Foo Theory

So the new Zune device line-up has been announced and is set to releases for the holiday season.  An 80GB hard-drive based model and 8GB and 4GB ultra-portable flash memory models are said to include the same basic feature line-up. 

Zune, using Zune Marketplace, will be able to download, share and play music minus the digital rights management software (DRM).  Other new features of note include updated touch controls, Wi-Fi sync and recorded TV content.  The Zune 80GB is said to be 1/3 the size with a 3.2 inch screen.

But here's the big news!

The Zune v.1 (yes - the old 30GB one) is getting the new Wi-Fi syncing, it's getting the updated sharing features, the new codecs, podcasts, and the new Windows Media Center TV on the go.  Hell - yes!!  This is a compelling story...

Tim posted something on Facebook recently (I think in the last couple of nights) hinting that some big news was possibly coming: I assumed it would be Microsoft related.  So this morning I looked through my feeds and both Scott Guthrie and Heuer dropped "it" in public.  The .NET Framework source code will be released under the Microsoft Reference License later this year. 

Holy c%@#!  Debugging without having to use Reflector is significant: yes, some of us have to do some spelunking from time-to-time.  And - well Scott says it best towards the end of his announcement:

Having source code access and debugger integration of the .NET Framework libraries is going to be really valuable for .NET developers.  Being able to step through and review the source should provide much better insight into how the .NET Framework libraries are implemented, and in turn enable developers to build better applications and make even better use of them. 

Great job developer division! 

Take a look at this file

Should have gotten the word out sooner, of course, but better late than never - right.  Bennie and I've been cramming a bit: it's been an exercise in designing an enterprise .NET 3.x reference architecture in about a week and a half while juggling the clients and family commitments.  Well - truth be told the: I'll have to do something special for the wife and boys after this...

What Are You Talking About?

Bennie and I are presenting the Building an Enterprise App with .NET 3.x  from the Application Design track at Desert Code Camp 2007 in Tempe Saturday morning.  As a heads-up we don't know how Acropolis and Silverlight got on the description: we won't be covering those technologies as exciting as they are. 

The track will, however, cover a vertical slice of a web based Events planning reference application.  The intent is to highlight how to go about tackling an enterprise or mission critical requirement given the mix of new technologies and tools in .NET 3.x and Visual Studio 2008 and to discuss recommendations for designing, developing and maintaining such applications.  We'll specifically discuss Windows Presentation Foundation (WPF), Windows Workflow Foundation (WF) used as a controller for the user interface as well as to control and represent the "long-running" sometimes complicated process, Windows Communication Foundation (WCF) which provides a message based services infrastructure that is easily extended, maintained and scaled, and the brand new .NET 3.5 feature Linq (Linq to SQL in particular) as the resource access layer within the service architecture. 

This reference architecture will be the subject of subsequent posts on Foo Theory and we've discussed, perhaps, collaborating with the broader community once it matures. 

Bennie will also be presenting (or teaching as they call it) two other tracks this weekend:

• The new C# Features behind Linq - Saturday
• Intermediate PowerShell - Sunday

Hope to see you at code camp.  It looks like the tracks are close to capacity so if you don't make it stay tuned at Foo Theory for more news. 

So I went to generate a service proxy using Orcas Beta 2's svcutil.exe and received the following error. 

After a quick web search it turns out that the this is a known Beta 2 svcutil.exe bug.  The MSDN forum moderator (LorenzP) describes that the svcutil.exe is not signed correctly and provides a quick and easy work around (also included here).

1. Open the VS command prompt.
2. Browse to the bin directory of the windows SDK.
   "C:\Program Files\Microsoft SDKs\Windows\v6.0A\bin" in my case.
3. Type: sn -Vr svcutil.exe

Good to go. 

The WCF Configuration Editor doesn't show-up in the right-click menu for WCF Service Host app.config files when the project wasn't created using a WCF project template.  You can, however, wire it up given a quick trick.  From the Tools menu choose WCF SvcConfigEditor and immediately close the editor.  Until you close the project the WCF Configuration Editor context menu will now be available when right-clicking the app.config. 

Why?  By design?  Dunno...

Follow-up

By design?  Yup.  Why?  VS startup performance.  Found this Indigo Forum post after initially publishing. 

