Software Testing

The Day I Quit

QA Hates You - Wed, 04/22/2015 - 08:37

When I was a lad, fresh out of the university with a degree in English and Philosophy and no actual career prospects, I worked as a produce clerk for a small off-chain produce and cheese shop. They had daily garbage pickup on weekdays, but nothing on the weekends, which were some of the busiest days of the week. As a result, on Sunday afternoons, the dumpster started to fail boundary analysis, at which time the store manager would order a clerk or two to climb up onto the pile and jump up and down to compact it so we could dump the last few cans of refuse into it. Come to think of it, I’ve seen the same philosophy applied to hardware resource management.

So as I stood and watched the younger kids jumping in the dumpster, I decided that if I was ever ordered to climb into the dumpster, I would drop my apron in the alley and never come back.

Want to know what would make me leave QA? Needing an implant of some sort to do my job:

PayPal is working on a new generation of embeddable, injectable and ingestible devices that could replace passwords as a means of identification.

Jonathan LeBlanc, PayPal’s global head of developer evangelism, claims that these devices could include brain implants, wafer-thin silicon chips that can be embedded into the skin, and ingestible devices with batteries that are powered by stomach acid.

These devices would allow “natural body identification,” by monitoring internal body functions like heartbeat, glucose levels and vein recognition, Mr LeBlanc told the Wall Street Journal.

Over time they would come to replace passwords and even more advanced methods of identification, like fingerprint scanning and location verification, which he says are not always reliable.

I’d rather not be personally, bodily on the Internet of Things unless there’s a compelling medical reason for it, and even then I’m going to ask my doctor to examine all the steampunk options first.

Categories: Software Testing

Web Performance News of the Week

LoadStorm - Mon, 04/20/2015 - 16:46

Google modifies the structure of URLs on mobile

This week Google made a change to the URL structure that is displayed normally from a mobile search, explaining that “well-structured URLs offer users a quick hint about the page topic and how the page fits within the website”. The algorithms were updated starting in the USin an effort to display names that would better reflect the page. This will structure the results using a breadcrumbs-like format, and use the “real-world name of the site” instead of the domain name. I checked search results on my mobile, but the change had not appeared to be made yet. Here is an example search Google used:

To complement the launch, Google also announced support for schema.org, a site that provides a collection of schemas that webmasters can use to markup HTML pages in ways recognized by major search engines. This intends to help webmasters signal both the website name that should be used instead of the domain name, as well as the URL structure as breadcrumbs.

Here are the examples Google pointed to in their documentation on site names and breadcrumbs.

The announcement garnered mixed reactions. While some people are calling the change helpful, arguing that the change encourages better site organization, others argue that by not displaying the actual URL, users no longer could easily verify the true identity of a site. What do you think about this interesting development?

 

Google adds mobile-friendliness as a part of their ranking algorithm

Tomorrow marks the official launch of Google’s new ranking algorithm, which will add mobile-friendliness as a ranking signal. Although Google held a live Q&A hangout a month ago, it was somewhat unclear whether or not the changes could be seen before the official launch date on April 21st.

So how do you know if your page is mobile friendly? The easiest way to check if a page on your site passes is by typing in the URL here.

This will test the way your site looks to a Googlebot. It checks for:

  • pages automatically sizing content to the screen
  • avoiding software that is uncommon on mobile devices
  • containing readable text without zooming
  • and placing links far enough apart to be selected easily

 

A few easy ways to see what your site looks like on an array of different devices, and to start making your site mobile-friendly, can be found in our responsive web design blog post.

 

Additional algorithm details to note include:

  • There are no degrees of mobile-friendliness; your page is either mobile-friendly or it isn’t. Google added that it will be the same with desktop search, and is not isolated to mobile searches.
  • The mobile-friendly ranking changes will affect your site on a page by page basis. So if half of your site’s pages are mobile-friendly, those half will benefit from the mobile friendly ranking changes. This is good news for anyone concerned that their site doesn’t make the cut, as they can focus on getting their main pages up to date first.
  • The new ranking algorithm will officially begin on April 21st. However, the algorithm runs in real-time, meaning you can start preparing your site for analysis now. It was said in the hangout that it may even take a week or so to completely roll out, so it’s not entirely clear how quick we can expect to see changes in site rankings.
  • Google News may not be ranked by the new mobile-friendly algorithm yet. Interestingly, the Google News ranking team has no plans on implementing the mobile-friendly algorithm into the Google News results.

Zuckerberg’s internet.org loses support, with companies siting net neutrality concerns

A facebook backed project intended to make the internet accessible for people in the developing world has lost the support of several prominent companies this week. In the midst of a fierce national debate regarding net neutrality, the New Delhi Television Limited (NDTV), the Times Group, a media company that owns Times of India, and Cleartrip, a travel website all removed content from Internet.org in India, which they had made available for free. Furthermore, the companies urged its competitors to withdraw as well.

“We support net neutrality because it creates a fair, level playing field for all companies – big and small – to produce the best service and offer it to consumers. We will lead the drive towards a neutral internet, but we need our fellow publishers and content providers to do so as well, so that the playing field continues to be level.” – statement from a Times official.

The companies sited concerns that the Internet.org initiative did not align with the net neutrality mission, arguing that Internet.org is set up to prioritize content from partner businesses who pay telecom companies for data charges.

Zuckerberg responded to critics, affirming that net neutrality and Internet.org “can and must coexist.” He detailed his stance on the controversy, proclaiming that “Internet.org doesn’t block or throttle any other services or create fast lanes — and it never will. We’re open for all mobile operators and we’re not stopping anyone from joining. We want as many internet providers to join so as many people as possible can be connected.”

The Internet.org initiative has reportedly expanded internet accesss to over six countries so far, including Columbia, India, Zambia, Tanzania, Kenya and Ghana. Still, some worry that the long term net neutrality implications outweigh the immediate access it provides to the developing world.

The post Web Performance News of the Week appeared first on LoadStorm.

How To Start A Challenging Test

Eric Jacobson's Software Testing Blog - Thu, 04/16/2015 - 15:58

Last week I started testing an update to a complex legacy process.  At first, my head was spinning (it still kind of is).    There are so many inputs and test scenarios...so much I don’t understand.  Where to begin?

I think doing something half-baked now is better than doing something fully-baked later.  If we start planning a rigorous test based on too many assumptions we may not understand what we’re observing. 

In my case, I started with the easiest tests I could think of:

  • Can I trigger the process-under-test? 
  • Can I tell when the process-under-test completes?
  • Can I access any internal error/success logging for said process?
  • If I repeat the process-under-test multiple times, are the results consistent?

If there were a spectrum that showed a focus between learning by not manipulating and learning by manipulating something-under-test, it might look like this:

My tests started on the left side of the spectrum and worked right.  Now that I can get consistent results, let me see if I can manipulate it and predict its results: 

  • If I pass ValueA to InputA, do the results match my expectations?
  • If I remove ValueA from InputA, do the results return as before?
  • If I pass ValueB to InputA, do the results match my expectations?

As long as my model of the process-under-test matches my above observations, I can start expanding complexity:

  • If I pass ValueA and ValueB to InputA and ValueC and ValueD to InputB, do the results match my expectations?
  • etc.

Now I have something valuable to discuss with the programmer or product owner.  “I’ve done the above tests.  What else can you think of?”.  It’s much easier to have this conversation when you’re not completely green, when you can show some effort.  It’s easier for the programmer or product owner to help when you lead them into the zone.

That worst is over.  The rest is easy.  Now you can really start testing!

Sometimes you just have to do something to get going.  Even if it’s half-baked.

Categories: Software Testing

Pages