Lately I’ve done some work on XSLT and search results from FAST. One of the features of FAST are navigators, they basically make it possible to do faceted searches. And the default XSLT that we were working on did not include counters for the navigators, making it harder for the user to guess/understand the impact of adding a new search term. So I had to add the counters (actually I’m working with a collegue on this). The navigators were outputted like this in the XSLT:

<xsl:element name=”option”>
<xsl:attribute name=”value”>
<xsl:value-of select=”@VALUE”/>
</xsl:attribute>
<xsl:attribute name=”title”>
<xsl:value-of select=”@NAME”/>
</xsl:attribute>
<xsl:attribute name=”label”>
<xsl:value-of select=”@NAME”/>
</xsl:attribute>
<xsl:call-template name=”trunctext”>
<xsl:with-param name=”param”><xsl:value-of select=”@NAME”/></xsl:with-param>
<xsl:with-param name=”length”>45</xsl:with-param>
</xsl:call-template>
</xsl:element>

So we changed it like this:

<xsl:element name=”option”>
<xsl:attribute name=”value”>
<xsl:value-of select=”@VALUE”/>
</xsl:attribute>
<xsl:attribute name=”title”>
<xsl:value-of select=”@NAME”/>
</xsl:attribute>
<xsl:attribute name=”label”>
<xsl:value-of select=”@NAME”/> (<xsl:value-of select=”@COUNT”/>)
</xsl:attribute>
<xsl:call-template name=”trunctext”>
<xsl:with-param name=”param”><xsl:value-of select=”@NAME”/></xsl:with-param>
<xsl:with-param name=”length”>45</xsl:with-param>
</xsl:call-template>
</xsl:element>

This would indicate to the user that at 12 documents should remain if that navigator was selected. And on our VPCs this worked as expected. It would ouput a listbox with the navigators and the counters would show up. So we commited the change and went home.

The very next day we checked our daily build, expecting the counters to show up. To to our surprise they did not! At first we figured it was just a deployment issue; maybe an old XSLT was causing problems for us? We navigated directly to the .xsl file and inspected it. Imagine our surprise when the XSLT file was the correct file. We fired up our VPCs and tried again; success on the VPC! Now we’re basically kicking into debugmode; look att differences! There might be a minor difference in the source Xml! We run diff tools but no difference is detected, the source is the same! A lot of head scratching and and ohing and uhing later I realize what the difference is (remember kids; when it looks the same, but behave differently; something is different!); on our physical machines we’re running Windows XP and IE6, but on the VPCs we’re running Windows Server 2003 and IE7/8… Our problem isn’t serverside, but a browser compatability issue! OMG! The problem is, IE6 ignores the label attribute! And IE7/8 will display the label value, but not the actual value (if label attribute is present). Once we realized this the solution was easy enough; we just added the counter after the “xsl:call-template” (trunctext) and removed the label attribute all together. Happy days!

Sorry about hte bad formatting…

I just discovered a cool new utility from Microsoft: Disk2vhd. It’s built by the Sysinternals team; Mark Russinovich and Bryce Cogswell. It’ll capture any disk on your system as a VHD; meaning you can transfer a physical machine to a virtual environment in a snap. I’m going to try to use it to get away from Hyper-V on my laptop, so I can install Windows 7 and run just Virtual PC. I’m thinking I can use this to make  a “running” backup of old computers as well. So when I’m ready to ditch a computer and get a replacement or pave it I can just capture the disk and store it on my Windows Home Server for later. Why would I do this? Because I always forget something; a document on my dekstop or some odd file in the temp folder… Or I can just boot it and see what that great litt utility I  installed 2 years ago on the old Dell laptop was called, because I can’t remember anymore…

In programming there are many, many principles. One of the more useful ones that I’ve discovered lately is DRY: Don’t Repeat Yourself. It’s simple really, never ever copy code and don’t write the same code twice. In the real world it might not be as easy though…

Today I was about to check in some code and I went through my regular check-in routine (you have one as well I presume?) and discovered that I was violating DRY! At least if you’re being anal about it… The code in question was a little like this: (slightly altered to protect the innocents)

   1: using (SPSite site = new SPSite(url)) {

   2:     using (SPWeb web = site.OpenWeb()) {

   3:         lang = web.Language;

   4:     }

   5: }

(I didn’t really return the language, but no need to include all of the source code)

So I was doing the double using in 4-5 places; a clear violation of DRY in my book. So what I did was this:

   1: private T OpenSiteWeb<T>(string url, Func<SPSite, SPWeb, T> implementingMember) {

   2:    using (SPSite site = new SPSite(url)) {

   3:        using (SPWeb web = site.OpenWeb()) {

   4:            return implementingMember(site, web);

   5:        }

   6:    }

   7: }

This method I can use like this:

   1: lang = OpenSiteWeb<uint>(url, (s, w) =>

   2: {

   3:     return w.Language;

   4: }

Pretty neat, no? It might be overkill, but fun code to write anyway!

