4 ms·

I'd also like to see a platform where discussions/debate can meaningfully continue beyond a day or two, but the problem inherent in long discussions is the form

by tux3 5y ago

I'd also like to see a platform where discussions/debate can meaningfully continue beyond a day or two, but the problem inherent in long discussions is the format results in terrible signal/noise ratio if you let it run for too long. Think of traditional forum threads with hundreds of pages. Closing discussions after some time is just a practical matter. When posts are organized by chronology, or even by vote count, discussion cannot scale beyond a few hundred posts without becoming very hard to navigate. Look at the incentive a new participant has when choosing whether to post in a new or an old thread. A post added deep down a reply chain, or at page 100 of a forum will have very little visibility, it'll be lost in the noise compared to a post in a fresh thread. I think the way discussion platforms organize information is far from perfect, and noise/visibility/organization is really what puts a practical limit on ongoing discussions.

BlueTemplar 5y ago
Instead of just looking at the front page, they're supposed to use the search feature before (necro)posting - most forum FAQs have this front and center.
- tux3 5y ago
  Many forums forbid reviving old threads, or proactively lock them after a few months. Searching is great when you have a common question (and when the search feature is vaguely usable!), but empirically it doesn't seem to work very well for anything broader than FAQ type content. Few people use search to navigate forums, and when you do the results aren't great.
  - BlueTemplar 5y ago
    They forbid reviving old threads for wrong reasons (in both senses of this phrase : see again : netiquette). A forum that would prevent continuing previous discussions wouldn't be a forum any more, just another useless waste of time social media (ok, HN isn't too bad, but it's due to the very high quality of the posting here). It is indeed annoying when the search feature doesn't have something so basic as multiple word search ("in quotes"). But in those cases it can be supplemented by third party search engines (query : "site:"). FAQs, by definition, don't need to be searched (well, except Ctrl+F I guess, but you know what I mean). Learning to use search is just basic web proficiency that everybody has to learn. And it does work very well in the case of forums, if I had a nickel every time I found some priceless information about an obscure problem in a web forum...
    - hilbert42 5y ago
      "A forum that would prevent continuing previous discussions wouldn't be a forum any more, just another useless waste of time social media (ok, HN isn't too bad, but it's due to the very high quality of the posting here)." Agreed. By observation and logic what you say has to be true. What's so stupid and absurd about this situation is that it's so wasteful of ideas and human effort. If there was some way of measuring the efficiency of forums then we'd find that it'd be very close to zero. If ideas are lost or lose promenance because previous discussions are halted then we cannot easily build on them. That's why we see so much stuff unnessarily duplicated across the web. Not only do posters waste time and effort through either not having seen an earlier identical post (it having already been lost in internet noise) or that they repost elsewhere to try to continue the previously lost thread. The trouble with reposting is that not all of the important or the most relevant posts from the previously closed forum will make it into any new one. Again, this wastes the time of both posters and those who'd benefit from reading the posts. In effect, the whole damn forum edifice has to be one of the most waseful and inefficient mechanisms humans have ever devised. As I've already said in another longwinded post much of the blame falls on the designers of forum software, which when looked at from a global perspective in what it's trying to achieve then we find that it's essentially utter shit. That's to say it's not fit for purpose, and given the evidence one cannot objectively or honesty come to any other reasonable conclusion. It's also hard to figure out why we haven't seen much more of a serious debate over the serious limitations of forums and forum software. What's happened with forums is that we reached a hiatus some decades ago when it comes to thinking about about how posters' information should be presented; it seems that developers then put any further improvements in the too-hard basket and there's been essentially no further development since then (likely because they can't see much of a buck in it for the effort it'd involve). We need to rethink forums in terms of information - subject, topic etc., and in much better ways of building on earlier information without losing the historical narrative whilst simultaneously not diminishing posters' efforts. When it comes to posters, it's important that the hierarchical order of their posts be maintained if for no other reason than to show their resonses - their progression in thinking etc. - for this too is both valuable and important information. As I see it, to accomplish even modest improvements in forum presentation then we ought to be rethinking the problem in radically difficult ways such as reformulating forum software in terms of a relational