Yea em dash with spaces looks better to me too, I find that it’s harder to read if the em dash is there without surrounding spaces. Looks too cramped, not separated enough.
I have never understood the classical rule of no spaces around em-dashes. If you’re going to use fancy dashes at all, an em-dash represents a clear pause, a break in thought — something more robust than a mere comma. Typesetting an em-dash sometimes literally touching the words on either side has the opposite effect, visually connecting those words rather than separating them, and unlike a lot of the typographical snobbery we sometimes engage in, that one is a well-known (at least to designers) effect of proximity. Personally I prefer a thin space rather than a full one in media where it’s possible, purely for cosmetic reasons, but I’d rather have a normal space than none.
How does Manifest v3 compare to Safari content blocker API in terms of ad blocking capability? Currently AdGuard for Safari (from Mac App Store, the one that uses native content blocking API) has 7 content blockers categories you have to enable, and each of them allow 150k rules, amounting to over a million rules total although you can’t count it like that really. V3’s 30k per extension and 330k total sounds like a lot less.
Is there something more to it than just the amount of rules, is v3 still better than Safari?
Yeah I hadn’t used a Windows machine for a year or so, and fired up Edge Dev just recently. The amount of adbloatware is truly baffling, it’s almost incredible.
Afaik Windows doesn’t ship with a default PDF viewer other than Edge’s, so they probably made a pretty sweet deal with Adobe to let them in to upsell their subscription.
For anyone interested (and using an M1/M2 Mac), whilst Amazing AI offers limited twiddling (just the step count and seed) it is native Apple Silicon, comparatively light on resources, and on my base model 2021 Air (8GB RAM, 7 GPU) it produces a 2048x2048 pixel 100 iteration image in under 5 minutes.
I know that headline number is slow compared to machines with beefy specs, but remember that this is 100 iterations on the lowest power M1 you can get, with 16 times the pixel count of the online 512x512 ones. And even with 40 iterations you can generate some stunning images with the right prompts.
Draw Things has a massive amount of configuration, is equally fast (it's also native) and can do a variety of output sizes.
Haven't noticed anything, but TBH I also haven't looked as I've had no need to question things - the rest of the OS runs fine alongside, with dozens of browser tabs, plus both Amazing AI and Draw Things running (though only one actually generating at once). And with about 50 iterations I'm losing around 1% battery every couple of images so it doesn't seem particularly taxing, whilst running cold.
Edit: It isn't obvious with Amazing AI but it may in fact be doing what Draw Things makes explicit in the options, which is generate at 512x512 and use your choice of upscaler to get to 2048x2048.
Yeah it’s been extermely slow with the updates. The Bear 2.0 is currently on Mac-only beta but not yet open (unless you join their beta forum and ask for access), but even then, it doesn’t have the web app. I’ve seen the Bear team say that they did a full rewrite (in C++ I think?) so that it would be easier to implement the web app in future but who knows how many years in the future that is.
I may be wrong on the features on Bear 2.0 but it doesn’t seem like a lot new features. A full rewrite in C++ may be why it’s taking so long but it definitely feels like they’ve quietly dropped the web version.
It’s fair enough if they want to focus on the Apple ecosystem but they let people think they were developing the web version for a while.
