Testing Methodology for the Cloud | Meryem Zid

[00:00:00] What do you think is the next big thing? It could be anything

[00:00:04] I think the next big thing on quality assurance will be focus on chaos engineering

[00:00:09] More sophisticated observability, edge computing testing methodologies

[00:00:14] Should be very precise and very understandable from the beginner in the QA field

[00:00:20] Have you ever encountered any incompatibilities or anything you have to be careful about when you use local stack?

[00:00:27] If you have all the configuration packs already set up, you can go from there

[00:00:31] And if you have to do it yourself, it's a little bit challenging

[00:00:34] When there's a new project and you come in and you look at quality assurance

[00:00:39] What would be like a perfect testing pipeline for like a QA pipeline from your perspective?

[00:00:45] What is necessary to do that?

[00:00:47] It's mandatory that we have the unit tests

[00:00:49] I mean, I encountered many other projects that don't have the unit tests

[00:00:54] And we have to do the automation and it's not that easy to do it internally

[00:01:03] You're listening to SimplyBlock's Cloud Commute podcast

[00:01:06] Your weekly 20-minute podcast about cloud technologies, Kubernetes, security, sustainability and more

[00:01:13] Hello everyone, welcome back to this week's episode of SimplyBlock's Cloud Commute podcast

[00:01:18] You know the spiel

[00:01:19] This week I have another incredible guest

[00:01:22] Again, a slightly different topic from what we've done the last times

[00:01:26] I'm trying to diversify here a little bit

[00:01:29] So, Miriam, maybe just introduce yourself real quick

[00:01:34] Thank you Chris for the invitation

[00:01:36] Well, Miriam, I live in Morocco

[00:01:39] I am a QA engineer with years of experience specializing in testing strategies

[00:01:50] Well, my background is in multiple industries from electronics to IT

[00:01:56] Then to fintech and then health field

[00:02:01] All right

[00:02:02] That was a quick one

[00:02:04] Much quicker than I expected

[00:02:05] Yeah, I can elaborate

[00:02:07] Oh yeah, feel free

[00:02:08] Well, I have another background of developing, of course

[00:02:13] So, I was developing testing frameworks to ensure high-quality and reliable software delivery

[00:02:21] So, I worked with diverse technologies, of course

[00:02:24] And before jumping into QA, I was a front-end developer

[00:02:28] So, it helped me a lot to shape my career into the QA testing field

[00:02:38] Right

[00:02:39] So, it's funny because normally you have to stop people

[00:02:43] Yes

[00:02:44] So, you worked as a front-end developer?

[00:02:48] Yeah

[00:02:48] Like mobile front-ends, web front-ends, any front-end?

[00:02:53] I worked for mobile, mainly in Java

[00:02:56] Using Java for Android Studio

[00:02:59] Because it was only available back then

[00:03:01] I'm talking 10 years back

[00:03:05] And after that, I had worked for other entities

[00:03:12] For web development, for front-end development also

[00:03:16] So, yeah, I didn't like both of them as much

[00:03:19] I can very much see that

[00:03:23] A back-end guy as long as I can remember

[00:03:25] I've done some web front, some web development

[00:03:29] I've done some mobile development

[00:03:31] It was all like now

[00:03:33] So, I always fell back to

[00:03:35] Let's go back to back-end

[00:03:38] At the moment you're at

[00:03:40] I'm not sure if I pronounced that correct

[00:03:42] Because I think it's a French company

[00:03:44] At Versio

[00:03:47] It's at Servio

[00:03:49] Oh, okay

[00:03:51] My bad

[00:03:52] Wrong note

[00:03:53] It's okay

[00:03:54] It's okay

[00:03:55] It's okay

[00:03:55] It's not your company, so it's all fine

[00:03:57] Maybe just

[00:03:59] Because I've met people

[00:04:01] Pronouncing the name of the company wrongly

[00:04:04] So, it's okay

[00:04:05] I hear that a lot

[00:04:07] Okay, fair enough

[00:04:09] It's fine

[00:04:09] Maybe just quickly tell us a little bit about the company

[00:04:12] Yeah, so

[00:04:13] So, at Servio Group is based now in France

[00:04:16] So, it's specialized in cutting edge technology consulting

[00:04:22] So, helping other organizations maybe optimize their software development