database structure which over time would be supplemented by AI - AI's purpose being to keep the most important and relevant posts as the most accessible, i.e.: the most easy and quick to find. Surely, I'm not alone in thinking this way as it's so damn obvious. You're right about HN. There's an enormous myriad of valuable information in the many thousands of HN's posts and it's a terrible shame that so much of this information has just gotten lost in internet noise by virtue of the way stories age (and that's just the beginning of the matter). Moreover, it's made worse by the fact that anyone searching for information that happens to be inside forum posts will find it difficult to do so given that search engines will have deranked the results by virtue that said information is already wallowing within the internet noise. It's pretty terrible that one usually has great difficulty in just finding a HN story after it's aged several days. I've often wasted inordinate amounts of time trying to locate HN stories after I've failed to either save or bookmark them first time round. Solving this ought to be a no-brainer for the HN team. It seems to me that both HN's story listings and its forums are far too valuable to leave as they are. We users of HN's valuable service need to find ways in which we can make it even more valuable, useful and user-friendly.
hilbert42 5y ago
There's little in what you say that I can disagree with. In fact, most of the difficulties were obvious to me before my post but I concentrated my comments on the higher issues rather than being concerned with the more mundane practical problems that you've rightly mentioned. As I see it, there are several levels to dealing with the 'messy' posts issue (topics where there are large numbers of replies). In essence, these 'levels' are more different classes of problem and they require different (more specific) solutions. First, there's the 'uncompleted reply' problem. This where one intends to reply and for various reasons the poster is unable to do so. Reasons may include the cutoff
- hilbert42 5y ago
  As happenstance would have it, somehow I accidentally posted this part of my next post whilst still editing it. The fact that I cannot remove this error because of inane web policies illustrates the arcane nature of most forum software. (note: HN is not alone in this regard, it's typical of most.) The relevant issue here is that the time of this post there is no other post in reply to this error and thus I ought to be able to remove it without disturbing any other part of the forum per se — ah but alas it's not possible as I've already exceeded some arbitrary time limit even though it has absolutely no relevance in this instance. My full post further amplifies my reasons.
hilbert42 5y ago
There's little in what you say that I can disagree with. In fact, most of the difficulties were obvious to me before my post but I concentrated my comments on the higher issues rather than being concerned with the more mundane practical problems that you've rightly mentioned. It seems to me that we need to have solid objectives about what we actually what to achieve before we can clear away the unacceptable dross that plagues the web. As I see it, there are several levels to dealing with the 'messy' posts issue (topics where there are large numbers of replies). In essence, these 'levels' represent different classes of problem and thus they require different (more specific) solutions. 1. First, there's the 'uncompleted reply' problem. This is where one intends to reply to an existing post and for various reasons one is unable to do so. Reasons may include the website's cut-off time for posts being exceeded or that the poster hasn't had time to reply to a post for various reasons. As I see it, this is a bigger problem than missing an initial post to a story because it can leave existing arguments in a specific thread to remain open and or be incomplete. On more than one occasion on HN I've said I'd get back to someone on a matter and missed the timeout deadline for various reasons. This unfinished action leaves that thread incomplete in that there is no logical outcome or conclusion in respect of the earlier commentary. The problem is exacerbated if the person to whom I'm replying has made new comment that requires a definite reply. Also, the absence of any reply from me is liable to make me look like an unreliable correspondent in the eyes of that person for seemingly ignoring him when in fact matters were outside my control. This may not seem important but I believe such etiquette is actually important, especially so if the previous poster has gone to the trouble of posting information that actually needs a reply. There are some simple solutions to this: a website can have a definite cut-off time, say 14 days, for new post to a story, but it could have a longer time for those who've already posted a comment and who later want to reply to someone who has already posted to his or her comment/post. 2. The Correcting of Errors and Typos Problem. For me, there's little worse than to revisit and reread one my posts several days after I've posted it only to find some glaringly stupid mistake that makes a nonsense of what I've written and that I can no longer correct it due to an inane website policy that stops latter editing. Mistakes such as the inclusion of wrong or inappropriate words easily happen when one misses, say, a wrong