——

9/5/08

We have arrived at Agile Summer Camp.  After claiming a bottom bunk in the rear of the cabin, I am sitting in my camp chair enjoying a beer while Brandon Joyce, funky town dancer extraordinaire, cranks the camp radio so we can listen to Chef (RIP) croon on a 70s station.  Sadukie is dead-tree blogging and Woody and Matt Werstler have gone for firewood.  We are anxiously awaiting the arrival of Mike Eaton who has recently called and is nearby. 

Suddenly, a car arrives.  Could this be Mr. Eaton?  Or a guy with a chainsaw (I hope he has bud light!).  It turns out to be Josh Holmes (who looks like he could have a chainsaw…).  He is followed shortly by Eaton, and then others.

There was a guest log in the back of the cabin packet that contained a riddle, as people arrived we asked for them to take it on, but no one could come up with an answer, see if you can:

We buried the loot where the fish used to swim

40 paces westerly of monitor, third stopper (not for bathtub) from the left.  one foot down right side.

1st Clue:  Are you good enough?

-John from Beaverton May 26, 2008

The evening ends with awesome discussions on community and a few already on the topic of agile.  Also me and Mike Eaton finish a fifth and a pint of Jack ourselves (okay Steve Andrews helped a little).

——

9/6/08

CLANK! CLING! CLANG!  I am roused from my fitful sleep by the sharp cry of metal to metal.  I change in my sleeping bag and make my way out of the cabin to find turkey sausage and eggs being cooked on the, now assembled, camp stove….

——

The rest of the journal devolves into generic notes, most of which can be found here (along with a lot more detail on the actual sessions).

Needless to say it was very educational and loads of fun.

Michael  Moore was a wonderful host and did an amazing job editing the conversation down to something tight.  It was an exciting opportunity to talk about IronRuby specifically and Ruby in general.  Hopefully it will help spark even more interest in it.

On a personal note, I’m interested in any feedback you might have, from the technical to the stylistic. It was my first time doing a podcast, so it’s definitely a learning experience for me.

Wednesday evening around 7:30, my wife dropped me off at Sadukie’s house. Brandon Joyce was already there, watching Family Guy.  After Sadukie finished gathering up her wares, we piled into her Durango and headed off to pick up Corey Haines. A quick trip across town and we were on our way to Toledo to catch the bus.

Well, more accurately we were on our way to Toledo to wait for the bus.  The bus apparently went to the wrong side of town and it had to turn around to pick us up.  This wasn’t entirely bad.  It gave us enough time to pick up beer and for Corey to have an impromptu parking lot concert. And what’s better then beer and music?

Eventually the bus arrived and we REALLY began our journey.  A journey of corny movies and lack of sleep.  After much tossing and turning and cursing of lack of leg room, we arrived in Tennessee about 9am.  At which point I collapsed in my hotel room and passed out. Waking up in the evening, I followed my instinct to the local hotel bar/restaurant. Later on, I ended up making my way to Alan’s room where a party was going down.  Guitars, Cigars, and Beers were being passed around with abandon until hotel security came to the door and asked us nicely to stop.

The rest of the time in Tennessee is really a blur.  I wish I could give a detailed synopsis of each day, replete with details and names.  But honestly, the days meld into one another, linked only by great conversations, new experiences, and wonderful people.  I can’t stress enough the amount of knowledge I gained, and the success of the Open Space format that was put into place.

Alas, it was over much too soon.  By Saturday night it was time to board the bus once again, and make the long, movie filled trip back to Ohio… Who knows, maybe this time I could get some sleep? I cuddled up to the window and slowly drifted…

“THUMP, THUMP, THUMP, THUMP!”  Okay, so sleep wasn’t going to happen.

A blown tire. Great, James Avery’s ribbing about “Everyone was screaming… except Cliff” went through my head.  It didn’t help that we passed a sign commemorating a 1988 bus tragedy (no joke). The bus slowly inched along, until we pulled into a gas station in English, Kentucky.  Our faces aglow in the soft light of nearby Waffle House.  We waited.  And Waited.  Cooter showed up, and proceeded to sheer three bolts off the wheel.  So we continued to wait.  But while we waited, a remarkable thing happened.  Incredible conversations took place.  On Comments, on SRP, and any number of things.  The Open Spaces continued into the most open space possible, the world beyond the conference. Then the acetylene torch came out.

