It's a disgrace. And should be a software case study.
PS: been playing MtG since 1995
Even with a properly managed team, moving 20+ years of edge cases and odd interactions for a turing complete game is going to be difficult. The fact that they need to come up with a usable ui on top of that only adds to the suffering. Its not unsurprising that what they currently have can only be considered working in the most generous and utilitarian sense.
Don't mistake me for covering for Worth: a top quality group led by management who knew what they were doing would be much farther in both correctness and UX.
And still, it would make a great case of analysis... I see it as one of the biggest -and saddest- failures in Digital Gaming (if not the greatest), of the last 20 years.
On the other hand, the actual cards just work. The hard part for them, it seems, is "operating a game that more than a few hundred people want to play", not the rules engine.
And that's the problem. Everybody sees these as "edge cases". And that leads to failure. You can't simplify the back end engine, each card needs to be fully general and you need to carry around all its characteristics. At that point, you can execute all the actions you need because edge cases aren't. But, if you take even a single shortcut, it ripples outward and causes bugs everywhere.
And, the less said about the GUI designers, the better. The GUI is without a doubt the worst gaming GUI I have ever experienced.