Superbowl 43 Friday, Jan 30 2009 

Pittsburgh Steelers Helmet

Pittsburgh Steelers Helmet


MindPulse 1.5 Avaliable Now! Wednesday, Jan 28 2009 

Just a quick note, Apple has approved MindPulse 1.5 and it is avaliable now at the iTunes App Store.

Somewhat related, I spoke yesterday with a reporter from The Washington Post about my BodyPulse and MindPulse apps.  The article should appear next Tuesday in the (free registration required) health section.

MindPulse 1.5 Monday, Jan 26 2009 

MindPulse (my MindMachine/BrainMachine/Brainwave Entrainment App for the iPhone and iPod Touch) has been updated to version 1.5 and submitted to Apple for approval.  The approval process for updates seems to take about a week; sometimes more, occasionally less.

Version 1.5 brings the most often-requested features:

  • Volume Control
  • Fade-out of the audio at the end of a session
MindPulse 1.5

MindPulse 1.5

In addition I’ve added a few things that haven’t been specifically requested, but that I felt were in order:

  • Updated UI
  • Smoother transition in and out of the visual stimulator
  • The app keeps track of where you are in a session (how many minutes are remaining), that way if you interrupt the session, you can pick up where you left off

There were a couple of additions I had planned for this release that did not make the cut this time, the main one was the ability to run the app in “audio-only” mode, to be able to jump back and forth from the visualizer without stopping the session.  I think this would be useful when you can’t partake in the full experience (for example, at the office), but you would still like to use the audio portion of the program.  It took me a long time to come up with a design that I was satisfied with for this function, but I wasn’t able to get it working satisfactorily before my code-freeze deadline.

Another feature that hasn’t been requested directly but indirectly is the ability to accompany the binarual stimulation with music from the iPod application.  I don’t like the approach other apps have taken of adding “canned” music or sound effects to the application because I think that it’s important that the user have a connection with the audio that accompanies the beats, and the chances of selecting a handful of tracks that will have meaning to the thousands of users of this application is low.  Instead, I would like to simply allow the iPod to continue playing while MindPulse runs, this way the user can select anything they like to go along with the entrainment program.  I don’t think there is a technical (or contractual) hurdle here, but I haven’t verified this because this feature was below the others in priority up to this point.  This is definately something I’ll be looking into for the next release.

The third item I’m seriously pursuing is customization to the visualizer.  The simplest form this may take is allowing for different colors of light to be used (currently the light is pure white), perhaps connected with a particular program or user-selected.  A more complex option would be to provide images in addition to colored light, perhaps even user-supplied images.  I’m not aware of any science that indicates the benefit of using images instead of pure light, but it has been requested and I don’t see the harm in it, as long as the priority of more compelling features are respected.

A few other loose ends reguarding MindPulse.  The first is that I wanted to mention that one of the biggest factors delaying the release of this update was my struggle with the user interface, and how to incorporate the changes without mucking up the cleanliness of the original interface.  I was never satisfied even with the original, and when I started working on adding controls (specifically the volume control) everything that I came up with frustrated me.

I’m not a graphic designer, but I’m a competent user interface designer; however, sometimes you need graphic design skill to implement a user interface and it is at this point that I struggle the most.  Someday I’ll simply engage a designer who I can work with collaboratively on these issues but in the meantime I have to try and struggle through it myself when the UI demands it and this is an area where vast amounts of my time are absorbed.

After several flawed revisions I finally came upon what you see here (actually, something slightly more sophisticated but I couldn’t quite put it together in time to make my deadline, but you’ll see it in the next release).  For many reasons I would have prefered to have released this update months ago, but the design simply would not present itself, and I felt in my gut that it was better to hold off until I had something that at least I was comfortable with.

One final note and then we’ll put this topic to bed until the update is actually avaliable.  I’m playing with the idea of releasing a “Lite” version of the app for free.  Something that would provide a limited taste of the experience to those who are curious about brainwave entrainment but not quite ready to invest financially in it.  What I have in mind is one-program (most likely Alpha) app that runs for a fixed amount of time (probably three minutes), enough for you to take the thing for a spin and see what it does for you.

If you have any thoughts on this I would encourage your feedback.

You can’t write a Killer App Thursday, Jan 22 2009 

Often the Holy Grail of independent software developers is to come up with the next Killer App.  Wikipedia defines a killer app as:

“…any computer program that is so necessary or desirable that it provides the core value of some larger technology, such as a gaming console, software, operating system, or piece of computer hardware.”

Often VisiCalc is cited as the first killer app, as the desire for this application was so great that many users bought an Apple ][ just for the ability to run it.

