Under the Hood: Insight from the Ironworks Technology Group

SharePoint on the iPhone: Implementation

In my previous post, we introduced the Ironworks SharePoint client for the iPhone. This client performs some basic tasks:

• Primarily, all data should be rendered natively on the iPhone if at all possible. This is not an "app" that does nothing but opens up the IDubHub website in Safari.
• The user should be able to browse through all the lists on the site and mark the ones they like as "favorites". These favorite lists would then be immediately accessible on one of the main screens.
• The user should be able to browse through the Ironworks employee list and look up details on employees. Furthermore, contact information should be able to be added as new contacts into existing contacts on the device's address book. In the case of the contact already existing, the new data should be merged into the existing contact.
• Basic search should be possible.

When I first sat down to write the app, I wasn't sure where to start. I've been so spoiled by the luxury of the .NET SharePoint object model. However, in order to make a completely language-agnostic SharePoint client, I would have to query SharePoint in a way that was totally new to me: by using the SharePoint web services.

Case Study: Implementing a SharePoint 2010 Intranet

As a Microsoft National Systems Integrator and a participant in the SharePoint 2010 Partner Evaluation Program (PEP), Ironworks has been hands-on with SharePoint 2010 since 2009. Most recently we launched a new corporate intranet for one of our clients here in the Richmond area. 

 

The Client

Headquartered in Richmond, Virginia, Luck Stone Corporation is one of the largest, family-owned and operated aggregates companies in the U.S. and is one of the largest producers of crushed stone in the nation. Luck Stone is also a leading architectural stone and tennis court supplier with operations in Virginia, North Carolina, and Maryland. 

 

The Business Challenge

With a business model spanning 4 distinct brands and more than 30 geographically dispersed offices and an enterprise content management strategy that was entirely dependent upon file shares, public folders, and email, Luck Stone required a platform for both content management and collaboration that could help the organization better leverage its existing data and expertise.

 

Luck Stone’s overreliance upon public folders had resulted in terabytes of document content that were completely unclassified and difficult to manage. The organization had no system in place for managing content review or expiration. IT resources were required to provision new folders, manage permissions, and assist with finding content buried in nested folders. 

 

Luck Stone was also in need of a platform for disseminating corporate news and announcements that could drive the organization’s values-based culture. As a communication tool, email was not meeting all of the organization’s needs. Information disseminated via email was not publicly discoverable or available for reuse through other communication channels. Additionally, the email communication model was not conducive to the two-way idea sharing that management hoped to foster with associates.

 

Finally, Luck Stone identified a need for implementing a consolidated platform for hosting line of business systems. Luck Stone envisioned a “one-stop-shop” where users could access business reporting tools and other back-end systems.

It is significant to note that Luck Stone had tried and failed to implement an intranet portal twice prior to engaging with Ironworks. Numerous factors had contributed to the failures including functional limitations and a lack of end user adoption.

 

The Solution
Luck Stone partnered with Ironworks Consulting, a Managed Gold partner and National Systems Integrator, to develop a corporate intranet built upon SharePoint 2010. Leveraging Enterprise edition features as well as Office Communication Server; the solution delivers a unified platform for content management and collaboration while also paving the way for significant business process improvements.

 

Implementing a Custom Taxonomy

In the area of content management, the solution leverages the new managed metadata service, document sets, and content type publication to apply a custom taxonomy developed by Ironworks. The content types and site columns implemented were the end result of a comprehensive taxonomy definition phase that included an intensive review of the organization’s content. The intranet capitalizes on SharePoint’s existing content management strengths to deliver versioning, information management policies, records retention, and publishing workflows.

 

Leveraging Managed Metadata

Delivering Findability

Ironworks leveraged SharePoint Search to build a comprehensive information access solution to reduce the time to find information and increase productivity and end user adoption. A combination of the Search Core Results web part and Ironworks’ suite of custom-developed search controls were used to deliver a robust associate directory, a directory of vendor companies and a custom utility for managing frequently asked questions. Additionally, standard configuration, the new Refinement Panel web part and use of federation services were leveraged to deliver a much acclaimed Search Center experience.

