One bit of feedback we frequently get is that PowerShell's learning curve has some steeper bumps than we would like. Or simply, is strongly affected by habits learned from other languages or shells.

Interestingly enough, many of these problems aren't new to us -- we just don't have a good way (aside from help) of exposing them to the user. This was something I touched on in the footnotes of a blog in 2005, and started implementing personally shortly after that. Here's an example of its output:

[C:\temp]
PS:14 > "c:\temp\has space\test.ps1"
c:\temp\has space\test.ps1

Suggestion: Did you mean to run the command in quotes?  If so, try using & "<command>"

The core of this is implemented by a script, Get-TrainingSuggestion.ps1. It retrieves the last item from your history, and runs a bunch of regular expression comparisons against it. If it finds a match in the list of training rules, it outputs the suggestion that corresponds to that rule.

##############################################################################
## Get-TrainingSuggestion.ps1
## Retrieve a training suggestion (if applicable) for the last command
## you typed.
##############################################################################

$history = get-history -Count 1

$suggestions = @{
   "/\?"="To get help on a topic, use -? instead of /?.`nAlternatively, use get-help <command>.";
   ".length"="Try using the .Count property instead of the .Length property.  Although .Length " +
       "often works, .Count is the more consistent way to get the number of objects in a collection.";
   ";[ \t]*$"="Semicolons are not required as line terminators in PowerShell.  Try your command without one.";
   "Regex.*]::Match"="PowerShell's -match evaluator is much more efficent than calling it through the .NET API.";
   "^`".*`"$"="Did you mean to run the command in quotes?  If so, try using & `"<command>`"";
   "start "="Are you trying to run the program associated with that path?  If so, try invoke-item <path>";
   "%.*%"="To access environment variables, try `"`$env:variable`" instead of `"%variable%`"";
   "ren[^ ]* [^ ]+ .*:\.*"="Rename-item takes only a name as its second argument.  Unless you want the " +
       "name to have those path characters, do not include them.";
   "dir.*/s.*"="To get a recursive directory listing, try `"dir . * -rec`".";
   "dir.*/ad"="To get a list of directories, try `"dir | where { `$_.PsIsContainer }`""
   "^get-childitem.*/s.*$"="To get a recursive directory listing, try `"get-childitem . * -rec`".";
   "^grep"="To search files for text patterns, use the Select-String cmdlet.";
}

$helpMatches = ""

if($history)
{
   $lastCommand = $history.CommandLine

   ## Get the suggestions from the baked list
   foreach($key in $suggestions.Keys)
   {
    if($lastCommand -match $key)
    {
        $helpMatches += "`nSuggestion: " + $suggestions[$key]
    }
   }
}

$helpMatches

To use this, simply update your PROMPT function to call Get-TrainingSuggestion.ps1.

function prompt
{
    $suggestion = Get-TrainingSuggestion
    if($suggestion)
    {
        Write-Host $suggestion
        Write-Host
    }

    "PS >"
}

One tension that sometimes arises in the PowerShell community is between the hard-core developers, and hair-on-fire administrators. As a broad technology, this tension and desire for balance drives our design decisions every day.

PowerShell is all about striking a balance between a range of experiences:

• Task oriented, and consumption-only. AKA “Copy and Paste Admin.”
• How do I use PowerShell to restart a computer?
• Problem oriented, and composition-based. AKA “Admin Scripter.”
• How do I collect a log file from a cluster of machines I need to parse out of a web page?
• Technology oriented, and interoperability-focused. AKA “Developer” / ISV
• How do I expose my product’s transacted object model to PowerShell users?

In general, the output of each category empowers the categories that come before it. As a product becomes PowerShell-enabled, admin scripters can then leverage the cmdlets and providers to solve their problems and scenarios. They might blog them, wrap them in a script or module, or otherwise share them. At that point, they become pre-packaged tasks ready for copy and paste.