[00:04:28] And also the testing strategy

[00:04:31] They are also particularly working with other

[00:04:35] For other projects in cloud and distributing computing systems

[00:04:40] So, they provide comprehensive services ranging from

[00:04:44] From this information to cloud migration

[00:04:47] And advanced quality assurance methodologies

[00:04:52] So, we're fundamentally improving how organizations think about implementing quality through the software development cycle

[00:05:05] Right, okay

[00:05:06] So, that also means I guess you're going to customers and trying to help them understand like how should

[00:05:13] Well, I think quality assurance can mean a lot of things

[00:05:16] It probably is simple testing like how you should approach tuning testing

[00:05:22] But it's probably also integration testing, resilience testing, regression testing, all that kind of stuff, right?

[00:05:29] Yes, yes

[00:05:31] And also we have to explain all the test strategies that we're going to work with for each project and with each company

[00:05:45] With each client

[00:05:48] So, other than that, we do also development and software maintenance

[00:05:57] It's all in all consulting

[00:05:59] So, we're going from development to cloud to all the things in between

[00:06:04] Right, okay

[00:06:05] So, that's a perfect bridge because that's exactly the reason why I invited you

[00:06:10] Because I know that you're all about quality assurance and testing

[00:06:14] And you're probably having a lot of, from my perspective, a lot of good hints or ideas or experience on how to approach that

[00:06:23] So, when we look into that and imagine that I would want to build an application, let's say, that is supposed to run on Amazon or on AWS

[00:06:35] And using some of the AWS services

[00:06:37] Or let's start simpler

[00:06:39] We just want to run it on AWS for now

[00:06:42] What would be like the common approach to make sure that I actually test stuff correct?

[00:06:47] Yeah, so, any application that will be running on AWS instance should use a built-in or a pre-built testing tool

[00:07:08] For example, and for instance, for AWS example, we can use local stack

[00:07:15] Or we can use Terraform based on the infrastructure we have

[00:07:20] And also the development lifecycle we have also for this solution

[00:07:28] So, for example, if we use local stack, it has many core concepts that we can be based on for the testing

[00:07:40] For example, we have the local stack that supports a wide range of AWS services

[00:07:49] Including the EC2, Lambda, messaging, the API, getways, the cloud watch

[00:07:57] And also, I mean, many more

[00:08:01] DynamoDB, the storage, S3

[00:08:06] It's a local cloud service simulation

[00:08:09] And we can develop and test inside the local stack

[00:08:15] It simulates locally

[00:08:19] It's also free for people who might ask if it comes with a price

[00:08:28] It includes units and integration testing

[00:08:32] So, if you want to test in the cloud and collaborate, for example, with other teams

[00:08:41] And make it perfect

[00:08:44] We can use Terraform for this example

[00:08:47] Because Terraform is AAC, infrastructure as a code

[00:08:52] And it doesn't support units and integration tests within Terraform

[00:09:02] You have to use other testing tools that are provided by Terraform to do this

[00:09:10] For example, if either you're tested in AWS or GTP or any other cloud provider

[00:09:17] You can use these tools

[00:09:20] We have TerraTest

[00:09:22] We have another one

[00:09:25] I mean, I only use TerraTest for now

[00:09:28] But I believe there are others that are very specific for every use case

[00:09:37] Interesting

[00:09:39] So, yeah

[00:09:39] I know about local stack

[00:09:41] And you kind of answered a question that I wanted to ask slightly later

[00:09:45] But we can come back to that, no worries

[00:09:47] Okay

[00:09:47] I've never heard about TerraTest

[00:09:50] Is that a testing for an integration testing framework on top of Terraform?

