Clicking an inline-react control for a comment shown in overlay can cause the picker box to appear partially off the edge, and (presumably due to something about how it’s not a normal top-level scroll context) I can’t (usually? always?) scroll horizontally/vertically to reach all the contents.
The “open stuff in a modal overlay page on top of the feed rather than linking normally, incidentally making the URL bar useless” is super confusing and annoying. Just now, when I tried to use my usual trick of Open Link in New Tab for getting around confusingly overridden navigation on the “Click to view all comments” link to this very thread, it wasn’t an actual link at all.
I don’t know how to interpret what’s going on when I’m only shown a subset of comments in a feed section and they don’t seem to be contiguous. Usually I feel like I’m missing context, fumble around with the controls that seem related to the relationships between comments to try to open the intervening ones, still can’t tell whether I’m missing things (I think the indentation behavior is different? How do I tell how many levels there are between two comments that look like they have a parent^N relationship?), and wind up giving up and trying to find a good thread parent to see in its entirety.
Cheers for the feedback, I apologize for confusing and annoyingness.
What do you mean by “makes the URL bar useless”? What’s the use you’re hoping would still be there? (typing in a different address should still work
The point of the modals is they don’t lose your place in the feed in a way that’s hard technically to do with proper navigation, though it’s possible we should just figure out how to do that.
And ah yeah, the “view all comments” isn’t a link on right-click, but I can make it be so (the titles are already that). That’s a good idea.
All comment threads are what I call a “linear-slice” (parent-child-child-child) with no branching. Conveying this relationship while breaking with the convention of the rest of the site (nesting) has proven tricky, but I’m reluctant to give up the horizontal space, and it looks cleaner. But two comments next to each other are just parent/child, and if there are ommitted comments, there’s a bar saying “+N” that when clicked, will display them.
Something I will do is make it so the post-modal and comments-modal is one, and when you click to view a particular comment, you’ll be shown it but the rest will also be there, which should hopefully help with orienting.
What do you mean by “makes the URL bar useless”? What’s the use you’re hoping would still be there?
The URL is externalized mental context on which article I’m currently viewing, and it’s also common for me to copy the link out to use elsewhere. Previously it would stay the URL of the front page—I think that’s been changed since I wrote that, though.
The point of the modals is they don’t lose your place in the feed in a way that’s hard technically to do with proper navigation, though it’s possible we should just figure out how to do that.
Yeah, though as a desktop browser user, I already have a well-practiced way of doing that, which is to open stuff in new tabs if I want to keep my place. I would imagine doing a pre-link-following replacement of the history state to include an anchor that restores my position on Back would allow true top-level navigation here? Or, stashing the “read up to X so far” state somewhere seems to be a common thing. (You’ve presumably thought of all that already.)
All comment threads are what I call a “linear-slice” (parent-child-child-child) with no branching.
Yeah, I figured it out eventually. It does seem tricky to get a good presentational design here; I don’t know of a great way to convey that difference in context, and I do feel like it’s awkward to remember the distinction or have to flip between the formats mentally when navigating around. Maybe if the visual frames were more distinct from the kind used in the nested-comments interface it’d be easier to remember that the chain isn’t siblings?
it recommended me drama things I hadn’t seen otherwise. I’d suggest that it should be biased away from drama things, eg by one of the ranking weight factors being dot product with some prototypical drama examples. I’ve had a lot of success with that on my https://www.graze.social/ custom feeds.
Your house, your rules, but there are two things I dislike about the way The Feed interacts with the way I use LW. I usually visit the site once a day and scroll down the front page looking at the titles and first words of articles. Those that interest me get opened in a new browser tab to read later, and I keep scrolling until I think I’ve seen most things both important and new.
The first thing The Feed does is replace the recent discussion on the front page, seemingly at random. Sometimes I get the usual list of articles, sometimes I get The Feed. I use a VPN, don’t keep browser history, reject cookies, and other such tinfoil hattery; so I don’t think the LW site has anything to base the decision on. Just before I started this comment I was getting The Feed, then closed and re-opened my browser and got the recent discussion. Something, somewhere, seems less deterministic than it should be.