We tend to walk a fine line with this goal, but I’ve grouped our 100+ new V2 cmdlets into three buckets: Admin, Developer, and Both. Here is what the comparison looks like right now:

PS >gc newcommands.txt | group { $_.Substring(0,1) }
Count Name                      Group
----- ----                      -----
   75 a                         {a Add-BitsFile, a Add-Module, a Checkpoint
   11 d                         {d Add-Type, d Debug-Process, d Export-Prox
   16 b                         {b Complete-PSTransaction, b ConvertFrom-St

Granted, we’ve spilled more blogging ink on the developer and scripting-oriented features (transactions, modules, eventing, etc,) but that’s largely because it will take some time for our SDK and help documentation to catch up.

In addition, PowerShell’s goal is always to soften the learning curve and blur the distinction between the buckets I mentioned earlier. For example, the scripts and functions you’ve come to know and love now have access to all of the power of compiled cmdlets. Or more simply, some scenarios that used to require .NET scripting (random numbers, split and join, etc) now have task-oriented cmdlets or language features.

Now, what about "constructive feedback?" Frequently, we'll find a blog post or comment saying something along the lines of, "PowerShell just isn't admin-friendly." Or, "PowerShell just isn't developer-friendly." Of course, we're always working to make the product better, but that comment is an ideal spot to help ensure we get this right. The best way to help us is to make your feedback constructive, and something we can act on. Tell us why you feel this way, and it has a chance of improving the product. Aimlessly grumble, and it does not.

... Quitting lasts Forever.                     ~ Lance Armstrong

I hadn't heard this before yesterday, but it makes a good quote for the whiteboard!

Once problem that often arises when trying to manage machines is when the management layer itself is the thing you need to diagnose. For example, trying to diagnose Remote Desktop connectivity issues when port 3389 is blocked, or using PowerShell Remoting when WSMAN is misconfigured.

Alternatively, you might not have the client you need to manage the machine -- such as an SSH client, or a version of PowerShell V2 installed.

One way to get around both of these problems is another communication channel. It may be of lower fidelity, but can help you get your job done.

One perfect example of an alternative communication channel is any of the many synchronization tools out there: Live Mesh, Syncplicity, FolderShare, etc. In addition to managing connectivity, they let you broadcast messages (by way of files) between connected computers. Let's use that as our communication protocol:

The core of this experience comes from a PowerShell Mesh Monitor. This script monitors a computer-specific directory for new scripts that appear. When it sees a new script, it processes it, captures the output, and then stores the output in a similarly named file. You can then retrieve the file, review its contents, and continue to manage the system this way:

#requires -version 2.0

## Assuming the script is located in the (Mesh-synched folder)
##    D:\Documents\Mesh
## and the computer name is
##     "MESH-CLIENT"
## We'll monitor D:\Documents\Mesh\Computers\MESH-CLIENT for PowerShell
## scripts to run.
##
## Run this script on any machines that you want listening for scriptable
## Mesh management commands:
##
## powershell -noprofile -file D:\Documents\Mesh\Start-MeshMonitor.ps1
##
## Commands that impact the shell's state (such as current location) are
## persisted for future commands. If you want to share data between scripts
## use global variables.

## Set up the environment
$host.UI.RawUI.WindowTitle = "PowerShell Mesh Monitor"
$myWindowHandle = (Get-Process -Id $pid).MainWindowHandle

## Determine the monitoring directory. Create it if we have to.
$scriptLocation = Split-Path -Parent $myInvocation.MyCommand.Definition
$monitorPath = Join-Path $scriptLocation "Computers\$($env:COMPUTERNAME)"
if(-not (Test-Path $monitorPath))
{
    [void] (New-Item -Type Directory $monitorPath -Force)
}

## Create a P/Invoke type that gives us access to the window
## management functions (show / hide)
$windowUtils = Add-Type -memberDefinition @"
[DllImport("user32.dll")]
public static extern bool ShowWindow(IntPtr hWnd, int nCmdShow);
"@ -name "Win32ShowWindowAsync" -namespace Win32Functions -passThru

