The overwhelming evidence of consumers shifting to mobile is forcing brands to quickly evaluate how to incorporate tablets and smartphones into their digital strategy. Nielsen research showed that 46 percent of US mobile consumers had smartphones as of Q42011, and that figure has continued to quickly grow. In fact, a recent study by IDC revealed that more people in the United States will access the web via mobile devices than via wireline computers by 2015. The drive to mobile begs the question: what's the best path to monetize mobile?

According to research by Adobe, tablet users spend over 50% more on online purchases when compared with smartphone users. Adobe analyzed roughly 16.2 billion online transactions from 150 top U.S. retailers in 2011 for the study. Additionally, the report concluded that tablet users are three times more likely to make a purchase than smartphone users. The findings are not surprising when considering that the browser on a tablet is more similar to one a PC than on a smartphone. For example, in-line videos on an iPad do not play within a native player as they do on an iPhone. Essentially, the display size on the smartphone, which is typically 3.5" to 4.7" does not provides enough room for a user to view a website. Thus, the smartphone is relegated to in-aisle product comparisons experiences whereas the full-form tablets are used for in-home browsing and purchases.

The study demonstrates that brands need to look beyond a one-size-fits-all for mobile screens. Instead, brands need to consider how to shift visitors from one screen to the next. Beyond simply moving smartphone visitors that are browsing products on-the-go to a full website experience on a tablet or a desktop, brands can also take advantage of in-app purchases to convert visitors into customers. But in the next two years, brands should expect to adjust their mobile strategies again as the display size on smartphone devices expand. For example, Samsung is rolling out the Galaxy Note in Q1 2012 which will feature a whopping 5.3" display. As these devices become more commonplace, the line between smartphones and tablets will blur and consumer will become more accustomed to completing purchases via the mobile phone browser instead of defaulting to a mobile application or a tablet.

Stackoverflow got me the answer to this.

If you're using Custom Google Search on your website and you have a small textbox on each page that is supposed to 'pass' that query to your search page (ultimately passed into your Google search post action), you should do the following: 

Instead of the code Google provided you with to embed, take your unique key provided by Google and place it in the '***my key***') setting of the first code snippet below.  What this is doing is creating a new instance of your custom google search with your unique key and drawing out the results (*cse*) and forcing it to submit the action (execute) with your parameters ($q). 

    google.load('search', '1', { language: 'en' });

    function OnLoad() {
        var customSearchControl = new google.search.CustomSearchControl('009045124056933145342:5wmgo53sugc');
        customSearchControl.setResultSetSize(google.search.Search.FILTERED_CSE_RESULTSET);
        customSearchControl.draw('cse');
        customSearchControl.execute(search_value);
    }
    google.setOnLoadCallback(OnLoad);

On my click event for my search text box on every page, I sanitized the query (this is a very basic sanitize, you should alter as necessary to protect your site!) and appended those parameters to the query string being sent to my search results page.

private String SanitizeUserInput(String text)
        {
            if (String.IsNullOrEmpty(text))
                return String.Empty;

            String rxPattern = "<(?>\"[^\"]*\"|'[^']*'|[^'\">])*>";
            Regex rx = new Regex(rxPattern);
            String output = rx.Replace(text, String.Empty);

            return output;
        }

        protected void Button1_Click(object sender, ClickEventArgs e)
        {

            Response.Redirect(
                String.Format(
                Page.ResolveUrl("~/{MySearchPage}.aspx?q={0}"),
                    HttpUtility.UrlEncode(SanitizeUserInput({mysearchtextbox}.Text.Trim()))
                    ),
                false
                );
        }

On my search results page, I remove those parameters with the below function and pass them into my $q variable.

function getQuerystring(key, default_) {
if (default_ == null) default_ = "";
key = key.replace(/[\[]/, "\\\[").replace(/[\]]/, "\\\]");
var regex = new RegExp("[\\?&]" + key + "=([^&#]*)");
var qs = regex.exec(window.location.href);
if (qs == null)
return default_;
else
return qs[1];
}

First off, I would like to thank Joe Gill for getting me started down the right path.  In fact, I'm going to suggest that everyone click that link so you can use his helpful screenshots to navigate where to put the JS I'm about to post. 

