The Internet of Things part 2 – a UKTMF update

The Internet of Things at the UK Test Management Forum.

Today (30th July) I was at the UKTMF in London, run by Paul Gerrard – soon to be run with assistance from a group calling ourselves ‘Friends of the Forum’.

Mike Bartley (CEO at TVS) did a superb talk on the Internet of Things, focussing on the functional testing side, whilst Declan O’Riordan talked about the security needs.

The IoT is going to be a phrase that we hear a lot about. In fact it would not be an understatement to say that it is going to revolutionise the way in which we live.

It’s definition is “Interconnectivity of uniquely identifiable embedded computing devices within the existing internet structure”.

Up until now, interaction with the internet is done by humans, via computers, smart phones, tablets etc. The IoT moves on a stage further, to devices which connect and pass information without human intervention.

A great example that Mike shared was a smart fridge. A bottle of milk will contain information regarding the type of milk, how much is in the carton and its sell-by date. The fridge will scan and send the information to a server. An app on a device will the receive a message from the server when the milk has reached a certain level, or has been used up, or has passed it’s sell-by date, telling the homeowner that he or she needs to buy milk. Of course it may be that an automatic order is placed instead, and milk is delivered before the person knew he or she needed it!

That’s just one example – another is the use of driverless cars (see for the story. It’s a great idea, but of course there will be concerns around safety, but by far the greatest concern for us as humans is privacy and security.

The BBC ran another story today about apps that control home devices being easy to hack into (see Nobody wants a burglar to hack into an app that contains information about the householder being on holiday! There are other considerations as well – if a fridge can tell you when to buy more milk, that can apply to other foods as well – e.g. beer. Could we be in a position where the app tells us we have drunk too much beer this week? Or passes this information on to insurance companies (who put up premiums), the health service (who may refuse treatment) and goodness knows who else, where we really will live in a 1984 style ‘nanny state’ with no privacy. 

The IoT has amazing potential, but sadly like any new invention, it won’t take long for someone to spoil it by using it for evil purposes. Look at how long it took from the first airplane flight to planes being used to drop bombs and kill people! therefore our privacy and security have to be baked in, and this is where Declan raised an important point – the usual way of doing something new is to build, test, release, get hacked, go back and add security, and play catch-up.

If we are going to do this properly and learn the lessons of the past, we have to build in security from the beginning, otherwise it’ll be yet another great invention that people do not feel they can trust. Remember, this will not just be about someone hacking into your Facebook account or worse, your bank account – this is information about the way you run your whole life will be going backwards and forwards. Where you shop, what you buy, your habits, your hobbies, your car and journeys you make, when you are on holiday etc. We really cannot afford to get this wrong!

It is a subject that for some reason has grabbed me, so I will be delving into this a lot more.

Mike Bartley –



Challenges when recruiting testers – number 6

Moving on to number 6 on my list of recruitment challenges is………….arrogance.

This might seem a strange thing to add, so bear with me whilst I explain.

A problem I find with candidates is to do with attitude, and it can come across just as much on the phone as it does in a face to face interview.

There is a fine line between explaining what someone has been doing in various projects and the contribution they made, and coming across as though they personally ran the whole test team and came up with every innovation ever implemented! A number of candidates I have spoken to think very highly of their abilities – which in itself not an issue, but if they can’t back up their opinion with cold hard facts, then perhaps this opinion is somewhat misplaced.

Luckily I do not come across too many candidates who speak ill of their current or previous employers (it’s a piece of advice that you can find within 5 seconds of searching for Interview Tips on the internet!), but I really do think that having a more humble attitude during interviews would not harm candidates who are over confident.

A side effect of this sort of attitude is the impression that I (as an interviewer) am lucky that they should deign to come to the interview, and quite frankly the questions are boring and beneath them. Some candidates are very easy to read – it’s all in the body language. 

Of course there is always the extreme example – there was a candidate who after every question I asked him, said ‘That’s a good question’, and then replied in a patronising way. That’s not the attitude I need in my team!

My advice to candidates is this: do not to make the interviewer feel like they are pulling teeth because you don’t give enough detail in your answers, and do not make it sound like you can single handedly save the testing world. Be realistic about what you can do/have done, have a little humility – there are times in business when it is needed, and if you can show it in an interview, it shows a depth of character to the interviewer and is a point in your favour.