So we waited a bit more.  Marriages came and went we waited so long. Eventually a new tire was on and we were off.  By 1pm I was home holding my daughter and reflecting on a fantastic conference.  I was thankful I ended up going, though some images will be seared into my brain, forever.

I’ll at least be able to say I shared a tour bus with a member of Ratt.

The best part is that the conversation continues at the Heartland Open Spaces Wiki. Be sure to stop by and join in!

Next, I’ll be at Agile Summer Camp, and then I’m home until CodeMash in January 2009.  To all my new friends, old friends, and friends to be, I hope to see you there!

There were a few test results that really stood out.  The lists test was by far the worst for IR, where it was 56 times as slow as MRI and 3 times as slow as JRuby.  But this wasn’t the only sore spot, there were other places that show room for improvement.  Namely, vm1_blocks in which IronRuby was almost 9 times as slow as MRI.  Also so_exception where IronRuby brought up the tail by being 10 times as slow as MRI.

The nitty gritty is that IronRuby is about twice as slow as MRI with JRuby being slightly faster, overall.  As the below graph illustrates:

There are a few high points in this test though.  Number one, is that IronRuby ran the majority of the tests.  There were only four failures out of forty, for a 90% success rate.  For comparison JRuby had 95% success rate. Number two, there were actually a few tests in which IronRuby beat JRuby and MRI.  No small feat for such a young implementation.

I think this mini-shootout illustrates that there is a lot of room for improvement for IronRuby.  However, a little perspective goes a long way.  JRuby was announced in 2001, IronRuby was announced last April.  The JRuby team has had 7 years to get to where they are today, whereas IronRuby has only had almost a year and a half! As the DLR improves IronRuby will receive those performance gains for free, and as compatibility becomes more stable the core team can focus on performance as well. The raw data can be downloaded here.

What a fantastic conference this was two weeks ago.  A great lineup of speakers, and fantastic after event gatherings.  And I missed the first day!  Josh Holmes has a great series of posts detailing the event, so I won’t repeat the play by play here.  But I can say that I will definitely be making the trip to Columbus for eRubyCon 2009. 

This was the first conference I have gone to where I actually presented.  My talk was originally geared more towards .NET developers interested in Ruby, thinking that there would be more Java and .NET enterprise folks there then it turned out.  After I arrived, as I listened to the attendees, I realized the audience was primarily people already doing Ruby.  I also noticed a certain level of distrust around Microsoft in general, even though the event was hosted in their office!  As I was going to be speaking on IronRuby, a Microsoft project, this concerned me a bit.  So I ended up putting in an opening section about some of, what I believe to be, the positive changes at Microsoft over the last half-decade or so.  I highlighted efforts like CodePlex and Port25.  A number of people seemed to copy down the Port25 address, so I hope I helped open some eyes in this case.  I also went over the current state of IronRuby, and compared it’s performance to JRuby.  I then asked who did .NET work in their day to day jobs and must have had about 5 people raise their hands.  Then I went through a short demo of opening up and changing a standard WinForms app using IronRuby. It seemed to go well, I’ve gotten positive feedback on it and a number of suggestions for improvement.

I can’t thank Joe O’Brien enough for the opportunity or confidence in me he showed by allowing me to speak.  The slides and demo used for my presentation are here.

The first episodes are based at the MVP summit that took place in April.  From a discussion on what interface a Chicken should implement to lay eggs (my vote is for IEggLayer) to extolling the Power of Twitter, each episode has been filled with warm humor and good information.  The passion they, and their guests, have for technology really comes shining through.  Leaving you with a contented, full feeling, similar to eating a good bucket of wings fresh from the deep fryer.

The production values are surprisingly high for such a new podcast.  Even though the podcast is only 3 episodes old (really two, the intro one almost doesn’t count ;) ), you would think these guys have at LEAST 10 episodes under their belts, minimum.  And there has been steady improvement. In the first episode or two, Woody sounded much quieter then Keith, by the third episode that’s been resolved.  The new intro from Elly Mae is top notch, leading you in that this is a serious (but not too serious) affair. The last episode shows a marked improvement in over all production (though you can tell they are still playing with editing, in the beginning there are parts where it sounds "clippy").  If better is yet to come, you’ll find me there!  And now that they have a sponsor, the future is looking bright.

So come on down, take yer shoes off, and pop a squat with the Paula Deens of Podcasting!

