Category Archives: Software testing

Posts about software testing in general

Standard

Posted by

Posted on

June 25, 2012

Posted under

Software testing, Test automation

Comments

5 Comments

The difference for test automation between cutting edge and legacy software

Within one of the LinkedIn groups (sorry, you need to be a member of the “QA Automation Architect” group to be able to read it fully) we started talking about the difference the state of project or product can make for test automation. In this post I will make a distinction between 2 states: new where no code has been written yet and existing  where application code has been written, but no test automation has been implemented.

Cutting edge

New So when creating a totally new product, life for the testers can be made easier by design, that at least is the thought. This does imply that testers, and not just the “manual” testers but all testers, including automation testers if these are a separate breed as some people seem to think, need to actively participate in the requirements phase of a product. With actively participating I do not mean to imply that they are normally not participating, I mean they need to look a bit further than just at what to test, is it testable etc.

They should also use their insights and ideas to help both product owners and software developers to understand what are the things that might make life easier for testing this new product.

When for example building a new web application, they might consider adding a simple REST api to the application, which in production can be closed off based on IP or firewall rules or something like that. A simple REST-API will make life a lot easier when creating your automated tests.

Another thing to make life easy might be ensuring clear and logical naming conventions to be used for all page object in order for the automation to use the Page-Object-Model. Not only is using solid naming conventions good for automation, it also makes maintenance on the application itself easier, since all objects are identifiable by their unique ID.

Legacy

How is existing code different from non-existent, other than that one is already in production and the other has to be created? As far as test automation is concerned, especially when talking about legacy software, it may turn out to be a lot more difficult to find proper hooks into the application for solid automation other than on the labels of buttons or fields.

When you have a fairly recent application it may be a website or a desktop app, both have the possibility that there are some sorts of ID’s for all objects. However when talking about true legacy software, such as 15 year old Delphi, it is quite unlikely the developers used WinForms, Win32 or SWT. Not having hooks like that into the application can result in having to scrape the UI for object labels, which is fine when testing one particular language, but if your software was localized things can get even more complicated.

Getting consensus within the technology group about new software is one thing, getting a “non-functional”, non-business related change about in existing software however is a whole different thing.

As long as the code is still “alive”, e.g. new features are still being added, bugs are being fixed and in general there are still developers working on the application, there is hope of getting some more “automatability” in the code.

First of all, while fixing bugs old code is touched, adjusted and retested, this is always an opening to talk to the developers resolving the issue about adding a small bit of extra “sauce” to make it easier to add this particular thing to the automated testing suite to ensure chances of recurrence are minimized, of course by fixing the bug you hope to completely obliterate this particular issue but it might cause new damage elsewhere in the application. So while talking to the developer about this function, try to convince him/her that adding a bit of extra to test not only for the fix of this issue, but also to verify the surrounding features.

While new features are added, this can be treated as “new code”, as long as you manage to get agreement on adding identifiers or a separate layer in these features to make test automation at least easier. If you achieve this, you are quite close to closing the majority of the gap. Refactoring is an excellent opportunity to again make minor changes in the application enabling test automation at a different level.

How do you get “automatability” in your specs?

Assuming you want to get your  product easy to automate and thus want to make sure it is thought through, how to get it in the specifications? And more importantly, how do you get it in there without adding things like:

  • unnecessary workload
  • unneeded and unwanted features
  • potential security holes
  • un-maintained code

Enterprise Architecture Layers with a "hidden test automation layer"One of the ways to go about it is by, in collaboration with the developers, enforcing a coding standard in which you ensure all objects receive an ID. Regardless of whether it is desktop or web based, most automation tools are looking for a hook into the UI, if there is one, and one of the nicest ways of doing that is simply by using the ID.

Alternatively you can have a “layer” put right underneath the UI, ensuring you can bypass the cumbersome UI while automating your tests. One of the issues with this option however, can be that you add “hidden” code which gets forgotten easily. It also is a potential risk for the security of your application, since you basically enable a man-in-the-middle hole.

If this path is taken, ensure that this “feature” does not end up being an opening for malicious code to reach your data. A relatively safe solution for this would be to put some (extra) form of authentication in the layer.

There probably are more options you can investigate, the two I mention above are fairly harmless and yet can make life in test automation a lot easier and predictable.