Challenges when recruiting testers – number 5

Continuing the series, and number 5 on my hit-list of recruitment challenges is using different fonts and sizes on a CV.

Let me clarify – it is absolutely fine to have a header in one font and the paragraph or bullet points in another, but the issue is where there is a mixture of fonts and sizes within one paragraph or section.

Don’t get me wrong – I do not specifically look for these sorts of things, they just jump out at me. It could be due to my many years of testing, looking for things that are out of sync, but from a hiring manager’s perspective it can be a good thing to spot…..

An example is where I have put out a job advert where we are looking for a specific skill, and I’ll use Selenium again. If I see that a CV has ‘Selenium’ added to a list of tools that the candidate professes to have experience of, and it is in a different font to the other tools, or the same font but a different size, it looks as though the candidate has added it to the CV just before sending off to apply for the role.

Again, it seems a bit picky, but when you add it to spelling mistakes and vague language, you can start to build up a picture of a candidate. If someone has amended their CV to better fit the job specification, that in itself is not a problem (as long as they are being truthful about their experiences, and not just adding bits to look good), as they could be doing housekeeping and removing irrelevant areas.

But it does lead to wondering whether they do have Selenium experience or are they just trying to get through the first screening by adding it in? Do I want to spend time on the phone asking questions only to find they don’t have this and I have wasted my time? Candidates can miss out on interviews just through carelessness.

Once again, attention to detail is expected of a tester, so font types, sizes and alignment should be spot on!

Challenges when recruiting testers – number 4

Continuing the series, and number 4 on my hit-list of recruitment challenges is spelling mistakes, or as I call them ‘Typo’s’.

Now this is a tricky subject – indeed whilst typing this I have made mistakes (I have the ability to type ‘the’ as ‘teh’ all the time) but I proof read and make corrections as I go. Actually the fact that there is a red underline also gives the game away.

So why is this a problem? Well Testers need to have good attention to detail. Period.

If I am going to ask someone to ensure that something works or doesnt work, then that requires someone to notice if there is a spelling mistake on a web page as an example. It is hugely embarrassing for any organisation to make errors, and in today’s world of Twitter, Facebook etc, these mistakes can ‘go viral’ (I dont like that phrase particularly but it seems appropriate in this instance). In fact research in 2013 suggested that 59% of Britons would switch websites if they saw obvious mistakes (see It looks shoddy, and a tester should notice these things.

So, I expect to see evidence of this on a CV. Am I being too picky expecting there to be no mistakes at all? Maybe, but that’s my prerogative. As I said before, a CV is the only thing a prospective employer has that represents the candidate, and if that candidate cannot be bothered to spot and correct typo’s, then it points to laziness. I accept that English is not always a first language, and I do make some allowances in those situations, but most people use MS Word and it automatically highlights errors. And there is always the option to ask someone else to proof read it. If I wanted a job in France, and wrote my CV in French, I’d make sure to ask a native French speaker to read it through for me!

I’ve seen some real humdingers in my time – probably the worst was a heading in bold type, using a larger font than the main text saying ‘Profisssional Expeerience’. It has to be my all time favourite!

It’s so simple to fix, takes just a short time to do and can make the difference between getting an interview or being rejected at the first hurdle, but it’s a shame that so many people do not seem to bother.

Challenges when recruiting testers – numbers 2 & 3

Continuing the theme on recruitment with challenge….. 

Number 2 – People are not always honest on their CV’s.

It’s a shocking statement I know, but sadly it is true.

I am not naive enough to think that people don’t embellish a little, but there are times I despair!

The main thing to look out for is where something you are looking for (let’s use Selenium as an example) is on the job advert. The CV contains the word ‘Selenium’ somewhere in the personal statement or list of skills, but strangely when you do a search for ‘Selenium’, it isn’t anywhere else in the CV. So basically it has been added so that whoever does the first screening finds it and sends it on to the next person.

Another is where the candidate is taking credit for something they didn’t actually do or only partly did. Words such as ‘We’ are a giveaway. ‘We implemented a new automation framework’. Hmmm, yes, great, but who is ‘we’, and what was your contribution? Ah, you watched and they did the actual setting up. Brilliant!

Do candidates really think that they can get through the door and not get found out? Or are they hoping to swot up the night before their new job???

Number 3 – Using words such as ‘Involved in’ or ‘Knowledge of’.