As a heads-up this post is non-technical and a bit raw but I just had to let it out. Turn away now if you’re turned off by that.

The Story

With the exception of when in a dessert I absolutely can't stand vanilla smelling stuff: soaps, candles, fill-in-the-blank. To me it's heavy, perfumy (is that a word?), old smelling: it's just not clean. 

So we took the kids up north today to have some fun. When we got home I started to take out the trash. I pulled the bag out of the can, started to cinch the ties and was out-of-the-blue assaulted by the curse of a day’s worth of a 5 ½ month olds diapers (not many really) mixed with the smell of trash and the new vanilla scented trash bags my wife bought. OMG! It may sound crazy but I can absolutely handle, as all dads and mommies must, that trash smell minus the vanilla scourge: but when together – spew!

The new 9-zillion-count case of vanilla scented trash bags will be replaced tomorrow never to return .

So I've been using Twitter lately not only as a public instant messaging tool but because it also updates my Facebook feeds: do I still need Twitter?  I could simply use Facebook Notes and not be limited to hyper-micro content.  And Facebook can be posted to using my windows mobile phone (via browser and/or text message) just as with Twitter.  One thing keeping me from going Facebook exclusively, though, is the inability to post notes publicly. 

Notes is Facebook's Twitter-like application.  Not that I have a significant number of Twitter followers, I don't, but that's not the point.  I publish on Twitter for myself as an online journal, for friends (which is why I wired it to Facebook) and in many cases to communicate ideas, places and experiences for public consumption.  If someone were to search the internet and find my micro-content interesting or helpful - good.  If I meet interesting people at the same time - cool. 

The question's further complicated by the fact that you're limited to 140 characters with Twitter.  I couldn't, for instance, publish this post using Twitter – I could, however, using Facebook Notes.  I know, I know - that's the point of micro-ya-da-ya-da versus blogging (I totally get it!) and I've embraced it like many other Twitter users.  Consider, though, that this post started as a Facebook note communicating a quick idea (only the first paragraph) as an experiment in FB Notes versus Twitter.  I reconsidered because it would only be seen by friends and only then it grew into this paragraph and beyond.  Hmm - walled garden you say?  Sure - but not all bad at the same time.  In fact, I like that I can configure content privacy in Facebook. At the same time I'd like the ability to make some of it public too.  In my mind this should be one of Facebook's highest priorities. 

What about the others? 

So, why do I like Facebook over other such sites?  It, at least for me and my friends, hasn't been a look-at-how-many-people-I'm-connected-with or a creepy-feeling singles portal like MySpace.  As for LinkedIn: I seem to have more LinkedIn friends than Facebook, at least to now.  I do use it though - even recently.  It's just not social networking (its professional networking) so perhaps it's not fare to compare.  Live Spaces - well it is just way late and, though, the Live team seem to be rushing to compete (and with good ideas) they need to deliver. Cohesion. Perhaps, an acquisition .  I'm still holding out hope here. 

I've held my comments on Plaxo as I consider it an area of special consideration.  I've got a lot of colleagues and friends that use it but many of us are allowing our data to get stale after a number of years of good service.  I'll share a long-time pet peeve with you: I absolutely abhor that Microsoft still hasn't nailed synergy, as they have with most other offerings, between the premier Outlook client and Live Messenger contacts, in particular.  Plaxo went half way there in the past: it offered a way to have each of us update our own contact information as we move from gig to gig - when our friends synchronize Outlook and Plaxo they get the updates automagically.  This is a wonderful tool for those of us transient types (consultants and IT professionals).  Enter Live Contacts, a step in the right direction - but where's the Outlook integration?  Argh... 

This being said I really see Live Contacts (the API included) and CardSpace as real opportunities.  In particular, among disparate tools like Facebook, Live and others like Community Server.  Based on some of Dare Obasnajo's recent comments and provided they stay the course we may see some good stuff here out of the Live group.  And I too hope that Facebook opens up and considers an open id like initiative (CardSpace(?) - even better!): though, I for one won't be holding my breath. 

Some closing thoughts

I'm gonna stick with Twitter and I certainly won't be giving up Facebook anytime soon.  I'm waiting for Facebook on the other hand to deliver on a couple of critical areas: 1) the APIs need to be expanded upon.  They're extremely limited at this point but a good foundation has been laid.  And 2) keep up the walls but provide users with doors that they can keep open.

