Interrupts are stateful on the server side; that creates problems in terms of scalability, both due to increased memory usage and by being less flexible (either you have each user "locked" to a single process, or you have to implement state sharing, which adds overhead).
It can also be extremely wasteful - if I leave a tab open for hours or days, you'll have to waste your resources and mine to keep pushing me stuff I won't see, while now I just hit refresh when I want to see new content.
In terms of usability, it's often jarring to watch content change when you're interacting with it - that's why even sites that implement real time notifications often have a link or button that you have to press to update the UI. In many cases, doing that completely negates the benefits of pushing.
Polling is a good default for the web; it fits most use cases (content that rarely changes) in a simple and economical way. WebSockets are useful for the exceptions.
It fits most cases.. except that web applications are _constantly_ polling for information. Which may be fine, but you can implement polling with interrupts. What part of that do you not understand? Are you aware that the OS is using an interrupt-driven system to poll the server? It's just hidden from you, so you have no choice. This is just dumb engineering. And to say that it's good for "most things" is just lazy thinking. "Most things" are that way precisely because of the crappy architecture. Rationalisations.
You haven't made any technical points. You just waffled nonsense about scalability. I see you're bowing out of the argument because you have no case. Once you have interrupts on a platform, you have polling. If you're stuck with polling you can't recover interrupts. Which is why we have websockets decades late. The truth is there is not tension between interrupts and polling. Interrupts are plainly superior, since you can opt out of them trivially. But the platform should not opt out ahead of the developer. This turns out to be inadequate, so we get polling implemented inside interrupts and then a separate mechanism for interrupts. And the interrupt driven stuff has stupid reload buttons and so forth on the GUI. It's dumb engineering because increases complexity and leads to bad results for the user. You have no case. Polling is dumb. And if you think software engineers ever needed "good reasons" to add a bunch of pointless complexity to things, then you simply have no clue on the history of software development.
I had not downvoted you, but I thought about it - because you are making ranty, ad hominem attacks ("lazy thinking", "dumb excuses"), and you appear to be ignoring or handwaving away all the actual points anyone makes. I have no interest in engaging you in any discussion for these reasons, and I expect that other people feel the same way: hence, downvotes and lack of replies. Feel free to do what you wish with this information.
