I dig some more and using my nifty programmatic scroll detection script found that they indeed have some logic to adjust and maintain the scroll position as they are adding new entry in their live blog:

```js
o.once("loaded", function(t) {
   return function() {
       var t, n;
       return e && o.$el.addClass("animate"),
       o.$el.removeClass("offscreen"),
       t = o.$el.outerHeight(!0),
       n = $(window).scrollTop(),
       o.$el.offset().top < n ? $("body, html").scrollTop(n + t) : void 0
   }
}(this)),

```

This basically measures the height of new item and tries to update the scroll position to compensate.
One wrinke is that it is doing so with a css transition:

``` css
.entry.offscreen {
    width: 100%;
    transform: translate3d(0, -2rem, 0) scale(1, 0);
    transform-origin: 50% 0;
    opacity: 0;
}

.entry.animate {
    transition: opacity 250ms linear, border-radius 250ms linear, -webkit-transform 250ms ease;
    transition: transform 250ms ease, opacity 250ms linear, border-radius 250ms linear;
    transition: transform 250ms ease, opacity 250ms linear, border-radius 250ms linear, -webkit-transform 250ms ease
}
```

So it has an element with scaleY of 0 which is then scaled to 1. So, it adjusts the scroll position and then
the element is grown which is why I see the jump in Firefox.

I suspect that the issue is that somehow Chrome does not move the content down as the scale is animated to 1?
That would make sense if I had scroll anchoring enable but now I am just confused.



[1] https://gist.github.com/majido/929eb68d37727d928455

majidvp@ did you happen to come across another example that is more reproducible as that blog is not quite live all the time.

I will close this for now. Majid can you open this again or file a new one when you find another reproducible example?