Essential Windows PowerShell Commandlets

Previously, we learned the basic commands: help and gal. I suppose now it's time to properly refer to these "commands" as commandlets or cmdlets following PowerShell's own set of terminologies. But just to note that whenever both commands and commandlets, I use them interchangeably for the same semantic meaning. Call me stubborn. Call it bad habit. But they really are all just commands to me. In any case, although I think of all of the commandlets simply as commands, we should still mentally note at the back of our mind that for "data-typing" purposes, PowerShell truly types these items or commands as Cmdlet.

Our first essential command after learning the basics is Get-Command

Get-Command

Get-Command outputs all of the available commands in PowerShell in a clean PowerShell installation. These are the commands that we really will use as the building blocks for writing PowerShell scripts. Through a process of trial and errors, let us try to see if we can perform the obligatory, introductory of any programming lesson: print the string "Hello World!" on screen (I promise I'll keep the trial and errors to an abridged version).

1. Type-in Get-Command at the prompt
2. Observe the output and see if any of the command contains descriptions that indicates it is a command for printing anything on-screen by using guess-words such as: "Print", "echo", "Write", "Response", or any other words from whatever your other experience may have been.
3. If nothing on-screen indicates as such, then you'll want to type in help [Cmdlet name] at the prompt to read the details one-by-one. Hey at least that's how I started learning PowerShell, by reading documentation as well as by trial and errors.

But, let's use my experience, shall we? Let's just type in the following simple command, and we'll simply get the result we want just as we'd see in Listing 1.

Listing 1: PS C:\Users\antonius> gcm | where {$_.definition -match ".*print*"} CommandType Name Definition ----------- ---- ---------- Cmdlet Out-Printer Out-Printer [[-Name] <string>] [-InputObject <psobje... ps c:\Users\antonius> gcm | where {$_.definition -match ".*echo*"} CommandType Name Definition ----------- ---- ---------- Cmdlet Set-AuthenticodeSignature Set-AuthenticodeSignature [-FilePath] <string []> [-C... PS C:\Users\antonius> gcm | where {$_.definition -match ".*write*"} CommandType Name Definition ----------- ---- ---------- Cmdlet Write-Debug Write-Debug [-Message] <string> [-Verbose] [-Debug] ... Cmdlet Write-Error Write-Error [-Message] <string> [-Category <errorcat... cmdlet write-host write-host [[-Object] ><object>] [-NoNewline] [-Separ... Cmdlet Write-Output Write-Output [-InputObject] <psobject []> [-Verbose] ... Cmdlet Write-Progress Write-Progress [-Activity] <string> [-Status] <strin... cmdlet write-verbose write-verbose [-Message] ><string> [-Verbose] [-Debug... Cmdlet Write-Warning Write-Warning [-Message] <string> [-Verbose] [-Debug...

From these output we can sort of guess which command we can use, but if in case we're still in doubt we could always do help write-debug, help write-error, etc. Soon after we go through this we'll find the command we want to use is Write-Output.

That's a rather "long" command to type, though. Let's try to see if there's a shorter version of it. For this purpose, let's try what's on Listing 2.

Listing 2: PS C:\Users\antonius> gal | where {$_.definition -match "Write-Output"} CommandType Name Definition ----------- ---- ---------- Alias write Write-Output Alias echo Write-Output

Now that we found what we want, we can try either command aliases just as in Listing 3

Listing 3 PS C:\Users\antonius> write "Hello World!" Hello World! PS C:\Users\antonius> echo "Hello World!" Hello World!

Sure enough, we found both yield the same result

We might wonder, "Why are there two aliases for the same command?" I don't really know the answer to that question other than perhaps the alias write is the built-in alias for Write-Output and perhaps the alias echo is simply a way to help those who are used to writing .BAT scripts to transition a little more easily to Windows PowerShell. Whatever the answer to that question is rather irrelevant; what's more important is that now we've gone through the basic exercises in learning each PowerShell command.

Wait wait wait!!! We're not done here!

What's all that business with the where command and the rest of the strings following it?

No, I didn't cheat :)

I simply wanted to shortcut all the trial and errors I went through in learning PowerShell without filling the content of this notes full of garbage. Trial and errors, after all, are things we each can do individually on our own time. As for the shortcut strung with the where command, I'll note an explanation of it another time.