Now if I can get more of my friends, family members and colleagues who've grown callused from those other experiences to try Facebook the semi-occasional email, phone calls and absolutely rare get togethers won't be all we have. 

I intended to post about this a couple of months ago and am finally getting around to it.  Good news though - my un-posted solution was already in the pipeline, it seems, and multi-targeting is now feature complete.  

Supported multi targeting in VS 2008 (Orcas) is no small thing: it's huge to customers with a mature or large codebase. 

Some History

Jomo Fisher's post was out there as early as Widbey Beta 1.  Though others provided additional insight over time we've been using a modified version of Jomo's targets file since Widbey Beta 2 at a long-term client to target the .NET Framework 1.1: yes - unsupported multi-targeting in Widbey.  Since, we've leveraged the targets file at a number of other clients: in particular where Microsoft Dynamics CRM 3.0 is used as callouts and workflow components (CRM business logic extensions) are required to be built to the .NET Framework 1.1.  There were certainly gaps with the approach like ASP.NET yuck but for middle tier components it worked fine in most cases. 

Widbey "multi-targeting" allowed this particular client to uptake VS 2K5 tools and the .NET Framework 2.0 early but to cross-compile down to the .NET 1.1 when needed.  This was significant - we had a large development team: MCS consultants (including Statera resources), body shop contractors and FTEs.  We implemented TFS in earnest (at Widbey Beta 2 as well) including continuous integration (product, unit test and package and deployment team builds).  It allowed us to almost entirely eliminate VS 2003 projects from regular team builds as we had little or no need for 1.1 except for "thin" (ahem) callout and workflow components that ultimately called 2.0 services.

So Scott Guthrie answers the next question in a post from June (like it or not - it's reasonable):

What about .NET 1.0 and 1.1?

Unfortunately the VS 2008 multi-targeting support only works with .NET 2.0, .NET 3.0 and .NET 3.5 - and not against older versions of the framework.  The reason for this is that there were significant CLR engine changes between .NET 1.x and 2.x that make debugging very difficult to support.  In the end the costing of the work to support that was so large and impacted so many parts of Visual Studio that we weren't able to add 1.1 support in this release. 

VS 2008 does run side-by-side, though, with VS 2005, VS 2003, and VS 2002.  So it is definitely possible to continue targeting .NET 1.1 projects using VS 2003 on the same machine as VS 2008.

I'd add one other consideration here and it may be tough to swallow: VS 2003 is not supported (but has been reported to work) on Vista.  Ouch!  Bottom line - get your clients to move along. 

Beta 2 - "Multi-targeting is Feature Complete"

So as I now have Beta 2 up and running I intended to follow-up on this post and get it out there.  I, again, opened the Community Server 2007 SDK solution to see if there was any change (my unpublished draft is below for context).  Hey - the conversion was different: it obviously upgraded the solution and project files but it didn't change the target framework as it did with Beta 1.  I like it - this'll work! 

Our clients will now be able to use the new toolset and features and be assured that existing projects won't incur unexpected test burden as a result.  In addition, they'll be able to use the tool to upgrade their existing projects and target netfx3.x as they get approval, as needed and when they have bandwidth (phased not forced).  Solid! 

Below is my original draft (couple months old):

So to test multi-targeting I attempted to open the Community Server 2007 SDK solution (Beta 1). 

I 'd assert that multi-targeting would be most often used when opening and maintaining existing solutions (created w/ VS 2K5).  I mean if you're using Orcas to create new projects you're likely to target the latest .NET framework - right(?).  I know - I know: this won't always be the case, but come on - 80-20 or 50-50 at least.  As consultants we're almost always using new (even Beta) technologies and need all the help we can get to "grease the wheels" for adoption. 

When the conversion wizard completed I had a considerable effort ahead to get things back to working - it not only upgraded the solution and projects but each of the projects target framework - hmm.  Ultimately, I was forced to make a decision to either make it all work with a new netfx version or go into each project and re-target 2.0.  The later approach requires changes to references and undoing the code changes (backups - you say). As this was an academic endeavor - fine.  However, clients would, and certainly have, delayed uptake as a result of forced test burden on larger projects. 

