A2Kers, where are you? Where is everyone?

man you love posting the same thing over again don't you

did you lose the link to your presidential historians report

Hi, Good post, I've been wondering this myself. It does appear that since the big format change that a lot of people are posting less. I for one am posting less because I've taken on more hours at work and do spend too much time on facebook scrabble.
I do peek in here and there to see if any new topics pique my interest though.

Been wondering how the economic crap has affected you (and a whole lot of other folk, too).

I hope you can squirrel away enough nuts to feast royally even if it is a long hard winter.

It's precisely the types of posts within this thread that has me visiting here less and less. Is everyone invited to comment or is this just a conversation between a few people???

I think this is a great idea and look forward to it.

When I joined Abuzz in February or March 2001, "Damn, Abuzz USED to be so much better" threads were all across the place. These threads are a bit like flu epidemics.

Personally, I don't the premise of MsOlga's question applies to me, because I don't think I'm on A2K less than I used to be. But as much as I like it here, A2K is not what I ultimately care about. Indeed, I don't ultimately care any particular internet site per se. What I care about is particular individuals that have become special to me, about a dozen of whom I happen to have met on Abuzz and A2K. (I think you know who y'all are.) Whether I talk with them on this site, on some other site, per e-mail, or in real life, is secondary to me. Indeed, if it was my choice, I'd prefer to have them all within driving distance in my real life. Maybe we can start up a commune together somewhere sometime.

Oh I don't know what, if any, threads are collapsed. But the ones I vote down are collapsed for me forever except in the very rare instance when somebody refers me to a voted down thread and I reinstate it. If you don't vote down threads, 'new posts' beocmes almost unusable to keep up with much of anything, but if you do vote down threads you think are of no interest, then there is no subsequent opportunity for somebody to peak your interest in that thread. The problem of unanaswered posts is sometimes inadvertent and sometimes just because the threads move off the page so quickly, folks who would have resonded to that post never see it.

This isn't a complaint. It's just an observation of why it is much less user friendly for somebody like me. How many 'like me's' there might be, I have no way of knowing. But I do think it is isn't conducive to building community and is likely why we have lost so many of the folks who used to visit regularly.



I appreciate that there are still bugs to work out of the system. And I appreciate that you or nobody else should accommodate me or anybody else with a random bee in his/her bonnet. But if we don't voice complaints and/or suggest ways to remedy what we perceive, then you have no way of knowing what the perception is. If I'm the only one seeing it that way, so be it. It wouldn't be the first time.



Maybe you're right. I don't know. I can only report my perception and how it impacts the way I use the board. Obviously viewership of the board is way up and, if that is where your bread and butter is, you can't afford to accommodate what some of us perceive.

I think everyone is invited to comment. This was started by a friend of ours in Oz, though, and it's Monday morning there, now, Marty, so they are likely on the way to work or already at work. In fact, at the time i'm posting, i think it's getting on for noon on Monday down under.

I sure would like to see the pm feature come back.n

When I first came here I intended on asking one question, getting my answer and then move on. That was 3 years ago.

Setanta I dare say msolga is still at work, time down under where she lives is 3.30pm right now.

I wholeheartedly agree with you martybarker re pm's, sure sorely missed at present.

Yes! And if could work like members who are put on ignore so you can check to see if there's anybody there that you need to un-ignore, that would be great. In other words, to be able to hide threads without having to vote them down would be a HUGE improvement to me. But I would like to be able to see the subjects of threads I am hiding just in case I want to rethink that action.

Let me see if I understand this then, you think the new software is less conducive to communities because on it you vote topics down? Why not just not vote them down.



This page works just like the old new posts did, with two key exceptions. 1) it doesn't reset every time your session dies, which was a big complaint on the previous software. 2) it doesn't have read tracking, which is a big deal that few people note how much affects them.

Those two things aside, the list works exactly like the last one did. So why exactly do you find this one to require the voting down of threads?



Call me crazy, but if your problem is that you are voting down topics then why not just not vote down topics?



How is this any different than it always was?



And there's the rub. There are plenty of legitimate problems with this site, and there's a lot we have to work on. But the most vocal people complaining about the software really don't tend to have any actionable feedback. If your problem boils down to how you are voting topics down and find this regrettable then just don't vote down topics. It'd work a lot better than blaming the software.