Links

• Deep Fried Bytes Homepage
• Deep Fried Bytes on iTunes
• Deep Fried Bytes on Zune Market Place
• Deep Fried Byte’s Twitter
• Keith’s Blog
• Woody’s Blog

Hope to see you there!

Getting The Source

As before, I recommend using TortoiseSVN to grab the source. Downloading will take a bit, after the 3.5MB of source is downloaded you should end up with a directory structure like the following:

Compiling IronRuby

Inside of the trunk directory you will find the IronRuby.sln file, double-clicking it will open the solution (note that the solution file is now in VS2008 format instead of the previous VS2005 format).  You may get a warning, you can select “Load the Project Normally” and uncheck “Ask me this for all projects in the solution.”

Once the solution is loaded, keep the solution on the “Debug” configuration.  Do NOT select “ExternalDebug” (you’ll get broken references if you do):

Next we will manually remove the “SIGNED” compilation symbol from all the projects, as of Rev. 113 you have to manually remove it from the IronRuby.Libraries, Microsoft.Scripting, and Microsoft.Scripting.Core (be sure not to delete the DLR symbol) projects.

Once those are removed you can build the source!

Running That Which You Have Wrought

Where has previously our compilation would output into a bin\Debug folder, the resulting files are now found in \trunk\build\debug, which should look like this:

Due to a conflict with rubinius, the rbx executable is now ir.

As before you can just run ir.exe and enter the wonderful world of ruby, or you can continue on to see how we roll in the .NET world.

Speaking C# With A Ruby Accent

The initial steps are basically the same as before, start a new C# console project and add references to Microsoft.Scripting.dll, Microsoft.Scripting.Core.dll, IronRuby.dll, and IronRuby.Libraries.dll.

We are going to create a simple console app that shows passing variables into and out of a ruby script.

The below program should be a good example:

using System;

using Ruby;

using Microsoft.Scripting.Hosting;
namespace IronRubyConsoleApp

{

    class Program

    {

        static void Main(string[] args)

        {
ScriptRuntime irruntime = IronRuby.CreateRuntime();

            ScriptEngine ir = IronRuby.GetEngine(irruntime);

            ScriptScope scope = ir.CreateScope();

            ScriptSource script = ir.CreateScriptSourceFromString(“puts \”Hello, #{name}!\”\ninput + 2″);

            scope.SetVariable(“name”, “.NET”);

            scope.SetVariable(“input”, 2);

            int x = script.Execute<int>(scope);

            Console.WriteLine(string.Format(“The Result was {0}”, x));

            Console.ReadLine();

        }

    }

}

You can see a clear hierarchy here:

We have a Runtime

That has an Engine

That has a Scope

That has a Source

We create a runtime, and we get our IronRuby engine into the runtime, we create a scope and load a script using that engine.  Set a couple of variables in the scope and then execute the script within the scope that we have set the variables in.

Hopefully this is enough to get you started, you should probably check out the other CreateScriptSource methods that the engine contains, you have a veritable cornucopia of options:

Ruby with a .NET Accent

Another of the popular activities with IronRuby, if not the most popular, is going to be interfacing with both the .NET framework and other .NET code.  The below example shows us interfacing with System.Windows.Forms and making a simple GUI app.

require ‘mscorlib’

require ‘ System.Windows.Forms, Version=2.0.0.0, Culture=neutral, PublicKeyToken=b77a5c561934e089′

require ‘System.Drawing, Version=2.0.0.0, Culture=neutral, PublicKeyToken=b03f5f7f11d50a3a’

Swf=System::Windows::Forms

Sd=System::Drawing

class RubyForm <Swf::Form

def add_button text, location

button = Swf::Button.new

button.Text = text

button.Location = location

self.Controls.Add button

button

end

def initialize

self.Text = “RubyForm”

@rbutton = add_button “Click Me!”,  Sd::Point.new(150, 150)

@rbutton2 = add_button “Click Me!”, Sd::Point.new(150, 100)

@rbutton.Click {|sender, e| Swf::MessageBox.show ‘Hello World!’}

@rbutton2.Click {|sender, e| Swf::MessageBox.show ‘Hello, .NET!’}

end

end

rf = RubyForm.new

rf.ShowDialog

As you can see above, when requiring items from the GAC we must include the Fully Qualified Name, including version and StrongNameToken if applicable. The above code inherits from System.Windows.Forms.Form and adds a helper class for adding buttons and wires up a couple of event handlers.  Again this is just a simple example to get you started.

