Dynatrace on Times Square

Ten Years of Blogging

This week ten years ago, I started blogging on software development. Here are the postings which drew most visitor attention (10k to 50k visitors):

#1: Top 10 Hibernate Performance Tuning Tips
#2: Tips for Lightning-Fast Insert Performance on SQL Server
#3: Hibernate: Joins without Associations
#4: How to improve Visual Studio .NET Build Performance
#5: MTU and the "Don't Fragment"-Bit
#6: When DataTable.Select() is slow, use DataView.FindRows()
#7: HashMap vs. TreeMap
#8: .NET 2.0, HttpWebRequest.KeepAlive and ServicePoint.ConnectionLimit
#9: Battleship Game Algorithm Explained (Part 1)
#10: Sorting Algorithms Visualized

Ruxit - Application Monitoring for Cloud Natives

Cool, ruxit was launched today. I can't say there are too many lines of code I personally contributed, but still I was involved into its development for something like 1,5 years...

Job Openings at Compuware Austria

Current Job Openings at Compuware Austria

• Software Engineer
• Senior Software Engineer
• Webbased User Interface Developer - Java 
• Technical Solution Engineer EMEA 
• Technical Product Manager 
• Online Community Owner 
• Creative Designer 
• Technical Inside Sales Services

How I Performance Hacked My Management Style

I really liked my CTO's blog post on leadership style.

Java 7 Update 65 Breaks Rich Client WebStart Apps on Windows

Hello Oracle, anyone at home?

Java 7u65 either breaks or affects many rich client (AWT/SWT) WebStart apps on Windows, because the previous WebStart launcher was replaced by jp2launcher.exe, which is lacking a DLL dependency to the Windows common controls library comctl32.dll. Button labels are missing, grid controls are unusable, stuff like that .

That happened one month ago. The bug has been filed for OpenJDK as well as in the Oracle Java bug database (see Eclipse bug report comment section and this Oracle froum thread, the bug is still not public), but no official statement so far. There even has been another update since (Java 7 Update 67), but this one did not address it either.

In the meantime, one can only try to prevent Java 7u65 / 7u67 rollouts wherever possible. And I used Resource Hacker to compile and store the missing dependency section to jp2launcher.exe's manifest manually, which works. But that's of course a hack...

This is the missing section, in case someone wants to try the same (do at your own risk!):

<dependency> <dependentAssembly> <assemblyIdentity type="win32" name="Microsoft.Windows.Common-Controls" version="6.0.0.0" processorArchitecture="*" publicKeyToken="6595b64144ccf1df" language="*" /> </dependentAssembly> </dependency>

Nine Things Developers Want More Than Money

Rob Walling on Nine Things Developers Want More Than Money:

• Being Set Up to Succeed 
• Having Excellent Management 
• Learning New Things 
• Exercising Creativity and Solving the Right Kind of Problems 
• Having a Voice 
• Being Recognized for Hard Work 
• Building Something that Matters 
• Building Software without an Act of Congress 
• Having Few Legacy Constraints 

Eight years later, this article holds true as ever.

Passing the Turing Test?

Oh come on! Read the transcripts of Eugene - the Turing test-beating 'human computer' - here.

UMDH is great for finding native memory leaks

When investigating native memleaks under Windows, Microsoft's UMDH (part of Microsoft Debugging Tools) is my tool of choice, combined with Application Verifier or gflags. Here is a short article explaining how to use it (it's actually pretty easy).