SauceLabs Acquires BackTrace
The first piece of news that came across my desk this week was the announcement by SauceLabs about the acquisition of Backtrace. If you're not familiar with Backtrace, they offer one of the best in class ever, monitoring and … Reporting solutions that enable organizations to quickly capture, prioritize. Reporting solutions that enable organizations to quickly capture, prioritize and resolve errors, thereby mitigating application risk, and BackTrace is very much like a safety net for your apps.
So this acquisition is great news and is also an indicator of where the market is going. The skills you're going to need in the future, obviously, SauceLabs saw a need for monitoring as part of the portfolio, probably because of the needs of their customers and that the BackTrace acquisition is really an awesome complement to the front end test automation capabilities that SauceLabs already has because the combination of both SauceLabs and backtrace is going to allow developers and testers rapidly deploy code into development with confidence and strong quality throughout all stages of the software development lifecycle.
Demand for Automation Testing Skills
The next thing that caught my eye around automation was this article on demand for IT. Contractors have been at the highest level since March 1998. This is a UK news article. However, I think it applies worldwide for sure. And what really struck my eye when I dug into this report was so if you look for the skills that are in short demand, are in short supply. One of the ones that they show was in such a shortage where automation testing skills.
So if you are involved in automation testing or you feel like automation testing is not a good skill to have, the study just sort of confirms that it is. And so it's just good news to know that the skills we have are in demand. I also like to look at skills that maybe you may not have, but might be a good thing to work on this year.
Codeless Automation
Also a way I think it probably will be a controversial topic, even though I'm not sure why I'm open to any type of technique that's going to help me build better quality software faster. And even though I know a lot of times vendors will just pitch something that's codeless automation, a lot of times it's not codeless. You do need to maintain it, obviously, but there are certain benefits to this type of technology. And I just see them getting better, especially as machine learning gets better over the years, I think these tools are actually going to become more and more viable.
So the recent article that caught my attention on codeless automation was so when I Google things. A lot of times these types of articles come up and I'm not sure how valid they are, how credible they are. But what I thought was interesting was this open paragraph.
And I think this is actually one of the reasons why we've seen a lot of companies acquiring other companies is they're trying to integrate several automation testing offerings under one umbrella. So that's why you see companies like SauceLabs buying a monitoring solution or you see other companies like or you see other companies like we saw last week, like Zebrunner acquiring Corina, which is a framework automation framework opensource, where you see some companies like Grafana buying a performance tool k6 or acquiring them as these companies more and more are trying to get like and solutions all under one umbrella. And if you're a big fan of Mercury, but I am a big fan of Mercury Tool's back on the day and I think they set the boilerplate they were seeing.
Now, having a company that comes in with end-to-end solutions for all your testing needs from the beginning, from requirements all the way down to production, thru monitoring and all that. So I think it's just a big trend we're seeing more of now. It's more with open source tools, but vendor-based companies are acquiring open source tools and plugging them into their ecosystem as well. So it's just something to be aware of. And codeless, as I said, a lot of times people are down on codeless, a lot of people were down on automation testing. Keep your eyes open. Don't be a snob about any type of testing tool. Give it a try, see if it has merit. If it does, it can help you to build better quality software, then use it. If it doesn't, then throw it away.
And as I always say, as always test everything and to keep the good. And that's one of my mottos. Is before you down something or before you put something down, check it out, see if it's for real. If it can help your team to help your team, then it's a winner. If it doesn't help your team, then don't use it.
Salesforce Low-code Solution
So speaking of codeless automation, that's a good setup for the next article I found. So I know Salesforce just announced a new release with a group that includes low-code tools for app development. So same type of concept codeless, low-code. The same type of family is really interesting, though, is the article talks about how they actually use it. And Salesforce also has a solution built into the sales platform called Trailhead. And the new platform includes lo code tools for interactive app development and automation as well as new elastic computing, AI data protection, identity management, and DevOps capabilities to help teams drive business outcomes fast.
Dangers of AI for testers
Ok, I was kind of on the fence about including this in this new show. It's kind of morbid, but I think it's really valuable that we think of this. It's not really automation-related. However, I think it has ramifications for testers as they help their teams develop better software, especially as more and more teams start leveraging AI & machine learning. Kind of a disturbing article. So it talks about an AI Wolf developed program, a character in a game, or an app that someone in China created preferred its suicide over eating sheep. So this brings into all kinds of ethical issues. If you think about ramifications or unintended consequences for software that we use every day. so Say you work for health care and your team start to utilize AI machine learning as a tester, you need to start questioning certain things that maybe people aren't thinking about. And I think this is the perfect place for a tester to be involved because they usually have a high-level understanding of the full and lifecycle of the software. And what users want especially requirements and what requirements could do, but also what maybe people aren't thinking about behind the scenes are what tests are really good at. So check out this article.
And so the reason why this happened is the way they programmed the A.I. wolves were awarded points based on the number of sheep… So basically the way they developed this is they train the machine, learning the AI. by giving points for the amount of sheep Wolf caught. But they noticed over time that. Hitting a boulder, if you couldn't get sheep in an amount of time, would give you fewer points off, so therefore it would encourage you to kill yourself rather than try to continue catching sheep. So in this example, the AI cared more about how to optimize its performance mathematically and as it's the case for almost all machine learning algorithms.
So even though this may seem like a trivial case, I think, as I think you could see as a tester how this can actually impact your software and real users and what important ways are for real mission-critical software. So something to keep an eye on as a tester to be aware of as you stand diving into AI machine learning type algorithms that you're testing software for.