Assuming I do get The Feed, the second thing It does is use very similar text for article titles and commenter user names. Similar enough that I have to read them to know what’s an article and what’s a comment. No disrespect to those making the comments, but I want to read the article first then decide whether to continue. If the article titles were in clearly different text than the user names (e.g. larger and seriffed just like they are in recent discussion) The Feed would work a lot better for me.
As of today, the feed decided to present me with five year old posts (with few new comments mixed in; checking if a pair of them were duplicates, I noticed that they weren’t and found out a comment-based poll in https://www.lesswrong.com/posts/QBhoBJAxDHHuC7BeH/poll-on-de-accelerating-ai). That was quite interesting, and down that way lay discussions about curi.
That may be unhealthy amount of drama highlighted, but it showed, like, depth of the LW community history, which I saw a bit shallower. Meta-point from back there (about optimally ending discussions) seems like it ties into the recent discussions...
Currently I see the feed as tool to see new content faster and, correspondingly, to see more in total. It means a deficit of closure, though, of some functionality like “this post tries to summarize / amend and supersede / highlight gaps in / agree with an earlier bunch of work” (the said body of work being more than one early post, to eventually conclude this graph).
Seeing this comment in the update feed, I clicked “read 592 more words”. It opened an overlay which scrolled to a mostly-collapsed (as in, single-line-per-comment display) thread, of which the only non-collapsed comment was a completely different one. After a few tries, I noticed that there was a temporary, subtle highlight of the left bar of one of the collapsed comments, which turned out to be the one I’d clicked on originally. I imagine this was meant to implicitly expand the target…
Maybe it should inherit the current choice of page suggestion algorithm at the top of the page? I generally only want to see the most recent posts and having 3+ yr old posts appear was very confusing to me until I realised what had happened.
There’s something narratively striking about a construction like this saving one’s place within the feed at the cost of losing one’s place within the post.
I spent several minutes trying to fit it into that one famous line from Star Wars but couldn’t make it work.
That last part’s not the interface’s responsibility though. That’s on me.
I just wound up in some additional super-confusing behavior. My best recollection is that “Angry Atoms” showed up in the feed display, I clicked it, it opened in the overlay view, and then when I tried to navigate from there using its own link to “How An Algorithm Feels From Inside”, seemingly nothing happened, except that I got a weird hunch and clicked the close-overlay left-arrow button, revealing that the front page had been replaced with the second post? And now there was no way to get back to the first one, and the browser Back button did… initially nothing, and then… maybe something? I wasn’t writing it all down while it happened and now I can’t easily reproduce it because the recommendation contents have naturally changed unpredictably.
I seem to be gradually picking up lingering emotional-experiential scrapes around the “yo I heard you liked navigation so I put navigation in your navigation so you can navigate while you navigate” school of site design.
To be clear, that last part isn’t just from here. Occasionally getting a Substack link that insists on surrounding the post with the rest of the feed/home interface and being unable to find how to get the same post without that has frustrated me before, and I think Reddit’s version of something similar has frustrated me, and maybe one or two other sites doing similar things… in general it seems like it’s a really easy source of “75% of it works, have fun whacking your metaphorical limbs on the edges”. Which is sort of still indirectly relevant as an influence on responses to the pattern even though some of it is not in your control. 🙂
I am curious what alternatives you’re looking at now!
Thanks for the extra context. I mean, if we can get our design right then maybe we can inspire the rest ;)
There’s a new experimental feature of React, <Activity>, that’d let us allow for navigation to a different page and then returning to the feed without losing your place. I haven’t tried to make it work yet but it’s high on the to-do list.
I’ve tried to get used to it, but it still feels significantly worse to me. Not quite sure why, I think the UI feels a bit busier which is part of it, but I think it’s mainly because it feels like it’s trying to control what I see in a way I don’t like.