auto-correction entry. Similarly, with other minor errors such as the wrong use of 'its' and 'it's', the same goes for possessives (the correct use of which I actually do understand but it doesn't stop me from mistyping them on occasions). As everyone in book publishing knows you cannot publish a book without an editor by virtue of the fact that we're all blind to many of our own mistakes, thus we simply cannot pick up certain errors no matter how many times we may proofread them. These days, through necessity, it's not possible to have someone sitting at our side correcting our errors whilst we type. That said, the problem of blindly missing our own mistakes hasn't gone away. In effect, even after 30-plus years, NO website has found satisfactory ways of accommodating/correcting typos after posting. What web designers are effectively saying to us is that they are 30 years or more behind the times and they could give a damn if they never caught up. Right, there's no incentive from web designers or makers of web standards to bring web publishing standards in line and level with normal book publishing—well that's the view from their perceptive and of course we poor hapless users have never have counted for much anyway. Frankly, it's an unmitigated disgrace, it's yet another instance of the unprofessional nature of the software industry - if you want actual proof then read this: https://www.researchgate.net/publication/247573088_Software's_Chronic_Crisis https://www.researchgate.net/publication/247573088_Software'... (despite this article being nearly 28 years old, almost nothing has changed to change its findings). There are many ways around this problem and they don't involve rocket science. There's the limited editing approach as well as the ability to edit certain words on an ongoing basis, especially so those involving the apostrophe as well as words such as 'them', 'then', 'whose', 'who's' and so on. I fully acknowledge that changing the wording or meaning of text no matter how small and insignificant the change after someone has already posted to or commented on said text is completely unacceptable and should never happen. Here too there are some easy solutions: if text is altered after a post has been made then it's dead easy to flag that post with say a dagger '†' whereupon clicking on said dagger would reveal the original text that the poster actually replied to. Moreover, many variations along these lines come to mind. Again, the real problem is that both those who run websites and who write the code for website comments don't actually give a damn about such niceties. Nor do they actually care about the fact that their laziness and ineptitude is contributing to the ever-increasing number of errors that we see appearing across the web not to mention the increasing general sloppiness of the web per se. Nor do they care or worry that their rotten workmanship is further contributing to increased web information entropy—or, as is more likely, they don't actually understand what increasing information entropy actually means. There are many things wrong with the web and we would be well served if those who actually ran the web set higher standards through example. Applying higher standards to accuracy and presentation is an easy point from which to begin. 3. I could not agree more with the fact that extended 'run-on' results in terrible signal/noise ratios. But again, little or nothing has been done to ameliorate this problem over the same 30-plus years. Moreover, the matter is made worse by websites allowing posters to reply to the first poster in order that their response appear high up in the posting hierarchy. This problem could be fixed in an instant if the typical tree structure that's used in most posts were modified to both better reflect the actual subject matter of the post as well reflect a more accurate relative temporal position in the posting list (both metadata from posts along with temporal data should be used to ensure that posts appear within the hierarchy in the correct and most relevant order). For this to happen it's likely that each thread in reply to any given post may have to split to accurately reflect both its metadata status as well as its temporal position relative to all other posts. I know that's a badly worded explanation but it'll have to for the moment. Again, I also need to point out that there has been precious little development in the way forums have been displayed over the last 30-plus years. Whether it's phpBB,WordPress, MyBB or whatever, none of them do a sufficiently satisfactory job that would ensure that 'discussions and debate can meaningfully continue beyond a day or two'. The fact that none of these forum programs has either the means or wherewithal in this regard simply means that those who have programmed them have failed dismally in achieving a workable solution. I repeat that the web is in an unholy mess: if we follow exiting example then we see that it inevitably leads upwards from the lowest common denominator and not as it ought from the highest of standards and from the moist excellent of ergonomic design downwards. I am strongly of the belief that for the web to improve significantly from its current ratbag existence then excellence should always begin at the top. Standards of excellence could start by websites actually agreeing on something as seemingly trivial as reaching an agreement on unified forum timeout procedures.

