If you’re interested in PowerShell then you have probably already seen the news over at the Microsoft PowerShell Team Blog, but just in case I’ve linked to the three recent posts that I think are pretty much just awesome. Go take a look, you can bet I will be ;-)

Jonathan and Hal (from the PowerScripting Podcast) will be hosting Jeffrey Snover for a Virtual PowerShell V2 Launch Party on Thursday, which is of course GA* day for Windows 7!!

Lee Holmes from the PowerShell Team has also announced that one of their internal tools has been released as a Codeplex Project.

The Cmdlet Designer makes it much easier for teams to concentrate on the design, naming, and consistency of their cmdlets, while also guaranteeing name registration and collision avoidance across Microsoft.

To sweeten the deal, it offers:

* Integrated help authoring
* Efficient bulk operations (parameter and cmdlet cloning)
* Generation of cmdlet code
* Full scripting support
* Automatic code-spec comparison and testing
* Role-based security, history logging, and more.

But wait, there’s more! It’s now available under MS-PL from http://www.codeplex.com/CmdletDesigner

And last, but in no means least, with the recent announcement of the Windows 7 Resource Kit book comes the availablity of the Windows 7 Resource Kit PowerShell Pack. There’s a detailed breakdown of the features in the PowerShell Team’s blog post, but in a brief look it contains: 10 modules with 600+ scripts. Great work James!

* General Availablity.

But he got back on the “tools” over the weekend, and posted a four part series on “Terminal Services Gateway and Terminal Services Web Access using Hyper-V”:
Part1
Part2
Part3
Part4

I wonder with that blog title, if his next move will be to Microsoft Marketing and product naming. *joke*

John has some excellent posts from his old days *grin* which are well worth a look while you’re perusing his site.

Hopefully, this latest burst in blogging signifies a return to form John?

This situation is very similar to when the Windows Installer (MSI) technology first came on to the scene around 2000-2001, and anyone authoring setup software was dealing with how to get it out into their environments. Everyone was including the redistributable with their software in case the machine it was going to be installed on didn’t have it in place already.

The complexity with PowerShell however is that there are two questions that need to be dealt with:1) Like MSI, how do you deploy PowerShell out in your environment? and 2) How do you manage your PowerShell landscape of Snappins?

I was lucky enough see an online demo from Magnus and Thorbjorn from Special Operations Software today. They have a new product expected to be announced at TechEd later this month, called SpecOps Command which I believe will deal with these questions, for starters.

SpecOps Command, seems to be a tightly integrated product between Group Policy and PowerShell. This combination has the ability to provide it with the best of both products: the ease and flexibility of PowerShell, and the centralised environment management of Group Policy.

The tool has loads of cool features including the ability to run PoSH scripts assigned in GPO’s, Undo scripts for when things fall out of scope, reporting, and the ability to target clients in a very granular manner e.g. Only apply to Dell machines running Windows XP.

Thorbjorn advised that SpecOps intend to release a couple of versions of Command, including a free version that should provide the core functionality including the abilty to distribute PowerShell out into your environment.

I think it’s great that ISV’s like Quest and SpecOps are adding value to the PowerShell community with free offerings, as well as their commercial products. It give them exposure to their intended market, and provides us with some cool tools to make life easier :)

Like Quest’s AD Cmdlet’s I think SpecOps Command is likely to have a big impact on the way we use PowerShell going forwards.

[Update 2 Nov]
Magnus just gave me the link to the SpecOps website for Command. This will be the product page when it releases.

The first thing that I like about this is that it will now officially support server deployments, as well as desktops. This gives you a good robust, free deployment mechanism for your environment if you don’t want to go heavy duty and get into SCCM (essentials etc).

I have a few scripts, which I use with BDD 2007, to perform things like configuring static IP addresses, and dealing with the XP/2003 Uber bug. Now I haven’t checked the Uber bug is resolved, but you can now configure static IP through the Workbench.

