I just spent the last 6 months on tour around the country at major collegiate hackathons with every type of hardware on the market meeting hundreds and hundreds of developers and watching them fail (and succeed) on a variety of devices. There is room for improvement, I have some opinions about what needs to change.
If I were to summarize it in one phrase: The next generation of devices must be "Hackathon Ready." Lets talk about what that means by first examining the problem areas with the current available devices.
1.) Wi-Fi is terrible, spotty, blocked by HTML login pages and inconsistent at most public spaces. When it does work, "Isolation Mode" prevents most consumer hardware hacking boards from being useful. Regardless, many devices REQUIRE Wi-Fi to be working in order to succeed.
2.) The ability to power devices efficiently in a way that lets users wear them and move them around without needing a plug or rats nest of batteries and wires leaves much to be desired.
3.) The #1 desire of most developers after plugging their device in is wanting to control it (trigger events) with their mobile phone. BLE is a natural choice for this but lack of widespread support and adoption (and, instead, reliance on Wi-Fi) is slowing down progress.
Solve these three problems and we are now "Hackathon Ready."
Here is what is being done right:
1.) BoneScript, Johnny-Five, Bleno: Allowing developers to write code (and use BLE) using a simple, higher level language like JavaScript is the future. Arduino code is simple and meant for beginners, but the syntax innovations being included in libraries like Johnny-Five, Bleno are an improvement over that.
2.) Web / Cloud Based Development / Wireless Deployment: Allowing developers to get code onto their device without the use of any wires is a good thing. This should be the future - no wires.
3.) Stackable portable battery units and modules: Allowing users to power their devices in an elegant way the way Edison can be stacked via the SparkFun kits. Powering devices should be easy and elegant, not a mess of wires and voltage converters.
Here is what I want to see in the next generation of devices:
1.) Eliminate Wi-Fi as a dependency to doing something cool. Because Wi-Fi is a complete and total failure at large public hackathons, it shouldn't be the basis of being able to deploy code to your "thing."
2.) Greatly enhance BLE support. I don't see any reason why code authored on a computer couldn't be transferred to a local device over BLE. Using BLE as the primary networking mechanism and means of deploying code would allow these new devices to succeed at hackathons where most of the current crop of devices are failing to deliver.
3.) Stackable battery packs for portability in every kit. This should be standard issue. I can't tell you how many times I have been at a hackathon and seen developers struggle to try to wear their "hack" in a graceful way. I wish this was default behavior.
4.) Instant control / ability to remotely, instantly trigger device events with an Android and iOS phone via BLE. Just like the Spark Core allows me to modify my Spark Core via the Wi-Fi, I want to see this being done via BLE.
5.) A Web based device simulator so that it will be possible to test javascript robotics code written using a tool like Johnny-Five can be simulated online, just as we do now with Native Android and iOS applications.
In short: No more Wi-Fi, there has to be a way to make this work with BLE.
More reasons to kill Wi-Fi as a requirement:
Hackathons are spreading like a highly contagious virus (according to TechCrunch, there were 1,500 this year). Nearly every major technical college now has one or several major yearly technical events where students come together and build projects. Many of the students I have met are having their FIRST project experiences with ANY technology at hackathons and will stick with whatever devices make them glimpse their first success at hacking. As a result, I believe that the "next big thing" will emerge as a result of success at these events.
I just spent the last 6 months on tour around the country at major collegiate hackathons with every type of hardware on the market meeting hundreds and hundreds of developers and watching them fail (and succeed) on a variety of devices. There is room for improvement, I have some opinions about what needs to change.
If I were to summarize it in one phrase: The next generation of devices must be "Hackathon Ready." Lets talk about what that means by first examining the problem areas with the current available devices.
1.) Wi-Fi is terrible, spotty, blocked by HTML login pages and inconsistent at most public spaces. When it does work, "Isolation Mode" prevents most consumer hardware hacking boards from being useful. Regardless, many devices REQUIRE Wi-Fi to be working in order to succeed.
2.) The ability to power devices efficiently in a way that lets users wear them and move them around without needing a plug or rats nest of batteries and wires leaves much to be desired.
3.) The #1 desire of most developers after plugging their device in is wanting to control it (trigger events) with their mobile phone. BLE is a natural choice for this but lack of widespread support and adoption (and, instead, reliance on Wi-Fi) is slowing down progress.
Solve these three problems and we are now "Hackathon Ready."
Here is what is being done right:
1.) BoneScript, Johnny-Five, Bleno: Allowing developers to write code (and use BLE) using a simple, higher level language like JavaScript is the future. Arduino code is simple and meant for beginners, but the syntax innovations being included in libraries like Johnny-Five, Bleno are an improvement over that.
2.) Web / Cloud Based Development / Wireless Deployment: Allowing developers to get code onto their device without the use of any wires is a good thing. This should be the future - no wires.
3.) Stackable portable battery units and modules: Allowing users to power their devices in an elegant way the way Edison can be stacked via the SparkFun kits. Powering devices should be easy and elegant, not a mess of wires and voltage converters.
Here is what I want to see in the next generation of devices:
1.) Eliminate Wi-Fi as a dependency to doing something cool. Because Wi-Fi is a complete and total failure at large public hackathons, it shouldn't be the basis of being able to deploy code to your "thing."
2.) Greatly enhance BLE support. I don't see any reason why code authored on a computer couldn't be transferred to a local device over BLE. Using BLE as the primary networking mechanism and means of deploying code would allow these new devices to succeed at hackathons where most of the current crop of devices are failing to deliver.
3.) Stackable battery packs for portability in every kit. This should be standard issue. I can't tell you how many times I have been at a hackathon and seen developers struggle to try to wear their "hack" in a graceful way. I wish this was default behavior.
4.) Instant control / ability to remotely, instantly trigger device events with an Android and iOS phone via BLE. Just like the Spark Core allows me to modify my Spark Core via the Wi-Fi, I want to see this being done via BLE.
5.) A Web based device simulator so that it will be possible to test javascript robotics code written using a tool like Johnny-Five can be simulated online, just as we do now with Native Android and iOS applications.
In short: No more Wi-Fi, there has to be a way to make this work with BLE.
More reasons to kill Wi-Fi as a requirement:
Hackathons are spreading like a highly contagious virus (according to TechCrunch, there were 1,500 this year). Nearly every major technical college now has one or several major yearly technical events where students come together and build projects. Many of the students I have met are having their FIRST project experiences with ANY technology at hackathons and will stick with whatever devices make them glimpse their first success at hacking. As a result, I believe that the "next big thing" will emerge as a result of success at these events.
If it is "Hackathon Ready," it is ready.