## Hide ourselves
[void] $windowUtils::ShowWindow($myWindowHandle, 0)

Set-Location $monitorPath

## Continuously go through all PowerShell scripts in the drop location,
## run them, and store their output back in the drop location.
while($true)
{
    ## Show the window if they create a file called "SHOW"
    if(Test-Path SHOW)
    {
        [void] $windowUtils::ShowWindow($myWindowHandle, 1)
        Remove-Item SHOW
    }
    
    foreach($inputFile in Get-ChildItem $monitorPath -Filter *.ps1 | Sort LastWriteTime)
    {
        $script = $inputFile.FullName
        "Processing input script $script"

        & { trap { $_; continue }; & $script } > "$script.output.txt" 2>&1 
        Move-Item $script "$script.processed"
    }

    Start-Sleep 1
}

With file synchronization as a back-end, you can now write a thin remoting client in nearly any language, and nearly any OS -- as long as it can read and write files, and is supported by your file synchronization client. Taking this same approach, you could also use email, FTP, web pages, or nearly anything else as a communication mechanism.

On top of that, you might want to build a primitive interactive remoting experience. Once Live Mesh completes their Windows Mobile client, I'll be doing that ASAP. Here's a PowerShell script that demonstrates the approach:

## Assuming the script is located in the (Mesh-synched folder)
##    D:\Documents\Mesh
## and the computer name is
##     "MESH-CLIENT"
## We'll drop scripts in D:\Documents\Mesh\Computers\MESH-CLIENT and
## process the resulting output files.
##
## Commands that impact the shell's state (such as current location) are
## persisted for future commands. If you want to share data between scripts
## use global variables.

param($computerName)

## If they specified a computer name, change to that directory
if($computerName)
{
    $scriptLocation = Split-Path -Parent $myInvocation.MyCommand.Definition
    $monitorPath = Join-Path $scriptLocation "Computers\$computerName"
    if(-not (Test-Path $monitorPath))
    {
        throw "$computerName does not have a Mesh Monitor directory"
    }
    
    Set-Location $monitorPath
}


$remotePromptText = "[MESH/$(Split-Path . -Leaf)] PS >"

## Invoke a command and retrieve its output
function InvokeCommand($command)
{
    $filename = [GUID]::NewGuid().ToString() + ".ps1"
    $outputFilename = "$filename.output.txt"
    $processedFilename = "$filename.processed"
    
    $command > $filename
    while(-not (Test-Path $processedFilename)) { Start-Sleep -m 100 }
    while(-not (Test-Path $outputFilename)) { Start-Sleep -m 100 }
    
    Get-Content $outputFilename

    ## Clean up our temporary files
    Remove-Item $outputFilename
    Remove-Item $processedFilename
}

## Get a command from the user, invoke it, and display
## the results
while($true)
{
    Write-Host -NoNewLine $remotePromptText
    $command = $host.UI.ReadLine()

    if($command -eq "exit")
    {
        break
    }
    
    InvokeCommand ($command + " | Out-String -Width $($host.UI.RawUI.BufferSize.Width - 1)")
}

If you have a PowerShell script that you are calling from cmd.exe, you might run into the following error:

Write-Host : The OS handle's position is not what FileStream expected. Do not use a handle simultaneously in one FileStream and in Win32 code or another FileStream. This may cause data loss.

This is bug in PowerShell, and happens when:

• a PowerShell command generates both regular and error output
• you have used cmd.exe to redirect the output to a file
• you have used cmd.exe to merge the output and error streams

For example:

PowerShell -Command '"start"'; Write-Error "Foo"; '"end"' > c:\temp\redirect.log 2>&1

One workaround is to use Start-Transcript for file logging (rather than cmd.exe) or have PowerShell do the error stream redirection.