We’re just now starting a huge Sharepoint project and this new edition of Isolator might be good for that project. I’ll need to try it out!

And they’re even giving it away for free! 50 licenses anyways. I’m hoping there’s one for me! For that to happen I have to include this text:

Typemock are offering their new product for unit testing SharePoint called Isolator For SharePoint, for a special introduction price. it is the only tool that allows you to unit test SharePoint without a SharePoint server. To learn more click here.

The first 50 bloggers who blog this text in their blog and tell us about it, will get a Full Isolator license, Free. for rules and info click here.

Announcing Isolator for Sharepoint with a free full license

So that’s cool! So, if I do happen to get a free license I’ll tell you all about it, right here!

for bloggers – ISerializable – Roy Osherove’s Blog

Holy cow! Silverlight 2 was just released and ScottGu mentions some of the improvements coming in the next major versjon; Silverlight 3.

Next year we will ship our next major Silverlight release — Silverlight 3.

Silverlight 3 will include major media enhancements (including H.264 video support), major graphics improvements (including 3D support and GPU hardware acceleration), as well as major application development improvements (including richer data-binding support and additional controls). Note these are just a small sampling of the improvements – we have plenty of additional cool features we are going to keep up our sleeves a little longer.

Looks like Microsoft is pouring a lot of resources into Silverlight. I’d have no issues betting the farm on Silverlight.

Update on Silverlight 2 – and a glimpse of Silverlight 3 – ScottGu’s Blog

So today I read two blog posts on Agile that talks about the failures of Agile. Basically what they’re saying is that Agile+SCRUM is HARD and a lot of teams are failing hard and fast with it.

There are a lot of teams right now failing with Agile. These teams are working in short cycles. The increased planning frequency has given them more control over their work and they’re discovering and fixing some problems. They feel good, and they really are seeing more success than they were before.

But they aren’t working in shared workspaces or emphasizing high-bandwidth communication. They’re don’t have on-site customers or work in cross-functional teams. They don’t even finish all of their stories by the end of each Sprint, let alone deliver releasable software, and they certainly don’t use good engineering practices.

I’ve always been a firm believer that the four most important aspects of agile software development are:

1. Short iterations
2. On-site customer voice
3. Best engineering practices
4. Process review

Ken Schwaber, one of the Scrum Fathers, said at NDC2008 that a SCRUM development process will never be complete, it should always evolve and become better. I think a lot of teams are forgetting this, they just churn out code and every 3-6 weeks they evaluate on what to do next. This is not scrum and it is not being truly agile. It’s just being sloppy if you ask me…

James Shore: Successful Software
The Decline and Fall of Agile and How Scrum Makes it Hurt More

So Rob Conery wrote a really cool post the other day that I just had to comment on.

Here’s my thought for you: What if you used an OODB for development ONLY and implemented SQL Server later, when you know what you need to create.

It made me wonder; how many applications really, really need a full on RDBMS (Oracle, Sql Server etc.) as backend? I wouldn’t know, but if I were to guess I’d say; not all. Vague, I know… But I do know this; my next project will use Db4O! I just have to try this stuff out!

Crazy Talk: Reducing ORM Friction : Rob Conery

I thought I did a good job, with 72 slides and 8 demos in 75 minutes and only one person said it felt rushed in the comments. Of course, I had about 9 hours of content, but I did prepare in specific ways in order to pull it off.

Scott, you did an amazing job! Unfortunately I was not able to attend PDF2008, but I watched you presentation, back to back; good stuff!

Also I’ll be using your tips for my “TFS Build – More than code” presentation @ NNUG tonight.

Scott Hanselman’s Computer Zen – Tips for Preparing for a Technical Presentation

I’m not much for shooter (excluding GoW) nor Horrors (excluding 7th Guest), but DirtyDivas short review of Dead Space was funny. It ended with this priceless quote:

The onslaught of games has begun. I need better time management skills. This whole work thing and that sleeping thing is not conducive to my gaming time.

Gamerscore Blog : Dead Space – A Horror Lover’s First Impressions

This looks interesting! I’ll have to follow the next n parts of this!

This system started small but gradually had increased demands in the form of new features clients wanted added. Pretty soon, when you clicked the save button multiple things were happening; a database row would be updated, an email would be sent, a new file would be written, another status file would be updated. These were all executed when the user clicked the button, therefore every action was a blocking action, the server couldn’t move on until it was done servicing that particular request.

The quote above is from the blog post and sort of sums up my experience with a CMS-like application (Synergi), where we once did a Save, Send Mail, Commit we ended up with something like this after a few years; Save, Mail, Mail, Mail, Update Table X, Update Table Y, Mail, Commit. So a solution, like the one proposed in the blog post, could work very well for Synergi. Interesting.

Mass Transit – Part 1 of N – Tim Barcz