1. No code completion
2. No auto formatting of code
3. No built in testing tools! <– this is the most important

The list goes on actually.  While Eclipse has its own share of problems, it is the best supported free IDE.  Eron Villarreal and I struggled to get the official Processing Eclipse plugin working in Eclipse Juno, but found the UI to be confusing and, worst, you can’t run your sketches from within Eclipse!  WTF, mate?

Luckily the Proclipsing project works great.

If you’re new to Eclipse, you’re going to have to be a little patient, though believe me it’s worth it compared to the Processing IDE*.  Eclipse is like the streets of old Boston; it must have made sense at some point when it was originally constructed.  For some reason to add plugins, you need to go to the Help menu.  To configure those same plugins, you need to go to the Window menu. I dunno.

Anyway, the Proclipsing getting started instructions are pretty good.

We found that for an existing project, it’s better to create a new Proclipsing Processing Project and copy-paste your code into the file generated by Proclipsing.

Two other notes:

1. Get used to right clicking on your project in Eclipse and hitting the refresh button.  We found that solved a lot of errors.
   
2. If you are using plugins for Processing (like ControlP5 for OpenBrewery), copy the .jar file out of the plugin zip file and drop it into your project’s lib directory. Then right click on the project in Eclipse, select Properties, select Java build path, open the libraries tab, and click the Add Jar button. See the picture below.
   

* I’m bashing Processing, but it’s a great application that makes it easy to get started using Processing.  I think that’s the way it should be– make it easy to get started, and if a project requires more professional tooling, switch once you are off the ground.

Saturday, May 18th 4:30pm, Electronics Stage

Sunday, May 19th 2:30pm, Electronics Stage

Where’s the electronic’s stage?

Oh, and the project now has a name: OpenBrewery.org.

I’ve created a BitBucket repo for a project which fixes this limitation and adds support for additional Arduino boards.  So far I’ve only tried the analog pins on the Mega and Teensy.  Just drop the files into your Processing project.  One additional argument, the board type, is required to create an Arduino object.

/*
 This example code is in the public domain.
 */

// the setup routine runs once when you press reset:
void setup() {
  // initialize serial communication at 9600 bits per second:
  Serial.begin(9600);
}

// the loop routine runs over and over again forever:
void loop() {
  // print out the value you read:
  for(int i=0; i < 16; i++) {
    Serial.print("A");
    Serial.print(i);
    Serial.print(": ");
    Serial.println(analogRead(i));
  }
  delay(5000);        // delay in between reads for stability
}

Eric had just gotten his Master Brewing certification after completing the Master Brewing Program at UC Davis, coincidentally where I got my Computer Science and Engineering degree years earlier.  We’d both been home brewing for years.  Eric had gotten the ball rolling with a Mr Beer kit while we were both undergrads.  The beer matured under his parent’s kitchen sink and it tasted awful, so we gave up.  In December 2005, the owners of Rabbit’s Foot Meadery in Sunnyvale, California let me sample their delicious brew and urged me to take up brewing again.  Ben Krasnow (see thirst extinguisher link here), Sheila and I signed up for the free Monday night class taught by Griz at SF Brewcraft in San Francisco’s Outer Richmond.  We called ourselves “SPB Brewery” after our first initials.  We brewed five batches before Eric and his brother also attended the class.  By the time we reached batch 7, Eric was past 10 batches.  When we reached 10, he’d lost count and had moved on to 10 gallon equipment while we were still making 4 or 5 gallons at a time.  This man was serious about beer.

SPB’s first beer label, named after the earwig we found that had somehow gotten into the beer

So I wasn’t too surprised when Eric signed up for the Master Brewers Program, although I did wonder what he would do with the certification.  Brewers are generally not well paid, and we both have more lucrative day jobs.

When Eric came back we ended up discussing how to make a good beer consistently.  The first beer out of Ben’s thirst extinguisher had been a crowd favorite, but we’d never been able to make it again, even though we followed the same recipe.  Repeatability comes down to control over the brewing process.  It’s not enough to start with the same ingredients– you need to add them in the right quantities at the right times and maintain the same conditions.  One of the reasons for moving to 10 gallon batches versus 4 or 5 (besides lots more beer to drink) is that small variations in a small batch have significantly greater side effects.  If you add a few extra pellets of hops to a 10 or 20 gallon batch, or the room temperature fluctuates, the effect is not as great as in a batch that is only a few gallons.  With larger batches, consistency is easier.

It would cost about $150,000 for the smallest “proper” brewery.  A brewery is sort of like a restaurant in that you need a menu worked out before you throw open the doors to customers.  Usually, brewers work for another brewery, creating their menu on the side before opening up their own operation. Well, that’s fine for a 23 year old fresh out of college, but we’d have to come up with something else.

This lead to the question: What would be the minimum equipment needed to make the same great beer over and over again?

If we could answer this question and build it, Eric could fine tune his recipes while working his day job.  We decided it was time to start meeting weekly.

We sketched out the major parts of a brewery and went over the process.  A brewery is quite a bit more complex than a homebrew setup, which is often just a large pot on a stove and a large glass jug to mature the beer in.  It became clear that the major impediment to operating the brewery was that it takes just about as much time to brew 10 gallons as it does to brew 1000– about six to eight hours.  There are a handful of steps during those hours, meaning someone has to be there to babysit the brew during the entire duration.  If you rush home from work to kick off a brew at 5pm you’ll still be there until 11pm to 1am.  Not really ideal.

No, we needed to be able to start brewing in the morning and the brewery would have to run all by itself while Eric was at work.  He could then cleanup once he got home.

We needed automation.  Luckily, for my day job I often test manufacturing equipment and I’d just built an automated beer serving machine called the Bierlitzer.  Between the two of us we should be able to pull this off, right? I drew up a project plan and estimated we should be done by April 2012 if everything went according to plan.  It didn’t.

To be continued…

While doing some project management research, I came across this more fancy algorithm, reproduced below.

PERT (Program Evaluation and Review Technique)

PERT is a variation on Critical Path Analysis that takes a slightly more skeptical view of time estimates made for each project stage. To use it, estimate the shortest possible time each activity will take, the most likely length of time, and the longest time that might be taken if the activity takes longer than expected.

Use the formula below to calculate the time to use for each project stage:

shortest time + 4 x likely time + longest time
———————————————————–
6

This helps to bias time estimates away from the unrealistically short time-scales normally assumed.