If you don't think it's conducive to building a community then don't do it.



It's not useful if it's not founded in reality. Since this site opened I've listened to people go on and on about what's wrong with it. Since the site began there have been people who are unhappy with it, and who leave.

Sometimes there's really good feedback, but a lot of times it's not actionable feedback and is not something that can be corroborated.



Large general forums are relatively poor bread and butter and the primary motivations I've had with able2know are not financial. Of all the things our company does this is by far the worst thing we can do from a financial perspective, but we genuinely like doing it, and want to build good community software. We formed our company so that we'd have time to work on projects like this that are technically and socially rewarding and I'm very much interested in "building community" and in extracting actionable feedback from criticism. Hell even though times are tight I've been paying for use case completion and feedback from Amazon Mechanical Turk (which is part of the reason there have been more "make money online questions" recently btw). I'm very passionate about doing this kind of challenge right and the notion that we're selling out to financial considerations just couldn't be further from the truth. This is a project we have on a pedestal, our magnus opus if you will and frankly we'd all be better off financially if we invested our time in more profitable, if less satisfying, pursuits.

As an example, 90% of the advertising on this site goes for under $0.50 per 1000 impressions. We have other sites where we earn as high as $2,000 CPM. Sites with a fraction of the traffic a2k gets are much more valuable financially to us, because the traffic is of a different nature in more profitable niches. We don't do this project for the money and we do want to get the psychology of the software right. That's why you've also seen fewer and fewer ads as this new software evolved. We want the site to be financially viable, but our aim is a vibrant community more than financial profit.

So I'm not rejecting criticism for the almighty dollar, but I've been around the block here, and have been listening to a2k criticism for years and I know there's a lot of it that I really can't do anything about. In this thread there are people who have claimed to be offput by the new software, and thusly using the software less when if you look at the numbers their activity is higher than it's every been in the history of the site.

There are claims of how the activity is dwindling, or how "so many" folk are leaving while the numbers don't bear it out and I've been listening to people say this since the first 6 months of the site.

You can't please everyone, and some people will always feel strongly against something about the site, whether or not their perception is accurate. I want to extract the useful feedback I can, but otherwise just have to do the best I can and learn to live with being unable to make everyone happy.

So when people give negative feedback I do pay attention. I've spent hours investigating some of the claims you and others have made, about how the ratings system is impacting your experience negatively. When people claim their topics, or the topics they are interested in, are not getting attention because of the voting system I dig into it to see what's wrong and how it can be made better. The voting algorithm is one thing we know needs work, we have reputation scores hidden while we perfect this and I spend a lot of my nights running the data against various mathematical algorithms finding ways to make it more socially useful. And when I see claims that the software is negatively impacting use I spend a lot of time and thought investigating and seeing if I can find a way to make it better.

And most of the time what I find is that what they claim is simply not happening. The writer complaining about being voted into oblivion didn't have a single topic that's collapsed, and was getting more views and responses than he previously did, and had voted their own work down hundreds of times in some form of strange self-depreciation. The people telling me how god-dammed horrible it is now can often be found telling me how god-dammed horrible it was before. The people saying they are unhappy and are leaving have always been here.

By now I'm used to it, and I need to measure things objectively. I have to find things that can actually be improved, and learn to have thicker skin for the less constructive criticism. There are always going to be unhappy people, and they aren't always going to be right about what makes them unhappy and we just have to do our best at trying to extract the useful feedback from both. We can't make everyone happy, but we're willing to try, and willing to listen to constructive feedback.

Well the good news is that they are pretty close to being launched. We actually got the basics working on our development site Friday night and unless real life continues to intervene I expect them to be released within the next 10-15 days.

And there's actually some feedback we were hoping to get from you guys about it. We took a very different direction than the previous implementation of messages. It would have been easy to just do it the same but we made a bold move trying to make them better that is sure to be polarizing and have some folks love it, and some folks hate it.

So let me explain. What we did is eliminate the single message inbox view and replaced it with threaded conversations. So instead of clicking a new message and seeing just one PM, you see a list of people you've exchanged messages with (with the newest received messages at the top) in your inbox, and clicking through to the message shows a threaded conversation, with all your messages (both sent and received) with that user sorted with the newest at the top. There's a quick reply box at the top and you can see the exchange all in one place.