 

Luck Stone's SharePoint 2010 Search Center

 

Leveraging Collaboration Features

Luck Stone’s goal of developing greater interaction and discussion among employees was enabled by developing a cohesive use of Blogs, discussion boards and wikis, all branded with a unified look and feel. In addition to these features, the new capabilities included with the My Sites as well as the tagging and commenting integration have proven especially popular with Luck Stone associates. Team sites have been leveraged to deliver a platform for project-centric collaboration, with unit-specific team site templates developed for targeted user groups.

 

Customizing the Look and Feel

Implementation of a custom look and feel for the Luck Stone site was accomplished through the use of SharePoint Designer 2010. Using a custom control and varied cascading style sheets, Ironworks delivered audience-targeted brand-specific design for each of Luck Stone’s core business units. In this way, associates from individual business units experience a different look and feel when accessing the same site. The solution has played a key role in fostering the high degree of end user adoption that Luck Stone hoped to achieve.

 

Audience-targeted look and feel

 

Enhanced Functionality through Custom Solutions

Luck Stone and Ironworks leveraged Visual Studio 2010, and took advantage of its dramatically enhanced support for SharePoint development. New SharePoint project templates helped to streamline the development of custom web parts, features and workflow.

The custom web parts developed included an Inbox Link control that ties into the organization’s OWA server to surface a link to the logged on user’s mailbox and a count of the user’s unread emails. Also developed was a suite of connected web parts that deliver weather, traffic, safety incident data, and driving directions specific to each of Luck Stone’s office locations. Embedded in the landing pages for each of the location web sites, the controls display the location-specific information based upon a connected page field filter. The location data is all managed centrally in a SharePoint list at the root of the primary site collection.

 

Issue Management interface for the Ironworks Newsletter Utility

Ironworks developed several custom features as part of the implementation. Ironworks’ Analytics feature adds a custom interface for managing Google Analytics integration to the intranet’s site settings page. Through the combination of a custom control embedded in the site masterpage and a user interface for setting Google application ID values at the site level, the solution facilitates granular management of analytical reporting. Ironworks also developed a newsletter utility that deploys as a custom feature. The custom application stores all newsletter content, including attachments, in SharePoint lists and comes complete with an easy to use administrative interface that supports different levels of access for authors and editors. The utility scales to support multiple newsletters and includes the ability to send out newsletters as HTML-formatted emails. This feature has played a significant role in Luck Stone’s updated communication strategy.

 

Audience targeting plays a major role in ensuring the intranet’s relevance to a highly segmented user base. To streamline the audience targeting process Ironworks developed a custom workflow which when associated with a particular list or library will automatically target the content in the list to a specified audience.

 

Custom Page Layout Content Type used to deliver FAQ functionality

 
 

Out-of-the-Box Business Process Improvement

The initial release of the corporate intranet includes the deployment and configuration of Excel Services which is used to surface key quarterly financial data to business users. The subsequent release of the Luck Stone intranet will include the development of a full complement of KPIs and dashboard components that leverage the PowerPivot and PerformancePoint tools now integrated into the Enterprise edition.

 

Delivering the Solution and Fostering End User Adoption
Strategy sessions related to the project began in August of 2009, with development beginning on the beta release of SharePoint 2010 in the fall of 2009. From the initial kick-off meeting through to final delivery, Ironworks worked with the Luck Stone team to build the case for end user adoption. Through inclusive strategy sessions with business stakeholders, the development of an internal “roadshow”, and day-long company-wide training sessions, Luck Stone and Ironworks paved the way for broad-based user adoption. With the release of the RTM build of SharePoint 2010 in April and the subsequent completion of content migration, the Luck Stone corporate intranet went live at noon on June 22, 2010, just 2 months after the RTM release of SharePoint 2010.