Using the space bar to scroll in overlay puts the next line of text underneath the slightly-transparent header so that I have to adjust backward to see it, often after reading a few words from the wrong spot and getting confused. (Edited to add: okay, obviously not so often after I recognized the pattern, but I’m prospectively amused that if/when this gets changed I’ll have to adjust the habit a second time.)
(Is this what frontend Web development is just like? Should I pawn off as much of it as possible on AI in the future? Will that work, or will it drive the AI mad? Or will I be insufficiently inured when I have to dive in and fix a bug that it can’t handle, and then I’ll lose a day gibbering, thus canceling out the time and sanity saved?)
Huh, I never scroll that way but I see what you mean. I’ll see what I can do.
Yeah, frontend web development is a lot like this. The current AIs are both stateless, already mad, and seemingly indefatigable though get a bit loopy the longer the conversation goes on. You feed them $$ and sanity points and problems get solved faster, hopefully. There’s sometimes sanity saved when debugging something gnarly, but in the regular course of things you (or at least I) am spending mine down. (It doesn’t matter how many times I ask it not to, Claude Opus 4 will revert to saying “you’re absolutely right!” about everything.) Move over autistic savant, we got alzheimers savant now.
(sorry for the snark, but I’m guessing leaving user emotions in during UX testing is valuable)
I just did the following:
Clicked on this answer to this post in the feed, which expanded the context to show the first part of the post itself.
Clicked on the title of the post in the expansion, which opened the whole thing in the post-over-feed overlay—or so I assumed.
Got terribly confused when the answer I’d originally gotten there from was absent. In fact, there was no indication of it being a question-type post—it was presented as though it were a basic article, with the comment section present but the answer section just missing!
Navigated to the original post in a new tab and facepalmed real hard.
This reeks of underlying fragility and has destroyed my confidence that the overlay is implemented in a way that will keep continued track of how posts are actually supposed to be presented. Until I see a “we refactored it so that it is now difficult to push changes that will desynchronize the presentation logic between these cases”, I’m going to have to assume it’s an attractive nuisance. 🙁
Edited to add: ah, I see dirk reported the same underlying issue—leaving this up in case the feeling/implication parts are still relevant.
Knowing user emotions is good! And it’s sometimes nice when people care enough to get mad. I’m working on fixing this now.
I’m afraid you’re right that the overlay does not automatically match the post page. Unfortunately it turns out a lot of complexity gets added because the overlway has its own scrolling context and I’m needing to make a lot of adjustment for that. Fortunately the core elements of the post page don’t really change, so once we’re matched it should stay matched.
I think this seems like it’s aiming addiction machine and it’s immoral.
EDIT: I just saw “Choose how much the feed prioritizes content based on your past engagement versus popular or new content.”—this seems quite good and make the thing much less immoral!
EDIT2: just saw that there’s much more ability to customize—updated to this being pretty nice, potentially
EDIT3: saw the ‘Incognito mode’ - this is very good, I like that a lot
Thread for feedback on the New Feed
Question, complaints, confusions, bug reports, feature requests, and long philosophical screeds – here is the place!
Clicking an inline-react control for a comment shown in overlay can cause the picker box to appear partially off the edge, and (presumably due to something about how it’s not a normal top-level scroll context) I can’t (usually? always?) scroll horizontally/vertically to reach all the contents.
Without having made much adaptation effort:
The “open stuff in a modal overlay page on top of the feed rather than linking normally, incidentally making the URL bar useless” is super confusing and annoying. Just now, when I tried to use my usual trick of Open Link in New Tab for getting around confusingly overridden navigation on the “Click to view all comments” link to this very thread, it wasn’t an actual link at all.
I don’t know how to interpret what’s going on when I’m only shown a subset of comments in a feed section and they don’t seem to be contiguous. Usually I feel like I’m missing context, fumble around with the controls that seem related to the relationships between comments to try to open the intervening ones, still can’t tell whether I’m missing things (I think the indentation behavior is different? How do I tell how many levels there are between two comments that look like they have a parent^N relationship?), and wind up giving up and trying to find a good thread parent to see in its entirety.
Cheers for the feedback, I apologize for confusing and annoyingness.