In the end, no matter which way you go, as long as you get both developers and product owners on board in working towards a higher “automatability” of the code life for you as a test engineer could become a lot more fun.

I am not very impressed with theological arguments whatever they may be used to support. Such arguments have often been found unsatisfactory in the past.

Alan  Turing

Standard

Posted by

Posted on

June 18, 2012

Posted under

Software testing, Test automation

Comments

24 Comments

Test automation in Agile and why it fails

It’s fairly safe to say that quite a lot of test automation efforts fail. It is also very safe to say that without test automation an agile team fails. So how can you make sure that while doing agile your test automation will not fail and thus your agile team will not fail? One of the ways to answer this question is by looking at why test automation often fails within agile environments.

When I am talking about test automation within this post, I am referring to testing that is done to reduce the amount of manual regression work, the so called functional test automation or automatic regression testing.

Moving target

Test automation quite often does not receive the attention it needs and deserves, also in agile teams. Quite some test automation efforts start off too late and without the appropriate preparation, resulting in organic test automation driven by a moving target. The moving target is the system under test which, in agile, is constantly in flux. Each sprint new features are added, bugs are fixed and quite often it is not clear at the start of a project where it is going to end up. Writing automated scripts against such a flexible environment which will stand the test of time, is difficult. It is even more difficult when the base on which automation is done is weak.

Quite often test automation runs behind on what is being delivered within an iteration, this is somewhat logical, considering that it is difficult to test, let alone automatically test what has not been built yet. Ideally while manually testing the new feature(s) as a tester, you’re already pondering how to automate it so that you do not have to do the tedious work more than once. Given enough time within your iteration, you actually might be able to automate some of the features, from what I have seen thus far, generally not all features will be covered in test automation within one iteration. So if these tests are not all automated, what happens to them in the next iteration? Are they omitted? Are they picked up and automated retrospectively?

If you do not keep track of what has been automated during an iteration for both your current iteration and your previous iteration, how can you rely on your test automation? You can’t be sure what exactly it is going through, so a bug can easily get through the net of your automated tests.

This moving target you are testing needs to be traced and tested solidly, repeatedly and in a trust-worthy way!

Definition of Done

In the majority of the DoD’s I have seen, one of the items is something referring to “tests automated”. The thing I have thus far not seen however, is the team adding as much value to the automation code as they do to the production code. Quite a lot of DoD’s refer to certain coding standards, however these standards often seem to not apply to functional test automation. Isn’t your functional automation code also just code? If so, why then should this not be covered in code reviews, be written according to some useful guidelines and standards and hopefully use a framework to make the code sustainable?

Test automation is just writing code

I have seen several automation efforts going on within agile teams where test automation was done without proper thinking having been put into it. A tool was chosen, based on what exactly other than members of the team having heard of it or having had good experiences with the tool. No base or framework to keep the code clean chosen. Since you are writing code, you should follow the same rules as the rest of the software developers. Don’t think your code, since they are merely tests, should not be hooked up to some form of framework. If you want to make your tests survive a few iterations, considering reuse of your code would be logical.

By the way, coding standards do not need to be too complicated. In 2009 “Agile in a flash” came up with a coding standard that could work for all languages and for most environments:

Coding Standards - agileinaflash.blogspot.com

All of the above mentioned points are “logical” when writing an application which is supposed to go into production. However when looking at a lot of (agile) projects, these logical “best practices” seem to be totally forgotten when it comes to test automation.

Succeed in test automation

So, how do you succeed in your test automation? How do you make it work? The answer seems clear to me: test automation is not like writing code, it is equal to writing code. Since it is the same, treat it the same way!

Do your code reviews, follow a form of a standard, use a (simple) framework to make life easier in writing tests, create reusable modules in your automation code. In other words, treat your functional test automation with the same respect as your production grade code. Who knows, you might want to run your tests against your production environment some day! In setting up your initial test automation environment and framework, don’t be shy and ask the developers in your team for tips, tricks and suggestions. They quite likely have gone through those setup steps more often than you have, so use their knowledge. Asking them for their insights and ideas not only helps you, it also helps them feel more responsible for doing their 5 pennies worth on the test automation side. They will get a clearer idea of what you intend to achieve, so they might also be more willing to help out keeping their code testable, they might even enjoy helping you write the testscripts!