Basic Windows PowerShell Commands

Windows PowerShell is relatively new shell command interpreter in comparison to the venerable CMD, CScript, and Windows Script Host.

Just like any shell script, it isn't really difficult to learn simply by trial and errors as long as we get the few basic commands down so we can drill further to other commands we seek to use.

The most basic command we should know is, of course, help.

Get-Help

As the command name suggests, it is the single command we use to see the details of a particular command we submit as a parameter to the help command. Listing 1 below illustrates the help command execution.

Listing 1: PS C:\Users\antonius> help help NAME Get-Help SYNOPSIS Displays information about Windows PowerShell cmdlets and concepts. SYNTAX Get-Help [[-name] <string>] [-component <string []>] [-functionality <string []>] [-role <string []>] [-category <string []>] [-full] [<commonparameters>] Get-Help [[-name] <string>] [-component <string []>] [-functionality <string []] [-role ><string []>] [-category <string []>] [-detailed] [<commonparameters>] Get-Help [[-name] <string>] [-component <string []>] [-functionality <string []>] [-role <string []>] [-category <string []>] [-examples] [<commonparameters>] Get-Help [[-name] <string>] [-component <string []>] [-functionality <string []>] [-role <string []>] [-category <string []>] [-parameter <string>] [<commonparameters>] DETAILED DESCRIPTION The Get-Help cmdlet displays information about Windows PowerShell cmdlets and concepts. You can also use "Help {<cmdlet nam e> | <topic-name>" or "<cmdlet-name> /?". "Help" displays the help topics one page at a time. The "/?" displays help for cm dlets on a single page. RELATED LINKS Get-Command Get-PSDrive Get-Member REMARKS For more information, type: "get-help Get-Help -detailed". For technical information, type: "get-help Get-Help -full".

It's so helpful that we can simply type help help at the prompt to display the details of the help command itself. As we examine the command output's "Related links" section, we see three other commands related to the help command. All the three new commands we found here: Get-Command, Get-PSDrive, and Get-Member are part of the important basic commands we will discuss more thoroughly a little later.

Get-Aliases

"Shallow Hal wants a gal."

The second basic command that is most useful to expedite our PowerShell command study is Get-Aliases also aliased as gal. First, we're gonna combine gal with the first command we learned earlier: help in Listing 2. Then, we'll try to see if sending the command twice to the shell prompt, one as the command call itself and another as the paramater to the command call gal gal in Listing 3.

Listing 2: PS C:\Users\antonius> help gal NAME Get-Alias SYNOPSIS Gets the aliases for the current session. SYNTAX Get-Alias [[-name] <string []>] [-scope <string>] [-exclude <string []>] [<commonparameters>] DETAILED DESCRIPTION The Get-Alias cmdlet gets the alternate names for cmdlets, functions, and executable files that have been established for the current session. This collection includes built-in aliases, aliases that you have set or imported, and aliases that you have added to your Windows PowerShell profile. If you specify one or more aliases, Get-Alias gets the alias object and displays its properties, including the object that was aliases, such as the full name of a cmdlet. This feature is made available by the Windows PowerShell Alias provider. RELATED LINKS Set-Alias New-Alias Export-Alias Import-Alias REMARKS For more information, type: "get-help Get-Alias -detailed". For technical information, type: "get-help Get-Alias -full".Listing 3: PS C:\Users\antonius> gal gal CommandType Name Definition ----------- ---- ---------- Alias gal Get-Alias

See! It's simple right? From Listing 2, we find the usage details of the command gal, and from Listing 3, we find that gal itself is an alias of the command Get-Aliases.

But, but, but... How did I know in the beginning that help and gal are commands we can execute? The answer is as ordinary as it could be—I guessed the first command: help presuming that a program usually has a variation of help, --help, --h, man or the old-school /? in some sort of form even without reading its documentation. As for the command: gal, after having found out that while the command help could be executed and yields an output indicating the actual command name: Get-Help, I simply sought in MSDN PowerShell site a reference entry pertaining to available command aliases.

Up next... We won't need to go through the details of each command, but will start our PowerShell exercises by using the three new "related" commands we learned earlier from Listing 1.