What do you mean by “makes the URL bar useless”? What’s the use you’re hoping would still be there? (typing in a different address should still work
The point of the modals is they don’t lose your place in the feed in a way that’s hard technically to do with proper navigation, though it’s possible we should just figure out how to do that.
And ah yeah, the “view all comments” isn’t a link on right-click, but I can make it be so (the titles are already that). That’s a good idea.
All comment threads are what I call a “linear-slice” (parent-child-child-child) with no branching. Conveying this relationship while breaking with the convention of the rest of the site (nesting) has proven tricky, but I’m reluctant to give up the horizontal space, and it looks cleaner. But two comments next to each other are just parent/child, and if there are ommitted comments, there’s a bar saying “+N” that when clicked, will display them.
Something I will do is make it so the post-modal and comments-modal is one, and when you click to view a particular comment, you’ll be shown it but the rest will also be there, which should hopefully help with orienting.
Thanks again for writing up those thoughts!
The URL is externalized mental context on which article I’m currently viewing, and it’s also common for me to copy the link out to use elsewhere. Previously it would stay the URL of the front page—I think that’s been changed since I wrote that, though.
Yeah, though as a desktop browser user, I already have a well-practiced way of doing that, which is to open stuff in new tabs if I want to keep my place. I would imagine doing a pre-link-following replacement of the history state to include an anchor that restores my position on Back would allow true top-level navigation here? Or, stashing the “read up to X so far” state somewhere seems to be a common thing. (You’ve presumably thought of all that already.)
Yeah, I figured it out eventually. It does seem tricky to get a good presentational design here; I don’t know of a great way to convey that difference in context, and I do feel like it’s awkward to remember the distinction or have to flip between the formats mentally when navigating around. Maybe if the visual frames were more distinct from the kind used in the nested-comments interface it’d be easier to remember that the chain isn’t siblings?
Thanks for continuing to try to improve the site!
it recommended me drama things I hadn’t seen otherwise. I’d suggest that it should be biased away from drama things, eg by one of the ranking weight factors being dot product with some prototypical drama examples. I’ve had a lot of success with that on my https://www.graze.social/ custom feeds.
I’m curious for examples, feel free to DM if you don’t want to draw further attention to them
Your house, your rules, but there are two things I dislike about the way The Feed interacts with the way I use LW. I usually visit the site once a day and scroll down the front page looking at the titles and first words of articles. Those that interest me get opened in a new browser tab to read later, and I keep scrolling until I think I’ve seen most things both important and new.
The first thing The Feed does is replace the recent discussion on the front page, seemingly at random. Sometimes I get the usual list of articles, sometimes I get The Feed. I use a VPN, don’t keep browser history, reject cookies, and other such tinfoil hattery; so I don’t think the LW site has anything to base the decision on. Just before I started this comment I was getting The Feed, then closed and re-opened my browser and got the recent discussion. Something, somewhere, seems less deterministic than it should be.
Assuming I do get The Feed, the second thing It does is use very similar text for article titles and commenter user names. Similar enough that I have to read them to know what’s an article and what’s a comment. No disrespect to those making the comments, but I want to read the article first then decide whether to continue. If the article titles were in clearly different text than the user names (e.g. larger and seriffed just like they are in recent discussion) The Feed would work a lot better for me.
I’m using Firefox 137.0 on Fedora.
As of today, the feed decided to present me with five year old posts (with few new comments mixed in; checking if a pair of them were duplicates, I noticed that they weren’t and found out a comment-based poll in https://www.lesswrong.com/posts/QBhoBJAxDHHuC7BeH/poll-on-de-accelerating-ai). That was quite interesting, and down that way lay discussions about curi.
That may be unhealthy amount of drama highlighted, but it showed, like, depth of the LW community history, which I saw a bit shallower. Meta-point from back there (about optimally ending discussions) seems like it ties into the recent discussions...
Currently I see the feed as tool to see new content faster and, correspondingly, to see more in total. It means a deficit of closure, though, of some functionality like “this post tries to summarize / amend and supersede / highlight gaps in / agree with an earlier bunch of work” (the said body of work being more than one early post, to eventually conclude this graph).