Proposed Solution

Simple - during the conversion wizard prompt the user to keep the same target or to upgrade.  

Technorati tags: .NET Framework, Beta Review, Community Server, Tools, Visual Studio

First, I want to welcome Shane Bishop the newest Foo Theory blogger!  Shane is one of those band members that worked at ********* with us back in the day but I didn't actually meet him until he came on at Statera: you know - different clients - we get busy...  He's a good guy, he knows his stuff and we look forward to his involvement.  Take a look at his Foo Theory blog here. 

So most of the people reading this blog are likely to have come across issues with prior Beta product upgrades.  In particular, when upgrading previous Visual Studio betas you would seriously consider re-imaging your machine.  In this post I'll share my experience of upgrading my Vista laptop which side-by-sides VS 2005 and Visual Studio Orcas Beta 1 to VS 2005 and VS Orcas (2008) Beta 2.  I'll point out that I, like many other, have a considerable amount of VS 2K5 extensions installed that I still rely on.  The Beta 1 readme indicated that you should be able to install later Beta versions but will require a repair to VS 2K5 after uninstalling Beta 1.  This indicated to me that the DivDev folks expected for the upgrade to be a reasonable undertaking and that they had some confidence that it would work. 

In general I'm less exposed than some as I didn't install a whole lot of add-ins like Orcas Silverlight tools and Orcas ASP.NET Futures.  I'd think this pretty typical, though, as there's a lot to digest in the Orcas line-up already.  I did, however, install (or configure) ReSharper (R#) for Orcas but I was familiar with the needs here as the JetBrains folks have been pretty helpful. 

In this post I'll track my upgrade experience start to finish: good, bad and ugly.  So let's get started: 

An initial lay-of-the-land looks something like the following (Programs and Features sorted by date).  Hmm - don't remember installing all that! 

Ok, Ok - yes, it took me a week and a half to pull down and install it.  Cut me some slack: I have a life too!   .. So ..

1)   Uninstall MSDN Library for Visual Studio Codename Orcas
2)   Uninstall Microsoft Visual Studio Codename Orcas Team Suite - ENU

Take a look at the installation/Uninstall components below.  Hey those look familiar. 

Below are the result of step 2, again, showing Programs and Features. 

Hold on: one thing stands out and it was installed on 5/12 the day I installed Orcas Beta 1 - Microsoft Visual Studio Web Authoring Component.  As I don't remember installing any VS 2005 or Orcas components that day (a few months ago) I did a Google search.  I found a post by Rick Strahl pointing out that he'd had a problem with the installation of Beta 2 as a result, ultimately, of not uninstalling this component.  Thanks Rick for providing a heads-up on this... 

3)   Uninstall Microsoft Visual Studio Web Authoring Component

As a note: Scott Guthrie's (ScottGu) blog confirms the Orcas Beta 1 release notes: you should run a repair on VS 2005 after uninstalling Beta 1. 

4)   Repair the VS 2005 installation 

The VS 2K5 repair ran without incident though it did take quite a long while to complete.  Make sure you don't need your machine for a while when you kick it off.  I'm one of those anal types that don't like to have a bunch of stuff running during critical installations.  Admittedly, though, I had Outlook up at times. 

Testing after the repair revealed that all of the VS 2005 extensions installed previously were working as expected including ASP.NET AJAX, Guidance Automation Extensions/Guidance Automation Toolkit, Enterprise Library and Service Factory.  There were two exceptions but not a big deal as I can re-install them later: ReSharper (as mentioned above) and Regionerate.  I'll provide an update if I find other issues after I publish this post. 

5)   Install Visual Studio Orcas Beta 2 (or 2008 Beta 2) - I chose to do a Full Install

During the installation Microsoft .NET Framework v3.5 required a restart of the machine.  In addition, the following warning was generated. 

After researching the above warning the only relevant information I could find was in the Orcas Beta 1 Readme where it mentions the following (at section 1.1.15): "Installing Orcas on Vista may prompt to close running applications such as Machine Debug Manager."  The resolution in the readme is to select Ignore and continue with the installation.  BTW: the Visual Studio 2008 Beta 2 readme does not mention this explicitly but does suggest killing all applications in the warning list.  My suggestion: review them both and "choose wisely".  So after killing 1088 and Ignoring MDM - I march on.

