Marcin Warpechowski

I am not giving up on it that easily :) Can Palindrom obtain the URL from [`xhr.responseURL`](https://developer.mozilla.org/en-US/docs/Web/API/XMLHttpRequest/responseURL) then?

This is actually a problem with JSON Patch shim but I think it is impossible to fix. In Firefox/IE/Safari, the changes are observed using a shim that triggers a dirty...

@mattiasmak has reported a similar problem with his app few days ago. Let's reopen this issue to collect some thoughts. The only idea that I have is to do something...

@mtornwall No, still to separate batches of patches you should better trigger a `mouseup` event, as described here: https://github.com/PuppetJs/PuppetJs#generating-patches-based-on-local-changes

Today there was a similar problem in Starcounter: https://github.com/Starcounter/Starcounter/issues/4044

I have added it to goals 2017 in September/October (as PuppetJs maintenance/fixes). This is very important issue, but others are more prioritised.

You are right, the proxy solves the problem of observing the changes in wrong order, which is what the OP says in https://github.com/PuppetJs/PuppetJs/issues/29#issue-41708810 But there is also another problem. Consider...

I think that good ideas are here: https://github.com/PuppetJs/PuppetJs/issues/29#issuecomment-170904351. But I agree, let's get back to this after proxies.

I am afraid that if the data change is performed asynchronously by some framework, we might be not possible to wrap changes in a function. Maybe we also should provide...

> Since until today we assume that jsonpatch is async, we can leverage this assumption and make proxies async too. I was thinking about this kind of throttling too, but...