However, if you don't have control over your logging, you can add the following snippet to any scripts that get launched this way.

Note: this is an unsupported workaround. It will almost definitely break as future versions of PowerShell are released.

V1

$bindingFlags = [Reflection.BindingFlags] "Instance,NonPublic,GetField"
$consoleHost = $host.GetType().GetField("externalHost", $bindingFlags).GetValue($host)
[void] $consoleHost.GetType().GetProperty("IsStandardOutputRedirected", $bindingFlags).GetValue($consoleHost, @())
$field = $consoleHost.GetType().GetField("standardOutputWriter", $bindingFlags)
$field.SetValue($consoleHost, [Console]::Out)

V2 CTP2

$bindingFlags = [Reflection.BindingFlags] "Instance,NonPublic,GetField"
$objectRef = $host.GetType().GetField("externalHostRef", $bindingFlags).GetValue($host)

$bindingFlags = [Reflection.BindingFlags] "Instance,NonPublic,GetProperty"
$consoleHost = $objectRef.GetType().GetProperty("Value", $bindingFlags).GetValue($objectRef, @())

[void] $consoleHost.GetType().GetProperty("IsStandardOutputRedirected", $bindingFlags).GetValue($consoleHost, @())
$bindingFlags = [Reflection.BindingFlags] "Instance,NonPublic,GetField"
$field = $consoleHost.GetType().GetField("standardOutputWriter", $bindingFlags)
$field.SetValue($consoleHost, [Console]::Out)

How does this work, and why does this happen in the first place? When PowerShell sends output to its output stream the first time, it keeps a reference to the output stream for future use. However, this output stream is really a wrapper around a lower-level stream. When cmd.exe writes to the output stream, it writes to the lower-level stream. This makes the .NET wrapper complain that the underlying stream has changed from beneath it.

This workaround modifies some private engine state to not keep a reference to the output stream -- but instead to re-examine the output stream on every use.

Once in awhile, we get questions about the best practice for PowerShell's execution policy when it is included as part of a standard desktop image.

This is one of the main reasons that PowerShell ships with a "Restricted" execution policy. Since you have to explicitly enable or install PowerShell, we've frequently been asked why we aren't more permissive by default. After all, you did just install / enable it! The reason is that separating the execution policy decision from the installation decision gives you a lot more freedom on installing PowerShell. It lets you push PowerShell to your whole enterprise via SMS (or include on a standard image,) and then later selectively configure the Execution Policy via Group Policy or other mechanisms.

When it comes to configuring these systems, the answer to this is largely determined by who will be using PowerShell. Is it primarily the end users, or as a target for logon scripts? The 80% rule is key here. You want to make a well-reasoned decision that applies to as many of your systems as possible, rather than have everybody make an un-informed decision on their own.

If it’s mostly for logon scripts, and you have a certificate you can sign these scripts with:

• Set the execution policy to AllSigned
• Add your signing certificate to the TrustedPublisher store

If scripting is likely to be extremely common (as with Exchange,) or you can’t obtain a code signing certificate, set the execution policy to RemoteSigned.

See here for some more guidance on this: http://www.leeholmes.com/blog/3rdPartiesAndPowerShellExecutionPolicies.aspx

As an brief caveat for logon scripts, some machines are configured to treat UNC paths as the same security zone as the internet (as opposed to the intranet.) This is Internet Explorer's "Enhanced Security Configuration." In this case, PowerShell responds the same as the Explorer Shell when it runs scripts from a UNC path: “While scripts from the internet can be useful, this script can potentially harm your computer. Do you want to run <script>?”

One way to fix this is by adding the source server to Internet Explorer’s Trusted Sites, or changing the “UncAsIntranet” configuration property (http://technet.microsoft.com/en-us/library/bb457150.aspx). This is also covered on page 341 of the (my) PowerShell Cookbook.

