Testing at Google

This is such a great post: this is in the top 1% of all must read material for both dev and QA engineers from James A. Whittaker, the former Test Director for Chrome and Chrome OS:

http://www.sqazone.net/modules/mylinks/visit.php?cid=6&lid=1026

DBUnit with Spring

rather interesting this integration of DBUnit with Spring: http://www.javacodegeeks.com/2012/04/dbunit-spring-and-annotations-for.html

High Risk Test Cases

Taken from: http://www.webperformance.com/load-testing/blog/2012/01/identify-and-test-high-risk-operations/

After testing and optimizing those, the next step is to identify and test the rarely-used but high-risk operations. These are operations that may have an impact on performance that is not proportional to their frequency. Even though these are not as common, they can have large impacts due to their system-wide nature and can cause sporadic performance drops during production hours that defy explanation.
Examples include:

• Complex searches
• Updates that touch many joined tables at once
• Mass changes to data that affect many rows
• Large or complex reports
• Administrative functions, such as data migrations, data cleanup or consolidation, cache clearing, etc
• Backups

Good Checklist for code audit

http://extreme-java.com/best-practices-list-in-java/

Must see video on UI Testing

Basically, does the page "look right?" http://www.softwaretestingmagazine.com/videos/web-consistency-testing/

Good post on Continuous Delivery

There is a good post on Continuous Delivery that can be found here.

 The point in continuous delivery practice, is that it enforces the development team, the QA team and the Managers, to invest heavily in automation. If each code change, is a potential production version, to keep up the speed, you have to automate everything. There is no time for manual regression testing, there is no time to write deployment manuals and there is no time to execute any of the steps needed to promote a software change to production, manually.

What I like about this post is that in a few line you know if CD is for you or not. It's definitely in our near future at Crowd Advisory

Nice post on BDD

Don't know the guy but this article points toward a nice post on using BDD with Jetty and Maven along with JBehave.

I'll try to follow along but by the way the post is constructed, it seem serious. Get the article here

Running QUnit JS tests in Maven

If this project is as good as it sounds, we need to get this running in our environment. If its not, we need to give the lead a hand 'cause that is absolutely something missing in our ecosystem.

jHiccup

This tool, jHiccup, seems like a very nice tool to had to the kit. You can read about it here

Reason to unit test

http://www.typemock.com/blog/2011/11/28/10-reasons-to-write-unit-tests gives 10 reasons on why doing TDD. It misses at least one reason: improved design.

Great post on .NET performance

Good pointer, good actionnable pointers!, can be found here.The reference section is good too!

Ten tests to add to the basic list when using cloud computing

Ten tests for software applications in the cloud list, well, 10 tests. I particularly like the tests under the "performance" umbrella:

• profile that is more challenging than the average performance profile
• Service provisioning/de-provisioning testing
• Distributed cloud testing
• Multi-tenancy testing 
• Graceful degradation testing
• Connected/disconnected operations testing
• Cloud portability testing 

Geb: very groovy browser automation… web testing, screen scraping and more

Found this tool that allow for web browser automation via Groovy and WebDriver. It has a very good documentation for such a young project (0.6.1).  Andre Steingress makes as a nice PPT here

Garmr: tool to automate security testing

Garmr is a security testing tool produced by the Mozilla QA Team. Although still at an early stage, the tool seems to be a good base on which to build a security baseline. You can read about it here

Test Scribe for MTM 2010

http://talkabouttests.com/  talks about a power tools called Test Scribe to enhance Microsoft Test Manager 2010. Test Scribe needs to be installed on TFS:

Test Scribe is a documentation power tool designed to construct documents directly from the TFS for test plan and test run artifacts for the purpose of discussion, reporting etc.

Basically, it exports your test plan to Word, which is nice when you want to communicate them to the client or a 3rd party

This is scary: Cautionary tale: Do you know where your customer database is?

[...]not just one, but TWO customer database files — one 4 MB .txt file and one 13 MB .bak file that was also plain text — comprised of ~11,000 users, each containing varying amounts of confidential information as noted above. From http://www.zdnet.com/blog/btl/cautionary-tale-do-you-know-where-your-customer-database-is/61811

It is at once sad and scary that this kind of mistakes would still happen. Come on! Governance and professionalism anyone???

Manual penetration testing

Taken from: http://www.esecurityplanet.com/network-security/penetration-testing-shows-unlikely-vulnerabilities.html

Overall, the goal of Spider Labs penetration testing efforts weren't just about seeing how far they could get, it was aslo about seeing what organizations were able to detect. Linn stressed that persistence in penetration testing is key in order to dig deeper just like real criminals would do.
"These types of vulnerabilities are not the things that an automated scan will find," Percoco said. "The things we find commonly through the manual process ends up getting us awesome results in the end."

Testing in production

http://searchsoftwarequality.techtarget.com/answer/Release-management-Software-testing-in-production raise the question of whether it is okay to test in production. Some say "never", the author is more nuanced in his answer. So far, I've never had the chance to work in projects where the pre-production or QA environment were conform replicas to the production environment and therefor I've always seen the need to test in production.

A new vulnerability is discovered in a library on the application server or the database server or IT has to make a change in the production environment, etc. There are a large number of reasons why we not only can but must test in production. It does need to be bounded by processes and communication streams more stringent than in the development environment so that tests don't triggers alarms (DoS for exemple) or don't skew the stats but planning for test in production is a must if one does not want to have very unpleasant conversation with the CSO...

Add automatic test to your bug report

Selenium is a very nice tool: it's a huge benefit to the software community. A good idea is to use it during exploratory testing and, when a incident is found, register the selenium script in the report so that it is easy to recap the steps to reproduce the problem before and after the fix.

It apparently is easy to export a script in different format, such as RSpec for Ruby. Have a look here: http://www.techdarkside.com/generating-rspec-tests-with-selenium-ide

Agile: Where does QA fits in

QA, along with architecture, is getting back into the equation for agile development. Good Ol' pendulum is swinging back to balance. So, to this simple question of the role of QA, or testing for that matter, we get this fairly simple answer from All About Agile:

So, our answer is that good professional testers must be in the Scrum team.  Ideally that means 100% allocated to one team.  At a minimum, 50% allocated to one Scrum team.

Where does QA fit in?  Well, usually the testers are aka QA people.  Sometimes QA means truly "quality assurance" per se, in which case the QA people look at the Scrum team (and the process elsewhere as well) to see if sufficient quality is being baked in in the best possible way.