One ID to rule them all!

Having an OpenID is a cool thing. In theory, I wouldn't have needed to register for any other account ever. Alas, most ideals are just that: ideals. No, I'm not saying anything bad against OpenID; I'm only saying that even with the best intention we still frequently are unable to realize the full potential of the idea's ideals.

Take SourceForge.Net for example, it's one of sites I frequented. Sure it does take my OpenID to authenticate me when I logged into the site, but it still requires me to bind to a local SourceForge account if I want to use it to access certain things such as its CVS repositories. C'est la vie.

The OpenID provider I chose for my use is MyOpenID.com. I like MyOpenID.com because it is one of only a few (as of this writing) strong authentication providers who uses SSL certificates and Windows CardSpace/InfoCards which enables auto-authentication. This is so that while I do have a password for my OpenID account, I won't need to really memorize it or use it to sign into any sites using my OpenID account.

To demonstrate:

1. Go to MyOpenID.com
2. Sign in (if you already have an OpenID account)
   
3. Choose a sign-in method
   
   Or
   
4. You'll be presented with a certificate you've previously created and install in your browser
   
   Or be prompted to send your Microsoft Windows InfoCard you've previously linked to your OpenID
   
5. Click sign-in
   
6. And swweeet!! I'm in... quick & easy.

How to write clean website layout using CSS (Part 2)

The pure <div/> approach: Positions

Creating web page layout using CSS positioning really is a very simple thing to do. All we need to remember is that an XHTML element's absolute position is always relative to its container. That is the one thing we usually forget. We are so caught up in the semantic definition of the word absolute position we forget that it isn't absolute only in the context of the web page itself (<body/>).

For a similar simple markup we wrote in Listing 2 of our previous notes, we now have the following Listing 1.

Listing 1: <body id="doc"> <div id="d11" class="d1"> <div id="d11-sub1"> <div id="d11-21" class="d11-2"> </div> <div id="d11-22" class="d11-2"> </div> </div> <div id="d11-sub2"> <div id="d11-23" class="d11-2"> </div> <div id="d11-24" class="d11-2"> </div> </div> </div> <div id="d12" class="d1"> <div id="d12-21" class="d12-2"> </div> <div id="d12-22" class="d12-2"> </div> <div id="d12-23" class="d12-2"> </div> <div id="d12-24" class="d12-2"> </div> <br style="clear: both" /> </div> <br style="clear: both" /> </body>

The styling commands we needed in Listing 2 felt simpler to do and easier to understand from the perspective of having to think about how the browser layout engine will interpret the commands we write.

Listing 2: /* Section: Content specific **********************************************************************/ body, #doc { border: 5px double rgb(255,0,0); width: auto; /* Use // to give a specific instruction to IE browser only **********************************************************************/ //padding-bottom: 100px; } #d11, #d12 { height: 100%; width: 920px; border: 5px solid rgb(0,255,0); margin: 100px; /* Set margin-{left, right} to &quot;auto&quot; to center the div box because * setting text-align to &quot;center&quot; at the container level only works * with IE but not with other browsers, in addition to having the * unintended effect of having the container's text actually centered * that is if we simply just want the layouts centered, not its text. **********************************************************************/ margin-left: auto; margin-right: auto; } div.d11-2 { border: 5px solid rgb(0,0,255); margin: 50px; } #d11-sub1, #d11-sub2 { width: 900px; height: 100%; border: 1px solid red; margin: 10px; position: relative; } #d11-21, #d11-22, #d11-23, #d11-24 { width: 100px; padding: 25px; } #d11-22 { position: absolute; top: 0px; right: 0px; } #d11-24 { position: absolute; top: 0px; right: 0px; } #d12 { border: 5px solid rgb(0,150,50); } div.d12-2 { border: 5px solid rgb(0,0,255); margin: 50px; } #d12-21, #d12-22, #d12-23, #d12-24 { padding: 10px; width: 100px; float: left; } /* End Section **********************************************************************/

I'd say it looks pretty straightforward without having spent lots of hours tweaking the style attributes to get the desired results.

