Things being what they are, After Effects is pretty screwed up these days. Not only is it behind the times in terms of technologies (notwithstanding of course some science-driven stuff like the Key Cleaner and Refine Edges effects) and performance, but also over the years many of its paradigms and tools didn’t turn out quite as powerful and useful in practice as when they were designed first. Throw on top of that the gigazillion bugs and some really piss-poor decision-making on changes in recent versions and you get the idea. Adobe are just lucky that nobody ever bothered to seriously come up with a high-profile competitor to that wretched little lady, so people keep using it for better or worse.
Just like everyone knows how the situation is, everyone also has ideas how to fix this and that. Usually this turns into endless frustrating and aggravating threads on forums, but sometimes people actually take the time to explain their position in more detail. This is what Sander van Dijk has done this time around. While he is to be applauded for working things out so detailed, to me it’s a very mixed bag.
Of course there are some very obvious things that make sense. I’ve long been saying that Shape Layers were never properly “finished”, so tapered/ variable width strokes and gradients on paths are “Your mother told you!” things. On the other hand it took 30 years to get such features in Illustrator… Similar things could be said for global expressions and the need for a built-in expression/ script editor and based thereon being able to generate script-based/ interactive output like SVGs for web use or JSON/ XAML packages for UI design.
Other stuff is what I would call nit-picky obsession over minutia. I’m not saying that e.g. the way keyframes are displayed is perfect or that blending behaviors in effects for Fill/ Blend with Original are always logical, but well, to me a keyframe is just some tick on the timeline and unless I dig into the graph editor or create a RAM preview, I never know how my layers will actually behave. In a similar vein one could argue that going crazy over icons or the layout of the layer switches is not a productive discussion, since the facade cosmetics won’t fix the deeper underlying problems like those layer switches being overloaded with functions or the program using system icons and file associations for source file types.
Finally there is stuff what I would call “convoluted nonsense” that misses the mark and doesn’t really offer a solution to some problems. A prime example of this are the various attempts to improve display of properties in the timeline by using tabular alignment . The real truth here in my opinion is, that After Effects simply shouldn’t be displaying so much stuff in the timeline at all. Wouldn’t it be wonderful (and much more stringent) if we had a proper attribute editor, a.k.a. a pimped effects palette, that would allow us to have some tick marks which property streams to display in the timeline, including base properties like position or rotation? After all, that’s how most 3D programs’ track/ graph editors work with their default “show only animated channels” option. In other words: We need more control over what we actually get to see, not just ways of making redundant information look good.
Similarly the proposed multiple Set Matte effects don’t really fix anything. The inherent danger is that you simply lose track of how things interact. You can easily try this already by stacking a layer with tons of Circle effects, using the Stencil Alpha mode, and then playing around with things like the Invert option and the effect’s opacity. I have done this lots of times for complimentary stuff to my lens flares and it can drive you completely nuts. Just changing a single effect at the bottom of the stack can completely destroy your work. In the grander scheme I do not feel that this offers a solution, though moving the matting to an effect without any of the fancies could still be useful to reduce UI clutter.
I could go on forever, but suffice it to say that to me many of those things feel only half-thought through. There is a clear bias on UI stuff and fixing some immediate issues, but I’m missing the broader vision. As I always like to say it’s not about trying to do things the way we do them currently, only slightly better, but rather some things needing to be re-thought from the ground up and I can’t see that here. Don’t misunderstand me – you can fall in love with some tiny things and they can greatly make things easier on a day to day basis, but you can only go so far. Conversely I feel that several of the proposed changes would lock users into other “bad” workflows like further overloading Alt/Option, Ctrl/Cmd and Shift keys with more “secret magic” based on what they are used for. This stuff has always has a way of haunting you later when you wish those combinations would do something different, after all.