Designers should code, but won't

This is very true. I have been experimenting with Bubble.is, where I can focus on programming instead of writing code – there is a big difference. With Bubble, you can actually create a living project that uses real data, repeating components, component styles and pretty advanced logic without writing any code. It's quite impressive and has been fundamental for us to build high fidelity prototypes that communicate our intent to testers and developers.

I had a Magic Mouse, and personally disliked it:

1. Magic Mouse 2 has the charging port on the bottom, making it useless while charging. Get used to charging it overnight, or else ..
2. You can never really relax your hand on it. If you do so, it will click.

Best mice I've used were the 2012 Razer Mamba and the Logitech Performance MX. Both feel super smooth and precise for designing, and work both wired or wirelessly. They are also very customizable via software, so you can easily fine tune the acceleration and sensitivity. Your experience may depend on how you grip your mouse – the MX is best for a palm grip (like the MX Revolution) and the Mamba is a little smaller, and works well for a claw grip.

If you haven't already, I would complement the mouse with a good mouse pad such as this one.

Before migrating to Sketch for UI design, I depended heavily on Specctr for generating specs in Illustrator.

Also, I use an assortment of scripts, especially Murdoch Carpenter's ExportAndroidResPNGs.

Illustrator is the man. Tried Photoshop, Sketch and Affinity Designer for UI design, but couldn't match the agility I have with Illustrator when it comes to blocking out flows, experimenting with layouts, creating form, finalizing assets and exporting specs.

I also use:

• Specctr add-on to generate UI specs for devs;

• Since 95% of my work is for Android, I use a slightly modified version of Murdoch Carpenter's Multi-res PNG exporter for Android. Adobe shipped the new Export Assets feature last month, but the file structure it generates is only good for iOS assets, not Android, so I'm sticking to the script.

Invision for quickly throwing together mid-res clickable prototypes, and also serving as context for translators to annotate their translated strings.

Pixate for certain high-res prototypes that require more advanced and specific behavior not covered by Invision's basic interaction options.

After Effects for complex animations, and also for simulating UI state changes. Linking certain parameters to sliders for example, can be a good way to design and preview in real time various different UI states that depend on certain conditions.

Asset Export is a useful addition for iOS design, but it's practically useless for exporting to Android, since it saves all the files in the same folder, and differentiates the dpi via filename suffix. Android projects have their resources organised in folders for each dpi, and the files for each asset should all have the same name.

People have been typing it all up into a spreadsheet: https://docs.google.com/spreadsheets/d/1lWnncM61FsDb47jMheTc2BP-c1YaWo6BYeblpd0Iiqk/edit?usp=sharing

Hard to attend, but worth looking out for the online content they may release afterwards: https://design.google.com/events/span/

I'm hoping for the best. Google has a well-documented design spec, and now it needs an official prototyping tool to complement it. I have been loving Pixate especially for how it can emulate Material Design's transitions and animations, but haven't seen significant improvements lately. I hope Google can help them push these updates faster.

Great observations and examples.

Just a heads up for the terminology – legibility has more to do with the actual typeface design, while readability is what is being neglected in the examples. You can see that all the examples use legible type, but when set over images and poorly contrasted backgrounds they render unreadable designs.

So yes, legibility ≠ optional, but in this case you're saying that readability ≠ optional, which is also very true.