I was pretty happy this turned out to be easy. But then just for kicks I wanted to find out "What if I did the other way around, strictly with floats and kept all the elements' positions at its default relative value?" "Would it be just as simple?" "Would it be just as quick to do?"

The pure <div/> approach: Floats

With the same base structural template as Listing 1 above, I found that achievement the desired result using float method is considerably longer. Sure if you look at the last CSS code-block of Listing 2, you'll see that I actually cheated by using floats because well... that was actually the simplest way to order the <div/>'s in simple column formation. But then, to figure out which of the <div/>'s in square formation need to be floated the right way took quite a bit of trial and error in the order of magnitude of 5. Yes. Five. Listing 2 took me only a couple of minutes to write and get right. Listing 3 took me almost 15 minutes to do due to tinkering with the float and clear attribute combinations.

Listing 3: /* Section: Content specific **********************************************************************/ body, #doc { border: 5px double rgb(255,0,0); //padding-bottom: 0; } #d11, #d12 { width: 920px; border: 5px solid rgb(0,255,0); margin: 100px; /* Set margin-{left, right} to &quot;auto&quot; to center the div box because * setting text-align to &quot;center&quot; at the container level only works * with IE but not with other browsers, in addition to having the * unintended effect of having the container's text actually centered * that is if we simply just want the layouts centered, not its text. **********************************************************************/ margin-left: auto; margin-right: auto; } div.d11-2, div.d12-2 { border: 5px solid rgb(0,0,255); margin: 50px; } #d11-21, #d11-22, #d11-23, #d11-24 { width: 100px; padding: 25px; } #d11-sub1, #d11-sub2 { border: 1px solid red; margin: 10px; //padding-bottom: 50px; } #d11-21, #d11-23 { float: left; clear: left; } #d11-22, #d11-24 { float: right; clear: right; } #d11-23 {clear: left;} #d12 { border: 5px solid rgb(0,150,50); } #d12-21, #d12-22, #d12-23, #d12-24 { padding: 10px; width: 100px; float: left; } #d12-21 {clear: left;} #d12-24 {clear: right;} /* End Section **********************************************************************/

How to write clean website layout using CSS (Part 1)

There are three well-known techniques in creating a website layout using CSS:

1. The <table/> approach.
2. The <table/> + <div/> hybrid approach.
3. The pure <div/> approach.
   1. The positioning approach of absolutists v. relativists; and,
   2. The clearing float (or is it the floating clear? <smile/>).

There are others of course; very advanced CSS masters like to further enhance the cleanliness of their XHTML code without using a single <div/> or <span/>.

As I jumped into refresher exercises, I wondered: "How easily would it be for me to do this (ref: screen)?" Or, "Is it even possible to easily do it for the four popular browsers today?"

---

Warning: At the time of writing, the following materials were only proofed on:

• Internet Explorer 7.0.6001.18000
• Firefox 3.0.6
• Safari 3.2.2 (525.28.1)
• Chrome 1.0.154.48

This article merely documents these variations and choices. We do not advocate one approach versus another as these choices are left to each of our individual circumstances and reasons. Although <grin/> I would soon give up doing the first approach.

The base skeletal structure for each of the tested approaches is the following Listing 1(with slight differences):

Listing 1: <!--xml version="1.0" encoding="UTF-8"?--> <!--DOCTYPE html PUBLIC "-//W3C/DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml1-transitional.dtd"--> <html> <head> <title>DIV</title> <style type="text/css" media="all"> /* Section: Initialization **********************************************************************/ * { margin: 0px auto; padding: 0px; outline: 0 border: 0 none; font-family: cambria, georgia, "Times New Roman", times, serif; font-size: 11pt; font-style: inherit; font-weight: inherit; vertical-align: baseline; text-decoration: none; /* Set min-height to "100%" and overflow to "auto" in order to set * the div box to stretch to the height of its children elements. This * setting bears no effect whatsoever on IE. **********************************************************************/ min-height: 100%; /* Use // to give a specific instruction to IE browser only **********************************************************************/ } } body { padding: 10px; } h1, h2, h3, h4, h5, h6, p, pre, blockquote, form, ul, ol, dl { margin: 20px 0; } ul, ol, dl {list-style: none;} table, tr, td { border-spacing: 0; } a img, :link img, :visited img, { border: 0 none; vertical-align: bottom; } code, .code{ padding: 10px; padding-left: 15px; font-family: consolas, monaco, "Courier New", courier, monospace; font-size: 9pt; color: rgb(100,100,100); white-space: pre; text-align: left; } /* End Section </style> </head> <body> ... </body> </html>

---

The <table/> approach

If we search the Internet for arguments against using <table/> tags for layout purposes, we will find them in abundance. They are all valid arguments and we should heed them. I simply could not bring myself to nest <table/> tags like it's 1999.