Seeing this comment in the update feed, I clicked “read 592 more words”. It opened an overlay which scrolled to a mostly-collapsed (as in, single-line-per-comment display) thread, of which the only non-collapsed comment was a completely different one. After a few tries, I noticed that there was a temporary, subtle highlight of the left bar of one of the collapsed comments, which turned out to be the one I’d clicked on originally. I imagine this was meant to implicitly expand the target…
Maybe it should inherit the current choice of page suggestion algorithm at the top of the page? I generally only want to see the most recent posts and having 3+ yr old posts appear was very confusing to me until I realised what had happened.
Click post to view it in overlay.
Read read read.
Ah, there’s a typo in it.
Select text to prepare to use the typo reaction.
The widget doesn’t appear.
Have to back out to open in new tab.
Scroll scroll scroll… sigh…
There’s something narratively striking about a construction like this saving one’s place within the feed at the cost of losing one’s place within the post.
I spent several minutes trying to fit it into that one famous line from Star Wars but couldn’t make it work.
That last part’s not the interface’s responsibility though. That’s on me.
I just wound up in some additional super-confusing behavior. My best recollection is that “Angry Atoms” showed up in the feed display, I clicked it, it opened in the overlay view, and then when I tried to navigate from there using its own link to “How An Algorithm Feels From Inside”, seemingly nothing happened, except that I got a weird hunch and clicked the close-overlay left-arrow button, revealing that the front page had been replaced with the second post? And now there was no way to get back to the first one, and the browser Back button did… initially nothing, and then… maybe something? I wasn’t writing it all down while it happened and now I can’t easily reproduce it because the recommendation contents have naturally changed unpredictably.
I seem to be gradually picking up lingering emotional-experiential scrapes around the “yo I heard you liked navigation so I put navigation in your navigation so you can navigate while you navigate” school of site design.
Oh no, that sounds no good at all. You might be relieved to I have on my to do to to explore overlay alternative design.
To be clear, that last part isn’t just from here. Occasionally getting a Substack link that insists on surrounding the post with the rest of the feed/home interface and being unable to find how to get the same post without that has frustrated me before, and I think Reddit’s version of something similar has frustrated me, and maybe one or two other sites doing similar things… in general it seems like it’s a really easy source of “75% of it works, have fun whacking your metaphorical limbs on the edges”. Which is sort of still indirectly relevant as an influence on responses to the pattern even though some of it is not in your control. 🙂
I am curious what alternatives you’re looking at now!
Thanks for the extra context. I mean, if we can get our design right then maybe we can inspire the rest ;)
There’s a new experimental feature of React, <Activity>, that’d let us allow for navigation to a different page and then returning to the feed without losing your place. I haven’t tried to make it work yet but it’s high on the to-do list.
I’ve tried to get used to it, but it still feels significantly worse to me. Not quite sure why, I think the UI feels a bit busier which is part of it, but I think it’s mainly because it feels like it’s trying to control what I see in a way I don’t like.
The feed should display the post, if it chooses it at all, above (or together with) the comment from there.
Ah yeah, that’s pretty silly (the cards are randomly sampled currently so just silly luck to have them ordered like this).
Using the space bar to scroll in overlay puts the next line of text underneath the slightly-transparent header so that I have to adjust backward to see it, often after reading a few words from the wrong spot and getting confused. (Edited to add: okay, obviously not so often after I recognized the pattern, but I’m prospectively amused that if/when this gets changed I’ll have to adjust the habit a second time.)
(Is this what frontend Web development is just like? Should I pawn off as much of it as possible on AI in the future? Will that work, or will it drive the AI mad? Or will I be insufficiently inured when I have to dive in and fix a bug that it can’t handle, and then I’ll lose a day gibbering, thus canceling out the time and sanity saved?)
I just coded a fix for this, will get deployed soon.
Huh, I never scroll that way but I see what you mean. I’ll see what I can do.
