Sorry this took me so long to crank through. What can I say. It was Valentine's Day and there were better things to do with my time.

But it's worth reading. Thanks. I disagree with his conclusions, though, and that's worth talking about, too.

The basic gripes of the piece are:

1) Don't change stuff in the name of progress unless it's actually progress

2) Don't change stuff without thinking about why you're changing it

3) Don't simplify stuff for the sake of simplifying it without accounting for why it's complicated.

As whipping boys for these arguments, the author uses three whipping boys: Twitter, Tumblr and KDE. And while I don't know KDE from my armpit, I know OS X and a lot of the KDE programs mentioned... and I know Twitter and Tumblr's latest foibles. And I think the author is griping about a symptom, not the underlying cause. Wanna hear a joke?

Q: How do you get dual monitor support on a 2008 Mac Pro under Mavericks?

A: Connect three monitors.

This is a documented problem across the Internet: Yosemite, Mavericks and El Capitan suck at multi-monitor support for older Macs. It's even been argued that it's an easy fix. It's a maddening one, too - you update your OS and all of a sudden your desktop is smaller. I've been dealing with it for a year now and it's become abundantly clear that Apple doesn't give the first fuck about older Macs. Why not?

- They're never making another dime off those computers

- Abandoning them makes their newer computers faster through streamlined code

- Actively screwing up your old shit forces people to adapt to new shit

And adapt I did, clear out of Apple's ecosystem. Which probably suits Apple just fine as they've effectively abandoned the pro market. If it isn't an iSomething they don't give a fuck. In this case, it's not that Apple "forgot" about their older systems, it's that Apple actively decided to burn those systems in the name of progress.

It's a profit-driven motive. If you aren't monetizing those systems anymore, why support them? Which gets to the other problem with the article: Engagement matters far more than efficiency in free software.

Yeah - Twitter is changing shit just to change shit. Thing is, their existing userbase doesn't fucking matter. Maybe if they tweak it they can get more people to look at it. And really - there's no way to measure productivity within Twitter because Twitter doesn't do a single productive thing. However, the more buttons you mash to get something done, the more metrics you're generating that prove you're "engaging" with the software. These are metrics Twitter desperately needs, as its stock price is swirling the bowl right now.

See also: Tumblr, which Marissa Mayer was trying to turn into Youtube. Yahoo's goal with Tumblr wasn't to keep existing Tumblr users happy, it was to bring new Tumblr users into the fold. They'll do anything to accomplish this. The whole dustup where they made the default image aspect ratio 16:9 rather than square was all about reblogging video at the expense of pictures, which is all about engagement. Tumblr actively making post editing difficult is the same idea: you're less likely to purge old media links that way, which increases their backlinks, which increases their SEO metrics, which increases their value. The fact that nowhere in here does the user matter is par for the course: Tumblr creates nothing of use to Tumblr except links and engagement, so links and engagement are what they're optimizing for. Tumblr also has a rep for controversial content and burning old users without a second glance if they're slightly controversial is exactly what Yahoo did to Flickr, which chased away a lot of their hard-core users and made everyone forget Flickr for half a decade. I'm sure it made sense for Yahoo. It didn't make sense at all for Flickr users.

And that's really the core issue: if you don't have a product whose utility can be measured by common-sense metrics, you'll use exotic, ridiculous metrics. That which is measured is managed and if you can't measure your customers' useful qualities you'll measure their useless qualities.

So really - I don't think this guy's beef is about UI. I think his beef is about the fact that he's defining "customer" differently than the companies he works with. And I think this is a realization we're all going to start having soon.