It's a beautiful rant, and I feel his pain. How does it happen that Google goes from having the best maps on the web to a broken toy? That Amazon goes from having fantastic search to returning lists of random items? That modern browswers can once again have visible lag when typing fast? Toward the end, he offers his solution:

  I think an interesting idea for development would be to 
  have a totally separate "finishing team". You get all new 
  developers and managers on at the end, so that they aren't 
  wedded to any of the ideas, they aren't building fences 
  around their work and unwilling to change anything, they 
  aren't prideful and unable to admit that it sucks. The 
  finishing team is only concerned with the *results*, what 
  the actual user experience is, and they can mercilessly cut 
  up the product to make it better. Politically the finishers 
  have no stake in keeping work, so they can even revert huge 
  chunks or just cut features.

Does anyone do this? Or would it just be too demoralizing?

I have approximately zero of these issues. But maybe using a modern operating system is also one of this guy's/girl's pet peeves.

It is unbelievably stupid to stay on such an old browser. Browsers are the main vector of attacking an individual user, and you can't sacrifice security updates in order to keep a UI that you seem to prefer.

> I have approximately zero of these issues.

I don't believe you.

Here's one example:

> > focus should never move without the user doing it", On iOS using Chrome: you visit a webpage. That page has an ad for an app. The app-store opens, which steals focus from Chrome. This is on a "modern" browser and OS.

Most of his rant is accurate and is not related to old browsers or old OSs but to the fucking stupid things that many web designers do.

Never happened to me. Like most people, I don't tap on ads (especially for apps).

So you're just stuck picking which way you want to be screwed over, eh?

If a product is developed correctly, it will please less than 100% of people. Some people will never be happy with what's out there.

I'm one of the whatever percent of people that is mostly satisfied with my browser. It stays almost entirely out of my way. My UX issues on the web are typically due to the web.

So to answer your question, some people may be making that choice, but I, like most people, am not.

The problem here is when someone is happy with what they have, but would be less happy with the new version because it sacrifices something which was important to them. Then they are stuck with a choice between allowing their computer to become less useful and allowing it to continue to be less secure. It's not clear that "upgrade anyway" is always the right choice.

Yeah, rounded corners. ouch. He is so screwed now.

He has to update to get security fixes. When updating also fucks perfectly good software, well, he has a point.

And it's something that many people have noticed and commented on. See ancient discussions about the version of Nero / ACDSEE / whatever that don't suck.

tl;dr CHANGE, I HATES IT! LET ME SET AN INSECURE PASSWORD GOD DAMN IT!

That guy has a rough life.

What he saying is this: he knows a grand total of 0 people whose accounts have been hacked because their password was too simple. It's almost always something else. Yet password management only seems to become more complex and cumbersome over time.

And it's not even debatable that major apps, sites, and operating systems ranging from Google Maps to Firefox to iOS have grown suckier and suckier over the last few years. This isn't old-man-yelling-at-cloud stuff -- it would easily show up in usability studies, if the companies involved actually ran valid usability studies. When I encounter the same daily frustrations as my non-computer-geek friends, something's wrong. That never used to happen.

Before salting became mainstream simple passwords were broken all the time when data was stolen

But that was a server-side issue, like virtually all modern attacks that don't involve trojans.

When it comes to security, a lot of misguided "best practices" seem to be geared toward making users pay for the sins of the admins.