VisiCalc on the Apple II

VisiCalc on the Apple II

Killer apps are good for developers and platform vendors as well for obvious reasons.  In addition, when an application has this sort of effect on hardware sales, it puts the developer in a position of advantage with the platform vendor and can lead to a level of collaboration (for example, unlocking undocumented features of the hardware) that would otherwise be unlikely.

So it makes sense that most developers spend time thinking about “the next killer app”; ironically, this ensures that they will never develop one.

This may surprise some of you but it should not.  There are very few, if any examples of killer apps that have been invented by software developers.  The reason for this is that the source of inspiration for a killer app is a serious need that is shared by many, typically in a area of specialty or a particular profession.

The Video Toaster sold Amigas to almost every American cable station

The Video Toaster sold Amigas to almost every American cable station

A software developer might come up with a killer app that solves a great problem in their own work, but it would only be useful to other software developers and while this might meet the definition of driving platform sales, the volume is not what we would normally associate with the general expectations that come with the term “killer app”.

The more common scenario is that a layman comes up with an idea in the process of their everyday work, and a way that a computer might solve that problem:

“The idea for the electronic spreadsheet came to me while I was a student at the Harvard Business School, working on my MBA degree, in the spring of 1978. Sitting in Aldrich Hall, room 108, I would daydream. “Imagine if my calculator had a ball in its back, like a mouse…” – Dan Bricklin, inventor of VisiCalc

They then engage a software developer (or more often, become one themselves) to construct the idea.  In the most successful examples, the inventor becomes enough of a programmer to build a usable prototype which is then used to lure in more specialized labor, management, etc. to turn the prototype into a product.

So if you set out to create “the next killer app”, the odds are definitely stacked against you, but if you insist I have a couple of thoughts on the subject:

1.  Build a killer app factory
Instead of trying to come up with the next killer app, build a software company that is optimized to fill the role of realizing the primitive prototype of one into a complete, marketable product.  The focus would be on recruiting ideas and on optimizing the product development, maintenance and support cycle.

2.  Diversify your interests
I would hope that most developers are interested in something besides computers.  Explore these areas of your life and how you might improve or alter them through software.  If you are able to do this, then you might just come up with an idea that meets a need for people outside the software development world and as a developer, you have a more concrete idea of the possibilities and limitations of the problems software is best suited to solve and those which it is not.

One final note on the subject: if we go by the definition of a killer app as an application that is so desirable that it justifies buying an entire platform just to run it, then your app should probably utilize a platform that isn’t in everyone’s hands yet, or at least not in the hands of your primary audience.

If you were to write an app for Windows, and the audience was corporate business users, while your application may be popular it is probably not going to increase hardware or operating system sales noticeably.  On the other hand if you wrote an application for Windows that made it easier for artists to reach their audience that might be another story.

The key point here being that most “killer apps” happen on emerging platforms that are yet to find where they fit in the software user universe and working with the platforms, even if only for inspiration, gives you an advantage in the quest to write the killer app…

…but you already know that’s impossible, right?

IRWIN Unibit Tuesday, Jan 20 2009 

Based on a mention in a project article I read I decided to pick up one of these “unibit” (this is Irwin’s term, there may be a more generic one) for drilling thin steel cases.

IRWIN Unibit 10231

IRWIN Unibit 10231

In particular, the article mentioned that these bits help to make cleaner holes that require less deburring.  I didn’t exactly know why this might be, and I’m saving up for a nice Cobalt Steel set of bits but I was at Fleet Farm anyway and knew I’d be drilling some of this steel that same afternoon, so I decided to give it a shot.

The only brand I could find was Irwin (maybe they are the only ones making these bits?), which is fine because I’ve been satisfied with the other Irwin bits I’ve purchased, although had I known the bit was available in Cobalt maybe I would have ordered the bit online and waited (probably not, since I wasn’t sure how well it was going to work yet).  There were three bits available on the shelf, the #1, #2 and #4.  I chose the #2, which has 13 steps and can drill (cut?) holes between 1/8″ and 1/2″.  The #4 looked cool, but was a lot more and the #1 with only 6 steps, wasn’t much cheaper.

As I found out, having more steps is better if you can swing it and when I buy a Cobalt version of this bit, I’ll opt for the 20-step bit.

First I tested the bit on an Altoids tin that I keep specifically for testing this operation.  This makes it easy to compare each method/technique.  For the drill I used a Chinese bench-top drill press that got from my father.  Instead of a vice I used this cool little “fence” that he made for it to keep everything stationary, this is important because in the past when I’ve used regualar twist-bits I’ve had these little boxes ride right up the flutes and take off like a helicopter.

