Subtle changes, bugs and feedback

Posted on
Page
of 312
  • Is the problem that people can't count?

  • ^ new page fail!

    #suchfail

  • So, sometimes when people post, their post appears at the top of a new page (in their view of the thread) and this upsets them, so they post "New Page Fail".

    Other people then point out that a) it isn't on a new page for everyone, depending on ignore lists etc and that b) so what, people reading are not morons.

    Anyway, that argument is not the suggestion just that it gave rise to the suggestion:

    If a user has page size set to N comments (e.g. 25) then always show N comments on the page they are viewing. Not fewer. Except on page 1.

    So on a thread with 2N+3 (53) comments, page 1 would show 3 comments and pages 2 and 3 would show N (25) comments.

    The rationale is really that in a multipage thread the last page is viewed all the time, the first page rarely.

  • Can we get tags back too?

  • It would be nice if we could advance or go back more than one page at a time (without having to type a number in to a box).

  • So on a thread with 2N+3 (53) comments, page 1 would show 3 comments and pages 2 and 3 would show N (25) comments.
    The rationale is really that in a multipage thread the last page is viewed all the time, the first page rarely.

    This doesn't make sense to me.

    Are you simply proposing that instead of the comments in threads being ordered by date descending (oldest first, newest last) that this should be reversed?

    For that is the only way to give you what you wanted... but would require reading UP the page.

    And even then... this wouldn't prevent the fact that pages show different things to different people depending on their ignore/blocks.

    The only way to truly solve the "problem" (if it is even a problem)... would be to allocate page numbers to comments when they are posted, and then only ever show the comments on the page in question.

    What this means is that a page would have a maximum of n comments per page, but may well have fewer comments. i.e. on a page of 25 comments if 6 comments were made by people who were ignored, then the page would only show you 19 comments.

    In many ways this proposal would be a lot faster for the database, and a would be more consistent for everyone... the first comment on page 17 of a thread would be the first comment on page 17 for all.

    But...

    If people delete things, or several people on ignore have a discussion, etc... it's possible under this proposal that there would be blank pages. Whole pages without a single comment.

    To avoid that... perhaps we show stubs... "there was something here, but it is deleted", "there is something here, but it is by someone you ignore".

    Then the problem is that the ignored stuff is still visible. This can actually be a real issue for people who can be a little triggered, especially if they're only ignoring a single person and they react to those people.

    There really is no beautiful solution to the "new page fail" scenario. But the one proposed has no benefits and is odd. The one I've proposed has benefits but also failings.

    The current system is about as good a set of compromises and usability as I can imagine... though if someone does solve inconsistent pagination when items may be absent, I'm all ears.

  • It would be nice if we could advance or go back more than one page at a time (without having to type a number in to a box).

    Would you like me to list all 265 pages of this thread as links?
    Or all 4,086 pages of porn?

    Would you even know which page you want to go back to if not the next or prior, first or last? And if so... typing it into a box is bad because?

  • If you want to go back a week or two, or a month or two, in a thread it's just easier if you can go back more than one page at a time. Even five or six would be quicker. And why is typing bad? it's not, it's just not as quick and easy as having a multi-page jump.
    It's not a major thing but I see it on other websites and just wondered if it was possible here. You sound like you're offended that someone even asked.

  • This doesn't make sense to me.

    Are you simply proposing that instead of the comments in threads being ordered by date descending (oldest first, newest last) that this should be reversed?

    Not at all.

    Just put the page breaks at different points:

    3, 50, 50 in date ascending order.

    Rather than:

    50, 50, 3 in date asc order.

    It would only look odd when viewing page 1. The advantage is that the latest comments have the maximum chance of having the appropriate context on the same page.

  • It would only look odd when viewing page 1.

    So...

    Make Page 1 odd for all (and it's the latest page under your scheme so everyone gets to experience this oddness).

    Even though this new page fail thing sounds like it only affects those who ignore others which make up an unbelievably small slice of the user base.

    Does it not sound a bit screwed up to inflict a sub par experience on all, for the sake of such a small minority when even within the minority only an even smaller minority have what they feel is an issue?

