In correspondence chess, there’s a good long time to decide what your move will be. Even playing online, the rules can allow for a move every three days or more, on average. Other web-based chess systems may be played in real time, but you’re facing a screen and not an opponent. After my recent post on my first game, and loss, one person commented on LinkedIn about the difficulty of knowing of whether the other player is cheating by using the computer for assistance.
There are two ways to address the issue. The first is to explicitly allow the support of computer chess engines, databases, and reference books. This is the approach taken by the ICCF. This makes the game somewhat more difficult since there are a number of strong chess engines available, many of them free. When playing these games, the chess engine provides an opportunity to evaluate potential opponent responses to your prospective moves. The reality of how chess engines are used is very different from the assumption.
The second way to address the issue is to forbid the use of supporting tools. The the CCLA and the USCF both prohibit the use of chess engines, while allowing the use of databases and reference books. SchemingMind uses the same rules. Books and databases are different from chess engines since they are not selecting moves, only describing prior games and theory. This is, ultimately, governed via the honor system, but it works. There are venues that allow computer chess engines and some that do not. Pick what you like to play.