[00:09:55] Yeah

[00:09:55] So, I think I can send you some documentation I was based on to use TerraTest

[00:10:03] Yeah, let's drop that

[00:10:05] You can send it afterwards

[00:10:06] I'll drop it into the show notes

[00:10:07] Interesting

[00:10:08] I mean, I use local stack

[00:10:13] That's why I know it

[00:10:15] So for the audience that is not 100% sure

[00:10:18] You kind of already explained a little bit

[00:10:21] It's basically a reimplementation of the

[00:10:25] Or not sure how many of the AWS APIs

[00:10:28] But quite a lot of them

[00:10:30] The AWS services basically

[00:10:32] And it gives you stuff like a local version of an S3 API

[00:10:37] It gives you Aurora compatible APIs and stuff like that

[00:10:40] So you basically spin it up and you can connect to it

[00:10:43] And use it as it would be like the official AWS version

[00:10:49] One of the questions I had

[00:10:51] Because I only did like very basic, very minor stuff

[00:10:54] I used it

[00:10:56] Or when I was still at Timescale

[00:10:58] I actually used it to build something that looked like the AWS event services

[00:11:04] But that was not something complicated

[00:11:07] Have you ever encountered like any incompatibilities or anything you have to be careful about when you use local stack?

[00:11:14] Because I think there's probably a lot of things which are like minor details still

[00:11:20] Because it's really important

[00:11:21] Yeah, well, the issue I have encountered while using the local stack for the first time was the configuration

[00:11:29] Because it's a little bit similar to where you configure a Docker image

[00:11:34] But I lacked some details and I had to call someone else to do it for me

[00:11:43] But now it's all fine

[00:11:45] I mean, if you have all the configuration packs already set up

[00:11:50] You can go from there

[00:11:51] And if you have to do it yourself, it's a little bit challenging

[00:11:56] Because it was the first time I didn't read the whole documentation yet

[00:12:00] To go through the whole process like smoothly

[00:12:05] But yeah, it was the very first step

[00:12:11] Which is the configuration

[00:12:12] And also the integration

[00:12:16] So back then we were using the ICD for the tab

[00:12:22] I didn't know that it was easy to implement

[00:12:28] Look a stack with GitHub Actions

[00:12:32] So I was working on each one separately

[00:12:35] It's a little bit hard working for me

[00:12:41] So yeah, I didn't know that

[00:12:43] So these are the challenges

[00:12:45] I mean, also

[00:12:48] It could be documentation for it

[00:12:54] When I was using the tool

[00:12:56] It was not that big and that rich

[00:13:00] It's very limited when it comes to local stack community

[00:13:05] But now, I mean, it's not the case anymore

[00:13:09] I mean, there are other tools we can use that can do the job as local stack

[00:13:15] But yeah, I mean, there is some concurrence in the market

[00:13:20] So, I mean, it's not the case like back then when I used it

[00:13:27] Right, what would be alternatives to local stack?

[00:13:30] Because I'm not aware of any

[00:13:31] It's literally the only thing I could find

[00:13:33] Well, I mean, if you want to test in the cloud, you can go fully Terraform

[00:13:41] It's not an ad or anything

[00:13:43] But they have tools for every cloud provider

[00:13:48] Right, okay, so in this case I would test against the actual services, right?

[00:13:55] Yeah, okay, okay, yes

[00:13:57] I try to avoid that for customer reasons

[00:14:05] When you, local stack, would you run this on test containers?

[00:14:12] Or would you literally, as you said, just like use GitHub action, start it up

[00:14:17] Run a separate bang and shut it down again?

[00:14:20] Yeah, you can use it locally

[00:14:24] Local stack, that's where the name came from

[00:14:29] So you can use it locally

[00:14:31] And also, you can install it in other instances in the cloud

[00:14:36] But that's where the challenge comes

[00:14:40] So it's limited to the environment of development and the configuration

[00:14:50] Now, what I mean is I used it as a module in test containers

[00:14:55] Which made it very, very easy to set up

[00:14:58] It actually took, I guess, most of the probably complicated configuration you mentioned

[00:15:04] Away from me

[00:15:05] And you basically, in test containers you say

[00:15:07] Hey, I want a local stack instance

[00:15:10] Enable the S3 API or whatever

[00:15:13] And it will do all the magic for you

[00:15:15] I had Oleg a few episodes ago here

[00:15:22] And we talked about that

[00:15:23] And local stack was one of the, like, from my perspective

[00:15:26] One of the most important things

[00:15:28] Especially when you work against cloud providers

[00:15:31] Or at least AWS

[00:15:33] Not sure if there's something similar for other cloud providers

[00:15:36] I mean, it depends on the use cases

[00:15:40] For everyone

[00:15:41] I mean, for me now

[00:15:43] If I was asked

[00:15:45] Which one I would prefer

[00:15:47] I would say 100% form

[00:15:48] Or build my own

[00:15:52] Testing script

[00:15:53] That I will use later for the pipeline, etc.

