Toyota Line Stop-Oye vey.
I read a post about the line stop at Toyota that was so complimentary. I just don't get it! They have cut so far they aren't even number 1 in quality anymore yet it seems everyone in testing is a panting fanboy for Toyota. Look at the product? Just not exciting and now not even slipping in overall quality for multiple years yet most people in quality want to be them?
Why? How about the fact that software isn't a car. It has a far shorter shelf life in most cases and it isn't a tangible item. Also, how about the fact that the actual quality is SLIPPING not improving over time?
Also, I'm sick to death of the process fanatics. I don't mean those who want to improve things, but I'm talking about this HUGE reality gap between what they say and what actually happens. Let's talk about Lean, Agile, Flavor of the Sell Me Process of the month. Some of it works really well. I like some of it. However, just the people who have these unrealistic ideals and are too far from the work to know that there is a major reality gap that exists. Sometimes I do read about the perfect land of unicorns and fairies and how that now that we're agile all of our problems are better. It's like a death march. All of our problems aren't over. We just make the same mistakes faster now and sometimes it helps and sometimes it just sucks. I don't know if it saves money or not. Why are we blindly following a car company with quality and reputation that is on the decline instead of trying new innovative ideas that actually APPLY to the kind of products we make?
On my team we just really hose stuff up faster, and sometimes that's good! Sometimes I get so flustered I wonder if this pace is sustainable. Why do our sprints feel like a death march? Isn't this supposed to help? We're still doing the trail of tears, just with a few more checkpoints on the way. Bleh. Anyhow, yes, lots of room between what the sales brochure of process improvement says and what really happens and although I think it is unfair to just pick on Agile, Lean, or XP as the problem as there are many good goals, I do think talking about problems in QUALITY needs to happen more. This "minimum viable product" idea is demoralizing. I'm against it. Really? How much less aspirational can you get than doing the bare minimum with a stripped down beaten down crew.
Also, why is testing and bug investigation and fix verification"Free" with no time dedicated to it. I feel like our "agile" just mean slots of reality denial and assumption of total success for everything we build with NO buffer. Reality gap number 947.
It should not be taboo to speak the truth about problems encountered on agile teams. It pisses me off to no end that no one ever considers integration testing as a valid backlog item and that our "agile" adoption has made my worklife balance suck and no one will talk about it. These are real problems!
On the other hand, the pairing ROCKS. It is going so well and producing so much better quality and I have actual confidence I really helped the team improve stability on our project in the last sprint. I like some things about process improvement, but fanatics do get a bit annoying.
In summary, I like many things about some of our process improvement, but I fear that some fanatics have lost all touch with reality and need to please come do my job for a week or at least admit that most places are struggling. What are they doing to address and help teams through the tough problems that arise? Are they coaching or just selling? I see some awesome coaches and that gives me hope that it can work. Company culture needs to be considered when proposing process change. Rather than selling the change, why aren't more people coaching humans through out so that the change can be successful?
Why? How about the fact that software isn't a car. It has a far shorter shelf life in most cases and it isn't a tangible item. Also, how about the fact that the actual quality is SLIPPING not improving over time?
Also, I'm sick to death of the process fanatics. I don't mean those who want to improve things, but I'm talking about this HUGE reality gap between what they say and what actually happens. Let's talk about Lean, Agile, Flavor of the Sell Me Process of the month. Some of it works really well. I like some of it. However, just the people who have these unrealistic ideals and are too far from the work to know that there is a major reality gap that exists. Sometimes I do read about the perfect land of unicorns and fairies and how that now that we're agile all of our problems are better. It's like a death march. All of our problems aren't over. We just make the same mistakes faster now and sometimes it helps and sometimes it just sucks. I don't know if it saves money or not. Why are we blindly following a car company with quality and reputation that is on the decline instead of trying new innovative ideas that actually APPLY to the kind of products we make?
On my team we just really hose stuff up faster, and sometimes that's good! Sometimes I get so flustered I wonder if this pace is sustainable. Why do our sprints feel like a death march? Isn't this supposed to help? We're still doing the trail of tears, just with a few more checkpoints on the way. Bleh. Anyhow, yes, lots of room between what the sales brochure of process improvement says and what really happens and although I think it is unfair to just pick on Agile, Lean, or XP as the problem as there are many good goals, I do think talking about problems in QUALITY needs to happen more. This "minimum viable product" idea is demoralizing. I'm against it. Really? How much less aspirational can you get than doing the bare minimum with a stripped down beaten down crew.
Also, why is testing and bug investigation and fix verification"Free" with no time dedicated to it. I feel like our "agile" just mean slots of reality denial and assumption of total success for everything we build with NO buffer. Reality gap number 947.
It should not be taboo to speak the truth about problems encountered on agile teams. It pisses me off to no end that no one ever considers integration testing as a valid backlog item and that our "agile" adoption has made my worklife balance suck and no one will talk about it. These are real problems!
On the other hand, the pairing ROCKS. It is going so well and producing so much better quality and I have actual confidence I really helped the team improve stability on our project in the last sprint. I like some things about process improvement, but fanatics do get a bit annoying.
In summary, I like many things about some of our process improvement, but I fear that some fanatics have lost all touch with reality and need to please come do my job for a week or at least admit that most places are struggling. What are they doing to address and help teams through the tough problems that arise? Are they coaching or just selling? I see some awesome coaches and that gives me hope that it can work. Company culture needs to be considered when proposing process change. Rather than selling the change, why aren't more people coaching humans through out so that the change can be successful?
Posted by Testy Redhead at 31 Jan 2010 
Categories: uncategorized
Tags: fanboys quality process automation worship testing
Categories: uncategorized
Tags: fanboys quality process automation worship testing
What about the difference between car design & development and car manufacturing processes?
Reply to this
Great point! What about it? I can't think of a more waterfall method than having the design totally complete before manufacturing, yet Lean enthusiasts point to Toyota. Interesting point. I'd love to read a blog on it. Link me to it if you have one.
Reply to this
When a car is being designed engineers and scientists actually build [working] models, perform usability and crush tests, experiment and investigate results... this could be compared to software development process. Both processes have a fully functional prototype at the end.
On manufacturing stage it's totally different.
Car manufacturing is a production process. An assembled product may (and does) deviate from the prototype. That is why a role of QC is as much as important as on development stage.
Software manufacturing is a digital replication process. A copy is absolutely identical to its original.
Reply to this
Interesting point! I recently switched to doing QC from mostly a job in testing during the design phase. It's been really interesting. We do lots more checking in QC.
I'm very fond of the idea of testing in production although it scares me to do it.
Reply to this
The main difference between designing (=coding) software and design a car is that manufacturing the end product from the design documents is so fast and cheap, that we can realisticly do it a hundred times a day. That allows us to use the real end product where other disciplines have to use simulations or models.
Reply to this
The first thing I tell people when teaching people about Lean and any business (software, law, whatever) is that any process can fail because processes are made of people. And people fail, it's what Tiggers do best.
They keys to good process for me are:
1. Is it as small as possible?
2. Does it create transparency?
I want everyone to know what's going on and I don't want them to work to find out.
Anything after that is crap put in to make people feel like the process is "real".
Reply to this