This discussion is good because a hastily designed solution can be worse than the problem.
But it seems like too many engineers, not enough product managers. :-)
Let us step back from the solutions and look at problems being solved.
Unfortunately, Ted or his link posts are seen as a problem by some. There is no technology solution to this other than an ignore button which I think is a very bad idea and sets precedent for discouraging discussions of all kinds of views as an unintended consequence. In most forums that implement this, the ignore becomes part of the argument to create cliques where people are encouraged to ignore what some don't like and so stifles views that are unconventional. At best, it creates echo chambers. In reality, this is just a form of hitting out at someone you don't like. Think about this very carefully before you implement. The solution is worse than the problem.
For people suggesting solutions, please get over your views on Ted if you want a practical solution or spend a week searching and culling posts to see how much work that is.
In my brief history here, I don't see this as a Ted problem but a usability problem that needs a technical solution.
I can see the tension between posts with links not all of which result in discussions and posts that have broad discussions. The avalanche of new posts with links may seem to suddenly create an end to a discussion as if it is time to move on. It especially punishes people who may have spent considerable amount of time to write a thoughtful post or comment only to see it banished to archives (not the first page) quickly. It also prevents discovery of threads with multiple views or comments when it is "archived" quickly. I think we can all agree that this is A usability problem to solve without bringing Ted into the picture.
This is not the first forum to have faced this problem. Reddit is probably the closest in terms of facing a similar problem while no two forums are alike. It has solved this with multiple tabs to filter by as you can see by visiting the site. News content sites also want to draw attention to latest news headlines while making articles that have an active discussion or are popular in some metric also be discoverable. Same problem at a high level. Most of these have converged over time to a multi tab display based on usage patterns NOT on tagging or categorization for some valid reasons.
@Accipiter's solution has some engineering merit but often good engineering design is not necessarily the best product design. My issues with it is:
It primarily tries to solve the usability problem by making it a Ted problem and "banishing" Ted to a different category. This will appeal to those with an antipathy to Ted but it will not work for the following reasons, not all of which are engineering reasons:
1. Ted is not going participate by being "banished" as much as his detractors might like. No one would. How is the enforcement of Off Topic category working out? Not very well. Besides, you cannot come up with any objective criterion that would force that "banishment". What if he posts a link with a question in it? Is it in the Link category? If not, then every link post will come with a question attached however trivial. The point is, you cannot have a viable solution that diminishes Ted in any way as long as he is critical to the vitality of this forum. Deal with it.
2. It solves a problem that isn't the central problem by making a weak assumption that I don't think is true. That people want to differentiate between discussions that arise from a link post and discussions that arise from a question posed. I could be wrong on this based on my own usage pattern, but I think the problem is really separating posts with discussion from posts that don't have any and making BOTH easily discoverable without one overwhelming the other. I don't think people care if a broad discussion came about from a link post or a question post. The tab solution and the link category differ in this very basic premise of the problem and I think you should arrive at a consensus of which problem you are trying to solve first.
3. Even if the problem to be solved is separating links and link-based discussions from other discussions, then the empirical data from the forum shows it to be a weak one. My quick scan shows almost 75% of the posts with discussion arise from a link post, not necessarily a Ted post. So the solution wouldn't really filter effectively and the link category will have the same problem we have now being almost as big as the All threads category.
@Anna's worries that threads without a follow up will not be found is a valid one but not a realistic one. It may seem realistic to those that view the solution as a way to avoid Ted but not otherwise. It is not some filter that will be the default and no one will see anything outside it. The user behavior will be the same as now with all discussions being the default but people can switch tab to see discussions with followups easily and participate without fear of that thread being archived quickly from an avalanche of new link posts.
I respectfully suggest that some of you with strong feelings against Ted (and I am not saying they are wholly unjustified, he is what he is) not confuse this as a way to avoid Ted or to get at him in some way and look at the solutions by factoring that out.
I don't have a personal solution to pitch. Just suggesting a solution that already exists for similar problems elsewhere and arrived at after much experimentation. No need to reinvent the wheel.