Summary

In this post we downloaded the IronRuby source, compiled it, and worked with IronRuby both from C# and by running a ruby script against .NET objects.  Hopefully this gets you going a little faster and onto the fun stuff quicker!

Links

IronRuby Homepage

IronRuby RubyForge Project

Ruby Language Homepage

John Lam’s homepage

Michael Eaton started it, and others took off and ran with it.  Now it’s a certified meme. So I guess I might as well join in the fray and relate the beginnings of my geekdom.

A Long, Long Time Ago…

How old were you when you started programming?

I was about 7ish when I first started. Read on for more details…

How did you you get started in programming?

When I was around 7, my uncle had an Atari 400.  I initially just played games on it, but one day I noticed a stack of Byte magazines in the corner.  Inside this magazine was source code, it said it was “BASIC” and needed to be typed in… hmm.. there’s this Atari-BASIC cartridge, and there’s a keyboard.  A light went off, I can make this machine /do/ something?  So I typed in the program, saved it to the cassette drive, and then tweaked it.

Next thing I know I’m begging my parent’s for a computer.  For my 8th birthday I got an Atari 65XE and a couple of BASIC games books, and I’ve been totally hooked ever since.  Time passed, and eventually I got into things like QBASIC, DOS Batch, and C/C++ (via an old copy of TurboC a friend loaned me).  Then in high school someone gave me a Linux CD and having a whole slew of development tools really helped.

I just kept hacking away…

What’s the most fun you’ve ever had … programming?

Staying awake for 48 hours straight and hacking an ANSI art viewer in C with Jim Balcomb.  We were probably 15 and 16 at the time and just didn’t need to rest when programming was to be had.  I think it’d kill me today.  Oh yeah, and then I delivered papers in the morning after being jacked up on coffee and Jolt and my father was convinced I was on speed…

Googoo gah gah?

What was your first language?

Well it was BASIC obviously, from Atari-BASIC, to BASICA, to QBasic, eventually to QuickBasic…

What was the first real program you wrote?

One that really stands out was a grading program I wrote in high school for the Learning department at the school.  It was all done in QBasic and had menus that were keyboard driven, grid data entry, and a printing subsystem.  I wish I still had that code…

What languages have you used since you started programming?

Define use? Here are all the languages I have written something of some complexity in:

• AutoIT (don’t ask…)
• Bash Shell
• BASIC (variations there of)
• C/C++
• C#
• DOS Batch
• JavaScript
• PHP
• Ruby
• TSQL (don’t ask)

..and [Insert Deity Here] knows what else…

Get a good job with more pay and you’re okay.

What was your first programming gig?

I’ve always loved programming, and have been doing it has a hobbyist since as far back as I can remember.  However, without experience (and never having finished college) I was never able to really get a foot in the door.  I worked in support at Harley-Davidson Dealer Systems for 6 years and had a chance to do some development work (I had written a number of utilities for support including SetPrinter) I jumped at the opportunity.  I actually took a “demotion” from Team Lead to Developer for the chance.

After a couple years of doing development work there I had pretty much tapped out all I could do and was stuck, that’s when I was introduced to PreEmptive Solutions and am currently doing development there.  My work at PreEmptive is much more like a “real” development job then HDDS.  At HDDS I was pretty much a cowboy left on my own.  Here I’m working with a true development team.

If you knew then what you know now, would you have started programming?

Oh hell yes, I truly believe I was born to be a computer programmer, I can imagine doing nothing else.

If there is one thing you learned along the way that you would tell new developers, what would it be?

RTFM.

Thanks Mike!

A big thank you to Mike Eaton for getting the ball rolling on this, it’s fascinating to see how my fellow tweeps and hackers got started.  I’m not going to “tag” anyone, because that’s silly, but I do encourage anyone who reads this post to do the same and link back to mike’s blog, he’s keeping a list of people responding!

$  history 1000 | awk '{a[$2]++}END{for(i in a){print a[i] " " i}}' | sort -rn
| head
114 cd
84 ls
79 irb
39 git
31 pedump
10 ps
10 objdump
10 e
8 ssh
8 dd

You can see a bit of my dual developerness there, irb and git for ruby, and pedump and e for the windows side (this is in cygwin).  What does your shell history have to say?