November 7, 2011 Leave a comment
A week or so ago, I was at my friend’s place and was playing with his IPad. I don’t have any type of tablet myself, but once I win the lottery, watch out. I am intrigued by the functional ability of such small form-factor devices (IPads, IPhones, IPod Touches, Android tablets, etc). I guess I’im looking at these things through the eyes of someone who once had a 286 with dual floppy drives. I don’t mean the nifty 3.5″ either. I’m talking the 5.25 drives. Remember using a paper punch to make your single-sided disk double sided? Yeah, those disks. That thing took up my entire desk.
I realize that I need to change my approach on functionality of mobile computing. Yes, the processor in my iphone is a monster compared to even desktops that are a few years old, but how it is used is different. Can it do image manipulation such as photoshop? Sure, but that would put strain on everything else such as memory and data storage. How do we unload such bulk from computing while keeping functionality? Instead of hording the latest and greatest technological architectures, let’s steal from the best technology of the 1960s: thin client.
Going back to my childhood, I remember being amazed that every checkout lane in the grocery store was able to know the price of a loaf of bread. I was amazed that whenever the price of that loaf of bread changed, someone would have to go around to each register and update the price in each individual register. Obviously, it doesn’t work that way. None of the registers have the ability to store the price of any given product. Instead, they are all connected to a centralized computer that does all of the pricing. Fast forward to present day: pixlr.com is a website that allows web-enabled devices to execute the functions of photoshop. Now any system with a web browser can edit photos without having photoshop A) installed and B) using system resources. The processing and storage exists on some server “out there” somewhere. The final result is sent back to the user and the device at hand (most literally) doesn’t even know what is happening.
Call it cloud computing, thin client computing, or voodoo for that matter, this gives us immense flexibility. All a device needs to be able to do at this point is connect to a resource that is designed to accomplish a task and then send the result back. Any new task that is designed can get its own existance out in the cloud. Once the user knows where to look for it, the mobile device has everything it needs to add the given functionality to its Batman-style utility belt.
Conceptually, this is exactly what I’m looking for. I could write code, store it on the cloud, and then pull it down when I’m ready to deploy it to the Arduino. The problem with Compilr, however, is that it doesn’t have an environment for Wiring, which is the language used by Arduinos. I could use the C++ environment, which would be close enough for my purposes, but I want something that is native. Furthermore, since there is no native support for Wiring, the application can’t be built on the cloud like it can with the supported languages. Ideally, a user should be able to plug an Arduino into the tablet in some way or another and then click a Build button on the top of the IDE screen and have it automatically tranfer to the Arduino automatically. I might be asking too much, but it is worth bringing up.
So here is another project I’d like to attempt. I want to write an ASP.Net web application that acts like the Arduino IDE, including the color coding for recognized terms. I also want to add in automatic indentation and click-to-add code (insertion of where/for loops and other skeleton code). It will also include a funtion to upload a zip file that includes new libraries.
Initially, I think that the user will need to copy and paste the final code into the Arduino IDE for compilation and uploading to the Arduino board, but later attempts will try to skip this requirement.