As another caveat, it is currently not possible to directly assign a .ps1 script as a logon script, since the architecture that enables this implies that you must be able to double-click on a PowerShell script to run it. We don’t want double-clicking to run a script by default for security reasons, so you can write a VBS script or batch file to call PowerShell with the correct arguments. The team blog goes into a bit more detail on this.

Last night, I flew my first half-hour as Pilot in Command of N8756C, a 1977 Piper Warrior. In flying, a pilot’s First Solo is a huge milestone. After hours of training and ground study, you finally take to the air on your own.

The most difficult part of flying is getting back down, so the First Solo is three takeoffs and landings -- about half-an-hour in the traffic pattern. It seemed to be the flight that would never come, though.

We first planned our pre-solo flight for Thursday, followed by the solo on Sunday. The pre-solo flight is a comprehensive review, so we booked the airplane from 5 to 7 PM. When we went through our pre-flight checklist, we realized that none of the avionics were working: I could hear myself, but couldn’t speak with my instructor. We tried everything we could to resolve the problem (including accidentally enabling the Emergency Locator Transmitter / Distress Beacon,) but couldn’t fix it. Dejected, we re-parked the plane and returned to base. I later learned that the squelch on the radio was wrongly adjusted. The squelch control determines the signal strength required for transmission. It is normally used to suppress low-level static, but in this case was set to exclude almost everything.

With that flight aside, we re-booked for Sunday – this time for both the pre-solo and solo flight.

Sunday came around bright and sunny, as did our 6 PM flight time. We patiently waited for our plane to arrive back from a previous lesson, filling time with some ground work in the meantime. We planned a flight to Thun Field, with detours for maneuver practice and instrument flying. Then, we waited some more. And some more. The plane finally got back 75 minutes late, but we went out hoping to get at least part of the lesson done.

When we got to the plane for our pre-flight checklist, we of course found it bone-dry. The gas truck took seemed to take its sweet time, but the real kicker came as we prepared to taxi out of the parking area. There was traffic to our left: a brand spanking new Boeing 737 being towed SLOWLY back toward the Boeing hangar in all of its unpainted glory. By slowly, I mean walking pace. This taxi would have taken 45 or 50 minutes, so we had to call off that flight, as well. We re-booked for Monday: 5 to 9 PM.

Monday (yesterday) finally came around, and my study work for Sunday night was to read up on how to use the VOR system for navigation, rather than use the primarily visual navigation we were previously planning. VORs are directional beacons that let a flight instrument determine the direction from the VOR ground station to your airplane. To find a specific point, you generally use two VORs at the same time to triangulate your position.

The flight was intense. It was 2.8 hours in the air and reviewed:

• Slow flight, where you fly and maneuver at near-stall speeds
• Power-On and Power-Off stalls, and stall recovery
• Ground reference maneuvers (Turns around a point, S-turns, rectangular courses)
• Steep turns and standard turns
• Emergency procedures (We “lost our engine,” so I had to pick and prepare for a landing)
• Go-Arounds (Calling a mulligan on un-safe or un-certain landings)
• 5 landings at Thun field, with my instructor having his eyes closed for the final 3 or 4

I was recording this on my GPS, so here are some highlights. First, the Thun Field landings:

Then, some ground-reference maneuvres:

On the way back to Renton airport, we also did 30 minutes of Instrument Training. For instrument training, you don a mask that prevents you from seeing outside the cockpit, and maneuver the airplane by instruments alone.

When we finally landed back at Renton airport, we practiced another 2 landings, finally coming to a full stop at the tie-down area. My instructor got out, wished me luck, and watched as went through the pre-flight checklist again. I alerted the tower as I prepared to taxi:

Renton Ground, this is Warrior 8756 Charlie at North Tower. Student pilot on first solo, requesting closed traffic with information Alpha.

