Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Pull to refresh. For the web

usehook.com

Post reply on HN
Print view View original

41–50 of 50 posts

Re: Pull to refresh. For the web

#41

Post by gojomo »

Encosia wrote: post #31
gojomo wrote: post #18

Earlier quoted context omitted.

In the best touchscreen implementations, you have to overscroll past the top by a certain amount to get the reload behavior. Further, while overscrolling less than the threshold amount, you get a visual hint that further pulling will reload, while letting go will avoid a reload. This Hook.js implementation doesn't yet seem to have that level of refinement... but if it could, I don't think the user expectations or con…

Unfortunately, there aren't any events to track overscroll with regular mouse wheel or trackpad scrolling. I looked into it for a bit, wanting to create something exactly like Hook, but with the expected UX. Once the viewport hits the top of the window, scroll and mousewheel events stop firing even if the fruitless user input continues. Touch is different, because you continue receiving touchmove events even if the p…

I'd guess desktop browsers will eventually get around to supplying similar overscroll-events. (Looks like Chrome already offers an elastic overscroll visual effect... so maybe there are events there, if you know what to look for?)

Re: Pull to refresh. For the web

#42

Post by trentfowler »

lowboy wrote: post #15

This is not a good idea because it breaks years (decades?) of expected behaviour on desktops. When a user scrolls a page back to the top, the expectation is that they'll see the top of the current page from the most recent pageload. With this, there's no way to revisit the top of the page in its current state and it's just gone when you scroll back to the top. This doesn't save much time/effort as a refresh is a comm…

This doesn't quite replicate the pull to refresh on phones but I think the right implementation wouldn't conflict with the expected behavior on desktops.

All it takes to refresh here is to scroll past the top (however barely) whereas on a phone you have to give it a little tug to make it refresh and if you don't pass the threshold on the tug it just bounces back, effectively taking you to the top of the page. I think this is really cool idea and a good implementation but the next iteration would benefit by requiring slightly more 'pull' to activate than you would use normally when doing a simple flick on your trackpad (or scroll on your mouse) to get to the top.

Re: Pull to refresh. For the web

#43

Post by aboutaaron »

This could interesting from development, but LiveReload exist, so I'm curious what the market is. As others have said, if you could scroll to the top without firing the reload immediately (and it worked on mobile) this could be a neat tool.

Re: Pull to refresh. For the web

#44

Post by oftenwrong »

If this was a browser extension (and of course it can be used as one), I think it would be much more sensible. As it stands, I think most users would be confused and surprised by it. If someone can make a popular site with it, maybe it will become a commonly-understood and expected feature.

Speaking more personally, this is why I use NoScript. I do not like dealing with this kind of odd, site-specific functionality. My browsers have refresh already.

Re: Pull to refresh. For the web

#45

Post by lowboy »

mbell wrote: post #39
lowboy wrote: post #15

This is not a good idea because it breaks years (decades?) of expected behaviour on desktops. When a user scrolls a page back to the top, the expectation is that they'll see the top of the current page from the most recent pageload. With this, there's no way to revisit the top of the page in its current state and it's just gone when you scroll back to the top. This doesn't save much time/effort as a refresh is a comm…

> This is not a good idea because it breaks years (decades?) of expected behaviour on desktops. I don't think thats really valid. How long have facebook and twitter (among others) been automatically updating the page with new tweets/feed content? It seems to me that the expectation of 'static page state' is already gone for pages where this would be useful - sites with fast moving feeds of data. I'd really love reddi…

Twitter on my desktop has a notification that there are new tweets, and I haven't seen if FB automatically loads more content on the top or follows a similar approach. The point isn't that sites need to be static for the life cycle of a page reload, it's that the site shouldn't trigger a browser-level reload on an unrelated user action such as scrolling to the top of the page. It's confusing, and should be considered an anti-pattern as others here have said.

Re: Pull to refresh. For the web

#47

Post by Encosia »

gojomo wrote: post #41
Encosia wrote: post #31

Earlier quoted context omitted.

Unfortunately, there aren't any events to track overscroll with regular mouse wheel or trackpad scrolling. I looked into it for a bit, wanting to create something exactly like Hook, but with the expected UX. Once the viewport hits the top of the window, scroll and mousewheel events stop firing even if the fruitless user input continues. Touch is different, because you continue receiving touchmove events even if the p…

I'd guess desktop browsers will eventually get around to supplying similar overscroll-events. (Looks like Chrome already offers an elastic overscroll visual effect... so maybe there are events there, if you know what to look for?)

I'd defer to someone on the Chromium team if they have any input, but Chrome is what I was using at the time (not too long ago) and there definitely weren't any events firing for overscroll. It would be nice if they'd add notification about that, but it would be very difficult at this point to disambiguate between true overscroll and the animations that some browser/os combinations are displaying now. For such a niche use case (that really doesn't provide much benefit on non-touch pages anyway), I wouldn't be surprised if we never see it added.

Re: Pull to refresh. For the web

#48

Post by taintlove »

This would work only on mobile-only web sites. I for one would be incredibly irritated if I were to, say, write a longass comment and then accidentally swipe up on my trackpad and lose 20 minutes of work. I've done it before in other stupid ways, it's very frustrating, and i just end up being too fed up to type up my thoughts again

Re: Pull to refresh. For the web

#50

Post by gte910h »

ErikAugust wrote: post #10
gte910h wrote: post #3

"Pull to Refresh" is a poor description for the end user. I literally clicked and pulled down as I would on the iPhone simulator to use pull to refresh.

You could add Hammer.js and achieve this.

I prefer how this works, honestly, but just thought the description poor
Post reply on HN