Re: Rampant Frontpaging

[grinder]

Another criterion for FPed nodes is that they should already have a certain number of offspring nodes (replies and replies to replies and so forth). That is, there should already be a good discussion going. Especially when there's a good argument of for and against and both sides have their pros and cons.

<update reply="dws"> vroom posting a new feature is the exception to the rule. New features deserve everybody's attention, and should ideally be the first category that appears on the front-page (of those logged in, anonymonks don't care). Anyway it's his toy and he has the root account. If vroom wants to FP something, we're not going to stand in his way. Oh and peoples, don't vote up vroom's posts on new features, he doesn't need the XP. If you think he did something cool, you know the drill.</update>

<update reply="dws" n="2">I see vroom's messages about new functionality very simply. They start with the word "new" and are posted to the PM Discussions area. And most of them are just minor tweaks to the system, nothing earth shattering, they just don't matter to most people, so why FP them? (again, vroom can FP them himself as he sees fit, but people shouldn't rush to FP a node simply because it was written by vroom). He's not the messiah, he's just a very naughty boy™.

Talks and PM meetings do not meet the FP criteria either, as the are not of planetary importance. When The Damian speaks in Berlin, it's good to know, but if you're really interested you're better off getting the details straight from the horse's mouth.

Short lead-time announcements should theoretically be posted to the News area. That that is hopelessly disfunctional at the moment is another matter for debate :).</update>

Then again, even though I raised this issue a while ago in Points system for frontpaging nodes (that lemming missed :), these days it doesn't bother me as much, as I use a custom front end to view the newest nodes, that threads the nodes with their responses, and I tend to visit all threads (unless they have a content-free title (beware!)). It also makes it easy to see those old nodes that get revisited.

From time to time and old (but interesting) thread is revisited and a flurry of replies appear. It would be useful to be able to bring that to the front, especially during a slow-news day.

I was quite surprised at the amount of votes showered on Is readdir ever deterministic?. I thought it was a good question, but when I saw the XP being racked up I was puzzled at first and then concluded that the node had been front-paged. (Then again, as I didn't look, maybe it wasn't and people just thought it was brilliant, I dunno). It's almost as if one needs a checkbox that says "Don't bother front-paging this".

Re: the argument of tying FP to rep, I don't think that front paging a node should be contingent on its rep, because we all agree that rep is only approximately related to a post's actual merits.

--

g r i n d e r

[dws]

Another criterion for FPed nodes is that they should already have a certain number of offspring nodes

A counter-example from earlier today: vroom posts to announce a new feature. Such posts are unlikely to garner many offspring nodes, other than the occassional well-deserved "you rock!" Should these be held off of the front page?

<update reply="grinder"> This is but one counter-example of many. Short lead-time announcement of things of interest (talks, PM meetings) are another class that often deserved to be front-paged immediately.

By the time you sort through the exceptions-to-the-rule, having a rule embedded in a mechanism (as opposed to in the culture) is very difficult. </update>