I was given clearance to taxi to runway 15 (150 degrees, roughly north to south,) and prepared for my first flight. I waited for a minute as a few other planes landed, and then was given clearance for takeoff. I had too much power on my first run around the traffic pattern, so ended up lining up with the runway a little too high. Rather than try to rescue the landing in a potentially nervous state, I let the tower know I was doing a go-around, and climbed back to the traffic pattern. The controller mentioned that the winds were fairly variable and gusty, approaching 8 knots at times. She offered the south to north landing on runway 33, which I declined (as the winds weren’t affecting my landings.) The second attempt was much better, marking my first landing as Pilot in Command.

Now for landing number two.

Take-off is usually a breeze, but the nose appeared unusually high as I gave the plane full power on the ground. I quickly brought the power down, and exited one of the taxiways to determine the cause. After checking everything (flaps, elevator, etc,) I realized it was just a mind trick: I had been in a basically nose-down attitude (for the descents) for the last 10 or 15 minutes, so it was a lot like Velocitization in a car. You get so used to something (speed in a car, or nose angle in an airplane,) that it becomes your mind’s new definition of normal. You combat velocitization once you realize you have it, so I requested clearance to taxi back on the runway and complete my landings.

In the air again, the controller mentioned that they were just about to close, and asked if I wanted her to stay on. I didn’t want to keep her late, so I let her know I’d make my next landing a full-stop. She was awesome, and said she had no problem staying so I could get the solo done, so I continued with my next landing – still OK, but with definite room for improvement.

My final takeoff is where it really kicked in. After lifting into the air, an enormous thrill washed over me as I took a relaxed breath of fresh air. It was physically impossible for me to NOT land a third time. I grinned as I made my final approach, and settled in for the smoothest and straightest landing yet.

As recorded by my GPS:

As I taxied off the runway, the tower controller congratulated me on my landings as I thanked her for her help. The landings were nothing for the highlight reels, of course, but it gave a satisfying end-cap to 30 minutes that I will never forget.

And the ripped shirt to prove it? Well, in-line with tradition, my instructor cut up the back of my shirt. Before modern electronics, in-cockpit communication was mainly through shirt tugging and hollers from the instructor that sat behind you. Cutting off the back of the shirt symbolizes this very important step -- as your instructor won't need to be yanking on it any more.

One question that comes up fairly often when dealing with (or writing!) secure cmdlets is how to properly handle usernames and passwords. The solution there is to use (or make) the -Credential parameter of type PSCredential. A PSCredential object helps ensure that your password stays protected in memory, unlike cmdlets that accept a straight username / password combination.

If a parameter is of type PSCredential, PowerShell supports several types of input:

• empty: If you supply no input to a mandatory -Credential parameter, PowerShell prompts you for the username and password.
• string: If you supply a string to the -Credential parameter, PowerShell treats it as a username and prompts you for the password.
• credential: If you supply a credential object to the -Credential parameter, PowerShell accepts it as-is.

This is great for interactive use, but what if you want to write an automated script for a cmdlet that accepts a -Credential parameter? The solution lies in passing a pre-constructed PSCredential object. The solution to this is covered by recipe 16.9 in the PowerShell Cookbook:

Securely Store Credentials on Disk

Problem

Your script performs an operation that requires credentials, but you don’t want it to require user interaction when it runs.

Solution

To securely store the credential’s password to disk so that your script can load it automatically, use the ConvertFrom-SecureString and ConvertTo-SecureString cmdlets.

Save the credential’s password to disk

The first step for storing a password on disk is usually a manual one. Given a credential that you’ve stored in the $credential variable, you can safely export its password to password.txt using the following command:

Recreate the credential from the password stored on disk

In the script that you want to run automatically, add the following commands:

These commands create a new credential object (for the CachedUser user) and store that object in the $credential variable.

Discussion

When reading the solution, you might at first be wary of storing a password on disk. While it is natural (and prudent) to be cautious of littering your hard drive with sensitive information, the ConvertFrom-SecureString cmdlet encrypts this data using Windows’ standard Data Protection API. This ensures that only your user account can properly decrypt its contents.