[00:15:57] But, yeah, I mean, I know there is an alternative for local stack

[00:16:02] And now I'm not on AWS projects for the moment

[00:16:08] But later, if I encounter another project on AWS

[00:16:14] I will most probably use it because I know it's the most effective for

[00:16:19] Yeah, I think it's probably like a two-stage integration test, right?

[00:16:23] You would run most tests against local stack

[00:16:27] And then you do a final

[00:16:29] I forgot the term right now

[00:16:31] There's, like, specific

[00:16:33] Oh, what is it?

[00:16:35] The smoke testing

[00:16:37] You run smoke test against the actual API

[00:16:40] Just to make sure that stuff actually works

[00:16:47] When there's a new project and you come in

[00:16:50] And you look at quality assurance

[00:16:53] What would be, like, a perfect testing pipeline

[00:16:56] Or, like, a QA pipeline from your perspective?

[00:16:59] What is necessary to do that?

[00:17:01] It will be interesting to know that the company already had a QA department

[00:17:08] Or QA team

[00:17:11] Because, you know, you'll not be working alone

[00:17:13] And win all the stuff from one to 100% alone

[00:17:22] So, yeah, it's interesting to know that you have some tasks to do

[00:17:27] For example, when we have a new project that we should work on

[00:17:33] And do the quality assurance from the beginning

[00:17:39] It's mandatory that we have the unit tests

[00:17:42] I mean, I encountered many other projects that

[00:17:46] Well, when we don't have the unit tests

[00:17:50] And we have to do the automation

[00:17:52] And it's not that easy to do it internally

[00:17:57] If we don't have the unit tests

[00:18:00] So, we have the benchmark of the project globally

[00:18:07] We do the test strategy that we're gonna follow for the project life cycle

[00:18:17] And from this test strategy

[00:18:21] We detect which parts are mandatory

[00:18:26] And we, of course, we work on sprints

[00:18:32] So, it's basically a sprint style

[00:18:36] So, we work basically on the tests we're gonna cover

[00:18:42] And the kind of tests we're gonna cover during this life cycle

[00:18:46] Like, we can't have all tests done

[00:18:49] We can't do black box and white box testing for example

[00:18:54] And smoke tests and integration

[00:18:56] And every kind of test in a whole project

[00:19:01] Like, we define the main focus and the main output we want

[00:19:07] And then we define which tests we're gonna focus on

[00:19:09] For example, unit tests are mandatory

[00:19:11] And we have the end-to-end tests

[00:19:14] Have you confused?

[00:19:17] Do you still see a lot of projects where there is no unit tests?

[00:19:20] Please say no

[00:19:21] Yes, yes, yes, yes, yes, yes, yes, yes, yes

[00:19:25] Well, uh

[00:19:26] We literally just had this conversation two days ago

[00:19:30] With, I don't remember

[00:19:33] I think Sebastian from Nowport

[00:19:37] And he was like, that's not a thing anymore

[00:19:41] I'm like, eh, I'm not sure about that

[00:19:44] Oh, God

[00:19:45] Okay, sorry, sorry, keep going

[00:19:48] So yeah, I was saying

[00:19:50] I was saying we have

[00:19:51] We know that we have unit tests at first

[00:19:54] And at the end we're gonna have the end-to-end testing

[00:19:57] And what's all in between depends on the project's output

[00:20:02] And the perspective of the client or the end user

[00:20:06] So we can have either integration, functional tests

[00:20:10] And automation at some point

[00:20:14] And if it's necessary, we have other kinds of tests to do within the other sprint

[00:20:23] Like regression testing and no regression testing

[00:20:25] So yeah, it depends on any, on each project

[00:20:30] But all the projects agree on one thing from my point of view

[00:20:35] That you have the unit test and we have the end-to-end testing

[00:20:38] Right

[00:20:39] So, one thing I wonder about

[00:20:42] I mean, we often say that stuff like

[00:20:45] Or like abstraction layers like Kubernetes

[00:20:47] Bring, make it easier for a developer to actually develop software

[00:20:52] Because you're closer to what it

[00:20:54] What the deployment infrastructure will look like

[00:20:58] Is that the same for quality assurance and testing?