In any case, I had not seen a full example of parsing a phone number in CRM 2011 as I had with CRM 4.0.  Since it is such a common need, I decided to make my own.  The following code will strip out all special characters, check and make sure it's a valid set of numbers, and then push it back to the field, parsed properly.  This is limited to 10-digit phone numbers at the moment, but the core structure should be easily expanded if you need different types of validation.

The nice thing is, this can point at any entity, and any attribute that holds a phone number.  I strongly recommend adding this to your standard JS library for all CRM 2011 projects.

function validatePhone(context)
{

var phone =context.getEventSource().getValue();
var sTmp = phone.replace(/[^0-9]/g, "");
phoneRegex = /^\d{10}$/;

if( !sTmp.match( phoneRegex ) )
   {
   event.returnValue = false;
   alert("Phone must contain 10 numbers.") ;
   }
else
  {
   var sTmpClean =  "(" + sTmp.substr(0, 3) + ") " + sTmp.substr(3, 3) + "-" + sTmp.substr(6, 4);
   context.getEventSource().setValue(sTmpClean);
  }
}

Today I had an issue where I had deleted a number of records from CRM during my loads of testing through Scribe.  After deleting these records, CRM updated those records to have a deletestatecode = 2.  Therefore, every subsequent scribe job would no longer work properly as the address still existed in the database.  Those of you used to CRM 3.0, there used to be a MSCRMDeletionService you could restart in your running services and it would force the job to run through all your records.  That is now wrapped in the Async Service of CRM 4.0 which you can restart.  For those of you who don't/can't restart that service for any reason, try this here:

USE MSCRM_CONFIG

UPDATE dbo.ScaleGroupOrganizationMaintenanceJobs SET
NextRunTime = getdate() -- Now
WHERE OperationType = 14 -- Deletion Service

This was found on an old archive site, posted by Aaron Elder.  Hope this helps others find it.  Be patient.  I fumed for a few minutes thinking it wasn't working but just give it some time. ;)

When all else fails.....

There are other ways going directly to the database level, but I cannot stress how bad of an idea that is and how much it should never be done.  Exhaust all other options CRM provides you before going to the database level and modifying it by hand.  'Because it would be faster' should never be a good enough excuse!!  If you have no other options of using CRM to it's full abilities and must do it by hand, something went horribly wrong and needs to be re-evaluated immediately after you clean up your data.

you're still reading....*sigh* ok but I warned you.

 CRM gurus correct me if I'm wrong, but from the understanding I have at this moment, every entity in CRM has two tables:

• [YourEntity]base - table that holds all the default out of the box attributes of your entity

• [YourEntity]extensionbase - table that holds all your 'new_' custom attributes for your entity

So if you have a 'Contact' with a 'new_foo' attribute and someone adds a new 'Contact' to your CRM instance, 

•  1 row added to your 'Contactbase' table - holds address1_line1, name, firstname, etc (all defaults)
  

•  1 row added to your 'Contactextensionbase' table - holds new_foo (all custom)
  

The BOTH have the same ContactID because they ARE the same contact.  So if you have to delete at the database level you MUST delete from the extensionbase table first and foremost.  The smartest thing to do is to first find out exactly which contacts you need to delete.  Whether its on IDs or a certain field state, get a select statement as correct and accurate as possible.  Once you have that select, I like to make a temp table of those records but this is purely optional.  The reason I like to do this is because sometimes the field I care to do my search on is a 'new_' attribute.  Where 'new_foo' = null.  But if I run my delete query on extensionbase where 'new_foo' = null, I just shot off my own foot.  The reason?  I just deleted all the rows who had any knowledge that 'new_foo' existed.  How do I know what rows I just deleted to now clear out the base table?   With a temp table, I can capture all that data and include a 'contactid' so I never am without a reference of what fields need to be deleted.

 As always, BACKUP before attempting this.  Run your delete on extensionbase.  Write down the number deleted.  Run your delete on base.  Match number deleted.  If it doesn't match, Panic.  Restore you database and refine your search until those numbers match.  Re-evaluate your process and NEVER DO IT AGAIN!

