Forum Discussion
Colin_McCrae I've worked both with HP's UFT (Unified Functional Testing - the new name they made out of QTP), as well as with IBM's Rational Functional tester, as well as SmartBear's TestComplete.
All provide solid bases for designing future proof, robust , data driven end-to-end test automation frameworks, and they differ (somewhat ;) in pricing regarding the license.
That said, I agree the concept regarding above tools are more or less the same.
I think Selenium has quite a different approach in the sense that it solely relies on scripting and requires much more effort to realize effective end-to-end validations.
And yet there is an economic pressure on the bottom line. While your points are valid when it comes to what makes good testing and good tools, unfortunately, many of the folks making the decisions on tool purchasing are not the ones who need to use them but are the ones who need to report to the shareholders as to how they are keeping costs low while maintaining a profit margin. And Selenium, effectively, does that from that narrow viewpoint.
Get a good tester with coding background in a shop that uses Selenium and it will SHINE like you wouldn't believe. If, however, you try and grab a good tester who excels at manual testing processes but cannot code themselves out of a wet paper bag... then you'll tank pretty quickly. This is why shops who are looking for Selenium test automators typically do not hire "pure testers"... someone who is a good tester AND a good test automator is a rarity... a rarity that, somehow, it seems that the SmartBear community has blown up the average on... lots of BOTH here.
- mgroen28 years agoSuper Contributor
tristaanogre So, basically, what you suggest that, looking at the market demands, all 'pure testers' should learn Selenium (if they want to become populair for businesses, get invited for job interviews, etc? And leave all the commercial tools (HP UFT, IBM Rational, TestComplete, Tosca, Telerik, etc) for what it is, because future will be Selenium?
- tristaanogre8 years agoEsteemed Contributor
Nope, not saying that at all. Case in point, the company I'm currently working for specifically did NOT want Selenium because a) it's web only and they have a mixture of web and desktop applications and want a SINGLE tool to do both and b) they want a tool that does not require coding skills to utilize but does support coding.
We are doing a kind of hybrid environment. There is a framework that is being created in code with Script extensions, SQL database, etc. There is a role here where someone needs to develop, update, enhance, and maintain that framework. That person needs experience with the tool and experience coding AND needs to know what makes a good test in order to support tests in that framework. However, there are other roles here that do the actual creation of individual test cases which does not require major coding experience and, with regards to the tool, doesn't need to be a major expert in the tool. So... we're utilizing multiple roles, multiple layers of quality assurance personnel to test the applications.
Yes, there is market pressure for Selenium but, as was the case a number of years ago with automated testing in general, I think we'll see a leveling out in the market place. It used to be that everyone wanted to jump into automated testing as "the thing to do" and it was seen as the silver bullet to solve all the testing problems of on-time delivery, bug free apps, etc. As James Bach wrote in an article back in the late 90's, there was a "snake oil" mentality many times put out by automated testing tool providers. "Buy our tool and all your problems will go away with simple click and record." When that didn't pan out, a LOT of companies that jumped on the test automation band wagon abandoned those projects. However, the companies that saw Test Automation not as a universal solution but as another stage in the software development cycle maintained their projects and now TestComplete and UFT both have some VERY robust tools that fit a variety of shops.
As for Selenium being the "end all" for test automation... not every application is "Web"... we have mobile application testing now. And there are STILL desktop based applications that need testing. Also, not every test automation environment supports low-level coding. And the whole "it's free so it's good" will level out for the very reasons you point out of support and general source maintenance.
Just stating the way the market is NOW is not necessarily endorsing a particular view of the future of the market. You're OP questions the footprint of TestComplete in the European business space. I've expanded that to point out that, in general, COTS tools like TestComplete and UFT are faced with a particular market pressure that comes from an economy that still hasn't fully recovered from the 2008 scenario. But these things move in cycles and I'd be willing to bet that, especially with some of the stuff that SmartBear has been doing (their new Environment Manager for one thing) is going to expand their share of the automated testing market and even make inroads into Selenium's share.
Related Content
Recent Discussions
- 19 minutes agoSvetaG