I think, it depends on the niche too. While it is a personal research or privately distributed service (like an Uni research where only the student and his professors have access), I think the danger-model is way more friendly than at a market product.
Also, how does the possibility influence your opinion that you may have no-money-cost access to a high-uptime hosting?
Topic: Desktop bots VS web applications (Read 634 times)
If I make the bot on web server and allow you to rent it , then the onus of protecting all your private keys and data lies with me. This may become a very risky and costly operation with time.
If it is a desktop application then the onus to protect your desktop from getting hacked lies with you.
If it is a desktop application then the onus to protect your desktop from getting hacked lies with you.
When a bot author runs the bots on behalf of other users, that's simply a web service. Of course, it requires additional trust from the user since they could potentially end up having to trust you to hold their money or private keys (depending on what the bot actually does).
Additionally, the cost of hosting enough instances, and the increased responsibility to maintain uptime (especially for trading bots and other time-critical operations) would move to the end-user when it's distributed as software rather than as a service.
Additionally, the cost of hosting enough instances, and the increased responsibility to maintain uptime (especially for trading bots and other time-critical operations) would move to the end-user when it's distributed as software rather than as a service.
Bots, without an exception, I have seen so far were all desktop software. Of course, you could put them into a server and run them from there, but why aren't they web applications from the start? Wouldn't it be better for creators protect their product if it was a web service?
Jump to: