Steven and Scotty have been working together to create a completely localized version of the collection for the sticks using the new genCollectionInterface tool. The biggest hurdle to overcome is presenting the user interface initially when the stick is inserted into the XO, and delivering the metadata before the book is opened in the Journal. In the meantime, we also developed an online version of the user interface that utilizes a streamlined version of the Internet Archive bookreader (also known as the embeddable bookreader). This bookreader interface works amazingly well on the XO and fits neatly on the screen. We also optimized the collection interface for viewing on the XO – we strongly encourage you to view it on a little green machine if you can (it looks a bit bare in a web browser – but wonderful on the XO … see the screenshots below!). We still have a bit of tweaking to do, but we are quite pleased with the results!

Additional Screenshots: [ 1 ] [ 2 ] [ 3 ] [ 4 ] [ 5 ]. We worked diligently to provide an “XO-friendly” interface that accommodates the native elements in the Sugar UI. We are still tweaking some navigational issues, and resolving screen resolution on the XO – but we are pleased with the results!

It is possible to generate either version of the user interface using the genCollectionInterface tool with minor modifications to the code. Right now, we are trying to reach consensus as a team on how much crossover there should be between the online and local versions.

Levi has polished up his public domain icon set, creating one that is Sugar-friendly. We are exploring how to finesse the resolution in Sugar, given the fact that the XO browser renders the pages using 134 DPI. We are also researching to see if there are any standard naming conventions for Sugar icons so we can appropriately name and format the icons for the various states (on, off, hover) so developers can get right to work. We invite other Sugar developers to use these icons in their applications to represent categories of public domain books, and we will be improving the collection in the future – including giving developers an outlet for feedback so we can improve the set. This will also be explained in our documentation. Stay tuned!

Hello! After a brief respite to organize a phenomenal art show and enjoy the changing seasons, we are back in action for the latest work from the Rural Design Collective! We learned from our demos that we gave at our Launch Party and have been ironing out a few bugs based on feedback.

After much deliberation, we have decided to package our previous summer work into a new tool called the “Collection Interface Generator” (genCollectionInterface), a python application developed by Scotty Auble. We think it is exciting in that it works on the XO plus it is accessible to any person interested in creating a localized collection of public domain books. For a sneak peek, you can check out the ongoing work we are doing in the UI Lab. The pages were generated using the program which made it easy to create of custom collection based on our UI work. The metadata for the tooltips came from a .csv metadata file that Steven McKenzie created using Internet Archive advanced search, and the covers were downloaded using the Open Library API. It is a work in progress – we will be tweaking the navigation and other UI details – but we think it is really starting to come together. Our release in October will be accompanied by documentation to explain the Collection Interface Generator, as well as additional information behind our overall concept and method which has developed over time during the 2009 Summer Mentoring Program.

Levi’s work on the icon collection is outstanding – he has developed a versatile set of pictographs that work in a broad range of visual settings. This is the hallmark of a strong icon set, and a great basis of images to release into the public domain. This will be documented along with our code.

The primary usability issue that we are encountering is how to load the DjVu books into the XO Journal as fluidly as possible. There is an interstitial step when the DjVu file is loaded onto the XO and transferred to the Journal that needs to be populated with metadata in order not to disrupt the user experience (you can see this screen here from the GIAB activity). One solution we are exploring is using the embeddable bookreader from the Internet Archive. This version of the bookreader works well on the XO and enables us to emphasize collections which is central to our theme – the only drawback is that it uses JPEG2 files which was not the format we agreed upon for the SoaS solution. We can provide links to the books at the Archive as an alternative format, however, for users with an Internet connection. We will be solving that problem in the weeks ahead – Stay tuned!