The Results
Early in the project, Luck Stone identified several key areas where they expected to realize a return on their SharePoint 2010 investment. These areas were as follows:

1)    Infrastructure Cost Savings: The transition away from shared drives and public folders built on top of fixed infrastructure to a virtualized environment that leverages SAN storage achieves a more flexible, scalable, and cost efficient solution for document storage.

2)    IT Resource Productivity: The introduction of self-service site creation and a role-based security model frees up IT resources that had previously served as the gatekeepers for provisioning public folders and managing access rights.

3)   Process Automation: Moving the dissemination of key data (quarterly financials) out of manually generated reports and into Excel Services (and eventually a comprehensive BI solution) will speed the organization’s ability to identify and respond to evolving business and financial trends.

4)   “Findability” Improvements: Implementation of a comprehensive enterprise search solution now allows users to find content that was previously lost in nested public folders and shared drives.

5)    Cross-Business Collaboration:The intranet now provides a means for discovering resources and expertise across the organization. Associates from different business units can easily share ideas and identify areas for mutual business improvement.

6)    Cultural Cohesion: Through the intranet, the business can better coordinate companywide communications, foster brand identity and build corporate loyalty.

Fixing User Profile Synchronization Error In SharePoint 2010 (An Update Conflict Has Occured, And You Must Re-Try This Action)

Symptoms:

When you try to start User Profile Synchronization (Full/Incremental) or start the "User Profile Synchronization Service" found under "Services on server" in SharePoint 2010, you receive an error "An update conflict has occurred, and you must re-try this action. The object UserProfileApplication Name=User Profile Service Application was updated by [User]"

SharePoint on the iPhone: An Introduction

Recently, I had some time between projects, and wanted to do something productive during that time. Ironworks has had an interest in mobile development for a while now, and I had been playing around with iPhone development at home.

So, I spent a little time creating an iPhone front end to our intranet portal, which we call "IDubHub".

How to create an External Content Type in SharePoint Designer 2010 using Business Connectivity Services(BCS) and fix issues that arise on the way

In this walk-through I will explain how to use SharePoint Server 2010 Business Connectivity Services(BCS) feature to access external business data (SQL Server 2008 in this example). This simple step-by-step will also help you fix the issues that you might encounter on the way.

Migraine Not Included: Removing Objects from a SPPropertyBag

In working on a site feature that we'll be using regularly, I was storing some small bits of information on a web's property bag. When the feature's event receiver gets called during feature deactivation, I wanted to be a good boy and clean up my mess. That is to say, I wanted to remove the item I added to the property bag.

Naturally, my first thought was to use the SPPropertyBag.Remove() method, which is inherited from StringDictionary. However, every time I tried that, and then called SPPropertyBag.Update() to commit the changes, the key/value pair came back.

After doing some digging, it became apparent that this was not an unusual thing. I eventually found this comment at the bottom of an MSDN page:

In some cases I have found that SPWeb.Properties.Remove("MyProperty") does not work even if you call update method after. In those case when I set the property value to null , it will be removed from the collection and SPWeb.Properties.ConstainsKey("MyProperty") does not return “True” which means it is not there.

Doing precisely that worked for me--by setting the value of my property to null, I was able to properly clear it. Hopefully this will save one of you from nearly causing yourself a migraine trying to explain the inexplicable, like I did!

Lessons Learned: Running Windows Server 2008 R2 VMs on Windows 7

So, you want to set up a virtualized SharePoint 2010 farm on your Windows 7 host.  I’m sure you’ve done your homework and secured a beefy host machine (minimum of 8GB RAM, multi-core 64-bit processor, and tons of disk space for your VMs on a dedicated spindle), and you’ve readied your install media for Windows Server 2008 R2, SQL Server 2008, and SharePoint 2010.  Now you’re ready to download your virtualization software to get started.

A word of warning – you won’t be using Microsoft Virtual PC for Windows 7 to host your VMs.

