignored (or not displayed) to the recipient upon playback. This was revolutionary – a important step toward dynamic edit points, that helped expose potential issues we would need to overcome (especially when providing for editor/recipient print of designs).
I set up the pages to take 0 – 3 photos each (all click to zoom), and so much text that it ultimately proved problematic for the Smilebox client. As it turns out, and many of you already know this, the rendering of text within Flash is incredibly processor intensive. As a result, we decided in future large amounts of text like this should be captured as bitmap data to trim down (wherever possible) the fat. We trimmed down (as much as possible) the file size of all the Maya renders, but they still proved somewhat heavy. The page turn masking done by ajellebean in After Effects and ported to Flash were light and gave us a great integrated feeling with the Maya renders done by Brian.
The most important lesson (and one that we’ve been aware of for quite some time) is this: when developing new technologies, especially ones that push the envelope, it is ideal to have at least as much time for testing and the refining of code as you’ve had for developing it. In this case, on a two week development cycle, it would have been ideal to allow for an additional two week refining period in order to address issues and improve the design. It is in this manner only that you can provide the best possible product for your user.
As such, I was making updates and changes to this design up until the very last moment before it was released. Not ideal. And the holidays being what they are, especially in this business, the process and effort necessary to get this design to a shippable shape was physically and mentally challenging — but worth it, as I move forward into the new year with lessons learned.
|Make a Smilebox greeting|