Every single time I see it, I have to re-read all the text and analyze exactly what my options are.

They have not changed this dialog even in 2007 with new UI improvements. Is this dialog clear to everyone else? I doubt it.

So I’m eager to test it out and see how well it works. My guess is that they are using a Verizon EV-DO wireless card hooked up to a router (something like a femto-cell). I might do some tests, but I don’t expect to do much more than just IM and browse the web. Don’t want to abuse the connection

9:17 - And we’re off. It’s interesting to see downtown DC on a weekend morning. I barely ever get this view because I have never lived downtown, nor do I usually wake up early enough to go downtown. My brother’s talking to his Danish friend (see Chris I mentioned you!) and hopefully we got some good tips out of it.

9:22 - The internet is not up yet. I am writing this blog post in textmate so it’s not too much of a big deal. But I’m going to wait for a good 30-45 minutes before I cry foul. It could just be one of those things where it takes 45 minutes to boot up the equipment. ;-)

9:27 - Another dude opened up his laptop - he’s got a macbook pro. It’s awesome to see more people using macs.

9:36 - We just got the Wifi password. I’m glad it’s password protected (although it’s WEP but who’s going to be anal about wifi security on a bus?) We also just took a vote on what movie to watch, I voted for Top Gun :-) Needless to say, a lot of people are not happy that they were overruled on the Top Gun vote. I just have to say… I don’t like you, Maverick. I don’t like you because you’re dangerous.
, and all
9:40 - We’re on New York Ave. Hee Hee!

9:43 - Just realized that I didn’t charge the macbook last night. I’ve got an hour. Damn!!!

9:53 - There are 3 of us on the bus using the internet (that I can see), and all 3 are on macbooks! We are slowly becoming the majority in these kinds of settings. I’m sure PC notebooks are still overwhelmingly popular in the enterprise, but they have to be feeling the pain when it comes to home users.

However, you might encounter an issue when trying to serialize your object to XML, especially if you use a StringWriter to serialize your object to an XML string instead of a file. Since .NET strings are always stored in UTF-16, your resulting xml file will have the encoding of UTF-16. One way to get around this is by creating a MemoryStream, creating a StreamWriter, and applying the UTF-8 encoding to that StreamWriter.

The easy way of serializing an object to xml as a string (not to a file):

XmlSerializer serializer = new XmlSerializer(typeof(YourObject));
System.IO.StringWriter sWriter = new System.IO.StringWriter(); 

serializer.Serialize(sWriter, yourObjectInstance);
sWriter.Flush(); 

string result = sWriter.ToString();

But since this will result in an xml file with encoding UTF-16, we have the following block:

XmlSerializer serializer = new XmlSerializer(typeof(YourObject)); 

// create a MemoryStream here, we are just working
// exclusively in memory
System.IO.Stream stream = new System.IO.MemoryStream();

// The XmlTextWriter takes a stream and encoding
// as one of its constructors
System.Xml.XmlTextWriter xtWriter = new System.Xml.XmlTextWriter(stream, Encoding.UTF8); 

serializer.Serialize(xtWriter, yourObjectInstance); 

xtWriter.Flush();

// go back to the beginning of the Stream to read its contents
stream.Seek(0, System.IO.SeekOrigin.Begin); 

// read back the contents of the stream and supply the encoding
System.IO.StreamReader reader = new System.IO.StreamReader(stream, Encoding.UTF8);

string result = reader.ReadToEnd();

Remember, the MemoryStream is just a stream of bytes and doesn’t have to be text. So you create a StreamReader to read it back and have to tell the reader what encoding to use.

The idea of strings represented as bytes out in the wild, and various encodings, can be a little daunting for newer developers, but as long as you keep your encodings consistent, you should have no problems. And in the case that you need to convert to a different encoding, there are built-in libraries that do that for you.

• hosting environment exclusively for web applications
• initially free up to a certain site load (500 MB storage, 10 Gb bandwidth a day, ~5 million pageviews/month — I wish that was a problem I had!)
• supports Python (initially) but will eventually support more languages
• they seem to have their own api and their own web application framework, I wonder if they’ll support more like Django (Python) or, of course, Ruby on Rails :-)

How this compares to Amazon:

• This is not a general purpose web hosting or grid service like EC2 — this is tailored specifically to web applications
• Free up to a certain amount of usage (probably only something Google/MS could afford to offer)

My Thoughts

The main thing that struck me at first was that it was based off Python. Not a surprise because Google is pretty much the largest promoter/user of Python on the internet, they even hired Python creator Guido Von Rossum. But apparently from the talk, Python is only the first supported language. More will be added over time.