C# Tricks Hiding in Plain Sight

While I do consider myself to have a pretty solid knowledge of the C# language, once in a while I'm shown something new which amazes me. Other times, I discover something on my own. Yet other times, like yesterday, I rediscover an old trick I had long since forgotten about.

As I'm developing, I frequently revise entire blocks of code as I realize that a block of code could be more efficient, more readable, or both. When I want to remove a block--without actually removing it--I typically don't comment it out. Even though highlighting the block and hitting CTRL-K followed by CTRL-C will automatically add // to every highlighted line, I don't like that approach.

Instead, I tend to do something like this:

foo.DoSomethingSmart();
#if false
foo.BreakEverything();
#endif
foo.DoSomethingSmarter();

Or sometimes:

foo.DoSomethingSmart();
#if true
foo.DoSomethingSmarter();
#else
foo.BreakEverything();
#endif
foo.DoSomethingSmarterStill();

I don't know why I prefer that. Perhaps it's because I cut my teeth on C++. Perhaps it's because I subconsciously feel like the compiler really ignores conditionally compiled blocks, instead of just sorta ignoring comments. Either way, it's what I typically do.

But what happens when you want to do that for a whole method?

I was creating a SharePoint feature receiver, and I had a method--the first one below--that was logging debug-style messages to the Windows event log:

void Log(string message)
{
    Log(message, EventLogEntryType.Information);
}

void Log(string message, EventLogEntryType type)
{
    if (!EventLog.SourceExists(EVENT_LOG_SOURCE))
    {
        EventLog.CreateEventSource(EVENT_LOG_SOURCE, "Application");
    }
    EventLog.WriteEntry(EVENT_LOG_SOURCE, message, type);
}

However, I didn't want the first method, which was being used only for debugging messages, to be shown in release builds. My first inclination was to leverage conditional compilation, as before:

void Log(string message)
{
    #if DEBUG
    Log(message, EventLogEntryType.Information);
    #endif
}

But then I stumbled on a much cooler and cleaner solution: the Conditional attribute.

This attribute allows you to specify that the method that has been decorated is only visible when a certain conditional compilation flag is set. If that flag is not set, the code still compiles cleanly, but the method/class in question is treated as one giant NOOP.

This is a much cleaner approach, in my mind, and more "C#-like". Thus, my method was quickly changed to be:

[Conditional("DEBUG")]
void Log(string message)
{
    Log(message, EventLogEntryType.Information);
}

I'm much happier with this cleaner code. I also quite like when I find useful nuggets like this. That's especially true for ones that have existed for a long time, but I either didn't know about or had--like this one--completely forgotten about.

If you have any fun tips along those lines, feel free to share them in the comments!

Reporting Services: Subscriptions, Linked Reports, & Default Parameters

This week I wanted to share some of my experiences with SQL Server Reporting Services and how to avoid cloning reports.  There may have been cases where you felt the urge to duplicate a report and provide different report parameters.  For instance, you may have the desire to run your report on a daily, weekly, monthly, and yearly basis.  The base-level report is similar in all cases but you need to change the report parameters to reflect the dates.  I had originally thought I could just remove all references to the report parameters having to do with begin and end dates and set up everything through a an automated subscription and set the date there. 

 

I thought that I would be able to set up daily subscriptions by specifying a begin date of =DateAdd(“d”, (-1), Today()).Date() and an end_date of Today().Date.  Then I could utilize the same report for weekly subscriptions by setting a begin date of =DateAdd(“d”, (-1), Today()).Date() and an end_date of Today().Date.  I found through trial and error that in inputting these formulas, the Report Manager would just immediately translate these dates into hard-coded dates.  For example, the begin date for the daily report would populate with 06/23/2010 and the end date would be today’s date of 06/24/2010.  This was not what I wanted, as the automation would just always send me data for that one particular day. 

Then I thought how silly would it be to clone the report for each case and specify different default report parameters.  That is when my colleague Jessica Moss pointed out I have several options around this!!!