Yeah, frontend web development is a lot like this. The current AIs are both stateless, already mad, and seemingly indefatigable though get a bit loopy the longer the conversation goes on. You feed them $$ and sanity points and problems get solved faster, hopefully. There’s sometimes sanity saved when debugging something gnarly, but in the regular course of things you (or at least I) am spending mine down. (It doesn’t matter how many times I ask it not to, Claude Opus 4 will revert to saying “you’re absolutely right!” about everything.) Move over autistic savant, we got alzheimers savant now.
(sorry for the snark, but I’m guessing leaving user emotions in during UX testing is valuable)
I just did the following:
Clicked on this answer to this post in the feed, which expanded the context to show the first part of the post itself.
Clicked on the title of the post in the expansion, which opened the whole thing in the post-over-feed overlay—or so I assumed.
Got terribly confused when the answer I’d originally gotten there from was absent. In fact, there was no indication of it being a question-type post—it was presented as though it were a basic article, with the comment section present but the answer section just missing!
Navigated to the original post in a new tab and facepalmed real hard.
This reeks of underlying fragility and has destroyed my confidence that the overlay is implemented in a way that will keep continued track of how posts are actually supposed to be presented. Until I see a “we refactored it so that it is now difficult to push changes that will desynchronize the presentation logic between these cases”, I’m going to have to assume it’s an attractive nuisance. 🙁
Edited to add: ah, I see dirk reported the same underlying issue—leaving this up in case the feeling/implication parts are still relevant.
This bug is at least fixed now! I await your next report, thanks.
Knowing user emotions is good! And it’s sometimes nice when people care enough to get mad. I’m working on fixing this now.
I’m afraid you’re right that the overlay does not automatically match the post page. Unfortunately it turns out a lot of complexity gets added because the overlway has its own scrolling context and I’m needing to make a lot of adjustment for that. Fortunately the core elements of the post page don’t really change, so once we’re matched it should stay matched.
Question answers don’t appear in the comments modal (or for that matter the entire-post modal).
This is fixed now. :)
True! That’s now next on my list.
I think this seems like it’s aiming addiction machine and it’s immoral.EDIT: I just saw “Choose how much the feed prioritizes content based on your past engagement versus popular or new content.”—this seems quite good and make the thing much less immoral!
EDIT2: just saw that there’s much more ability to customize—updated to this being pretty nice, potentially
EDIT3: saw the ‘Incognito mode’ - this is very good, I like that a lot
In the modal, when I’m writing a comment and go to add a link, hitting the checkmark after I put in the URL closes the modal.
This should be fixed now.
Seems to be working! Thanks for the fix :)
Oh, indeed. That’s no good. I’ll fix it.
I click “(read more)” and it seems to have no function. On Windows 11/Chrome, reproducible.
Huh, that’s pretty odd and not good. I’ll look into it. Any other buttons or interactions that do nothing?
Just tried all other buttons, none on my end!
All buttons within the feed don’d do anything, but other buttons on the site do? That’s very strange.
*Every non-”(read more)” button works as intended for me; i.e. none of the other buttons do nothing. Apologies if that was unclear.
The font for text in the New Feed is like 2pts larger than the feed on the rest of LW on mobile, which feels somewhat jarring
I agree it’s a stark difference. The intention here was to match other sites with feeds out of a general sense that our mobile font is too small.
If you wanted to choose one font size across mobile, which would you go for?
I prefer smaller font but I’m also young, like information density, and don’t have presbyopia, so idk how representative my preferences are
Also, I don’t like that if I click on the post in the update feed and then refresh the page I loose the post
Oh, very reasonable. I’ll have a think about how to solve that. So I can understand what you’re trying to do, why is it you want to refresh the page?
i.e. to see fresh comments
It might be that I have smth wrong with the internet but this beige widget isn’t loading
Oh, that’s the audio player widget. Seems it is broken here! Thank you for the report.
The new feed doesn’t load at all for me.
Hmm, that’s no good. Sorry for the slow reply, if you’re willing I’d like to debug it with you (will DM).
It’s working now. I think the problem was on my end.