The crmsvcutil program has some lousy code in it that breaks when you are using "Claims based authentication" (which is used for Internet Facing Deployments - IFD).

This error shows up as:

" Exiting program with exception: The logon attempt failed

Enable tracing and view the trace files for more information."

The fine folks at adxstudio have provided a dll that you can use to work around the problem.  http://community.adxstudio.com/Default.aspx?DN=9a7499fb-4e9a-408c-8096-6d658f9509a2

This is specifically for the 5.0.3 version of the SDK, which is the one currently available today at http://www.microsoft.com/downloads/en/details.aspx?FamilyID=420f0f05-c226-4194-b7e1-f23ceaa83b69

Presumably MS will fix this in the next SDK release.

Put the dll into the bin directory that has your crmsvcutil, and then add to your command line:

/metadataproviderservice:"MetadataProvider.IfdMetadataProviderService, MetadataProvider" /codecustomization:"Microsoft.Xrm.Client.CodeGeneration.CodeCustomization, Microsoft.Xrm.Client.CodeGeneration"

For example, a working command line is:

C:\projects\crm2011\sdk5.0.3\sdk\bin>crmsvcutil /url:https://CRM.YOURCOMPANY.COM/xrmservices/2011/organization.svc /out:xrm.cs /u[username] /p:"[passwordgoeshere]" /metadataproviderservice:"MetadataProvider.IfdMetadataProviderService, MetadataProvider" /codecustomization:"Microsoft.Xrm.Client.CodeGeneration.CodeCustomization, Microsoft.Xrm.Client.CodeGeneration"

The root cause is the MS crmsvcutil included this code:

if (orgServiceConfig.AuthenticationType != AuthenticationProviderType.LiveId)

which doesn't account for IFD using AuthenticationProviderType.Federation

(bug found in http://social.microsoft.com/Forums/en/crmdevelopment/thread/10bfc1ca-3cac-46d9-99b5-8f997e4b1ec9?prof=required)

I found this tool out on the web.  It’s a bulk delete tool for CRM 4.0 since it doesn’t have a bulk delete option.  (As far as I’ve been told, CRM 4.0 Online does, but CRM 4.0 does not).  Anyways, you run the executable and it gives you a very simple UI.  All you need to do is create a view for the entity with all the records you want to bulk delete.

1.       Run the executable

2.       The bottom corner shows you’re not connected.  Click the drop down and ‘Create a new Connection…’

3.       You will get the connection information screen

4.       Give it all your information and click ‘OK’.

5.       You’ll now see you’re connected at the bottom of your screen.  And in the future, you will have that connection information available to you in that small dropdown as shown below:

6.       NOTE: for this to work as you want it to, you need to create a view on that entity for the records you want to delete.  If this is going to be routine, perhaps a CRM job plugin might be better, but we’ve had several instances where we loaded a CRM instance with a test data and want to dump it all, so here’s a fun tool for it.  Create your view. (I’m assuming here you know how to do that).

7.       After your view is created, enter your entity in the textbox and click ‘Retrieve Views’.  You’ll now get a list of those views. 

8.       You can schedule it to run in the future or recur but, again, I think a plugin would be more beneficial for that.  You can also have it send you an email after the job is done deleting.  It’s not perfect but it’s out there. 

Enjoy.

Posted from Amy’s Sitecore Adventures (a little late)

Breadcrumbs have been covered by just about everyone so there are lots of examples that all seem to do things a Little differently. With that in mind I’m going to keep this short with my example and two other examples I’ve found that might also meet your needs and cover the basics for just about every xslt breadcrumb example you’ll find.

The general idea: you’re at item c, in your tree the path is something like:  /sitecore/content/a/b/c and you want to display a pretty html list for a » b » c anywhere on your site.  You’ll always be dealing with the ancestors of your current item so you’ll be making use of $sc_currentitem/ancestor-or-self::… somewhere.

You’ll need to go through each ancestor item and display it, probably checking to see if you’re at the last item so you don’t display a ‘»’ after the final item. You also need to make sure to not display unwanted ancestors in your breadcrumb, being /sitecore and /content in our case.

