Do you test in production?
https://youtube.com/shorts/lBA1gJqjf_g?feature=share

Woah, this is cool! 🔥
In a study of "production failures" conducted 10 years ago, the researchers found that:

A majority of the production failures (77%) can be reproduced by a unit test.

77%!
That is a lot! Traditional sentiment in many companies (I think) is that the majority of the failures come from the "integration" of blocks, not from the "internal logic" of blocks.

Give it a read, this is a very insightful article (link to a video in comments):
https://www.usenix.org/system/files/conference/osdi14/osdi14-paper-yuan.pdf

My article got the "most popular" title in the "Software Testing Weekly" newsletter. 🥇
Awesome? Awesome!

I genuinely never thought this will ever going to happen...
If you missed it, give it a read, it is a gold-winning article after all (or the most clickbaity one??)! 🤣

https://twitter.com/QANewsletter/status/1833289939204685905

That is a cool idea that everyone can try in their team:

"Days since last..." dashboard

Cool way to reflect and not lose track on the stuff team considers important.
https://www.linkedin.com/posts/letsimprove_have-you-heard-of-dsl-metrics-they-can-activity-7236367812787675138-BvfA?utm_source=share&utm_medium=member_desktop

I didn't know that before, assuming you also might not be aware.

You can create a mock server in Postman 🔥

Of course, there are other tools that will achieve the same result. But if your team is using Postman already and doesn't like to search/introduce new tools/libraries, then it's a "no-brainer" option!
https://x.com/mjovanovictech/status/1833785734848749914

Do you know what "cumulative layout shift" is?
Do you pay attention to it when you test?
I look for it when performing my testing!

Why? Because it is annoying! 😅 When it happens, it is a usability issue... You don't want your users to be annoyed, do you? ;)

If you don't know or want a refresher on what "cumulative layout shift" is, read this article:
https://web.dev/articles/cls

I just came across this super cool site for accessibility testing! If you’re building or testing websites and not sure how to test for a11y, go and check it out!
I am not kidding, it has a nice UI and good examples!

Of course, it is not comprehensive, so don't assume it can replace all WCAG checks.
But it is amazing if you are new to accessibility! ❤️
#a11y #accessibility #accessibilityTesting #testing #qa #softwareTesting

(Screenshot in comments)
https://www.magentaa11y.com/web/

Ok I have been convinced to FINALLY do it! 🤣

Join tomorrow for a livestream!

Don't forget to like/repost the LinkedIn post for the algorithm discovery purposes!
See you tomorrow! ❤️😱

https://www.linkedin.com/events/gettingstartedwithuitestautomat7244846473924816899/about/

Here's a link to the broadcast!
27th of September, 4 PM, New Zealand Standard Time.
SEE YOU THERE, USERNAME!
https://youtube.com/live/fqK0LhySn4c?feature=share

Anyone keen to join me? ;)
https://www.linkedin.com/posts/ivanandcode_testautomation-testing-pairprogramming-activity-7248776746190184448-kUoT

🚨New article alert! 🚨

Reporting bugs? Why do I even ask, we all do from time to time!
But how many details do you need to add? Sync or async? And what does this have to do with culture?!

You'll find some answers and some food for thought in the following article!
https://www.linkedin.com/pulse/bug-reporting-how-important-details-ivan-karaman-tfjsc/

Remember the "anyone keen to join me" post?
I found a few people and arranged an event!
But... almost forgot to announce it here! This time the topic is:

Generating test data

Come to the stream, chat with me, stay for the learning and fun (haha!)
———————————————————————————
➡️ Friday 11th, 4 PM, NZST New Zealand Standard Time—-
———————————————————————————
https://www.youtube.com/watch?v=QAbIFJy-oPQ

Here's "yet another horror story" of how software can RUIN THE LIVES of the people. This one has a happy ending, but this ending was possible because the Twitter/X thread went viral.

For any detectives here, it is very likely that the answer to the question "Who is the murderer?" is "AI", but humans FAILED to control and stop it.
Periodic reminder that if you are working with SOPHISTICATED AND BIG software systems (like eBay) you need to TEST THEM BETTER.

(assumptions zone) The team that ships the "AI flagging feature" might not be aware of how exactly the customer support operates, and vice versa... So it is possible that AI flags the profile and there is no working mechanism to stop it from doing so or mark it as a "false positive". As a result, the "end-to-end" user journey ends up being a complete mess!

But hey, all tests were passing when you shipped it, right? 😅

Link to the original thread: https://x.com/CassieCeleste/status/1843702724090835178
For those without an account, PDF in comments ↘️

A reminder that the "test data generation" livestream is happening today!

...and 2 more streams are in the pipeline!
(link in comments)

New stream is scheduled to go live on Wednesday, 4 PM New Zealand Standard Time!
Come over and chat with me and James 😎
https://youtube.com/live/9_PfMYR-W98

Surely most of you reading this done some sort of "estimations" in some shape or form.
Then this is a good watch for you. It is not a testing/quality related per se, but useful.

TLDR:

Estimates are waste 🗑

What to use instead? Projections!
Turns out, if you just count "how many stories we have" and "how many we finished" the projection line for the future is pretty accurate (and more agile)!

There's a lot more good wisdom nuggets in it!
Watch the video and show it to your team (and especially your managers)! 😜

https://www.youtube.com/watch?v=QVBlnCTu9Ms

Reminder that I'm live streaming with James Espie tonight!
4 PM NZST (in 7 hours)
https://www.youtube.com/watch?v=9_PfMYR-W98

VERY NICE mnemonic and visualisation (screenshot inside) for different types of test: UI/Integration/Unit!

The SMURF mnemonic is an easy way to remember the tradeoffs to consider when balancing your test suite:
- Speed
- Maintainability
- Utilization
- Reliability
- Fidelity

Different types (or layers) of tests will look differently on each of those 5 "dimensions"
See screenshot in comments and read full blogpost here: https://testing.googleblog.com/2024/10/smurf-beyond-test-pyramid.html

Periodic reminder that "automated tests/checks" should be only part of the bigger picture and testing strategy, not the WHOLE/ONLY THING.

"Automate everything" or "only automation" mentality leads to bad testing.
Bad testing leads to bad product...
Bad product leads to "we need more automated tests"...

...and the circle continues 😅

Sometimes it is worth to stop, look around, and reflect!
https://medium.com/@vrytinki/your-problem-isnt-test-scripts-741adf79bf84

Very good rant on AI and how it is used and abused by many professionals! 🤣
TLDR summary:
1. Faster != Better
2. AI! AI! AI! At the conferences (everyone is sick of it)
3. AI content has no soul

Give it a read!
https://www.ontestautomation.com/i-am-tired-of-ai/

One more link that seems to be great resource for accessibility testing. Now it is "WCAG requirement -> explanation + how to test".

Check it out and bookmark for future:
https://github.com/alphagov/wcag-primer/wiki