Option 1: Data Driven Subscriptions

If your subscriber data is dynamic, you’ll want to consider setting up one of these.  It provides flexibility – you can send reports to a subscribers list that may very from week to week.  Or you may send reports out to users only if they meet a certain criteria.  In order to set up these subscriptions, you must have The Evaluation, Developer, or Enterprise edition of SQL Server.  For a step-by-step tutorial, please visit http://msdn.microsoft.com/en-us/library/ms169673.aspx?PHPSESSID=o1fb21liejulfgrptbmi9dec92.

Option 2: Linked Reports

Linked Reports are based upon an existing report but they use different settings.  These reports still maintain the base-level layout and data source properties. 

Option 3: Setting Default Parameters

A third option is to alter your stored procedure so that the dates are handled through Transact SQL code.  On the subscription side you can set a parameter to a default value such as “Weekly” or “Monthly”.  If the parameter is sent in to the stored procedure, it will know what dates to set and return the appropriate data for that date range.

In summary, there usually is a much more elegant way of solving your reporting needs whether it be through using the features available through SSRS or by varying your stored procedure slightly.

Microsoft CRM real-world application - tracking partner competencies (part 1)

Microsoft has recently released their Partner Competency Guide, which lays out the requirements for organizations like Ironworks to meet in order to be considered a partner.  Keeping track of these requirements can be a full-time job, and then keeping track of your organization’s progress within that framework can just be overwhelming.  Couple that with keeping track of basic skills such as C#.NET and T-SQL, and then blow that out beyond Microsoft’s walls and well, someone in your organization is pulling their hair out on a consistent basis.

Enter CRM (or more appropriately in this context, xRM), a platform that is great at managing these various entities, their relationships, and the reporting necessary to pull it all together.  What’s great about xRM is not only the fact that this can all be done out of the box with just a little configuration, but can then be integrated with your company’s intranet (in our case, a MOSS 2007 portal) as well as managed through a common tool within the Enterprise – Microsoft Outlook.  What does this mean?  Typically it means less time actually learning the tool and more time getting the most of out of it.

This is going to be a multiple-post blog entry as this is a work in progress.  We started off looking for existing solutions within this space.  One of the great things about CRM/xRM is the vibrant community and the solutions that are provided by that community.  Microsoft provides a tool that allows you to find partner-built solutions that work with CRM/xRM to meet your needs.  There are some great vendors and solutions to be found within this database.  Additionally, Microsoft provides a set of CRM accelerators, or add-on solutions that provide good examples of how the platform can be configured for various business needs.  These are provided free of charge and can be extended even further, or modified to meet your needs.  There are also various vendor solutions and add-ons that are provided at a cost.

Unfortunately, our needs were not met by any pre-built solution, so we began building out our prototype by mapping out our entities in Visio just as we would a relational database, which is important to note.  CRM/xRM is a great platform for managing structured data, or data that maintains relationships between entities.  In contrast, a platform like SharePoint is a great platform for managing unstructured entities such as documents, which are typically not related to other documents (which changes in SharePoint 2010 with documents sets – which is another topic entirely). 

This is important for two reasons.  First, you do not want to select a platform like SharePoint for a solution such as the one we are trying to build out – it just doesn’t make sense.  It’s important to realize this so that you are not trying to fit a square peg into a round hole.  Secondly, you should also take into consideration that CRM/xRM may not fit every need that you have, whether that need is to retain documents or to display information to end-users, which is where a product like SharePoint does fit nicely.  The point is that these two products are meant to live side-by-side, and complement each other’s capabilities nicely.  That’s not to say that CRM will not stand on its own, as it very often does, but for a more robust and fully featured solution, CRM can, and often will integrate with other platforms.

I’ve provided our relational model below.  This is really a first cut, as I expect this to grow over time.   Next up is creating the entities and relationships within CRM, which will be the next post.  Stay tuned!