So onto the examples!

First is the Sitecore breadcrumb xslt example: http://sdn.sitecore.net/Articles/XSL/Breadcrumb%20Example.aspx
  You will need a login to sdn to view this, but the magic is that it does a for-each across ancestor-or-self::item and then has an if statement to make sure that position()> 2 (this avoids the sitecore/content portion) and avoiding folders and there is the requisite check for position()!=last() so that we do not have the extra » after c.

Next up is Brian Pedreson’s breadcrumb example http://briancaos.wordpress.com/2009/02/09/breadcrumb-in-sitecore/
Similar to the above, however in this case he only selected items which had the appropriate template and a check to make sure the items should be in the navigation at all:
select="$sc_currentitem/ancestor-or-self::item[sc:IsItemOfType('mypages',.) and sc:fld('ShowInMenu',.)='1']"

I am certain there are many many more breadcrumb examples, but these were the first I was able to find easily to share.

Finally, my contribution to the breadcrumb party:

  <xsl:template match="*" mode="main">
    <div>
      <ul class="breadcrumb">
        <xsl:variable name="ancestors" select="$sc_currentitem/ancestor-or-self::item[ancestor-or-self::item/@template='home page']" />
        <xsl:for-each select="$ancestors">
          <li>
            <xsl:choose>
              <xsl:when test="position()=last()">
                <sc:text field="Breadcrumb Title" />
              </xsl:when>
              <xsl:otherwise>
                <sc:link>
                  <sc:text field="Breadcrumb Title" />
                </sc:link>  »
              </xsl:otherwise>
            </xsl:choose>
          </li>
        </xsl:for-each>
      </ul>
    </div>
  </xsl:template>

To satisfy the condition of not showing /sitecore or /content, I am only grabbing ancestors (or self) who have the ancestor (or is the item) that has the special template for the Home item. This excludes anything above the home item, but includes the home item too so that it can be listed. And we end up with a » b » c.

Hope this can be helpful if you are getting started with breadcrumbs!

For each of the properties in your template you can set a source for it, this isn’t always used but can improve user experience drastically when done throughout a site. The source field comes in to play whenever you are using any of the following fields: Droplink, Droplist, Droptree, File, Grouped Droplink, Grouped Droplist, Image, Multilist, Treelist, Rich text field and a number of others.

There are various ways of setting these up to achieve different results – but in general you are using the source to limit the set of items that can be used, and this requirement can also help you determine what kind of field to use. For example, if you have a Set of items all split down into sub folders and want the content editor to make use of the tree, you could use a TreeList or Drop Tree, but if you just want a set of items without the opportunity to see where those items are – multilists or droplinks are the way to go. For Images you’re generally just specifying where to look for and put the images within the media library, and for Rich text fields the source determines the type of editor to use (if not the default).

There are a number of options for setting the source property but not all can work with every field.

Setting the Root Node: To do this you just give the full path to the item you want to use as the ‘root’. This works with just about every field that pulls options and you can easily grab the path from the content pane when you select your desired item(looking at the Item Path). Ie: if you have a treelist and you only want to show the categories item and its children, you’d put in the path to that item: /sitecore/content/Data/Categories

Sitecore Query: This applies to a smaller set of fields (the List Fields), but gives a lot of power and is what I’ll mainly be going over. The fields you can use this with are: Checklist, Droplink, Droplist, Grouped Droplist, Grouped Droplink, or Multilist. You can also use the fast query: this has some limitations over just using sitecore query but gives benefits of better performance and using less memory.

Treelists are also a little special and can use parameters(I think some other fields will work with this as well, but I most often end up using the query on other items so I don’t have a ton of experience with this), you can specify the root item, what templates/items to display or ignore and which they can select, for example:

DataSource=/Sitecore/Content/Home/Root/Node&IncludeTemplatesForSelection=desiredTemplate1,desiredTemplate2&ExcludeTemplatesForDisplay=secretFolder&ExcludeItemsForDisplay=secretItem&AllowMultipleSelection=true

The above sets the root node to /Sitecore/Content/Home/Root/Node and allows the user to select items using desiredTemplate1 or desiredTemplate2, it also excludes secretFolder and secretItem from showing up in the list as well as allowing the user to choose more than one of the same item.