The other thing that struck me was that they seem to have chosen to develop their own web application framework. I wonder why. There are plenty of good ones out there (like Django for Python). I hope this isn’t just a case of programmers always wanting to implement their own version of a certain app just for problem-solving satisfcation and the joy of learning something new.

But again, if they allow other languages to be supported, I’m sure they’ll eventually allow various frameworks. If I had to guess, I’d think their framework probably handles caching and database persistence really well with their backend, which makes it the most ideal. But knowing Google, they will probably open up all the protocols being used to interact with their infrastructure, and allow the community to develop mappings in other languages. Nothing wrong with this inherently open source approach.

The final thing to note is that this is a preview release, and a lot of my worries will most likely be put to rest by the time they release a final version.

This is quite promising for developers. These are very interesting times for small companies. It is growing ever easier and cheaper to have an IT infrastructure that supports more and more users. There are startups that run entirely on Amazon S3 and EC2, and don’t own a single server. This is just possible because of the internet and a lot of its supporting technologies becoming cheaper.

Now that I think about it, it’s actually a very Apple-like approach to this problem. We give you the whole ecosystem where all the pieces work together perfectly, and as long as you use our tools and technologies, your app will work beautifully. It has limited appeal.

Now I’m waiting for Yahoo and Microsoft’s reponses :-) They better be working on something worthwhile to compete. This is the future of startups.

Overall, I think Amazon is still king of this space. This thing got a luke-warm reception from myself (and apparently most who attended). Hopefully it’ll mature into something more useful for a larger number of people.

So we decided to show attendees information that is immediately useful to them, such as the conference event schedule.SO this weekend, Michael Herndon from OSC, and myself, started to build what would become CID.

We decided from the beginning to open source the project and host it on Google Code, so that others can benefit and learn from our experience, and also so that it could be extended easily in the future. This series of posts outlines some of the things we learned in the process.

WPF

Firstly, a short intro to WPF. WPF is short for Windows Presentation Foundation, is a set of Microsoft-based technologies (namely, XAML and a .NET 3.0 Presentation Framework) that allow developers to build rich media applications which take advantage of graphics hardware that’s common in today’s computers. For a better overview of WPF, go to WPF.

Goals

There were three goals to this project:

• Bring people to the OSC Booth and make them talk about it
• Show off OSC’s skills in cutting edge technologies (like WPF)
• And most importantly, it gave us an excuse to play with WPF. ;-)

Tools we used

Not much of a surprise, but we used entirely Microsoft-based tools, since they were optimized for building WPF applications and writing XAML.- Visual Studio 2008 (with .NET 3.5)- Expression Blend 2 February CTP

The App

Click above for a screenshot of the final app (final as of this weekend). In the coming days, I’m going to outline how we did various things with CID and WPF. I want to cover the following topics:

• Part 2: Layouts
• Part 3: Databinding (Relative, Elements, DataContexts)
• Part 4: Animations, Per-character Text Effects, Calling Storyboards from C#
• Part 5: Code-behind events, programmatic animations and storyboards, and wrap-up 

So stick around! :-)

In a nutshell, it’s a kit to build your own electronic devices by piecing together blocks in a Mr. Potato Head style. You can then write code that controls how all these pieces talk to each other.

This is a very exciting new platform especially for programmers that don’t have the electronics know-how, but have a lot of ideas for creating their own gadgets. I had attempted my foray into electronics by researching AVR microcontroller kits, but I chickened out as soon as I read about BUG.

The guys behind this new startup seem to genuinely want to make the (geek’s) world a better place. Their mission is littered with good intentions and the execution seems top notch. I’m really excited to get my BUG and start playing with it.

My first idea is to build a fitness gadget. I’m going to use the accelerometer as a pedometer, and the GPS receiver to track my position. Hopefully this will result in a full-fledged treadmill-in-a-box.

I’d also like to use the accelerometer/pedometer to calculate the rate of my stride and maybe adjust the tempo of music according to it.