Our thinking is that for the use case of organizing messages in saved folders, or as notes this implementation is not as good as the previous one, but for those who use it for it's more traditional purpose (of exchanging messages where the context and narrative are more relevant than the control over organization) it is more useful. For example, those who use it as a private chat are likely to be better served by the new format.

We expect some to love it and some to hate it, but would like to get feedback from this early on, as we shape the UI of the message area.

So what do you guys think? How do you typically use the private messages, and how would that impact your use (both positively and negatively)?

Thank you for that explanation Robert Gentel, much appreciated.

I don't use them much.....but your idea sounds a very positive and useful change to me.

Very much so.

Well have a closer look at two other things we think will have a bigger impact than that:

We have been designing a change where after voting a topic down, you'd see the options to ignore the user, or the whole tags. Right now, there are a lot of users who vote down every word game, and instead of having them impact the voting on each topic it would be easier for them, and more useful to those who enjoy those niches, to just filter out the user or all the topics with the tag they are uninterested in.

And the Bayesian algorithm is another big one. One of the fundamental rating problems breaks down to a conflict between volume and rating.

For example, if one person rates something up, it has a 100% rating and an aggregate vote score of 1. If a topic is voted up 130 times, and voted down 100 times, it has a rating of 56.5% with an aggregate score of 30.

So in a simple percentage system a small group of votes have too much impact, and in a simple aggregate score system a relatively pedestrian percentage rating can be expressed as large popularity.

And these examples work in negatives as well. Right now, if 100 people vote something up, and 110 people vote something down it would collapse. But if some spam had 7 votes down and none up it wouldn't.

This is a fundamental problem of trust, where you can't trust the percentage rankings at low volume, but can't just rely on high volume without failing to represent some of the smaller niches.

A Bayesian rating system takes both into account, and ratings need both to reach extremes. So a topic that is only a bit unpopular or a bit popular won't move much at all. And a topic that is about split but has large volume won't skew one way or another. I'm not explaining it well, so have a look at some very simplified out put:



In the first example is a topic with healthy up votes, but enough down votes to be collapsed in our current simple aggregate. If our Bayesian rating cutoff is 25% or 20% (the range I think it would be in) it would be well away from being collapsed.

Next is a topic that would appear very popular on our current system, but with a Bayesian system would reflect the split in opinion with more nuance.

The next two examples are the most important, the first is a typical flame/spam kind of interaction rating, where only the topic author has voted up. The second is also under the current collapse threshold, but shows how factoring in the percentage rating would bring the topic out if only a couple of other people do like it.

The last two just show what it does to a handful of votes either way.

It has the effect of making things tend to stay closer to average ratings, requiring volume (which can be trusted more than no volume) combined with the percentage to cause significant deviation.

Personally, I think those solutions may be good enough not to need a separated hide, which has its own usability and UI downsides.

You want more drama monarchs?




Setanta said:



Don't think we didn't see that.

Because we did.

Not long home from work on Monday afternoon (Oz time) & I've read (very quickly) through the whole thread.
I am really pleased that this thread didn't turn into an all out attack on A2K, as so many other such discussions have, in the past. It was a relief to read so many genuine & serious comments on our current "situation". (genoves excluded)
My impulse for starting this thread in the first place, was a nagging feeling that something was happening here and I couldn't figure out quite what it was. I just kept seeing the numbers of folk "online" & those posting fall & fall. Which worried me because, apart from the A2K "boredom" factor, which many of us have been felt recently, there was also a nagging, pragmatic concern that, if this continued, A2K could find its self in trouble supporting itself. I was worried about advertising revenue, as much as anything else ... which (in a time of recession spending cut-backs) could well be affected by low site participation rates. I want the people employed to run the site to feel secure in their employments as much as I want the A2K to be interesting & engaging for all of us who really value this site . It matters enormously to me, as it obviously does to so many other "regulars", that A2K remains "healthy" & viable.

But then, this may well be a temporary glitch. Who knows? As others have mentioned, A2K goes through it's heights & troughs. This might well be just another of our troughs, for all I know.

When is it apparent that the community has spoken and what it is is?