This is a classic! Such careful use of words implies that the applicant knows all of the things that are on the CV, but is trying to vary the words to avoid being repetitive (every line saying ‘Extensive use of…..’ would be boring).

Ok, fair enough, but I have seen too many CV’s that list skills that I am led to believe are something that the candidate has done/can do, only to probe in a phone interview and find they actually haven’t done it at all. The responses are things such as:

  • I paired with someone and saw them writing the tests,
  • I looked into this one evening/one weekend but havent used it at work, 
  • I read about it somewhere before I came here today…

The list goes on.

I have knowledge of the men’s downhill bobsleigh, but only because I have seen it on the TV. It does not make me qualified in any way at all.

Golden rule….DO NOT LIE! It just wastes people’s time – the recruiting manager’s and the candidate’s.

Challenges when recruiting testers – number 1

Recruiting testers….

I find recruiting testers to be somewhat challenging! I don’t know about you – it may be easier for you, but there are a number of issues that I have dealt with over the last few years and I felt it was time to share my recruitment experience with others.

Recruitment is a lengthy process from the point at which you identify that there is a position to be filled through to the successful new hire walking in the door on day, so at the UK Test Management Forum in April I shared a talk with a group of Test Managers and we had some interesting discussions.

(I wasn’t aware of that many people doing talks on recruitment, although after meeting Stephen Janaway last week I found out that  he has a blog on recruitment which is well worth a read!)

Part of my April talk was around my Top Ten challenges with recruiting, and I decided it is time to put them up on the blog and invite your comments.


Number 1 – The ‘Scattergun’ approach.

By this I mean getting in CV’s for a role where the applicant is not suitable.

Now, to be fair, this may not always be the fault of the applicant, as many CV’s come through from agencies as speculative, but it is a time wasting pain, no matter where they come from.

When recruiting, I know what skills I am looking for in a role, and I expect to receive CV’s that contain those skills. There are compromises to be made of course, so if the requirement is for C#, Selenium & Jmeter, if a CV is sent through with two out of the three, then that’s ok, as sometimes a candidate may have similar experience (substitute Java for C# for example). 

BUT – and this is the issue – I get CV’s with none of those! Ideally a 70% fit works as it gives the candidate some room for growth in the role, but a zero % fit? Really?

Would you apply to be a driving instructor if you could not drive? No – you have to learn to drive first!

So why are I.T. jobs any different?

It may seem trivial, but the number of CV’s that are sent in for any job can run into the hundreds, and someone has to sift through looking for those elusive CV’s which actually fit the role. I would say that a good 90% are rejected during the initial pass. Imagine what we could spend our time doing instead…….

Challenge number 2 coming soon!

The Internet of Things

Last Thursday (26th June), I was involved with the Unicom organised Next Generation Test Conference at the Lancaster Hotel in London. My role was to enjoy the day, and then at 4pm take part in a Q&A session with four other test managers in a ‘Topic Guru’ panel session which was to run for just over an hour. I must admit to enjoying the questions, and the time flew by – luckily the other panellists were very open and I think we shared the answers well between us.

But there is one thing I particularly want to mention, The day got off to an interesting start with a talk about ‘The Internet of Things’, an expression I hadn’t heard before, but the whole theme of the day was looking to the future (of testing and other things too).

I need to do some more research around this, but what I took away from it is that we need to stop thinking about systems and applications, and start thinking about people – yes, human beings like you and me! It’s about the way in which we interact with the internet to gain access to the amazing amount of information available to us. It makes perfect sense, as although we write systems to perform tasks, run batch jobs, do uploads, downloads, schedule reports etc, the end result is that data in some form or another is going to be looked at by a person, and a decision made on the basis of that data, whether it be a business or commercial decision, or a personal one, such as which insurer you should choose based on the cost of insurance provided by a man dressed in a seafarers uniform, a nodding dog or a cute furry animal!

As testers it’s easy to look at a user story or requirement and think ‘Ah yes, the system should do x when the user presses y’. Fine, but we should be questioning what the user needs to get out of this interaction in the first place. We design systems that are complex, clever and look great, but often it’s the least technical person who comes along and asks why they need to press 5 buttons just to create a simple report or add in some data, and they would prefer to do this with as little interaction and navigation as possible.

I’d be interested to know your thoughts on this subject – I feel it is one that can run, and run, and run……