by tux3 5y ago

BlueTemplar 5y ago

Instead of just looking at the front page, they're supposed to use the search feature before (necro)posting - most forum FAQs have this front and center.

tux3 5y ago
Many forums forbid reviving old threads, or proactively lock them after a few months. Searching is great when you have a common question (and when the search feature is vaguely usable!), but empirically it doesn't seem to work very well for anything broader than FAQ type content. Few people use search to navigate forums, and when you do the results aren't great.
- BlueTemplar 5y ago
  They forbid reviving old threads for wrong reasons (in both senses of this phrase : see again : netiquette). A forum that would prevent continuing previous discussions wouldn't be a forum any more, just another useless waste of time social media (ok, HN isn't too bad, but it's due to the very high quality of the posting here). It is indeed annoying when the search feature doesn't have something so basic as multiple word search ("in quotes"). But in those cases it can be supplemented by third party search engines (query : "site:"). FAQs, by definition, don't need to be searched (well, except Ctrl+F I guess, but you know what I mean). Learning to use search is just basic web proficiency that everybody has to learn. And it does work very well in the case of forums, if I had a nickel every time I found some priceless information about an obscure problem in a web forum...
  - hilbert42 5y ago
    "A forum that would prevent continuing previous discussions wouldn't be a forum any more, just another useless waste of time social media (ok, HN isn't too bad, but it's due to the very high quality of the posting here)." Agreed. By observation and logic what you say has to be true. What's so stupid and absurd about this situation is that it's so wasteful of ideas and human effort. If there was some way of measuring the efficiency of forums then we'd find that it'd be very close to zero. If ideas are lost or lose promenance because previous discussions are halted then we cannot easily build on them. That's why we see so much stuff unnessarily duplicated across the web. Not only do posters waste time and effort through either not having seen an earlier identical post (it having already been lost in internet noise) or that they repost elsewhere to try to continue the previously lost thread. The trouble with reposting is that not all of the important or the most relevant posts from the previously closed forum will make it into any new one. Again, this wastes the time of both posters and those who'd benefit from reading the posts. In effect, the whole damn forum edifice has to be one of the most waseful and inefficient mechanisms humans have ever devised. As I've already said in another longwinded post much of the blame falls on the designers of forum software, which when looked at from a global perspective in what it's trying to achieve then we find that it's essentially utter shit. That's to say it's not fit for purpose, and given the evidence one cannot objectively or honesty come to any other reasonable conclusion. It's also hard to figure out why we haven't seen much more of a serious debate over the serious limitations of forums and forum software. What's happened with forums is that we reached a hiatus some decades ago when it comes to thinking about about how posters' information should be presented; it seems that developers then put any further improvements in the too-hard basket and there's been essentially no further development since then (likely because they can't see much of a buck in it for the effort it'd involve). We need to rethink forums in terms of information - subject, topic etc., and in much better ways of building on earlier information without losing the historical narrative whilst simultaneously not diminishing posters' efforts. When it comes to posters, it's important that the hierarchical order of their posts be maintained if for no other reason than to show their resonses - their progression in thinking etc. - for this too is both valuable and important information. As I see it, to accomplish even modest improvements in forum presentation then we ought to be rethinking the problem in radically difficult ways such as reformulating forum software in terms of a relational database structure which over time would be supplemented by AI - AI's purpose being to keep the most important and relevant posts as the most accessible, i.e.: the most easy and quick to find. Surely, I'm not alone in thinking this way as it's so damn obvious. You're right about HN. There's an enormous myriad of valuable information in the many thousands of HN's posts and it's a terrible shame that so much of this information has just gotten lost in internet noise by virtue of the way stories age (and that's just the beginning of the matter). Moreover, it's made worse by the fact that anyone searching for information that happens to be inside forum posts will find it difficult to do so given that search engines will have deranked the results by virtue that said information is already wallowing within the internet noise. It's pretty terrible that one usually has great difficulty in just finding a HN story after it's aged several days. I've often wasted inordinate amounts of time trying to locate HN stories after I've failed to either save or bookmark them first time round. Solving this ought to be a no-brainer for the HN team. It seems to me that both HN's story listings and its forums are far too valuable to leave as they are. We users of HN's valuable service need to find ways in which we can make it even more valuable, useful and user-friendly.

hilbert42 5y ago

hilbert42 5y ago
As happenstance would have it, somehow I accidentally posted this part of my next post whilst still editing it. The fact that I cannot remove this error because of inane web policies illustrates the arcane nature of most forum software. (note: HN is not alone in this regard, it's typical of most.) The relevant issue here is that the time of this post there is no other post in reply to this error and thus I ought to be able to remove it without disturbing any other part of the forum per se — ah but alas it's not possible as I've already exceeded some arbitrary time limit even though it has absolutely no relevance in this instance. My full post further amplifies my reasons.