More on Sitecore Query

Using the query option over simply setting a root node improves the experience for your clients as well as helps to keep your data accurate. If you don’t want them to be able to pick a certain template or value or need to select something dependent on a specific axes – using sitecore query will make it possible.

In the source field, your queries need to begin with query: followed by your query. I’m going to go over a few examples I’ve found useful, but for a more detailed explanation of using Sitecore Query take a look at the Data Definition Reference, there is a Sitecore Query section that explains all the details!

So, a few useful tidbits:

* grabs all the children of a node: query:/sitecore/content/home/dessert/*  <--  the * grabs every child and / followed by text denotes the exact item name you’re looking at. So you could also mix this up a bit and return query:/sitecore/home/*/pie/* <-- this will grab all the items that have a parent pie from any item under home with the pie grandchild.

. references the context item, this can be handy if you need to find the ancestors, children or anything along the axes (if you need just the parent, use: .. ): query:./pies/* <-- if we were the dessert item, this would grab all the grandchildren with the parent pie.

// is the descendant axis – this should be used Very sparingly but can be done the following way: query:/sitecore/content/home//pie/* <-- this grabs all the items with a parent pie under home. So that would include, home/pie/*, home/anything/pie/* and so on. The fear with using this is that you’re going to return the whole tree or a whole section which might have thousands of items, so when considering using the descendant axis or any query, be mindful of the result set you will get.

@ denotes a field, and @@ denotes an xml attribute for the item, you’ll probably be using mainly @@templatename or @@templateid.

For example: query:/sitecore/content/home/dessert/*[@pastry=’1’] <-- this grabs all the items under desserts which have the checkbox ‘pastry’ checked (so it might return pies).

Sitecore query also supports the xpath axes, allowing you to use things like ancestor-or-self, following and preceeding (siblings) and so on:

query:./ancestor-or-self::*[@@templatename='Site']/Data/Touts/*

For a more practical example in a multisite solution, the above takes the current item, finds an ancestor or itself that represents the top ‘site’ item and then find the Data folder and then grabs all the children of the Touts folder.

Logical operators can be used to combine options as well, so we could look for query:./*[@@templatename='template1' or @@templatename='template2'] or something more like query:./pies/*|./cakes/* <-- this would give both children of pies and cakes instead of choosing one or the other.

There are also functions you can use (here’s a link with a listing of a number of useful functions), primarily I end up using: position(), last(), and contains()

To use them, you’d do something like query:./*[position()=1] <-- grabs the first item

query:./*[position()=last()] <-- grabs the last item

query:./*[contains(@ingredients,’apple’)] <-- grabs the items with ‘apple’ in their ingredients field, this could also be written as: query:./*[@ingredients = '%apple%']

To test out any query, you can always open up the developer center and then open the xpath builder. You do not need query: before your query but you do need to include fast: if you want to use the fast query.

These queries can become pretty complex depending on your needs, but that initial work can leave content editors with a very easy to use and understand set of items and fields. Leave a comment if you know of any good query examples or are wondering how to form a query to meet your needs (I’ll try to help)!

Reddit had a 6 hour downtime that was caused by them running a database backed by Amazon's cloud disk storage product (EBS Elastic Block Store), but EBS is both unreliable and doesn't flush writes to disk when told to.  This led to corrupted data and disagreements between the master database and slave databases, causing the slaves to not be usable while the master was down. 

Modern SQL databases are not written to work right on hardware that does not have a reliable flush to disk operation. Correct functioning of a modern database absolutely requires that "write back" caching can be shut off, so that if a disk reports a write succeeded that the write actually succeeded.  See for example http://www.postgresql.org/docs/current/static/wal-reliability.html ; this is not at all unique to postgres, I believe that every sql database requires committed writes to actually commit to disk.

The law of leaky abstractions says that as we virtualize more we can do things that appear to work but don't really actually truly work exactly how we think they do, and Murphy says it will hit us with public downtime.