[00:21:02] Or does that, like all the layers that Kubernetes brings to the table

[00:21:06] Actually makes it more complicated because you have way more stuff to test?

[00:21:10] Well, it's not the case

[00:21:13] I would still say that

[00:21:16] That the testing is

[00:21:18] I mean, the quality assurance is

[00:21:20] It's a little bit more challenging

[00:21:23] And if we

[00:21:25] If we want to compare it with

[00:21:27] What we have in Kubernetes

[00:21:30] It's not that easy

[00:21:31] To be honest

[00:21:32] I worked

[00:21:33] I worked

[00:21:35] I worked on

[00:21:36] On

[00:21:37] On a GCP project last year

[00:21:39] And it was

[00:21:41] Well

[00:21:42] It was

[00:21:43] It was

[00:21:45] It was using all the GCP instances

[00:21:47] I mean

[00:21:48] From BigQuery to Airflow to Kubernetes

[00:21:51] To using the

[00:21:52] The

[00:21:53] Console

[00:21:54] I remember

[00:21:57] And

[00:21:58] And

[00:21:58] Manipulating

[00:21:59] All

[00:21:59] All

[00:22:00] All

[00:22:00] These instances

[00:22:03] All at once

[00:22:04] To

[00:22:05] Sort out

[00:22:06] Some

[00:22:06] Some automation testing from it

[00:22:08] Because

[00:22:09] You have

[00:22:11] In

[00:22:11] In one script

[00:22:12] You have

[00:22:13] The

[00:22:14] The SQL commands

[00:22:15] And you have

[00:22:16] The

[00:22:16] The console commands

[00:22:18] And you have

[00:22:19] A lot of other

[00:22:20] Stuff

[00:22:21] I don't remember

[00:22:22] I don't want to remember

[00:22:23] To be honest

[00:22:24] But I

[00:22:26] If I have to work

[00:22:28] In

[00:22:28] In

[00:22:29] In GCP

[00:22:30] Or in a GCP project

[00:22:31] In the future

[00:22:32] I will focus on

[00:22:33] One thing

[00:22:34] Or

[00:22:35] Or

[00:22:36] One thing at a time

[00:22:37] Not all in one

[00:22:38] So

[00:22:41] So yeah

[00:22:42] I mean

[00:22:43] From

[00:22:43] From an apprentice

[00:22:46] Like me

[00:22:47] I mean

[00:22:47] I didn't have

[00:22:48] Much knowledge

[00:22:50] In working

[00:22:51] With

[00:22:51] GCP

[00:22:52] Like

[00:22:53] It was the biggest project

[00:22:54] I worked on

[00:22:55] I worked on

[00:22:57] Many other

[00:22:58] But

[00:22:59] Not as big as this one

[00:23:01] It's like

[00:23:01] Your

[00:23:03] Your

[00:23:03] We're saying

[00:23:05] The Netflix of the projects

[00:23:06] You have a lot of instances

[00:23:08] All at once

[00:23:09] But

[00:23:11] And

[00:23:12] You have

[00:23:13] You have to do the job alone

[00:23:14] Because

[00:23:14] I was the only

[00:23:16] QA

[00:23:17] Affected to this project

[00:23:20] You said scripts a few times

[00:23:23] Yes

[00:23:25] What language do you build them in?

[00:23:26] Is that like

[00:23:27] Actual Bash?