Resources

Some informational resources where you can find some ideas on how to setup the test automation framework:

Standard

Posted by

Posted on

May 9, 2012

Posted under

Software testing

Comments

4 Comments

How do you test for SEO

In this post I mention SEO and Search engine optimization several times, I am referring here to the optimization of a website for natural search, so without paying for it showing up high in the search result lists.

While on holiday I spent some time talking to a local entrepeneur. He makes his money through his own website, we got to talking about his website and about the translations of his site in particular. Since this site was translated from Greek into English, French, German, Italian and Russian I had a quick hunch that his meta-keywords would not be in order for all the separate languages (which does not apply to all translations he made by the way).

When asked to test a site specifically for SEO, what are the things to look at? As I mentioned above, there are a few tell-tales when you start your testing, especially when the site has been translated:

  • lang – this should be set to the actual language of the page you are testing
  • meta-keywords – these should be in the same language as the lang set in the header
  • meta-description – this should be in the same language as the lang set in the header
  • Alt-text for images –  – these should be in the same language as the lang set in the header
  • page specific URL’s – should be in the same language as the lang set in the header

Please note that this is just a sub-set of what needs to be looked at when testing a site for SEO optimization.

Based on the before mentioned website I will give some examples of what to look for when testing for search engine optimizations.

HEADER

Looking at the header of the Russian version I indeed saw exactly what I assumed I would see:

<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="ru-ru" lang="ru-ru" >
<head>
<base href="http://www.corfu-villa.gr/ru.html" />
<meta http-equiv="content-type" content="text/html; charset=utf-8" />
<meta name="robots" content="index, follow" />
<meta name="keywords" content="corfu villa, villa corfu, seaside villa corfu, pool villa corfu, villa with sunset view corfu, villa rental corfu, corfu villa rentals" />
<meta name="description" content="Corfu villa. Two elegant seaside pool villas located in Chalikounas Corfu with amazing panoramic sunset view. See photos check 2010 availability and book online with our paypal secure system." />
<meta name="generator" content="" />
<title>Корфу Вилла | бассейн вилла с потрясающим видом на закат | villa.gr Корфу</title>

If you look fast and are not sure what to look at this looks fine, however from a Search Engine point of view this header is a bit of a drama.

The language is set to Russian in the first line, the title is in Russian, in the Cyrillic alphabet. The SEO issue however is in between the language declaration and the title: keywords and description are in English. When a Russian is trying to find a “seaside villa in Corfu” he will probably not use the English words for it, instead the keywords used will quite likely be “вилла на Корфу с видом на море“.

Just for fun, here are the result pages for the two searches; the English search on Google.ru and the Russian search on Google.ru. On the Cyrillic search the first page doesn’t have any links to our test site. On the English search however, the site is the first to surface, underneath the paid links. Problem I see with that result page however is, the url we get back is the main URL rather than the Russian URL.

The header of the homepage of the site is just the beginning of testing it for search engine optimization.

IMAGES

This being a site aimed at renting out a villa with amazing views in a fairly decadent location, it is quite visually driven. As a tester you might not pay too much attention to the images, however when testing the search engine optimization, the images should be looked at as well.

Sticking to the example of the Russian version of this site I grabbed another piece of the source code:

<div title="Corfu Villa Boxes" id="boxes">
<div title="Corfu Villa Gallery. Click to view more photos" id="left_box">
<div class="module">
<table border="0" cellspacing="0" cellpadding="0" width="100%">
<tbody>
<tr>
<td><a href="/ru/gallery.html"><img src="/templates/corfuvilla/images/gallery.jpg" border="0" alt="Corfu Villa Gallery" /></a></td>
</tr>
</tbody>
</table>
</div>
</div>
</div>

Within this snippet you’ll notice again an issue similar to that in the header meta-tags. The page is supposed to be in Russian, yet the title of the div is in English. The alt-tag of the image is also in English. Both of these are supposed to be in Russian in order for this page to properly be indexed by the search engines in that language. If you really want to make an effort, the image name should also be in Russian.

The useage of the images themselves, on the rendered HTML page already gave an indication that SEO and translation were not well thought through or at least not fully implemented.

Corfu villa gallery image

The text on the image should of course also have been translated. When, as a tester, you see mistakes like this on a website, this should quickly give you an idea that the SEO has not been done properly, nor quite likely will the translation of the site have been done properly. Ideally you would want the text on this image to be configurable in the CMS and be attached to the language the page is in.

