At what point is an extra feature 'bloat'?

Summer 2005 saw Invision and vBulletin launch new versions of their extremely popular board softwares (2.1 and 3.5 respectively). These two market leaders were faced with the same problems faced by any company releasing a new product or service at the same time as a rival company: how do you make sure people buy your product, and not the rival one?

Invision jumped the gun, releasing first but having to issue a series of new versions to fix bugs not fixed or discovered in time for the release. It's now on v2.1.2, with 2.1.3 imminent. vBulletin released 3.5 two weeks later, and is yet to release a bugfix update - although, given how full its bugtracker is, one suspects they know the moment they do release a bugfix, they'll lose the reputation for thorough pre-release testing they're currently enjoying.

Appearances may be everything, but I dunno to what extent their customers are happy about having these bugfixes withheld from them because of marketing pressures.

Headline-grabbing with new features is a second strategy that's probably more important to the market (that's you and me) than release timings. A lot of the development this summer, both on IPB and vBulletin, was AJAX implementation, which makes forum pages more dynamic by doing stuff like being able to post a quick reply or edit a post without reloading the page (on vBulletin) or checking that a username isn't already in use at the same time as someone types it into the registration form.

Very nifty, certainly; it's certainly made life easier for moderators, who can edit titles and topic descriptions at a click of the mouse, as if they were files on a desktop (in Invision). AJAX has been used a lot - after all, it makes the software look real swish to the pros like you and I, and definitely boosts the attractiveness of the featurelist.

But the acid test for a feature, one every cynical admin should be asking himself when reviewing a featurelist, is 'Will this make my board more successful?'

At the end of the day, that's what a feature's value always boils down to. If it's a feature that will make it easier to keep the site well moderated, there's little value if your site is already well moderated; for a forum with threads in ALL CAPS the AJAX one-click title editing will be handy, sure. But has adding two edit modes (for the poor scared user to choose from) really added value? I say no. Adding complexity and pointless features like vCards will just confuse the user or divert them from doing what your site is there for: participating in the community. If your community is so specialised that it would actually gain from a vCard feature, add it as a mod!

Cliquey Community?

In previous posts I've spoken about the importance of community spirit to the success of your message board - good advice, no doubt, but sometimes it can go too far. Communities can get so close (e.g. many members being on first name basis) that you can have serious trouble turning lurkers and newbies into active members.

This problem is more likely to hit small, young forums - on a large message board there's usually enough diversity between the subforums that there are plenty of opportunities for lurkers and newbies to integrate without problems or fear. On a young forum however, if a 'clique' of 'veteran' members forms, it could easily 'choke' your community.

Interestingly enough, the risk of a 'closed community' (which is intimidating and/or hostile to outsiders like lurkers or newbies) getting established on your boards is greatest right at the start. Let me explain. When starting up a new site, the lack of activity can be extremely frustrating, and there's a big temptation to get all your friends to sign up and post. In the short term, you get masses of activity, and the stats rise nicely. That's what I experienced on Rockforums.net when I started up - all my friends from uni signed up and started posting, and a few of their brothers, and brothers' friends, too. The growth of my forums was gorgeous, the posts were rolling in.

The thing I hadn't noticed, or realised the danger of, was that a lot of these posts people were making were like 'Bob, what are you doing on the weekend? I'm going over to Andy's'. This, of course, meant absolutely nothing to the other users of the site, who just felt excluded as a result, and never felt like hanging around long enough to become a part of the community. So the problem here is, don't rely exclusively on your friends for posting and growth, or else their posts, quite naturally, will just be a string of in-jokes, similar opinions, and generally threads and posts which other members have absolutely no desire to be a part of.

What's worse, your friends will often break the board rules just for a joke, to see how strict you're prepared to be. You may understand it's a joke, and be lenient, but others may just see blatant rulebreaking and the admin turning a blind eye - setting a terrible example. Deciding how strict to be with your real life friends, and friends of friends, is an extremely hard thing to do!

So the moral of the story: whilst you can get great short-term boosts by enlisting the help of all your real-life friends to get a site started, or more active, there's a big risk that the long term effect of that strategy will be that the community won't be open enough to new members, and growth in the future will be affected, as spam and crap, uninteresting discussion goes rife because you don't want to annoy your friends by deleting their threads and posts.

My tip to someone starting up a forum is: get a few of your friends, maybe three at most, to help out getting things started, but you're safer looking elsewhere on the Internet for new members and more posting.

A challenger comes?

UseBB. Terrible name, no doubt about it; it seems like every free bulletin board under the sun these days has a name that ends in -BB (obviously that's untrue, look at SMF and MyTopix for example - but my point stands). But that's just a name - whilst they'll have a harder time getting people to notice them, it doesn't actually affect he software.

And boy, what fantastic software. For a few months now I've been struggling to make Invision more search-engine friendly. Installing the Friendly URL mod helped a bit; so did spattering some rel="nofollow" tags around on the links I didn't want spiders following. But Invision still insists on sticking bit onto links like /boards/index.php?showtopic=1280&view=getnewpost . That's fine for users that want to see the latest posts, but Google & co are treating that as a separate page from /boards/index.php?showtopic=1280 - and because it sees two different pages on your site with the same content, you get penalized. Not fun.

UseBB, very wisely indeed, just uses anchor links like /UseBB/topic-post10.html#post10 to achieve exactly the same thing for users - but Google realises that's the same page as /UseBB/topic-post10.html and so you don't get penalised for duplicate content!

See the difference in the tidyness of search results for yourself:
http://www.google.com/search?q=site:http://www.usebb.net/community
http://www.google.com/search?q=site:http://forums.invisionpower.com

I use Invision as an example because it's terrible handling of search engine spiders has been frustrating me for a while now, but I notice vBulletin has exactly the same problems.

Add to that the fact that the XHTML code UseBB puts out for all its pages is extremely clean and 100% XHTML strict (wow!) valid, and you've got a recipe for producing threads that are going to go straight to the top of Google results and drive the growth of your board for years to come! There's no excuse for the established boards to be so bad at getting the great content we have in our forums properly into the search engines. Forums often have way more content than the average website - if we admins find it so hard to make the kind of money you'd expect from content like that, it's because we can't turn that content into hits well enough.

There's a lot missing from UseBB though - polls, private messages and a fully functional admin backend. A better backend is essential (and in the pipeline for 1.0), but I'm not convinced of the need for polls and PM's - users can email each other from the boards, and as for polls I've always thought it would be better for the community spirit of a board for people to voice their opinions, rather than anonymously vote. After all, we're here to share opinions and make friends!