While keeping a password secure is an important security feature, you may sometimes want to store a password (or other sensitive information) on disk so that other accounts have access to it anyway. This is often the case with scripts run by service accounts or scripts designed to be transferred between computers. The ConvertFrom-SecureString and ConvertTo-SecureString cmdlets support this by allowing you to specify an encryption key.

Note: Due to limitations in Version 1 of PowerShell, passwords encrypted with a specific encryption key can only be successfully decrypted by the same instance of PowerShell.exe process. Trying to decrypt the passwords with a different PowerShell.exe process will not be successful. To encrypt the passwords to disk in a way that can be read by other processes, use the .NET Encryption APIs: http://poshcode.org/116.

Although the solution stores the password in a specific named file, it is more common to store the file in a more generic location—such as the directory that contains the script, or the directory that contains your profile.

To load password.txt from the same location as your profile, use the following command:

To learn how to load it from the same location as your script, see “Find your Script’s Location.”

For more information about the ConvertTo-SecureString and ConvertFrom-SecureString cmdlets, type Get-Help ConvertTo-SecureString or Get-Help ConvertFrom-SecureString.

After what seems like an eternity, I finally got to start my official flight training. As a Canadian (AKA "Alien,") applying for flight school also means getting your fingerprints taken and being put in a special registration system. My training permission cleared late last week, so I booked my first lessons as soon as I could. In fact, for a government agency, the flight registration process was surprisingly efficient. Finding a place to get TSA-approved fingerprints outside of work hours? Well, that's a different story.

Anyways.

One of the things I really missed about my intro flights was a good understanding of exactly where I traveled during the flight. I remembered some landmarks, but didn't really catch on to many streets or things I knew from my ground travels. Recording my tracks through a GPS system seemed to be the best approach, so I finally settled on a Lowrance Airmap 2000C. Its features rival most of the top-end aircraft GPS products (aside from XM weather and traffic, which I'm not in the market for,) at a fraction of the price. For about the same cost, I would have been able to use some computer-based GPS software on my Tablet PC, but messing around with a Tablet PC in the cockpit seems to be a lot more trouble than it's worth.

I took it on a long camping trip during the Memorial Day long weekend, and it was fun watching it trace my path around the Puget Sound. It warned me plenty as I passed through the restricted airspaces (McChord airforce base, SeaTac airport, etc,) so luckily I wasn't actually flying.

On Thursday, I went for my first real lesson, flying AcuWings' Piper Warrior PA28-151. This is the kind of plane I want to do most of my training on -- as the Cirrus SR20 (while newer and nicer) felt like it might prevent me from learning some of the subtle basics -- like rudder control, or use of the nose-wheel while on the ground.

In the introductory flights, the instructor does almost all of the ground work, aside from giving you a brief overview of what they are doing. For your first training flight, though, you start to take over the technical preparations: aircraft examination, engine start procedures, pre-takeoff work, and the post-flight run-down. We spent about 45 minutes on these preparations, learning the how and why behind 80+ items in the pre-flight and post-flight checklists.

In the air, we practiced straight and level flight, ascents, descents, and (level, ascending, descending) turns. Like the flight-preparation work, non-introductory training means that you begin to take over much more of the landing preparations. During this flight, we did 2 landings, one being a touch-and-go. For both, I flew the traffic pattern, lined up the approach, and helped guide us toward the runway. I did a little less of the actual landing than I had done during my introductory flights, but I think that was probably because of higher wind. This flight gave me 0.8 hours of qualified dual-instruction to apply to my flight training.

On Sunday, we put in another 1.1 hours of dual instruction, with the learning curve continuing just as steep. Since I had been taught the checklists, pre-flight this time was all my own – including the inevitable splashing of leaded, high-octane Avgas on my hands while testing for fuel contaminants. As we prepared for takeoff, I was shown how to talk with the tower, request permission to taxi, take-off, and practice landings. That's a massive system of verbal cryptography that will take a while to get used to, but you feel like part of a club when you're doing it.

