Quantum of Anti-Trust

This week’s Crux has lots of (in)security news, valuation shenanigans, and a smidgen of antitrust. There’s also predictably bad behaviour from people we always knew were terrible, but a bunch of other positive news about hilarious brand names and improved security postures.

And a little bit of quantum.

Things to note

In news that surprised nobody, Uber “broke laws, duped police, and secretly lobbied governments”. We have to expect that this is how well-funded competitors will function, given the lack of consequences for such behaviour, and plan accordingly.

The whistleblower, Mark MacGann, doesn’t exactly have clean hands, either, but better late than never, I guess.

STMicro and GlobalFoundries are setting up new chip factories in France. There’s likely to be a bit of a capacity glut once these new fabs outside China all come online.

Broadcom’s software president, Tom Krause, is leaving now that the VMware deal is done. VMware people seem happier that Hock Tan will be running things rather than Krause.

Credit bureau Experian is abysmal at security. Maybe don’t use them any more if security is important to you.

Klarna is showing its investors how to buy now pay later with an 85% lower valuation after its latest fundraising round. I’m unclear on the difference between Klarna and a loan shark but it probably involves some kind of app.

Red Hat and ABB have started a global partnership for process automation and industrial software. ABB makes a lot of the stuff that makes embedded industrial systems go, so this is a big deal for Red Hat getting their stuff out into the wider ’edge’. It also means bugs in OpenShift will help us pwn things that explode, so that’ll be fun.

In other Red Hat news, Matt Hicks has donned the reddest of hats to become President and CEO. This seems a good appointment to me. Hicks is generally well regarded.

‘Cloud-native’ database SingleStore has raised $116 million in new money. There’s a lot of database funding action at the moment, particularly ones that can even claim a bit of cloud-native-ness. I’m still digging into why, or rather why now. The data trends aren’t much different this year compared to last year.

Kubernetes policy engine Kyverno is now a CNCF incubating project. Policy management is growing as Kubernetes matures and people start deciding that just throwing lots of random stuff in a pile isn’t quite the model of governance they want.

PyPI is insisting on multi-factor authentication for maintainers of ‘critical’ Python packages to better secure the software supply chain. Some maintainers have had a sook, and this writeup by James Bennett patiently explains why they’re wrong to do so, which I agree with. I’m friends with a bunch of the Python folks so I’m not entirely unbiased here. They work incredibly hard to make things as inclusive and painless as possible, so the sooking really feels like misplaced ego to me.

Matthew Garrett has an important writeup on the stewardship of the UEFI secure boot ecosystem. Microsoft are being dicks here, but we need to figure out a better way of doing security than letting a handful of tech companies decide what it should be.

A bunch of new ransomware groups have sprung up because there continues to be lots of money to be made doing ransomware. You get more of what you reward, so we’re going to keep getting more ransomware.

I reckon Fermyon’s WebAssembly microservices Platform-as-a-Service thing is probably going to be a big deal. Also CEO Matt Butcher is the most ridiculously likeable person I’ve met in ages.

Australian retailer Woolworths bought data analytics company Quantium and have decided to call the combination ‘wiq’. We assume the brand brief was drafted in vim and they couldn’t figure out how to quit before the ad agencies started work.

Amazon is trying to head off EU antitrust investigations by claiming it will stop using sellers’ data for its own competing retail business and private label products. The Everything Store wasn’t a subtle hint, folks. Pay attention.

Advisor to many IT buyers Gartner says that many IT buyers experience buyers’ remorse. It seems 56% of organisations had a “high degree of regret” over their largest tech-related purchase in the last two years. Irony buyers are welcome to send me their thoughts by reply email.

And lastly just because it’s incredibly cool: Physicists have done amazing ’time reversal’ quantum things to measure vibrating atoms. Quantum mechanics is so weird. I love it!

Longer reads

A nice paper on Governing data and artificial intelligence for all: Models for sustainable and just data governance. “With a particular focus on artificial intelligence (AI), this study identifies and examines policy options for the EU’s data governance framework that align with a data justice perspective.”

Weekly Tip

Because technology tends to deal in physical things or glorified mathematics, we tech folk like to think we’re deeply rational people. This is despite the abundance of evidence to the contrary, but it colours the way we talk about tech.

A lot of time is spent talking about easily quantifiable things like database gigabits-per-microfortnight or how many acres the turbo encabulator covers. Speeds and feeds it’s frequently called. But this ignores a lot of what actually motivates human beings to do things. Lots of decisions we make are for less-than-rational reasons, even by tech people who believe they’re purely rational beings of unparalleled logic and reason.

Developer experience is a good example, because very little of that is about things that are truly quantifiable in the same way that network bandwidth is. How do you explain why you prefer to write code in Ruby or Python or Perl and not C or Typescript or Powershell? And other people prefer different things?

I like to use this Reward/Experience Framework table to remind me of the other, non-rational benefits that motivate people to want to do something.

Note how only one column is about rational benefits, and three are about other things? And how some benefits aren’t actually about the way you use a technology at all?

I have a working theory (backed by a bunch of evidence and research literature) that the tech industry has a lot more in common with the fashion industry than we like to admit. I’ve been researching it for a while, so if this interests you, get in touch and I’ll be happy to share some of my citations and ideas.

For now, maybe consider talking about something other than the size of your RAM?