Re: PR luvvie
"But spaceX still have to do the inflight abort test, and prove that the strut issue is behind them by doing a few more launches."
Sure, they have to get some successful flights off the ground, but there's more to it than that. The last flight failed because their quality assurance wasn't up to scratch. SpaceX said as much. If they've done nothing to change that, and merely decided to take a closer look at that strut, then no one knows what other problems are lurking in their manufacturing build chain. If they had poor quality assurance on struts, what else were they not looking at properly?
Effectively the next few flights are a test of the processes they have for building a rocket, not really a test of the design itself. Once you've had a single successful launch (which they have), the design is proven. Every failure after that is down to the inability to control the build of that design, which is seemingly where SpaceX have been for some time now. And if they've still not got the right build quality assurance process then their rockets will continue to fall out of the sky.
Of course, SpaceX aren't stupid and I'm sure they know this as well as anyone else. I hope that they have got on top of their process problems. If they haven't, someone will eventually have to call a halt to their activities if they haven't gone bust first. You can't keep launching rockets over people's heads without a satisfactory approach to quality control, not in this day and age.
Russians
The Russians are still learning things about the way they build their 1960s design. The routing of a pipeline was thought to be non-critical until quite recently. Then they built one where this pipe, purely by chance whim of the construction engineers on the day, shared a bracket with another. The fact that one of them was carrying a cryogenically cold fluid and would freeze the fluid in the other (the heat being conducted by the bracket) was something they discovered only after it failed and crashed... Needless to say the pipeline routing has been rethought.
Arianne
The first Arianne 5 suffered a navigation failure due to dodgy re-use of Arianne 4 software (the spec was fine, but the actual implementation didn't quite meet the spec. Very embarassing). It was blown up by the range safey system. They've had only 2 out of 83 outright launch failures, and those were both a long time ago in 2002. They've been sending them up regular as clockwork for years now, and from what I hear they very rarely suffer even 1 second of launch delay (except for weather).
Commercially speaking Arianne is a rock-solid proposition; slightly more expensive than, say SpaceX, but your $billion satellite will get launched on time into the right orbit with the cheapest insurance premium in the business. That commercial certainty is worth a fortune to the companies that commission and operate satellites, far more than the actual cost of the launch (thought to be €100million for Ariane 5).
All the time your satellite is sat on the ground waiting for the launcher guys to get their act together you're losing a monumental fortune in lost revenue, interest payments and programme costs. A cheap launch that's 6 months late is hugely expensive for the satellite owners.