While in the air, we practiced slow flight (near-stall conditions,) stalls, and recovering from stalls. These are the most common "emergency" situations in flight, and are relatively easy to recover from as long as you know what you are doing. If you haven't practiced them, then you'll just panic and donate some iron to the earth's mantle.

After practicing those, we started working on landings. As a skill in flying, landings are a unique breed. They are a skill within a skill: a technique that gets lavish attention in the context of flying, but completely independent of it. Pilots obsess about landings the same way that cigar aficionados gush about about the tower of ash emanating from their cigar, or how developers delight in "Perl Golf,"

It's a journey that's just beginning -- I can't wait for more.

Once in awhile, you'll see a PowerShell screenshot that is still black and boring like the traditional cmd.exe prompt. Other times, you'll see the fancy blue window you've come to love. Why the difference? And if you're suffering from black window syndrome, what can you do about it?

Before shipping Version 1, our marketing and design teams set to make the PowerShell window a unique and marquee experience. What form did this take?

• Blue. Specifically, Red: 1, Green: 36, Blue: 86. -- In hex, 012 456. That number is much too cool to be a coincidence, but I haven't done the social archeology to find out why :)
• Lucida Console. Lucida Console is infinitely more readable than the raster fonts, and ships with all versions of Windows. Consolas is a beautiful alternative, but isn't broadly available.
• Window Size. Long gone are the days of 640x480 monitors being the most common, so PowerShell's default window gives about 800x600 of usable space.
• Screen Buffer. Likewise, long gone are the days of 256mb of memory being the most common, so PowerShell's default window gives a dreamy 60 pages of history buffer. This is a shell you want to live in. And when you live in a shell, you  want to be able to see what you've done.
• Quick Edit. Quick Edit mode lets you copy and paste easily to and from the console window. It is the setting that most console users enable once they realize it exists, and long gone are the days of mouse-driven console applications.

Now, even with all of these improvements, you still see screen shots of PowerShell prompts blinking desolate in the confines of a traditional black console window. Why is this?

PowerShell customizes its window through the shortcut properties in its Start Menu link. When you launch PowerShell through the Start | Run dialog or some other app launcher, these shortcut properties don't apply. Surprisingly, Windows does not support a mechanism to make these console customizations machine-wide. Windows supports per-user console customizations in the the HKCU hive of the registry, but there's usually no "Current User" during a PowerShell install. And even if there were, any customizations would not apply to user accounts created after PowerShell was installed.

So how do you get PowerShell's Noble Blue if you prefer Start | Run? Here's a script from the PowerShell Cookbook that does exactly that -- for your current user account.:

## From Windows PowerShell, The Definitive Guide (O'Reilly)
## by Lee Holmes (http://www.leeholmes.com/guide)

Push-Location
Set-Location HKCU:\Console
New-Item ".\%SystemRoot%_system32_WindowsPowerShell_v1.0_powershell.exe"
Set-Location ".\%SystemRoot%_system32_WindowsPowerShell_v1.0_powershell.exe"

New-ItemProperty . ColorTable00 -type DWORD -value 0x00562401
New-ItemProperty . ColorTable07 -type DWORD -value 0x00f0edee
New-ItemProperty . FaceName -type STRING -value "Lucida Console"
New-ItemProperty . FontFamily -type DWORD -value 0x00000036
New-ItemProperty . FontSize -type DWORD -value 0x000c0000
New-ItemProperty . FontWeight -type DWORD -value 0x00000190
New-ItemProperty . HistoryNoDup -type DWORD -value 0x00000000
New-ItemProperty . QuickEdit -type DWORD -value 0x00000001
New-ItemProperty . ScreenBufferSize -type DWORD -value 0x0bb80078
New-ItemProperty . WindowSize -type DWORD -value 0x00320078
Pop-Location