I worked on a Y2K project once where it was full of QA consultants.

I’m not even sure what a QA person is anyway, it certainly isn’t anything to do with what I do which is to test software till it works. With these QA people, I wasn’t even convinced they had that much experience, as they appeared to have better knowledge in other areas of business rather than IT. But then what do you expect. It was Y2K time and all companies were panicking a bit and taking on anybody to test software and paying way over the odds.

These QA guys did push out a hell of a lot of documentation regarding the generation of quality systems, and I mean a lot. But they were a bit reticent about actually testing any software i.e. they didn’t actually want to get their hands dirty.

Now the cynic in me might say that they didn’t want to actually test software probably because they’d never actually done it before and maybe never actually worked in IT before and therefore didn’t know how to do stuff. Which did make me think that most of them probably had just read a very dry QA manual, rejigged their CV a bit and voila, instant expert QA consultant. Of course me in the middle of all of these non-practical experts was interesting. I did enjoy some of the conversations though.

One in particular, where somebody tried to convince me that he could prove, via documentation alone, that a system was a quality system. I progressed the argument a bit and eventually the expert admitted that you could have a quality system with quite a lot of significant bugs still in it.

Eh. A quality system riddled with defects. What does their version of the word ‘quality’ mean? Whatever was their interpretation, it doesn’t seem of much use.

Seems that the word ‘quality’ had been hi-jacked and was something that I never wanted to be involved with. 

Hell. My last post was a week ago. Is this somehow connected to the fact that I started seriously kicking some code last week. Who knows.

Did some really nasty back-end edits, ran the software and watched bits drop off here and there.

Raised a dozen defects and I’m not sure some people really like it. In fact I’m supposed to be writing some high-level test plans right now and not testing at all. Apparently we just don’t have time to do all this testing. Hmm… and this is software that is due to go live next year and has been produced by a software house that is infamous for developing very buggy systems.

The one thing we do have is time. We have 5 months to test this system to death, to find every defect coded, to test every avenue, every requirement, everything. We might then stand a chance of not being embarressed.

History: In one of my previous posts I mentioned a Y2K testing project where the Y2K testing barely happened. We had a young chap on the project who was an administrator, not a systems admin person, but an office administrator. We are talking about administrating contractors timesheets here. The large project management company that ran the project gave him a new job description – systems administrator – and changed his work terms from being an office temp to being a software contractor and tripled his rate. They were then able to make him one of their consultants and therefore able to make him very expensive to the end client. The one-time temp was very happy and now new that the streets of London are paved with gold. True story.

Quote of the day

‘I don’t want any yes-men around me. I want everybody to tell the truth even if it costs them their jobs’ Samuel Goldwyn (1882-1974)

In 1998/9 I was working on a Y2K (remember that) project for an insurance company in London. The project, which like most Y2K projects, wasn’t so much an IT project more of a management project.

It was managed to death by a large and successful management company. There were dozens of us on the project. Old developers and systems analysts that hadn’t been in work for years, very essential logistics people, sudden test ‘experts’. The caboodle. And above that motley crew the management company, managing away like mad. It was quite impressive.

These guys could have meetings about anything. And very professional they were too. Didn’t have much relevance to what we were supposed to be doing i.e. making sure old bashed-up software wouldn’t collapse at 1 sec past midnight 31st December 1999, but boy they were high-profile and did the insurance company love it? Yep.

There were guys on this project, who were managing the testing of complete suites of systems. Not a couple of applications, we are talking complete personnel systems, various insurance systems, bill payment systems etc. And guess what, these managers didn’t actually have any team members. They had to do all this work themselves. Most had no testing experience and when I talked to them it was obvious they were completely out of depth. I have no doubt those systems were never system tested at all. Just as well the programming was ok.

Another curious thing was that about June and July 1999 the management company started having meetings with a different slant, they just started talking about the Y2K system testing in the past tense and about ramping down the project. The whole project was managed beyond belief but barely any IT was performed. If anyone had actually found a defect and reported it I don’t think the management company would known how to cope.

Quote of the day:

‘In a few minutes a computer can make a mistake so great that it would take many men many months to equal it’ Unknown