[00:23:28] Yes

[00:23:29] Bash Scripting

[00:23:30] Shast Scripting

[00:23:31] And Python

[00:23:32] The main ones

[00:23:33] I used

[00:23:36] I still use them

[00:23:37] By the way

[00:23:37] I love Python

[00:23:38] And

[00:23:41] It's my go-to

[00:23:42] It's my go-to

[00:23:43] Programming language

[00:23:44] For everything

[00:23:45] And

[00:23:46] And yeah

[00:23:47] For

[00:23:47] For scripting

[00:23:48] I use Python mainly

[00:23:50] For

[00:23:50] For the Bash

[00:23:52] It's shell scripts

[00:23:54] For the

[00:23:55] CICD pipelines

[00:23:56] Because it's easier to implement

[00:23:58] And

[00:23:58] Every

[00:23:59] Every

[00:24:01] Every

[00:24:02] Environment supports

[00:24:04] The shell script

[00:24:05] Like

[00:24:05] Yeah okay

[00:24:06] That's

[00:24:06] That's

[00:24:07] Very fair

[00:24:07] Whether you're in GitLab

[00:24:09] With BigPacket

[00:24:10] Or GitHub Actions

[00:24:11] They all support the shell

[00:24:13] Right

[00:24:13] Right

[00:24:14] I think Python is probably also very very

[00:24:17] Well

[00:24:18] Situated for that

[00:24:19] Before we come to the last question

[00:24:21] Because we almost hit the

[00:24:23] 25 minutes

[00:24:26] There is this like

[00:24:27] Big boom of AI

[00:24:29] Yay

[00:24:30] AI

[00:24:31] AI systems

[00:24:33] Whatever

[00:24:35] And

[00:24:35] A lot of people

[00:24:37] Claim

[00:24:38] That it will make

[00:24:39] Well

[00:24:40] Our job go away

[00:24:41] Which I don't think

[00:24:42] But

[00:24:43] There's also

[00:24:44] A big bunch of people

[00:24:45] That say

[00:24:45] That it will at least

[00:24:46] Make the job easier

[00:24:48] How do you see that

[00:24:49] For testing

[00:24:50] Especially with unit tests

[00:24:51] I think that

[00:24:52] You might be

[00:24:54] Able to generate

[00:24:55] Quite a bit of unit tests

[00:24:57] But then the

[00:24:58] Complicated edge cases

[00:24:59] Might be a problem

[00:25:00] Well

[00:25:01] A little bit

[00:25:03] A little bit of disclaimer here

[00:25:05] If you

[00:25:06] If you want to use AI

[00:25:09] To generate unit tests

[00:25:10] You have to use AI

[00:25:11] To develop

[00:25:12] From the beginning

[00:25:13] Because

[00:25:14] I mean

[00:25:16] AI is interesting

[00:25:19] They are transformative

[00:25:20] But not

[00:25:22] As a replacement

[00:25:23] I mean

[00:25:25] They can assist

[00:25:27] Of course

[00:25:28] To help

[00:25:29] Identify potential

[00:25:30] Potential edge cases

[00:25:32] Or

[00:25:32] Unit tests

[00:25:33] You should write

[00:25:34] Or

[00:25:35] Improve

[00:25:36] The test coverage

[00:25:38] Also

[00:25:41] Enable

[00:25:42] More

[00:25:42] More sophisticated

[00:25:43] Anomaly detection

[00:25:45] But

[00:25:46] For

[00:25:47] For me

[00:25:48] It's not

[00:25:49] They will not

[00:25:50] Replace humans

[00:25:51] Because we made them

[00:25:52] As I believe

[00:25:53] So

[00:25:55] For

[00:25:56] For

[00:25:57] For the testing

[00:25:58] Field

[00:25:59] I believe

[00:26:00] That humans

[00:26:01] They will still be needed

[00:26:03] Of course

[00:26:04] Because humans

[00:26:05] Define the strategy

[00:26:07] This design

[00:26:08] They understand complex

[00:26:10] Business logic

[00:26:11] Of course

[00:26:12] They interpret

[00:26:14] Very nuanced

[00:26:16] Test results

[00:26:17] So

[00:26:18] Yeah I mean

[00:26:19] For me

[00:26:20] It can

[00:26:21] It can

[00:26:22] Be

[00:26:24] As I said

[00:26:25] In the beginning

[00:26:26] It's transformative

[00:26:27] But not

[00:26:28] Truthfully

[00:26:29] You said

[00:26:30] Germans

[00:26:31] Or

[00:26:32] Humans

[00:26:33] Germans

[00:26:34] Humans gonna

[00:26:35] Define the strategy

[00:26:37] Don't let a German

[00:26:38] Define your strategy

[00:26:39] Bad idea

[00:26:40] Humans define the strategy

[00:26:41] We mean

[00:26:42] Like

[00:26:42] Is there more strategy