Now I will caveat my observations to say that I’m still pretty new to BDD/Dep4 and the stuff that I’ve seen could just be a PICNIC error ;)

1) There appears to be no keyboard navigation through configuring the details of your Computer/Role/Make and Model/Location object.
On the details tab, which contains your “unattend.txt”-esque configuration options, there doesn’t appear to be anyway of actually navigating your way down the page. It’s a case of mouse in one hand keyboard with the other as you set Display Resolution etc.

2) No multiple assignment ability for driver groups.
When I point the “Out-Of-Box” drivers wizard at the Dell PowerEdge driver CD, it imports about 50 drivers. Great. Now I want to assign them to, say, “Dell PowerEdge Servers” driver group. As fas as I can see this has to be done one driver at a time!

3) Static IP assisgnment, one server per Task Sequence. Really?
I added a new task to a “standard server task sequence” object that I created (formerly known as “builds”). Here you have the ability to add configuration objects, for what appears to be per-NIC. You can give it a name, and then the usual IP data (DHCP, DNS/WINS/Static data). What strikes me as odd is that this appears to be the IP address data for this Task Sequence. That makes the TS only useful to one server. I would have thought that this should be configured under a lower node, more specific to the Computer/Location etc?

4) A new deployment doesn’t appear to be able to use “SkipWizard=YES”. In a quirk (which I suspect I might have read in the release notes to be fair) results in the information given when creating a new Task Sequence (Product Key, Local Admin Password etc) not being retained when you deploy a machine. If you then try and deploy a new machine and have selected to Skip the Wizard in PE, everything just kinda stops after wpeinit.

It works if you step through the wizard (Next, Next Next *yawn*), and you’ve also populated the missing data (Product Code, Local Admin Password) under the details tab, but still, why?

Now, as I said before this could be just me (really, please someone tell me I missed something here), but if it’s not then this seems pretty strange stuff to be experiencing in a Beta 3 release IMHO. Well point 4) is, the rest of course, could just be “By Design”.

Don’t get me wrong I actually really like BDD/Dep 4. The ease of getting a build up and running to deployed is excellent. And I love the way I can throw almost anything at it to get the job done – .Reg, VBS, MSI, EXE, PS1 it will do anything you need it to do – so far ;)

I’ve submitted the above as “feedback” on Connect so I guess we’ll wait and see. I’m hoping some bright tech out there will post in the comments that I’m a muppet and it works fine if you do …. ;)

At work today one of the guys told me that he’d had an Xmod and replaced it due to a few hassles. Namely on his 64-bit Windows machine it had caused numerous BSOD’s. That would have been sufficient for him to pass it on, but his other two criticisms were that it required external power to run, and drivers installed to get it operational under Windows.

Suffice to say I got home tonight, less than expectant about my new toy. I figured I’d give it a try on my Ubuntu machine, and if it didn’t live up to my expectation – namely no driver hassles, and power drawn off the USB alone, I’d send it back and pick up a Terratec Aureon instead.

Well, my first impression was good. I liked the packaging and the iPod-esque colouring. I’m not an Apple Fanboy by any stretch, but you have to give them points for style!

First up I connected it to my Windows XP laptop, and sure enough I got the New Hardware dialogue. Not a good sign!

I hooked the unit up to the Ubuntu laptop, and opened up a terminal session to see if the OS recognized the device.

Now I’m not brilliant with configuring sound on Linux – despite my work mates labeling me as the token tree-hugging-sandal-wearing Nix-Fanboy ;) I do however know how to configure Amarok to use a different card, so this is where I started. Under Settings | Configure Amarok select the Engine settings from the left side of the dialogue, as below.

The output from the cat command previously gave us the name of the audio device – Xmod. So this is the name we use.

A quick restart of Amarok, had “Blink182 – Feeling This” pumping down my earphones :)

So far so good, but I still had no system sounds, or any of other other media apps (Totem, or VLC etc). What I needed to do was set the Xmod as the default audio card in the system.

A quick look around the OS, and I found – asoundconf. Typing the command gave us a list of expected switches, of which two looked useful: list and set-default-card

As you can see, stepping through the switches appears to have set the Xmod as the default card. A quick reboot, and, I suddenly have system sound, and all the media apps are playing nicely!