hilbert42 5y ago

There's little in what you say that I can disagree with. In fact, most of the difficulties were obvious to me before my post but I concentrated my comments on the higher issues rather than being concerned with the more mundane practical problems that you've rightly mentioned. It seems to me that we need to have solid objectives about what we actually what to achieve before we can clear away the unacceptable dross that plagues the web. As I see it, there are several levels to dealing with the 'messy' posts issue (topics where there are large numbers of replies). In essence, these 'levels' represent different classes of problem and thus they require different (more specific) solutions. 1. First, there's the 'uncompleted reply' problem. This is where one intends to reply to an existing post and for various reasons one is unable to do so. Reasons may include the website's cut-off time for posts being exceeded or that the poster hasn't had time to reply to a post for various reasons. As I see it, this is a bigger problem than missing an initial post to a story because it can leave existing arguments in a specific thread to remain open and or be incomplete. On more than one occasion on HN I've said I'd get back to someone on a matter and missed the timeout deadline for various reasons. This unfinished action leaves that thread incomplete in that there is no logical outcome or conclusion in respect of the earlier commentary. The problem is exacerbated if the person to whom I'm replying has made new comment that requires a definite reply. Also, the absence of any reply from me is liable to make me look like an unreliable correspondent in the eyes of that person for seemingly ignoring him when in fact matters were outside my control. This may not seem important but I believe such etiquette is actually important, especially so if the previous poster has gone to the trouble of posting information that actually needs a reply. There are some simple solutions to this: a website can have a definite cut-off time, say 14 days, for new post to a story, but it could have a longer time for those who've already posted a comment and who later want to reply to someone who has already posted to his or her comment/post. 2. The Correcting of Errors and Typos Problem. For me, there's little worse than to revisit and reread one my posts several days after I've posted it only to find some glaringly stupid mistake that makes a nonsense of what I've written and that I can no longer correct it due to an inane website policy that stops latter editing. Mistakes such as the inclusion of wrong or inappropriate words easily happen when one misses, say, a wrong auto-correction entry. Similarly, with other minor errors such as the wrong use of 'its' and 'it's', the same goes for possessives (the correct use of which I actually do understand but it doesn't stop me from mistyping them on occasions). As everyone in book publishing knows you cannot publish a book without an editor by virtue of the fact that we're all blind to many of our own mistakes, thus we simply cannot pick up certain errors no matter how many times we may proofread them. These days, through necessity, it's not possible to have someone sitting at our side correcting our errors whilst we type. That said, the problem of blindly missing our own mistakes hasn't gone away. In effect, even after 30-plus years, NO website has found satisfactory ways of accommodating/correcting typos after posting. What web designers are effectively saying to us is that they are 30 years or more behind the times and they could give a damn if they never caught up. Right, there's no incentive from web designers or makers of web standards to bring web publishing standards in line and level with normal book publishing—well that's the view from their perceptive and of course we poor hapless users have never have counted for much anyway. Frankly, it's an unmitigated disgrace, it's yet another instance of the unprofessional nature of the software industry - if you want actual proof then read this: https://www.researchgate.net/publication/247573088_Software's_Chronic_Crisis https://www.researchgate.net/publication/247573088_Software'... (despite this article being nearly 28 years old, almost nothing has changed to change its findings). There are many ways around this problem and they don't involve rocket science. There's the limited editing approach as well as the ability to edit certain words on an ongoing basis, especially so those involving the apostrophe as well as words such as 'them', 'then', 'whose', 'who's' and so on. I fully acknowledge that changing the wording or meaning of text no matter how small and insignificant the change after someone has already posted to or commented on said text is completely unacceptable and should never happen. Here too there are some easy solutions: if text is altered after a post has been made then it's dead easy to flag that post with say a dagger '†' whereupon clicking on said dagger would reveal the original text that the poster actually replied to. Moreover, many variations along these lines come to mind. Again, the real problem is that both those who run websites and who write the code for website comments don't actually give a damn about such niceties. Nor do they actually care about the fact that their laziness and ineptitude is contributing to the ever-increasing number of errors that we see appearing across the web not to mention the increasing general sloppiness of the web per se. Nor do they care or worry that their rotten workmanship is further contributing to increased web information entropy—or, as is more likely, they don't actually understand what increasing information entropy actually means. There are many things wrong with the web and we would be well served if those who actually ran the web set higher standards through example. Applying higher standards to accuracy and presentation is an easy point from which to begin. 3. I could not agree more with the fact that extended 'run-on' results in terrible signal/noise ratios. But again, little or nothing has been done to ameliorate this problem over the same 30-plus years. Moreover, the matter is made worse by websites allowing posters to reply to the first poster in order that their response appear high up in the posting hierarchy. This problem could be fixed in an instant if the typical tree structure that's used in most posts were modified to both better reflect the actual subject matter of the post as well reflect a more accurate relative temporal position in the posting list (both metadata from posts along with temporal data should be used to ensure that posts appear within the hierarchy in the correct and most relevant order). For this to happen it's likely that each thread in reply to any given post may have to split to accurately reflect both its metadata status as well as its temporal position relative to all other posts. I know that's a badly worded explanation but it'll have to for the moment. Again, I also need to point out that there has been precious little development in the way forums have been displayed over the last 30-plus years. Whether it's phpBB,WordPress, MyBB or whatever, none of them do a sufficiently satisfactory job that would ensure that 'discussions and debate can meaningfully continue beyond a day or two'. The fact that none of these forum programs has either the means or wherewithal in this regard simply means that those who have programmed them have failed dismally in achieving a workable solution. I repeat that the web is in an unholy mess: if we follow exiting example then we see that it inevitably leads upwards from the lowest common denominator and not as it ought from the highest of standards and from the moist excellent of ergonomic design downwards. I am strongly of the belief that for the web to improve significantly from its current ratbag existence then excellence should always begin at the top. Standards of excellence could start by websites actually agreeing on something as seemingly trivial as reaching an agreement on unified forum timeout procedures.