I’m happy that my last post generated some discussion. I’m not sure about the rest of the blogosphere, but for some reason I seem to get half or more of the comments privately instead of through the blog comment system. Something I heard from a few people was this comment (anonymity retained).
Why don’t you list the forward thinking that testers should be concerned with? A list of topics may help advance things.
If I think if I had the full list of where software testing was heading, that I’d certainly include it. But because that would mean I also had the ability to predict the future, first, I’d buy a lottery ticket.
But they probably(?) didn’t mean that. I hope the questions were more about what I thought some areas to explore may be. I’m not sure if I know that either – but I suppose I can share some random ideas (that I may or may not explore in future blog posts).
- I think we have a bit of an issue with data. On one hand, we don’t know what to do with the data we have. I’ll be stronger – we don’t have a freaking clue what to do with the software related data we have today. But we don’t have enough data to make good decisions. There are two issue here – we need more data, and we need a way to interpret that data in a way that lets us make great decisions.
- We really have no idea (again, in general) how to design tests. We don’t know which set of tests are relevant for a given software context, we don’t understand which tests should be automated or performed manually, and we don’t know how to design robust and reliable tests when we do automate. I know that this doesn’t apply to some of you (or you think it doesn’t apply), but testers do a huge amount of under-testing and over-testing. Finding a much better balance is a huge challenge.
- I’m concerned with the way most test teams (and many software teams in general) are organized and managed (I mentioned this in a comment, and Catherine Powell added her thoughts in a blog post). Software creation is knowledge work – yet we manage teams like they’re making widgets. That approach is idiotic at best.
What I’m most worried about is what I stated originally – I’m worried that not enough people are worrying about this stuff.
But then I wonder if I should worry at all? Maybe software will always sort of suck, and average quality software is enough. If that’s the case, then little or none of this matters – it’s all a waste of money. I’m serious – one possibility is that people just don’t care enough about software quality to pay for it. Is it a viable option to stop (or massively reduce) software testing and pass the cost savings on to customers?
Something else to explore later I suppose.
“One possibility is that people just don’t care enough about software quality to pay for it.”
I think this is a significant possibility. I believe I’m seeing evidence every day which suggests that people are satisfied at rather low quality levels, as long as the price is right.
I’d love to see you explore this further. Heck, I’d love to see *me* explore this further! (Where can I purchase some free time for writing – cheap?)
Yes – I was serious when I said it…which has caused some deep soul searching on my part on what I do…more to come.
>> But then I wonder if I should worry at all? Maybe software will always sort of suck, and average quality software is enough. If that’s the case, then little or none of this matters – it’s all a waste of money. I’m serious – one possibility is that people just don’t care enough about software quality to pay for it. Is it a viable option to stop (or massively reduce) software testing and pass the cost savings on to customers?
I wonder if somewhere in Detroit back in the 70’s someone was saying something similar about quality in the auto industry…
At the place where I cut my hair, there is a sign where you can read something like: ‘The sweetness of the low price fades soon and only leaves the bitter taste of dissapointment because of the low quality you’ve paid for. ‘ And those folks are the 5th generation of hairdressers.
To stop testing can’t be the way to go, I agree with Ron on that.
Alan, as always, the questions generate more questions!
I had my shower floor retiled recently, and he came in and said this:
“Whoever installed this put standard sheetrock behind it, which was wrong. The sheetrock is wet and is rotting. So I can do just your shower floor, and it’ll cost you about $1000. Or I can take the whole thing out and redo it with duraboard behind it so it’s waterproof, and that’ll cost you about $4000. Or I can take the whole thing out, put in duraboard, put in a waterproof lining, and then redo the tile,and that’ll cost you about $6000. The cheap way will last about 6 months before it leaks, the middle one will last you about 8-15 years, and the expensive one will last you about 15-18 years.” And here’s the thing: once I get over a certain bar of quality – won’t leak any time soon – I’m set. I don’t need to keep paying more for quality that frankly I can’t really comprehend (redoing the shower in 15 vs 18 years? I can’t really distinguish between those two).
I think it’s a bit of a misnomer to say that software always sort of sucks. It’s completely true, but more pessimistic than necessary. That’s like saying that my shower will always suck because eventually it will always leak eventually. That’s true, but there’s a broad range of sucking available, and we can fall anywhere in those shades of grey.
Should we worry about doing better? Absolutely. Do we need to worry about keeping up? Definitely. Do we need to worry about handling the bigger/stronger/faster things they keep throwing at us? You bet.
But let’s recognize those for what they are: an opportunity for continuing education. I think it’s a positive sign that we’re being challenged to handle bigger better stronger more complex systems. They wouldn’t ask us to do it if they didn’t think we could find a way. Now we just have to worry about finding a way (or more likely several ways).
Fantastic points, and just the sort of discussion I was pushing for. Thank you.
Compare Iphone and Windows Phone (Windows Mobile 6.5). You can not say that apple did a better testing that Microsoft did, but they had a better design, a better idea, maybe a better development… and people pays way more money for an Iphone than for the Windows Phone. My point is that we’re not alone on this. We do need smart people having bright ideas about what is the future like, and superb developers with quality in mind, because you will never success as a tester in a situation you got lame specs and poorly implemented software.
We need help, from the world outside to success as testers!