Despite my earlier worries, the device works well with by Ubuntu machine. No extra drivers required, no extra power needed, and it even works with the hardware sound control!

Tomorrow, there shall be music :)

I just saw this and thought it worthy of passing on.

I’ve not had a play so I can’t tell you how good/bad it is, but I’ve always liked the Snippets functionality in PrimalScript. It’s a nice way of building functions up into a library of re-usable code :)

They add right into SAPIEN PrimalScript, and let you type any VBScript function or statement name, press Ctrl+J, and see how that VBScript feature is performed in PowerShell

Nice :) I’m a big advocate of learning by trying to achieve objectives when it comes to scripting, but I can see this would be useful for seeing what Cmdlets are the VBScript equivalent. Coupled with the Microsoft VBScript to PowerShell Conversion Guide I can see both as being useful tools to help the learning curve of PowerShell.

Admittedly I couldn’t justify the expense of purchasing a copy for my PowerShell scripting alone, as the Tab Completion from MoW, and the built in Shell Intellisense is pretty impressive in itself.

I’m keen to take a look at the Beta to see how PrimalScript is going, and see if it’s time to re-evaulate my postion ;)

There’s a feature list on their website. If you want to take part you need to keep an eye on the Sapien blog to find out when the limited beta group becomes available.

If I get into the Beta group, NDA’s permitting etc, I’ll let you know how it goes ……

As people start to become aware of the power of PowerShell, we can see vendors starting looking to leverage this also:

Lists taken from PowerShell Home page

I can see that as each product team and vendor take up this new technology they are going to create their own cmdlets to make administration tasks easier. Now I think this is a good thing, the adoption, creating tools to make life easier. However I see a problem that could arise from this.

We are going to see the PowerShell landscape changing from machine to machine depending on what products are installed. Are we going to start seeing scripts fail because they have been written with cmdlet dependencies?

So what’s the solution to this? Should cmdlets be deployed as addons to all PowerShell installations to keep the landscape uniform? What would be the impact on this?

If I’m on a non-exchange 2007 server, and I have the exchange cmdlets installed should they work regardless, so long as they can reach the exchange server in question?

Or is the solution to have our scripts smarter, and exit nicely when the cmdlet isn’t available? This still leaves us with a fractured landscape, and I can see as more products adopt PowerShell a larger difference from installation to installation. Or am I just looking at this all wrong?

Now I can see their concern, to a point. Google has had rapid growth over recent years, and I believe they have taken the lead from Microsoft in the innovation front – think Gmail, Google Earth, Maps, Reader and SketchUp to name a few. Microsoft have Hotmail, Microsoft Virtual Earth and Popfly.

Now Popfly isn’t released yet, but seems like it might have some cool potential there. Comparing the two company’s they have similar offerings but is it me, or do Google’s seem to be more popular? When it comes to web applications then is Google the new Apple?

I don’t know, maybe I still buy the whole “Do no evil” thing.

There’s an ad campaign here in the UK at the moment from ask.com where they are trying to brand themselves as “the other search engine”. How many people use Google as there search engine out of habit? What search engine would you use if you didn’t use Google?

Personally I like the Google home page, it’s crisp and clean of clutter. I find the Yahoo Home Page too cluttered.

Now we all know I’m no coder, and I’m so busy with automation and build stuff, or Linux stuff that I don’t usually get chance to play with “cool” stuff, but I have to say Google Sketchup is pretty damn cool!

The LFM House in Sketchup:

[edit]
Updated this morning to finish the MS Vs Google application comparison. I wrote this late last night, and completely missed that I hadn’t finished the thought. Obviously not a good idea to post when you’re tired ;)

Recently I moved to using Google Reader because I can use it wherever I happen to be. It’s also platform independent, which is quite handy.

Well, one of the cool features of using Google Reader is the ability to share posts that I come across that might be of interest to, well, you guys I guess ;) So there’s a funky widget in the Sidebar. Google call it a Clip. Let me know what you think.