Welcome to RiotACT 3.5

In regards to the NCA Works Approval for the proposed Geocon development at 70 Allara Street: the traffic implications for local residents as well as the impacts this build will have on commuters using that end of town, have either been ignored or overlooked … it is very disappointing and not because of lack of enthusiasm and effort by locals. We took every opportunity over the last eighteen months to present to the NCA, a number of development, traffic and engineering experts who explained to the NCA the reasons why they should reconsider the Geocon plan as presented. It will be cold comfort to know, that in a couple of years we were right.

ya go away and have a son for a minute and come back to more rioting newness – haven’t explored much yet, but will…

The quote / blockquote display has been sorted. Might also have a look at left justified comment & post text rather than the full.

The comments disabled is funny bug we are seeing on some posts. We very, very rarely lock down comments on a post, so it just depends how long it takes for one of the editors to notice and fix it. Anyway, should have it sorted soon!

As for the stars and ratings, we have been toying with the idea of just a thumbs up, thumbs down scoring system – which would address the negative scoring and rating weighting issues.

Thanks all for your positive comments and feedback!
Keep coming with the bugs…(and yes, there are more that I am still working on)

Can we have comments text be left-aligned and not justified? Justified text is harder to read.

I’ve noticed an article posted today with comments disabled… was this a new feature?

Or was it functionality carried over from the old site and hardly used?

I’ve liked that all stories are up for comment in the past, it would be a shame to lose the ‘right of response’ this site offers if this option is made available to posters, rather than being an admin function of the Riot team.

+1 / -1 works for voting too.
Still make it Bayesian weighted though, since you want a believable rating system that avoids having a new article with 1 vote immediately escalated to first place, as its rating would then be 1.00.
Bayesian rated, its starting rating with a single positive vote would be a little bit higher than the average rating of all items.

The more votes an article has, the more your audience can trust the rating. If it has only few votes, each vote, positive or negative, will only have little power to influence the rating away from the average rating of articles.

As in:

Front Page Article Rating =
((Average_article_votes_quantity*Average_article_votes_quality)+(This_article_votes_quantity*This_article_votes_quality))
/
(Average_article_votes_quantity + This_article_votes_quantity)

Where:
Average_article_votes_quantity: The average number of votes of all items that have votes_quantity != 0
Average_article_votes_quality: The average rating of each item (of those that have votes_quantity != 0)
This_article_votes_quantity: number of votes for this item
This_article_votes_quality: the rating of this item

Avg_article_votes_quantity is an auto-adapting weight. The higher its value, the more votes it takes to influence the Bayesian rating system.
ie: If people are keen on voting on articles, the more people who use the voting function, the harder it becomes to out-game it.
If you really wanted to, you could finetune by setting an upper limit on it to make it easier to game or to prevent some potential super-article from overbalancing the rest of the voting system.

If you introduce a time factor so that nothing stays on the frontpage forever, too.

On the star ratings, this is a great article on why star ratings end up being useless.

To be of any use, the tag cloud should probably only look at posts from the last N months, rather than forever. Election08 is still right up there but it’s hardly the topic of the moment.

i like it, well done

Nice work!

PPS: Some features I like, huzzah for the codemonkey. Get that man\woman a pint, stat!

Bayesian weighted voting would be a nice fix on an ongoing broken feature, though.
(or you end up with one five-star vote outweighing 100 four star votes, which is bad)