The <table/> + <div/> approach

For the following simple layout in Listing 2:

Listing 2: <body> <table> <tr> <td id="t11-21"> <div id="d11-21" class="d11-2"> </div> </td> <td id="t11-22"> <div id="d11-22" class="d11-2"> </div> </td> </tr> <tr> <td id="t11-23"> <div id="d11-23" class="d11-2"> </div> </td> <td id="t11-24"> <div id="d11-24" class="d11-2"> </div> </td> </tr> </table> <table> <tr> <td id="t12-21"> <div id="d12-21" class="d12-2"> </div> </td> <td id="t12-22"> <div id="d12-22" class="d12-2"> </div> </td> <td id="t12-23"> <div id="d12-23" class="d12-2"> </div> </td> <td id="t12-24"> <div id="d12-24" class="d12-2"> </div> </td> </tr> </table> </body>

We still had the following styling code in Listing 3

Listing 3: /* Section: Content specific **********************************************************************/ body, #doc { width: 100%; border: 5px double rgb(255,0,0); //padding-bottom: 100px; } table { width: 948px; border: 5px solid rgb(0,255,0); margin: 100px; /* Set margin-{left, right} to &quot;auto&quot; to center the div box because * setting text-align to &quot;center&quot; at the container level only works * with IE but not with other browsers, in addition to having the * unintended effect of having the container's text actually centered * that is if we simply just want the layouts centered, not its text. **********************************************************************/ margin-left: auto; margin-right: auto; table-layout: fixed; } /* TR styling doesn't work on IE ************************************************************************/ tr { width: 928px; border: 5px solid red; display: block; } td { vertical-align: top; } #t11-21, #t11-23 {width: 100%;} #t11-22, #t11-24 {text-align: right;} #t11-21, #t11-22, #t11-23, #t11-24 { //width: 50%; border: 1px solid blue; } #t12-21, #t12-22, #t12-23, #t12-24 { width: 25%; text-align: center; border: 1px solid blue; } div.d11-2, div.d12-2 { border: 5px solid rgb(0,0,255); margin: 50px; } #d11-21, #d11-22, #d11-23, #d11-24 , #d12-21, #d12-22, #d12-23, #d12-24 { text-align: left; width: 100px; padding: 25px; overflow: visible; } #d12-21, #d12-22, #d12-23, #d12-24 { padding: 10px; //margin-left: 0px; //margin-right: 0px; } /* End Section ************************************************************************/

Yes, this approach yields a slightly cleaner code result than only the <table/> based approach. But it isn't without its own problems. There's that <tr/> rendering issue in IE. And of course, I found that the CSS implementation differences between the various browsers render this approach virtually indefensible, as in saying: "If we have to resort to using various CSS tricks/hacks anyway, why not simply do a Table-less Layout with its pros and cons already weighed by the web design community at large, especially when we already have ample samples written by masters such as those from A LIST apart and Zen Garden—two sources of my favorites reading materials where I found a wealth of knowledge shared to us all common netizens."

Need I say more?

Up next: The pure <div/> approach (Part 2).

How to write clean X/HTML document

---

Warning: XMP tag is an officially deprecated tag as of HTML specification 3.02. So follow this advise at your own risk.