[00:26:44] Than I want an error-free

[00:26:46] Resilient application

[00:26:50] Well

[00:26:51] Let's

[00:26:52] Daydreaming

[00:26:52] I know

[00:26:54] I know

[00:26:56] Alright

[00:26:57] We're

[00:26:57] We're running

[00:26:59] Slightly

[00:26:59] Towards the end

[00:27:01] So

[00:27:02] What do you think

[00:27:03] Is like

[00:27:03] The next big thing

[00:27:04] It could be anything

[00:27:05] Related to your job

[00:27:07] Or in general

[00:27:08] A lot of people's

[00:27:09] Would

[00:27:09] Actually bring up

[00:27:11] As

[00:27:11] As you said

[00:27:13] A very transformative thing

[00:27:14] So that's fair as well

[00:27:16] Whatever you like

[00:27:18] Okay

[00:27:18] So I think the next big thing

[00:27:20] On quality assurance

[00:27:21] Will be

[00:27:22] An increased focus

[00:27:24] On chaos engineering

[00:27:26] More

[00:27:27] More

[00:27:28] Sophisticated

[00:27:29] Observability

[00:27:30] I mean

[00:27:31] I love chaos engineering

[00:27:32] It's not

[00:27:33] It's not something

[00:27:35] I would

[00:27:36] Not want to do

[00:27:37] Next

[00:27:38] Next thing

[00:27:39] So the next big thing

[00:27:42] I guess

[00:27:43] Also would be

[00:27:46] Edge computing

[00:27:48] Testing methodologies

[00:27:49] Which would be very

[00:27:51] Very

[00:27:53] Very precise

[00:27:54] And very understandable

[00:27:56] From each

[00:27:57] Each beginner

[00:27:58] In the QA field

[00:28:02] Yeah

[00:28:02] I would say also

[00:28:04] The

[00:28:05] Test strategy

[00:28:06] From

[00:28:07] Quantum computing

[00:28:08] Point of view

[00:28:11] Oh you're

[00:28:12] You're aiming high

[00:28:13] Yeah

[00:28:14] Well

[00:28:15] We dream big

[00:28:16] I don't know

[00:28:17] That's interesting

[00:28:18] I've never thought about that

[00:28:20] To be honest

[00:28:21] I

[00:28:22] Can't wrap my head

[00:28:23] Around exactly how

[00:28:24] Quantum computers work

[00:28:25] In general

[00:28:26] Also

[00:28:26] That

[00:28:26] I think

[00:28:27] Trying to understand

[00:28:29] How you would actually

[00:28:29] Test that

[00:28:31] Is

[00:28:32] Really interesting

[00:28:33] I never thought about that

[00:28:34] It's very challenging

[00:28:35] From

[00:28:36] From

[00:28:36] From

[00:28:37] From

[00:28:37] From

[00:28:37] From

[00:28:37] From

[00:28:37] From

[00:28:37] From

[00:28:37] From

[00:28:37] From

[00:28:38] From

[00:28:38] From

[00:28:38] From

[00:28:39] From

[00:28:40] From

[00:28:41] From

[00:28:41] From

[00:28:42] From

[00:28:42] From

[00:28:43] From

[00:28:43] From

[00:28:43] To

[00:28:43] To

[00:28:44] From

[00:28:47] The

[00:28:48] The other thing you mentioned is

[00:28:51] Chaos Engineering

[00:28:52] I

[00:28:54] Know that there's like

[00:28:55] Two big players

[00:28:56] Any

[00:28:57] Any specific tools

[00:28:58] You really like

[00:29:00] Gremlin

[00:29:00] Okay

[00:29:01] That

[00:29:02] That was the wrong one

[00:29:03] The other one is

[00:29:04] Ex-colleagues of mine

[00:29:07] No

[00:29:07] But

[00:29:08] I think Gremlin is fine

[00:29:09] I mean because

[00:29:10] Because it was

[00:29:11] The

[00:29:11] It was the second one I used

[00:29:13] After

[00:29:14] After

[00:29:15] I

[00:29:15] I don't remember

[00:29:16] It was the second one

[00:29:17] And I liked it

[00:29:19] And

[00:29:19] Whenever I

[00:29:20] I'm faced to

[00:29:21] To do chaos engineering

[00:29:23] For

[00:29:23] For any

[00:29:24] Use case

[00:29:25] I

[00:29:25] I opt for

[00:29:27] I think chaos engineering

[00:29:28] Is actually a really good topic

[00:29:31] I

[00:29:32] See something coming up in the future

[00:29:34] I didn't think about that yet

[00:29:36] But

[00:29:36] It is a really interesting topic

[00:29:38] And

[00:29:39] And

[00:29:40] And I think a complicated one as well

[00:29:41] All right

[00:29:43] So last question

[00:29:44] Is there anything else you want the audience to know?