In the end the install was a long one but seemed shorter than the VS 2K5 repair: actually fell asleep during the repair last night.  So the installation went as expected from here and resulted in a successful install.  Almost there.  Almost there.  Stay on target ... 

Oh @$%! - did I just nerd out in public?  Pull it together - Matt.  ... So ...

You'd be crazy to not install integrated MSDN documentation IMO.  As there's a lot to uptake with Orcas, .NET 3.0 and 3.5 new designers, ya-da-ya-da: MSDN documentation will be a launching point.

6)   Install Product Documentation - optional (MSDN)

Finally, there are some follow-up tasks.  Scott Guthrie provides additional context on these steps in his VS 2008 Beta 2 post (in the section "Important Installation Notes - PLEASE READ!"):

7)   If you have existing projects using ASP.NET AJAX 1.0 and built using VS 2K5 extract and install the script described in step 3 of the Beta 2 installation instructions.  More information on the script can be found here.  Note: the script must be run as administrator. 
8)   If you've installed previous version of VS Orcas/2008 (CTP/Beta 1) reset VS settings by executing "devenv /resetsettings" from the VS 2008 command prompt. 

That's it: you're ready to roll! 

So I've been using Orcas now for a few months almost exclusively for EDD and POC development.  I'll be posting over the next bit on my experiences - good and bad.  Heck - good and wish list.  All-in-all it was a smooth beta upgrade that most should be able to live with: thanks VS team (DevDiv) for taking the time. 

Have fun! And hope this helps...

Shane and I were talking about how I wired my Foo Theory weblog posts to my Facebook profile yesterday.  It's actually fairly easy to do - so here's how it goes...

1.  Login to Facebook and from your Profile menu or from Applications choose Notes.

2.  On the right under "Notes Settings" choose "Edit import settings".

3.  Enter your weblogs RSS feed Url and save.

You can visit my Facebook public profile here.

So Social networking apps and services are maturing and proliferating at the speed of light.  The new kid on the block, from Kevin Rose of Digg fame, Pownce has some promising features and for an alpha product is getting some great press (billed the "next" Twitter killer).  As in the early days of Gmail Pownce is providing new accounts by invite only right now and as a result you find craziness like invites being bid on at eBay.  I'd hold on this - Pownce is no Gmail. 

Thanks to Tim, though, I've been checking out Pownce tonight and I'm initially impressed with features like file sharing, links, and shared event calendars: none of which Twitter provides - "yet".  Pownce also has a simpler reply and directed post implementation and larger message size.  The Twitter implementation, however, still provides something that's key in my mind to this type of networking - the ability to post via text message (mobile integration).  There's something to be said about micro-posting while at an event, some landmark or simply while not at a computer.  I'll be interested to see how the Pownce-Facebook application and Twitter mashups will play in proliferation: integrated experience is important. 

Hold the presses

With facebook's implementation of mobile text posting recently and with facebook clients emerging (nothing too exciting yet) Pownce and Twitter may already be "running on fumes".  I'd be interested to hear what others opinions are on this.

The Mothership's Live groups (Messenger, Hotmail, Groups and Spaces) need to take note.  It's crazy to me how disjointed these offerings still are.  It'd take me a lot (no a lot) to give up on Live Messenger but I'd really like it to be my one client for IM, micro-posting (public and targeted), contacts, calendars and an access point into things like facebook status, pokes, ya-da-ya-da... 

It'll be interesting to see how the social networking landscape looks in 6 months/year. 

Pownce Invites

So I have 6 Pownce invites - if you want one contact me.  I'll send them to the first 6 that request them: if you don't get a reply/invite assume they're gone.  Please do comment to this post if you'd like but (heads-up) I won't be monitoring comments for the invite requests.

I'm one of those OCD-developers that has to have my source artifacts formatted and organized for readability.  So when I got a glimpse of Regionerate I was thrilled to see that there are like-minded folks in the open source community helping those of us out. 

This kick-$%@! simple open source add-in for Visual Studio (2K5 and Orcas) and #develop applies automatic layout rules to your C# code.  The site has a set of tutorials that explain how to use the tool, how to start applying your own custom rules (code layouts) and how to integrate it with NAnt to run and apply your rules during automated builds.  Note to self: getting it to run with Team Builds (WIX) would be cool.  The site also provides a gallery where people can upload custom layouts. 