---

Use <xmp/> tag instead of the using many <br/> tags to pre-format ourmultiline texts inside a <p> tag. We should use <xmp/> tag for plain text and demonstrated code listings instead of using escaped stuff (entities) such as &lt; and &gt; Escaping characters are such tedious chores and we are all better off with using <xmp/> tag. If we want to have the HTML content of our tags still interpreted as HTML content then we choose to use the <pre/> tag instead of <xmp/> tag (ref: Listing 1) Listing 1: <pre> <b><u>HI THERE!</u></b> XMP is sweet. I don't know why such as useful tag is deprecated. </pre> Listing 1 will still be interpreted as HTML content and render as the following: (notice how the rendered text includes the invisible 'tab' characters). Let's compare it to what we would have to do in Listing 2 below if we don't use <xmp/> Listing 2: &nbsp;&nbsp;&nbsp;&nbsp;<b><u>HI THERE!</u></b><br/> &nbsp;&nbsp;&nbsp;&nbsp;XMP is sweet.<br/> &nbsp;&nbsp;&nbsp;&nbsp;I don't know why such as useful tag is deprecated.<br/> Which codeblock is easier to work with? Just for writing:

	HI THERE!
	XMP is sweet.
	I don't know why such as useful tag is deprecated.

We could alternatively use CSS to control our content like the following: Listing 3: <p style="white-space: pre;"> My multi line content inside a paragraph &lt;p/&gt; tag </p> to render the following:

My multi line content inside a paragraph <p/> tag

Although the sample in Listing 3 takes the white-space character and carriage-return/line-feed character, it does not take the tab character and other invisible characters that may be inside our text content-not to mention more typing compared to using <pre/> tag. If it is still chosen, IE will fault us (if we are reading this using IE, we will see the above example above as a single-line, instead of what it should be). Using the <xmp/> tag and/or <pre/> tag helps us reduce the mess that other developers would see when they view the source of our program. What mess am I talking about? Think about this: How often do we use &nbsp; like the above example? Yep, that's what I'm talking about :) So what if we want to stylize an <xmp/> content?

All we have to do is use a CSS class as we would in any other case:
For example: we could do the following:
Listing 4:
<xmp style="color: blue; font-weight: bold;">Content</xmp>

Such that the result renders as the following: Content We should be careful and test the use of <xmp/> tag carefully since this tag has been officially deprecated as of HTML standard 3.02 and is replaced by the <pre/> tag (see: WARNING at the top of this page), despite the fact that the <pre/> tag is available, it doesn't completely offer all the simple escaping facilities offered by the <xmp/> tag.

Confidence == Self.Comfort in {Knowns, Unknowns}

Confidence != Perfection;

Confidence != Arrogance.Hide(Inferiority);

What do we do when a user asks for a facility for a matter of convenience and where the cost of providing such convenience far outweighs the value of the time saved from the convenience?

There's a dictum that states: "Customer is King." If we want to be responsible subjects to our king, it is our duties to say to the King: "Your Majesty, if you wear this 'clothes' you will be naked; is that what you really want?" and instead of: "Yes Your Majesty, we will use the finest invisible silk thread to match the invisible royal garb you ordered."

Information is to knowledge as...

Bit is to Byte.

Byte is to data.

Data is to information.

 

A computer is such a nice tool to collect, prepare / process, and consume knowledge. But nice is different from easy. The level of ease in using computer for the aforementioned activities however are usually inversely proportional to one another. For example, if it could be considered easy to collect data, it can be reasonably said (and observed) that the data processing is relatively more difficult than the its collection; or so the logic dictates, for what it's worth.

Fast forward to preparing cubes for SQL Server Analysis Services, I found myself trying to figure out how to prepare an ease of processing for my work product's audience, starting with its setup. After a little bit of struggle through the various binaries in my machine, I found the following steps helpful for those of you trying to do the same:

Local environment variable path setup:

1. Add to your local environment path an entry pointing your VS.2005 IDE location (screen Evp1.1 - 1.4).
2. Add to the same local environment path another entry pointing to the location where you find Microsoft.AnalysisServices.Deployment.exe, if it isn't already there. (Notes local environment path entries are separated by semi-colon).

Screens:

Evp1.1 Evp1.2

Evp1.3 Evp1.4

 

The lazy version:

1. prompt:> devenv [solution filepath and filename] /deploy

Well... sure you might run into a couple of pop-up dialogs, but just a couple of OK clicks without knowing what they mean will put you back on the processing track. No worries though, even when you don't click the buttons until the end the show will go on until completion with the exception that the thread is still be hanging until said clicks are made.