[00:29:47] Yeah, if we

[00:29:49] If we're talking about

[00:29:51] Quality assurance

[00:29:52] And what the audio should know

[00:29:54] That

[00:29:54] QA

[00:29:56] Department

[00:29:57] Is

[00:29:58] It's a culture

[00:29:59] So

[00:30:01] It's not something you have to

[00:30:03] To be afraid of

[00:30:06] It's a culture

[00:30:07] That

[00:30:09] Embrace continuous learning

[00:30:11] And

[00:30:12] Of course

[00:30:12] For the audience

[00:30:13] I want you to

[00:30:14] Stay curious

[00:30:15] Learn new stuff

[00:30:16] And be ready to

[00:30:19] To challenge

[00:30:21] The existing

[00:30:22] Testing

[00:30:23] Fire

[00:30:23] The existing we have

[00:30:25] The future

[00:30:26] We don't know

[00:30:27] Where it's gonna be

[00:30:28] Where it's gonna lead us

[00:30:29] But

[00:30:29] I

[00:30:30] I'm confident

[00:30:32] That the future of

[00:30:34] Software

[00:30:35] Testing

[00:30:35] Is collaborative

[00:30:36] Intelligence

[00:30:38] And

[00:30:39] Do your homework

[00:30:40] Write unit tests

[00:30:43] Do at least

[00:30:44] The bare minimum

[00:30:45] Please

[00:30:45] I mean

[00:30:46] A friend

[00:30:47] A friend told me

[00:30:48] Many years ago

[00:30:49] That a developer

[00:30:51] That respects himself

[00:30:52] He does

[00:30:53] He does write the unit tests

[00:30:55] That is

[00:30:56] That is

[00:30:57] A very nice way

[00:30:58] Of ending the episode

[00:30:59] I think there is nothing

[00:31:00] To add to that

[00:31:02] Well

[00:31:03] Thank you

[00:31:04] Thank you Chris

[00:31:04] Yeah, no

[00:31:05] Thank you for being here

[00:31:07] It was a really good chat

[00:31:09] So

[00:31:10] I think we come back

[00:31:12] To the chaos engineering part

[00:31:13] Maybe

[00:31:14] I could see that

[00:31:15] Maybe being like

[00:31:16] An interesting episode

[00:31:17] With like

[00:31:18] Maybe two or three guests

[00:31:20] So let's

[00:31:20] Let's see

[00:31:21] Like

[00:31:21] The two fighting over

[00:31:23] Best tool and you're sitting

[00:31:24] In the middle or something

[00:31:28] All right

[00:31:29] Yeah, thank you very much

[00:31:30] For being here

[00:31:31] It was a great pleasure

[00:31:32] You're welcome anytime

[00:31:33] And for the audience

[00:31:35] Normally I would say

[00:31:37] See you next week

[00:31:38] But this time

[00:31:39] We'll be at

[00:31:40] KubeCon

[00:31:41] Salt Lake City

[00:31:42] So I'm gonna

[00:31:43] Skip one week

[00:31:44] And I see you back

[00:31:46] What is it?

[00:31:47] Like the third week

[00:31:48] Or fourth week of

[00:31:50] November

[00:31:51] With another incredible guest

[00:31:53] Thank you very much

[00:31:54] And hope to see you again

[00:31:57] The Cloud Commute podcast

[00:31:59] Is sponsored by

[00:32:00] SimplyBlock

[00:32:00] Your own elastic block

[00:32:02] Storage engine for the cloud

[00:32:03] Get higher IOPS

[00:32:04] And low predictable latency

[00:32:05] While bringing down

[00:32:06] Your total cost of ownership

[00:32:07] www.simplyblock.io