Shouldn't Hottest Forums List Be Based Upon Replies?

"Hot" topics at ATS have traditionally been based upon the number of replies, since the term was first applied and implemented. This was as opposed to "Top" topics which are based on number of Flags.

So I am kinda curious why in the "Hottest Forums" list, to the right of the recent posts page, we have deviated from that self standard to make the priority based upon number of new threads posted into each forum?

The only reason I question this is because I feel it is not accurate, that is, according to the ATS definition of "Hot."

It would stand to reason that the BAN forum would take that top spot, because there is always fresh content to report on daily. That as opposed to say, the 9/11 forum where new topics at this point are hard to come by.

So I wonder what that forum list would look like if all the posts were added up instead of the new topics posted? Would it be roughly the same? Or would some things change around a bit?

Justa wonderin.

According to the page about flags;

www.abovetopsecret.com...

we are instructed to "flag threads you feel are important or interesting". This to me is quite different to replies, as the number of replies might just indicate a heated argument is going on, which may not necessarily mean the thread is actually important, just that it's a subject that polarizes people or something like that.

reply to post by jimminycricket


Umm, I think you may have totally missed the point of the OP.

Currently the Hottest Forums list, to the right of the recent posts display, is sorting the list by the number of new threads posted in a given forum. All I am asking here is why is that so when a measure of total posts in each forum, over the given time period, might be a more accurate indicator. And it would be consistent with the hot topics display.

One forum may have 150 new threads posted, but only, say 2500 total replies, while another forum might only have 100 new threads posted, but 3,000 total replies. According to ATS's own definition of "hot"- based on total replies- then really the forum that had the 3,000 replies should take precedence, and be listed first.

About the only reason I can fathom is that combined with the new drive on flag content, ATS is focusing on thread contributions. If that's the case, then maybe another word instead of "Hottest" might be considered, so that we are consistent with the definition, that's all. Not bitching, just saying...

No, I don't think it should be based on replies.
Or flags.
Or stars.

It should be based on views.
If a topic garners more views than an other it should top the list.

Let's face it, page views run the advertising revenue on a web site.

reply to post by anxietydisorder


Well that's actually a great point!

Except for one slight little problem...

reply to post by TrueAmerican
...they're not tallying views any longer, but views should be included with posts and other factors to determine if it's hot or not (IMO). How do you know what the hottest stocks are...participation...even if just one buyer aquires all of the at&t stocks to be had. It was interesting to see see the ratio of views to actual posts. I don't think there is anonymous posting any longer either. The evolution of this site would be interesting to read about in a comprehensive format written by anyone who's been around from the beggining...Simon himself could probably write a book about this that might be a valuable addition to at least a couple of different genres if he could squeeze out the time and inspiration to produce such a work. I'd buy a copy

reply to post by plainmike


Hi Mike,

Well, that actually brings a question to mind. ATS just went through a pretty good transition to some heavy-duty hardware. Seems like now more than ever SO has the juice to bring back the views column. Of course the way it works now, if a thread has no replies and no flags, it's either got to be a new thread or an uninteresting one. I have my share of those. But I guess the catch is you have to go view it to find out. I'd vote for the views column back, meeself, if we had the chance.

Basing it on views is so misleading. A good catchy title gets you views but could be totally worthless in its content and importance.

Something really important should be flagged or starred whether you care to comment or not. Sometimes there just isn't anything to say that hasn't already been said so you star it or flag it.

Or anyway I thought I understood how ATS worked with flags and stars until that stupid CIA avatar topic got like a gazillion flags and I commented and got lambasted for it. Now I'm all confused too.

I guess all the guidelines in the world are worthless unless everyone clearly understands and executes them exactly the same. LOL

reply to post by TrueAmerican


Yeah, view counts would be priceless when it comes to working out what kind of thread topics and in which forums people click on and what catches their eye.
Probably meaningless in the big scheme of things, but it would help those of us who aren't big thread starters to work out what we should be looking for and what people are interested in reading.

Originally posted by TrueAmerican
Well, that actually brings a question to mind. ATS just went through a pretty good transition to some heavy-duty hardware. Seems like now more than ever SO has the juice to bring back the views column.

There's one very-good reason views are not tallied, even though we may be in a horsepower position to tally them... the number would be meaningless.

Because of our size and search prominence, roughly half of all our daily traffic is from bots and automated spidering programs... and that's on average, very often, it's more than that. We average about 500,000 page-views from "human" visitors every day... when I logged views from known-bots for two days, it exceeded 600,000 a day. We get roughly 180,000 a day from Google alone.

Any thread-view tallying function would naturally record any hit on a thread... and since half (or more, especially for new threads) are non-human, it makes no sense to do it.


As for "hot forums," to me it makes perfect sense to gauge which forum has the highest heat factor based on new threads... which defines an overall interest level in the umbrella topic of the forum.

Originally posted by SkepticOverlord
Because of our size and search prominence, roughly half of all our daily traffic is from bots and automated spidering programs... and that's on average, very often, it's more than that. We average about 500,000 page-views from "human" visitors every day... when I logged views from known-bots for two days, it exceeded 600,000 a day. We get roughly 180,000 a day from Google alone.

Any thread-view tallying function would naturally record any hit on a thread... and since half (or more, especially for new threads) are non-human, it makes no sense to do it.

Well great. Since you have been kind enough to endow us with that information, we simply half the number of views, or maybe take 40% of the number of views, and then we have a rough idea of what they really are.

Personally, I have another reason I'd like that column back, which may become evident one of these days when I'm ready.

As for "hot forums," to me it makes perfect sense to gauge which forum has the highest heat factor based on new threads... which defines an overall interest level in the umbrella topic of the forum.

Ahh yes, back to the OP topic. The problem with that is it doesn't equalize the forums for available new content- and of the two choices, replies or new threads started, replies equalizes the forums a bit better- PLUS, you still haven't addressed the inconsistency with the "Hot Topics" designation. To do it by replies kills the two birds with one stone, as they say.

But ok, I have an idea... Why not sum the total replies and total new threads and then base it on that? That way we take both into account. As a matter of fact, you could even sum flags into that as well- seeing as flags are important too. It would be kinda curious to see where the most flagged threads come from (what forum).

Originally posted by TrueAmerican
we simply half the number of views, or maybe take 40% of the number of views, and then we have a rough idea of what they really are.

No... not really.

The newest threads tend to get the most bots/spiders, while older threads do not... unless they've been bumped with a reply... or unless they're linked from other sites and a bot follows from there (there are hundreds of thousands of links to ATS out in the wild).

And my tally was simply based on the identifiers of known bots... there's plenty more than those that are known.

"Views" on large/popular sites are so inaccurate as to be not worth the effort. Sorry.

The problem with that is it doesn't equalize the forums for available new content

Correct. The focus is on available new topics.