If you're bothered by the only seemingly semi-autonomous run (since you still may be needed for those two manual clicks), here's another way to do it... properly... starting with the same local environment path setup steps above, then followed by these steps:

1. Build the .asdatabase file from your solution file using prompt:> devenv [solution filepath and filename] /build
2. The deploy using prompt:> microsoft.analysisservices.deployment [solution filepath]/bin/[project name].asdatabase /s:[deployment log filepath and filename]

A word of caution: Don't be alarmed if your screen seemingly freezes during the deployment; this will likely be the scenario depending on your hardware configuration and the volume of data in your database that Analysis Service engine processes.

Why do we want to go through this when we can just load the solution file in VS.IDE then with just a few clicks to the above easily without having to memorize seemingly arcane and puzzling commands and steps? There are several different answers to this, but my favorite one doesn't have anything to do with it as knowledge that reliable technician needs to have; rather, it has to do with the pursuit of knowledge itself.

If our knowledge is the sum of our experience, and if we postulate that human beings learn more from difficult experience than an easy one, then it follows that by doing the above steps we will retain this knowledge longer that had we just done the usual mouse-clicks. And that my friend is what knowledge is to an application...

Joy is...

A smile in spite of what's around.

You know... these days I sensed a lot of unhappiness around me—not from me, mind you—around me. So I took a few minutes out of my day and scoured the Internet to see if I could find some quick, inspirational statements people say about how we can reach happiness since I don't seem to be doing well in inspiring happiness for others.

Here's what I found:

From Dennis Wholey:
Quote #1: Happy people plan actions, they don't plan results.
Quote #2: Expecting the world to treat you fairly because you are a good person is a little like expecting a bull not to attack you because you are a vegetarian.

Here's another from our favorite genius, Albert Einstein:
Quote #3: If you want to live a happy life, tie it to a goal, not to people or things.

From an unknown smartie:
Quote #4: Happiness is not the absence of problems but the ability to deal with them.
Quote #5: Some pursue happiness; others create it.

How are these useful for me? Eeh, they aren't new to me, really; they're what I already naturally use for me. And I'm happy in a happy-go-lucky-sort-of-way-kind-of-guy; see if these could be as useful for you as they have been for me.

To be gainfully employed is...

A rather challenging act to perform.

It's not just about having a job nor about the money that the job brings.

It is a constant juggle to balance our personal and professional lives.

I've shared with many people I encounter in my careers a personal attitude that may help reduce one's own stress in a work setting and be able to feel secure on and of our own: that work is just a vehicle for us to meet new friends and earn income to support our family.

Professionally, the level of quality of my output and my team's output is much less driven by our individual talents, but more about the principles we go by and use in our teamwork:

• Kanban: Just-in time (classic production understanding in manufacturing industry of what's now called Agile in the software industry).
• Kaizen: Continuous improvement; our work is never done, even when the output has shipped. This translates personally to set a goal where we can pace ourselves according to our health and ability and professionally to always understand that the best we deliver today should hopefully be better tomorrow.
• Gandhi engineering: A combination of irreverence toward established way with a scarcity mentality that spurns superfluities while at the same time maintaining consistency of applying the principles in play.
• Einstein tolerance: That can't be paraphrased, but simply understood in what Albert Einstein himself said: "Great spirits have always found violent opposition from mediocre minds. The latter cannot understand it when a man does not thoughtlessly submit to hereditary prejudices but honestly and courageously uses his intelligence and fulfills the duty to express the results of his thoughts in clear form", to help us maintain composure in an adversarial environment. This also relates to what I wrote earlier in my post: The 道 of Writing, the sublime of Work Ethics...

A customer is NOT always right

1. If he is belligerent and makes my team and employees unhappy.

2. If a customer makes me choose between unreasonably serving his business or my relationship with good colleagues and employees. My employees come first. (This is along the line of: "If an employer makes me choose between unreasonably serving his business or my devotion to my family. My family comes first.)

3. If his business is bad for my business: we all have a finite amount of resources serving a customer that does not further our own goal. For example, serving a bad customer who takes so much more time and efforts where consequently we are unable to provide an equal or better time and service for a good customer, we'll have to say goodbye to the bad customer.

