Inline event handlers are still viable approaches for managing event capturing, except for two limitations: maintenance and mashability.
Maintainable Event Handling
If event handlers are added directly to page elements, and later there's a change to the function name or parameters, the web developer has to hunt down every occurrence of the load handler text and modify it. This approach of adding event handling isn't efficient, and with today's more complex and dynamically generated web pages, it's infeasible.
An exception to this premise is dynamically generated content, where the code to create the content is located in one file. Though the generated content can stretch across many pages, you still only have to make changes in one place.
This is also a viable solution, but another problem arises when you assign a function to an event handler. In the example just given, assigning someFunction to the onclick event handler for the document object overwrites whatever other assignment has been made before this script is run. If you're incorporating external Ajax libraries or merging multiple Ajax libraries into an application, you can't assume that you "own" the event handler. Your code has to play well with others, unless you want to restrict all code to your code and your code only.