I think most of you might know about Breadth First Search (BFS), in which you print nodes of a tree from root to leaves. Now there is a small change to the way the nodes are printed. Lets say given a binary tree, I want the nodes to be printed level order but from leaves to root
Eg : consider the binary tree at this locations http://www.cs.mcgill.ca/~jeromew/comp252/images/Binary_tree.pn
Now you need to print : 5 11 4 2 6 9 7 5 2
Today I have attended a tech-talk for the first time. The tech talk is by Amazon Chennai’s team, on Kindle – From Ink to E-Ink.
When I first came to know about the tech-talk and noticed that one of the speakers is my ex-manager, Ashok Nadar, from Trilogy, I was very excited and very eager to meet him, after a gap of an year. He is one of the most brilliant guys I have ever come across. And I used to feel it was a privilage to work with him in my early stage of career :). I used to admire him a lot. The talk has been held in Taj Residency.
I reached there at around 6:00 PM, there were guys from Amazon who were interacting with every attendee and gave every one to get a hands-on experience on Kindle. It has wireless connectivity, an experimental browser that supports a sub-set of web sites, you can use annotations etc etc. Then the actual presentation started at around 6:30 PM. The main concentration is on what is Chennai team’s role in Kindle Development process, and a slight taste of few of the technical challenges they faced while digitizing the content from Ink to E-Ink that can been read on the hardware device, Kindle. As per my understanding the Chennai team has been involved in the process of showing the content on the device. The technical challenges included starting from scanning the text from the hard-copy or a note book, collecting information from blogs, periodicals etc and then going through some pipelines to display the same content on Kindle as the author/publisher of the book/blog/periodical had intended to be. While process the text you might encounter various different fonts of text (including the text being italicesed, bold, underlined etc) and you need to replicate it similarly, some times the scanned text has many overlapping of the characters etc. A slight picture of what kind of algorithmics used are kind of slightly touched.
It followed with a Q/A session, then I had a small talk with my manager followed by the dinner :-). And by the way we got a T-shirt too!
PS: This is for the first time I am blogging/writing about a technical session, so don’t mind if the content had not been properly organized 🙂