I opted not to use a block behind the hole.  I usually do so, because it allows me to use more pressure on the drill without denting the steel, but I wanted to see what this bit could do unassisted.

Taking my time and watching the material, the first test (to the first stop on the bit) went very well.  I took the box out of the press and examined the hole.  The denting problem was minimal if not non-existant.  I’m sure that this won’t be the case on a larger piece, but it’s good to know.  Overall the hole was very clean, with some burring on the exit side but better than a regular bit.  I think the deburring action of the bit is due to the cutter of the next step up beginning to bite which removes a little material from the entry side of the hole.

Next I took the bit down another step and then another.  The sound changes and the following holes require less effort than the first (something to note if you’re only trying to go a few steps).  By stopping the drill it’s easy to see what size the hole is and what’s next using the legend on the bit.

Once the testing was over I used the bit to drill the speaker mounting holes in a lunchbox that I’m turning into a guitar amplifier.  Overall it worked very well, I didn’t even “dimple” the steel to guide the bit (although this can be attributed as much to the use of the press as to the bit itself).  The holes required little cleanup afterward and overall I was very impressed.

Next I tried the bit freehand as the holes for the amp’s controls were going in the side of the box and it was too big to fit in my press on its side.  For these holes I used an awl to dimple the steel slightly to prevent the bit from wandering and chucked the bit up in my 18v Black & Decker cordless.

Using the variable speed control of the drill, I kept the speed low until the bit penetrated the steel.  Controlling the pressure of this drill is a lot tougher than the press and I didn’t want it to shoot through and accidentally make the hole too big.  Once the first hole size was cut however I increased the drill to maximum speed and cut the hole to size for the gain control without problems.

Drilling the remaining holes I noticed a couple of things about the bit.  The main thing is that the larger the hole, the less deburring you have to do.  If you drill just to the first stop, there is usually a big chunk of burr on the exit side of the hole, but if you keep going this falls away and leaves a very clean hole.

The second thing I noticed happened when I was drilling the many holes for the speaker grill.  I drilled the holes to the first stop but allowed the next stop to just touch the entry hole to deburr the face of the amp and while this worked, it does dimple the hole a bit.  For this application the result is favorable, giving the grill a sort of “bullet hole” look, but I could see this being a problem if you are going for flat, smooth holes.

Also I tried using the bit on the reverse side of some of the holes to try and deburr them the way the entry holes, but this didn’t work all that well and I wouldn’t recommend it, at least not in light gauge material such as this.

Overall I think this bit is a great addition to the toolbox.  It was a bit more expensive than I would have expected at about $17.00 and I imagine the Cobalt version (mine was just high speed steel) is probably twice that, but since it would get a lot of use in hard material (and I have no idea how to sharpen it), I would recommend going with the Cobalt, as well as a model with more “steps”, if you can afford it.

Sun Shack Monday, Jan 12 2009 

Since the article in The Journal, I’ve had a lot of questions about the solar power system we put together for the cabin. Here’s a diagram of the current setup:

We started with the a 45-Watt “kit” from Harbor Freight, added a sealed deep-cycle battery from Wal-Mart, tested this for awhile, found that the kits charge controller wasn’t working right, replaced it with a much better but still inexpensive one, and then tweaked the wiring, etc.

The system performs well with plenty of room for improvement which is excellent. The controller that we selected has enough capacity for at least another battery of the cheap HF cells, and that coupled with a more appropriate battery would provide far more power than we could use at the moment. As I mentioned in the article the ultimate goal is to be able to run a microwave, but I believe this has less to do with our system’s capacity and more to do with the fact that the inexpensive inverter we’re using doesn’t produce a proper sine wave. There are inverters that remedy this, but they are very expensive. Instead, I have an idea for “improving” the wave produced by this inverter that, if it works, will be a lot less expensive.

Feel free to post questions, I’ve deliberately left out a lot of the details.

Fit to print Monday, Jan 5 2009 

A couple of updates:

I was interviewed for a “Maker” article that appeared in the 77Square section of the Wisconsin State Journal last week.  Always good to see the word getting out on the streets.

the workshop

Also, we decided to re-release DV Awareness to help raise some money for the shelter and to gauge interest in an “extended” version of the application.

On a final note, I’ve submitted the latest app “BodyPulse“, a compliment to our MindPulse application.  If it makes it through the approval process, I’ll post more details later.

New Year Thursday, Jan 1 2009 

Just submitted the latest app to iTunes, we’ll see how that goes.

I’m also bringing back DV Awareness, try and help generate a bit of revenue for Jamie’s shelter.

More to come…