  • You sound like you're offended that someone even asked.

    Nope. Not offended, but any more link elements... even just an extra one or two to jump back 5 or 10 pages... would break the layout for a lot of mobile users.

    When I did the original work on what pagination to use, and looked at the usage of the vBulletin site. I did see people trying to navigate using multi-page jumps... and was amused how many jumps it took to achieve reaching their destination. And for this, vBulletin would put in these huge navigation controls, which they rendered in smaller fonts because they wouldn't fit on a desktop screen otherwise.

    We're mobile friendly by default, and you may have a high-res mobile screen or phablet, but a lot of users do not. In designing for this I went back to thinking how to offer the multi-page or any page navigation (the other thing I saw were random jumps - I suspect people were editing query strings in the URL bar). And the controls for these cannot be in a reduced font as they need to be buttons for fat fingers on a touch screen.

    This is basically why we only have a few buttons for the 98% common actions, and page numbers for the 2% actions.

  • having a multi-page jump.
    It's not a major thing but I see it on other websites PornHub

    ftfy

  • Make Page 1 odd for all (and it's the latest page under your scheme so everyone gets to experience this oddness).

    The first page is not the latest page. It is the first page, with the oldest posts on. There is no change of date order. Just that under this suggestion it may have fewer posts displayed on it.

    Even though this new page fail thing sounds like it only affects those who ignore others which make up an unbelievably small slice of the user base.

    The suggestion is not aimed at being a fix for new page fail. At all. It just came out of that conversation.

    I just felt that it sounded intriguing as an idea.

    Long threads very rarely have their first page viewed. They often have their last page viewed. Why not make that page always "complete"?

    All of your replies so far have misunderstood the suggestion. Probably my fault.

  • Log out and read the forum. See how many first pages you experience.

    It's the default experience for guests.

  • You're saying the last page always has the last 25 posts on, with the 26th to last falling to the previous page, keeping the order the same otherwise. The final page always has 25 posts on, the first would be the only one with less until it reaches 25?

  • Ah yes, guests. Our silent majority. I had forgotten them.

    Can't we, you know, um, ignore them? The silent fuckers?

    WinkyFace

  • @snottyotter, yep that's what I'm saying.

    For logged in users, who have stateful reading positions maintained across sessions.

  • They'd still see the same thing. Not that I'm for or against this suggestion.

  • Not ignore like "add to our ignore list".

    Ignore like "fuck those losers, who gives a crap about their needs and desires?".

  • the ongoing, slightly odd, battle of people who think that "new page fails" are a thing

    Be careful what you wish for, or @Velocio might implement infinite scrolling :)

  • For people who can't do math, what about allowing to input +X or -X and jump to current +/-X page (or the first, last page if that comes before)

  • Ah yes, guests. Our silent majority. I had forgotten them.

    Can't we, you know, um, ignore them? The silent fuckers?

    That does include Googlebot... which would suddenly start seeing the same content on different pages as people would be replying to the end of the conversation and earlier comments may not appear on different pages.

    Google treat duplicate as a spam technique deserving de-listing.

    I mean... you seem to imagine that the current system lacks thought. It doesn't.

    The only people affected by this non-issue are those who ignore others. A miniscule numbers of users, who by their own action create a slightly different pagination for them. Yet navigation wise this doesn't even break things... all links to comments will calculate which page to send you to and get you to the right place.

    And @mdcc_tester infinite scroll is possible. Always has been. But if we did it, then on image heavy threads the amount of RAM that browsers would take up would be huge.

  • I'm not sure that's the issue. I thought it was that people saying "new page fail" is just annoying at any time, so making pagination dynamic would put an end to it.

  • making pagination dynamic would put an end to

    For a small number of users, by breaking it for everyone else

  • Post a reply
    • Bold
    • Italics
    • Link
    • Image
    • List
    • Quote
    • code
    • Preview
About

Subtle changes, bugs and feedback

Posted by Avatar for Velocio @Velocio

Actions