Even eye bleed causing assets take some time to produce and a game may need a fair few for the result to be practically playable. When a project is small perhaps no one on the team (maybe a team of just one) has the time and/or they all hate the idea of the task and they have the original assets initially to hand anyway so the time and effort is likely better spent elsewhere. OpenTTD didn't always have its own asset sets available.
There may be a fear of greater copyright issues. The clean-room defense for clone code isn't going to work as well for graphics and sounds, because they have seen the originals and there are not many ways you can draw an inter-city-125-a-like (to use a TTD example) if you desire to maintain the same overall feel for the clone as the original.
And bad imitation art may put off more potential players than having to extract the original assets does.
> There may be a fear of greater copyright issues.
To be clear, I’m not suggesting they actually attempt to mimic the assets of the game you’re cloning. Rather, symbolize them.
If you’ve ever seen the game Baba is You, the text-tiles in that game are perfect examples of “symbolic assets” you can create to stand in for the real assets of a missing asset pack. Creating such a symbolic asset pack should be the work of five minutes.
The result will not-at-all resemble the real game; but it will also obviously be “not the way the game is intended to look”, so players won’t think that the game is just “a game with art made by programmers” like SuperTux/TuxKart/etc. It’s clearly “the game in a state where it’s missing something.” But it’s still playable in that state!
And, intriguingly, creating a fallback asset pack like this, and showing it to people by default (if just for a minute), will also get into people’s heads the idea that these assets are skinnable. So this will implicitly encourage creatives to look at how asset packs for the game are made (with your symbolic pack as an example), and maybe make one—or dozens!—of “real” packs.
Even eye bleed causing assets take some time to produce and a game may need a fair few for the result to be practically playable. When a project is small perhaps no one on the team (maybe a team of just one) has the time and/or they all hate the idea of the task and they have the original assets initially to hand anyway so the time and effort is likely better spent elsewhere. OpenTTD didn't always have its own asset sets available.
There may be a fear of greater copyright issues. The clean-room defense for clone code isn't going to work as well for graphics and sounds, because they have seen the originals and there are not many ways you can draw an inter-city-125-a-like (to use a TTD example) if you desire to maintain the same overall feel for the clone as the original.
And bad imitation art may put off more potential players than having to extract the original assets does.