How cool would that be?

One is a Zone. A zone is like the “area” of interest. There are two zones involved in doing this:

• The location of your plugin (the dll files, etc)
• The location of the Word/Excel file being opened (*.doc, *.xls, etc)

These zones usually match up (if your plugin is installed on your computer, and you’re accessing an Office document on your hard drive, both locations are the “My Computer” zone).

For my situation, I had the plugin installed in my computer, but the file being accessed was over an untrusted network share, hence the file was placed in the “Internet” zone.

I had my assemblies in C:\Program Files\InstalledAppDir\*
My Word Doc was in: \\networkserver\sharename\something.doc

It triggered the following error:

The customization does not have the required permissions to execute.

************** Exception Text **************
System.Security.SecurityException: The customization does not have the required permissions to execute.
at Microsoft.VisualStudio.Tools.Applications.Runtime.DomainFactory.CreateDomain(Uri evidenceUri, String domainName, String codeBase, String configFileName, IAddinSecurityManager secman)
at Microsoft.VisualStudio.Tools.Applications.Runtime.AppDomainManagerInternal.CreateCustomizationDomainInternal(Uri uriFullDocumentDirectory, Uri uriFullDocFilePath, String documentName, IHostServiceProvider hostCallback, IAddinSecurityManager secman, AppManifest& applicationManifest, Boolean& manifestDirty, IAppInfo& appInfo)
at Microsoft.VisualStudio.Tools.Applications.Runtime.AppDomainManagerInternal.CreateCustomizationDomain(String applicationDomainPath, String documentName, IHostServiceProvider hostCallback, IExecuteCustomization& executor)
at Microsoft.VisualStudio.Tools.Applications.Runtime.AppDomainManager.CreateCustomizationDomain(String applicationDomainPath, String documentName, IHostServiceProvider hostCallback, IExecuteCustomization& executor)
The Zone of the assembly that failed was:
MyComputer

I followed this guide and tweaked the following policies a little and it now works:

MyComputer zone:
“FullTrust” permissions to the url: C:\Program Files\InstalledAppDir\*
Internet zone:
“Nothing” permission to the url: \\networkserver\sharename
“FullTrust” permissions to the OfficeDocumentMembershipCondition

Make sure you add the membership collection to the internet zone, even if you are accessing a network share. The easiest way to know the zone (and how I found out) is to look at your status bar when you go to the network share. It’ll tell you what security zone that path is seen as.

We’ll be creating a lot of compilation and linking across directories, so it’s good to keep it all contained in one place. So let’s start by creating a directory in the user’s home that’s going to contain all this.

mkdir ~/svncompile
cd ~/svncompile

Before you really get started, download the following files and extract them:

Subversion (all releases latest as of Sep 13, 2007)

curl -O http://subversion.tigris.org/downloads/subversion-1.4.5.tar.bz2
tar jxf subversion-1.4.5.tar.bz2

Apache Portable Runtime (apr):

curl -O http://www.mirrorgeek.com/apache.org/apr/apr-0.9.16.tar.gz
tar zxf apr-0.9.16.tar.gz

Apache Portable Runtime Utilities (apr-util):

curl -O http://www.mirrorgeek.com/apache.org/apr/apr-util-0.9.15.tar.gz
tar zxf apr-util-0.9.15.tar.gz

If you do an ls, you should have the following directories extracted:

drwxr-xr-x 44 ashish ashish 1496 Sep 3 21:43 apr-0.9.16
drwxr-xr-x 34 ashish ashish 1156 Sep 3 21:46 apr-util-0.9.15
drwxr-xr-x 29 ashish ashish 986 Aug 14 06:18 subversion-1.4.5

Next, we’re going to compile the dependency libraries, apr and apr-util. We configure, then make, and finally make install:

cd apr-0.9.16
./configure
make
sudo make install
cd ..

apr-util depends on apr, so when we configure apr-util, we need to point it to apr:

cd apr-util-0.9.15
./configure --with-apr=../apr-0.9.16
make
sudo make install
cd ..

Using –with-apr tells the utils to use that instance of apr located in the directory we just built.