Row ID #770 - Bob submits article about puppies.  Master says it commits, so data is sent to slaves.  Master lied, data was actually in a cache somewhere, write later actually fails in the master - but succeeded in the slaves.

Row ID #770 - John submits article about kittens.  Master now has kittens, slaves have puppies, dogs and cats living together mass hysteria reddit is down for 6 hours migrating master data to new hardware and manually hacking up rebuilt slave tables. 

I don't know how this works with nosql systems and eventual consistency.  Is cassandra ok to run on ebs disks but postgres not at all?  Reddit says their solution is they are going to move to using local ec2 disks; is that actually a solution, or does it just make hitting the problem less likely because ec2 local disks are more reliable than ebs?  Do they still do write back caching?

Meanwhile, Netflix has pointed out that they have actually moved most of their functionality into the cloud.  Specifically, most everything that scales with customers and streaming usage is now served from clouds (although movies come from CDNs, not Amazon's EC2.)

Netflix has posted some really interesting information about the testing they did on EC2: http://perfcap.blogspot.com/2011/03/understanding-and-using-amazon-ebs.html

And their lessons learned is a great place to start when considering working in the cloud at scale: Netflix 5 Lessons We’ve Learned Using AWS

The upshot is, scaling by working in the cloud leads to a whole new set of challenges.  You have to invest more in writing your software to handle hardware failure, you have to test failure scenarios more, you may have to go so far as to redesign network protocols to be less chatty because you have unpredictable latency from shared systems, and you have to expect problems from when abstractions leak as layers of complexity are added to what used to be a simple operation like “write this to disk”.  If you’re at the point where your hardware costs from scaling exceed your software development costs, or if you truly need to be able to handle rapid customer growth faster than you can expand traditional data center use, it can make a lot of sense to tackle these challenges.  But it’s not a no-brainer no-effort proposition – development and testing are going to get harder to handle new scenarios as you switch to using a larger quantity of less reliable resources.

Reddit's explanation:

http://blog.reddit.com/2011/03/why-reddit-was-down-for-6-of-last-24.html

Netflix uses simpledb, hadoop, and cassandra.

http://nosql.mypopescu.com/post/2981945438/why-netflix-picked-amazon-simpledb-hadoop-hbase-and

http://techblog.netflix.com/2011/01/nosql-at-netflix.html

-David

A while back I wrote to ArkeGlobal about windows 7 blocking my files.  It really should go here, so here it is with some updates.

 Original Sept 10, 2009

I downloaded an external dll source for a web application and adding the reference to my project yielded this error:

“System.Security.SecurityException: Request for the permission of type 'System.Web.AspNetHostingPermission, System, Version=2.0.0.0, Culture=neutral, PublicKeyToken=b77a5c561934e089' failed.”

As it turns out Windows 7 security was blocking my downloaded dlls from running as a trusted source.  Going into the file system itself, I checked the permissions on the actually dll file and a message at the bottom says:

“This file came from another computer and might be blocked to help protect this computer.”

There is an ‘Unblock’ button right next to it and once I clicked it, it solved my issue.

Updated Oct 27, 2009

Since then, I ran into a similar issue but because of the way I had to build the project, every time it ran, it re-added that security back onto those dlls, so I found this: http://www.petri.co.il/unblock-files-windows-vista.htm

Look at solution 3 and/or 4.

 Updated Mar 2011

This last encounter is for those of you that have large, and I do mean large files to copy down.  I had a particular instance where I had to download an upgrade zip file for DNN to lay over a pre-existing one.  When I downloaded and extracted all the contents of it to the client's machine, I had some blocked files.  I grumbled and unblocked it, tried again and got the same error.  Then I realized, Windows had blocked *every* file inside that zip.  I immediately started googling and every single article basically said, 'window's does not provide a way to unblock in bulk'.  So I stared at over 1700 files I had to right-click and unblock by hand with no one providing a solution.  I was about 1 hour in when I had an idea: 

"Right-click -> Properties" on the ZIP folder

It will have the same 'unblock' feature that all it's children are inheriting from.  Once I unblocked the zip, I extracted and all my files played nice again.