Of course there are more things you have to look at when testing for SEO, however I will stop here for now.

Standard

Posted by

Posted on

April 24, 2012

Posted under

Software testing

Comments

2 Comments

Why test “by the book”

The other day I was reading a blog post on agile and why agile will fail in many instances. One of the comments got my specific attention, the comment states the following:

” a process with little Agility due to the remains of the “old process”. ”

This is why by-the-book scrum is so powerful. Too many agile consultants try to fit agile into the existing org structure and processes, thereby allowing existing dysfunction to remain, or worse, covering it up. They try to modify everything right out of the gate, instead of just choosing scrum.

This got me thinking about why so many methodologies seem to get followers who treat “their” methodology as a religion. I have been pondering about the different development and testing methods, such as XP, Scrum, Lean for development life cycles and ISTQB, TMap and such for testing in particular.

In religions it is generally considered bad to be extreme in following the rules, hence the term extremist, whether this is an orthodox, katholic, jewish or islamic extremist, they are always considered to be dangerous to society. Isn’t it the same in software development and testing? Aren’t the people who go to extremes to follow the rules as dreamed up by some author also extremists that seem to lose sight of the context?

Thus far the best implementation of any methodology I have seen, is a form of hybrid, or as the Dutch would call it a “polder model”, where you make a compromise between “the book” and “what actually works for us as a team or organization”.

Are methodologies best practices then? Aren’t methodologies meant to help people get a frame of reference and fill that in for themselves, by thinking about the frame of reference, critizing it, adjusting it to their needs. Shaping the method in such a way that it works optimally for you, in this situation, in this particular context. When moving on to a new task or assignment, you can take these learnings with you and see what of it works for you within this new context and adjust whatever doesn’t work.

Maybe it is time for a new methodology, which ties in well with a solid development method by Zed A. Shaw. A possible working name could be “Testing, fuckwit!”.

Standard

Posted by

Posted on

April 20, 2012

Posted under

Software testing

Comments

7 Comments

Is testing the dumping grounds of IT?

The other day I was talking to a few developers I was on an assignment with about getting testers added to their scrum team, and the response I got from them disturbed me. They told me that in their experience most testers do not work together in the team, they work against development, trying to get everything fully tested, despite them knowing this is not a feasible thing, and with that delay projects. On top of that they told me, most of the testers they have worked with, are part of the dumping grounds of the IT industry. And with that they meant that in their view most testers are not good enough to be a developer, so they decided to become testers instead (<sarcasm> cause, come on, testing is not that difficult anyone can do that! <\sarcasm>).

I was shocked to hear there are still a lot of developers out there who believe that testers are the dumping grounds of the IT industry, but I was even more shocked of their experiences with testers, working in an “us versus them” modus operandi instead of working in a team, part of a joing effort with a shared focus and goal.

What is it that still makes testers often work against developers instead of with them?

Most testers I have worked with over the past years agree that working side by side with development is the most effective and efficient way of working, this way you both keep track of your joint goal: get the software out on time, on budget and according to what your customer (or end-user for that matter) wants and needs. Together you try to add value to the software.

So is it indeed true that there are still a lot of testers out there in the field who are indeed not seeing the big picture and are trying to prove their worth by working against dev and looking for bugs that are not relevant, e.g. just looking for bugs for the sake of finding one, no matter what the value of that bug is to the end-user/customer, just so they can triumphantly point to a developer that indeed, “see! There are bugs in your code, you did it wrong!” Unfortunately I fear there are still too many testers out there that think and work this way, not to even mention all the developers out there that seem to not understand the added value of a good tester to the team and to the developers work!

Fortunately there is a wonderful contrast out there as well, in the form of this blog post by Nathan Lusher who shows that there indeed are good testers out there, who weigh in on a project and prove the value of testing and with that show that testers are not (or at least not everywhere) the dumping grounds of IT.

In my experience, there are a lot of very good, inspired and knowledgable testers out there, who see the added value of working together, in a team with a shared goal, a shared approach and shared respect. If testers want to get the respect of developers, I believe it is up to quite a lot of the testers to start by showing respect to the developers and where needed, increasing their technical knowledge in order to be able to counterbalance a developers viewpoint. You get what you give!