4. If he is unreasonable despite our best effort to serve him/his company with a good service, and that they have been receiving good attention and services.

5. If we happen to get a glimpse, somehow, that he doesn't treat {another person | his own employees | general public} fairly and with dignity as it is a telltale of what is to come for us as an outside vendor.

Holy chocolate fudge!!! Quick! Show the window pronto!

The first month of 2008 came and went.

We've gone through 31 days of a new year so quickly, we barely had a chance to reflect on the past year due to piles of backlog tasks that seems to increase faster than we can keep up. It's a blessing and a curse.

Speaking of a curse, I was in the middle of debugging an application last night and was unable to find the Immediate Window in my VS2005 IDE. It had been there before; I sought for it in all of the menu and sub-menu items and couldn't find it. I still don't know what caused its disappearance. Perhaps it was a recent VS2008 IDE installation that might've affected the 2005 environment settings; perhaps others unknown reasons.

Suddenly, a simple thought sprang to mind : go to Tools/Customize/Debug and simply add it back-in to the menu bar and my shortcut toolbar. Et voilà! My indispensable Immediate Window returns; and all is well.

What a great {insert a nerdy title here} feels:

• I don't need to prove I'm 1337 to know I am one.
• I don't understand the end-user, but I'll always try to.
• My end-user is the reason I have a job.
• Patience really is a virtue.
• Nobody likes complicated stuff.
• It's better to be perceived as inexperienced but a quick study, than experienced but a slow poke.
• Knowing the subject well is better than knowing how to code well.
• If I don't know how to code it well, give it to someone who does, then learn from it.
• $1/day for 365 days is better than $10 for only 30 days. Long term builds trust, character and wealth; short one brings false sense of confidence.
• Each new job feels like my first job; done with great excitement, else it could be my last one.
• Get completely in or get out, capiche?

2008

A few days into the new year, a sleeping sick baby and an offline HQ—I found a few moments of peace.

So here's a piece of advice: When trouble brews, be calm and proceed to the nearest exit, brew a cup of hot cocoa, lounge, and read.

If your trouble is specifically of the message:

"The current identity (NT AUTHORITY NETWORK SERVICE) does not have write access to 'C:\WINDOWS\Microsoft.NET\Framework\V2.0.50727\Temporary ASP.NET Files'. This error may be cause by having installed visual studio C# or by group policy settings."

Please, oh please don't manually assign the permission to the folder through the folder property; you should instead run the following from a command prompt to wish your trouble away:

prompt:>c:
prompt:>cd \windows\micro*\frame*\v2.0*\
prompt:>aspnet_regiis -i -enable

Happy New Year!

The 道 of Writing, the sublime of Work Ethics...

Be it software, technical documents or emails.

If any man wish to write in clear style, let him be first clear in his thoughts; and if any would write in a noble style, let him first possess a noble soul.
~ Johann Wolfgang von Goethe

Learn as much by writing as by reading.
~ Lord Acton

I too am an apprentice, always—learning as much by failing as by succeeding, as much by following as by leading.
~ antonius

So don't despair that no one agrees with you and point your mistakes, for "... when you're screwing up and nobody's saying anything to you anymore, that means they've given up on you."
~ Randy Pausch

Software Entropy

Software entropy refers to the amount of disorder present in a system.

Entropy usually creeps in by way of logical inconsistencies at the thought-level, even before any tangible work begins to implement said system. As the system and its parts are developed, these logical inconsistencies manifest themselves in various forms of errors of all sizes. The smallest ones considered negligible are typically the clichéd primordial stew of chaos in software systems.

When coupled with what's frequently unfortunate timing of most software development schedules, the hacks we are forced to place into the system here and there by necessity of meeting some established deadline further exacerbate the condition.

Now plagued with such level of anomalies any additional issues encountered are prevented from meeting their resolutions simply because it's no longer feasible.

They say, only two things are certain in life: Death and taxes; well... in software Chaos reigns supreme if unchecked and snipped in the bud. Entropy is contagious.

What's the moral of the story kids?

 

冰冻三尺，非一日之寒
Three-feet of ice didn't result from one day of freezing weather.

 

Do the job right the first time or don't do it at all.
Better yet, get out of the way of anyone who's trying to do the job right the first time.