Building SVN

And finally, we can build svn. :-)

First, we go to svn and run configure, but we point it to apr and apr-utils:

cd subversion-1.4.5
./configure --with-apr=../apr-0.9.16 --with-apr-util=../apr-util-0.9.15

Subversion might tell you that you don’t have BerkeleyDB installed. I’ll have to modify this section for that.

Now build and install subversion:

make
sudo make install

And that’s it! Assuming all went well, you should be able to type “svn” and get an svn usage message.

Cleanup

There are two cleanup options, depending on what you want to do. If you just want to get rid of all of the source, the easy thing to do is:

rm -rf ~/svncompile

And we’re done. If you want to hold on to the source, I always prefer keeping the source directory clean because that saves space if you’re not actively developing and also keeps your directories, well, clean. For that, just run make clean on each of those projects:

cd apr-0.9.16
make clean
cd ../apr-util-0.9.15
make clean
cd ../subversion-1.4.5
make clean

And you have a clean source repo. :-)

To put it all together, here are all the commands you need to fully install Subversion 1.4.5 on Mac OS X 10.4 (Tiger):

mkdir ~/svncompile
cd ~/svncompile
curl -O http://subversion.tigris.org/downloads/subversion-1.4.5.tar.bz2
tar jxf subversion-1.4.5.tar.bz2
curl -O http://www.mirrorgeek.com/apache.org/apr/apr-0.9.16.tar.gz
tar zxf apr-0.9.16.tar.gz
curl -O http://www.mirrorgeek.com/apache.org/apr/apr-util-0.9.15.tar.gz
tar zxf apr-util-0.9.15.tar.gz
cd apr-0.9.16
./configure
make
sudo make install
cd ..
cd apr-util-0.9.15
./configure --with-apr=../apr-0.9.16
make
sudo make install
cd ..
cd subversion-1.4.5
./configure --with-apr=../apr-0.9.16 --with-apr-util=../apr-util-0.9.15
make
sudo make install

Hope this helps!! Leave me a comment if you have any problems following the guide.
TODO: mod_dav_svn and BerkeleyDB support.

Check out our entry, Fish4Brains: http://takemeaway.railsrumble.com and vote for us!

I’ll have more details about this project, the underlying technologies, and stories from working on a 2-day project.

Some might read this and think “Ok, that happens to me all the time.” If it does happen to you often, consider yourself lucky.

The History

I have been writing a particular piece of software for about 6 months, and we are finally planning to expose this software to a subset of its prospective users for Acceptance and User-Interface testing.

For the 6 months that I’ve developed it, I have been the only tester, and the only person demonstrating it. I knew all along that at some point, I would have to unleash it on others and get a new set of eyeballs to really expose bugs, but it wasn’t at a stage where I was comfortable enough to do so.

But this past week, I proudly proclaimed in a meeting “yes, we can start user testing!” It resulted in quite a reaction from the room, almost one of disbelief. So we decided to have the user testing a week from that meeting, and have an internal dry run with our staff on Friday.

The Dry Run

Ah, the dry run. How to describe it? It was successful. They destroyed my program in minutes. Why is this success? Because that’s the point. The point is to find flaws before you show it off to the world. But we still have the main testing scheduled for next week. So I decided to work parts of the weekend to fix as much of the code as possible to have it running in a more stable manner.

The Team

What was extraordinary about this was the amount of help the team offered. If it was the type of team I am more used to working in, they would’ve casually offered help and then not done anything, especially since I was (until recently) the solo writer of the software. But this team spent most of their day helping work out kinks in my software and the overall environment in which it executes, plus they even offered to be on standby for the Labor Day weekend just to test software.

I can’t be too specific but I even got offers from colleagues to do some of the work that I would normally be tasked to do. And I gladly took them up on some of the offers. I understand that they are all doing it in the capacity of a job, but it was just something I had never experienced before, which makes me truly value the team.

I’ve been working in IT for about 8 years now. For most of those 8 years, I’ve worked by myself as a consultant, and handled all aspects of a project. Part of this was because I had worked with bad teams and had built up a distrust of working in teams in general. This was until I started working in a team full of totally competent individuals, who seemingly have zero issues of politics, who all get along, and who all seem to share an ethic of pulling together for a common cause — truly “taking one for the team.”

This experience, though small, definitely made an impression on me, and changed my views on generally working in teams. Yes, it was that easy. I know now what the true value of the cliché of a “team player” is.