The creator Omer Rauchwerger has done a great job with this one.  A+++ recommendation here...

Clinton provided some feedback to the prior post so I'll reply here and will continue to explore the question aloud. 

Clinton Jones said:
Out of the gates I would say that Agile and scrum are NOT a good approach for over-arching ERP and CRM implementation HOWEVER it is a good approach for Business Warehouse/Data Warehousing (in fact almost unavoidable) and is also probably not a bad fit for UI development for ERP systems however if you need to set up the standard modules and fundamental ERP functionality it is almost impossible to do this without going through a standard waterfall approach.

I suppose what I'd point out, and perhaps what Clinton refers to, is that ERP implementations require significant analysis on business processes as they relate to the package's implementation of a particular requirement/module. This isn't necessarily the output normally associated with agile/scrum sprints (thinking while writing: is this really something that scrum/agile discourages though?). I submit that you need to do significant analysis when building custom applications too: even if you're using agile. Perhaps not to decide how to use (or not to use) existing functionality but so that you don't end up with a mess that doesn't fit together, doesn't perform well, doesn't provide production/management support, doesn't scale, ya-da-ya-da. I'd also say that BI and UI development are not the only places where scrum/agile have proven their worth.  Bennie and I both come from a commercial software architecture (tier 1 supply chain) space where scrum has by many accounts made a positive difference. 

I say it often, and true – I believe: the business of doing (core) business is generally the same. But, the package you decide on certainly has its own realities, lest there'd be only one. Ok, don't point out that Oracle SCM; for instance, is the "one" as this would certainly turn into a religious debate (Dynamics AX ;-)) and that's not the intent of the conversation. If you invest in a canned package you should have made that decision based on the fact that some percent of your business processes can be facilitated by that package out-of-the-box so-to-speak. Does this make implementations more difficult? Not inherently. Does this affect the decision to use agile/scrum for these type of projects categorically? I've not come to that conclusion yet.

I'm not willing to assert that a waterfall approach is always necessary. Ok – let me get out in front of the next obvious rebuttal: there are significant costs associated with specialized skills such as ERP consultants (or FTEs) and implementations – so defining what is to be solved when and estimating the cost of doing so is of critical importance to the folks signing the checks. Right!? Scrum, however, does not impose that you don't provide a plan including cost estimates. Nor does it edict that baselining and tracking actuals against the plan isn't done. The value or accuracy of the estimate, however, certainly comes down to the experience the consulting or internal organization doing the planning/implementation.  Not to mention the ability (or investment) by the client to provide enough detail, in some cases, to derive a solid estimate. But these reality don't change with the adoption of either project management approach.

In addition, many (no the vast majority) of these projects, though primarily implementation, also derive some custom requirements: be they out-of-the-box customizations (supported), bolt-on or invasive, UI, BI or integration (for instance). So to this point: those custom requirements could and should be done, in many cases, using scrum/agile project management techniques. Follow-up: experienced ERP/CRM consultants should warn (no scream) against invasive changes as rev-locking the organization, as a result, is a significant problem.

But to bring it back: does the fact that the output of each iteration, say if it were early and often, is in some cases analysis and in others some partial implementation through the life of the project a bad approach toward driving through your requirements? I don't think so. Is embracing change and other agile principles as important in an implementation project? It feels like – yes: why not?

So there's risk associated with painting yourself into a corner, not exclusively, for expensive canned package. However providing a Dev Sandbox environment should reduce the need to hit a home run at each iteration and provides a way of iteratively tracking towards an end deliverable as in the case of custom app dev. So, why can't agile's inspect and adapt techniques be applied given this?

I believe I'm leaning here. But I continue to pose the question: is agile/scrum a good fit for ERP or CRM implementation projects?

AZGroups is bringing Scott Guthrie back to the valley "for a little reMIX event" at the Symphony Hall in Phoenix, AZ, Wednesday June 27th (9am-5pm).  1-day.  Free. 

ScottGu is a prolific blogger and uber-GM within Microsoft's Developer Division and Stefan is a manager on the web platform and tools team at Microsoft.  If you missed MIX '07 - don't miss this one! 

Visit AZGroups site to register for the event and to see bios for Stefan and Scott.