Mapraider 2.0

Enhancing the browsing functionality.

Browse maps?

Yuck.

I mean really — what was I smoking when I implemented that portion of the site?

In all honestly I think it just boiled down to wanting to get the thing done and pushed live to the world. The current browse system was intended to evolve into an intelligent browsing system that would progressively show you maps based on your click-path through the thousands of potentials in the database.

So much for grand ideas.

Funny thing though — being able to browse the entire database of maps is probably one of the most crucial features of the site, if not for one reason:

It gives search engine spiders access to every map in the database.

And why is this such a wonderful thing? Because it helps improve the site's ranking in search results, which results in more traffic to Mapraider and potentially more users, more ratings and more map submissions.

So, keeping this importance in mind, how do we make it so that it doesn't suck for users on the site? Great question!

Back in the days of the first internet boom, I had the privilege of working with one of those huge dot-com agencies (now defunct) where I really got my feet wet in interface design and usability. It was fun at the time, but most of what we did back then was so "90's web style" that I had put a lot of it out of mind since then.

It dawned on me when I first started attacking the browsing feature that one of the things we worked on years ago — a dynamic, "intelligent" resturant finder where you filtered results without feeling like you were laboriously searching — was a perfect solution to my map browsing delimma.

Essentially, it works by simply providing a list of categorical links that, when clicked, filter the list of maps into more finely grained buckets. It's nothing more than a progressively disclosed, hierarchical tree structure — though in my implementation, it's not purely a forced linear experience like you might expect.

Not only would it provide simple, clickable, faster and meaningful filtering through all of Mapraider.com's maps, but it would also push the names of the game systems that we host maps for, which are important search engine keywords, into more prominence on the site. Right now, the only place these game systems are listed is in the game system pull-down of the search toolbar.

I also discovered that this structure became an awesome foundation for the site, providing automatic, direct access points to each gaming system and their associated maps. It allowed me to simplify the site's navigation, provided logical locations for featuring other content such as RSS feeds, and almost eliminates the need for manual text searching (don't panic ... search is not going away).

It may not seem like much, but this dramatically new approach to the site's organization is probably one of the most exciting changes in the site, and I simply cannot wait to launch it.

Categories:

So the next version of Mapraider.com will have a simplified rating system. Great. What about rating moderation?

Well, if you've spent any time reading this blog, you probably remember the great suggestion for improving the moderation system made by Virgo47 that I blogged about.

So of course I have to ask...

With the simplified rating system, is moderation necessary?

In Virgo47's defense, at the time he made those suggestions I hadn't revealed that Mapraider would officially have a different rating system moving forward.

Obviously, you could have a good debate with whichever side you may choose. My gut is telling me that having moderations for a 5-star rating system is overkill just in terms of the difference in rating scales. That is to say, it's already confusing for some members on how to moderate on a 10-point scale (as evident in the Mapraider forums). I imagine that it will be even more challenging to do it fairly on the new 5-point scale.

Speaking of scale, and economies of... if you look at the history of the site in terms of ratings and moderations, there are two other arguments for removing the moderation system that really jump out at me:

• At any given time, there are thousands of ratings waiting to get moderated and most of them simply "time out" and get moderated as fair by default (as I write this, there are 4,000+ ratings pending moderation)
  
  This essentially means that most members don't bother moderating ratings, aren't logging in often enough to do so, or are simply not rating enough to see the pending moderations.
  
  I've already pointed out in a previous post that the number of members actively rating maps is low.
• The amount of database storage space to store the ratings, the moderation status and their multiple moderations, along with the CPU power required to access all that data increases exponetially as the site grows
  
  Granted, we're not talking huge amounts (yet) for the current, but the return on investment for this allocation seems minimal and since I do pay for the hosting out of my own pocket, size matters

Over the years I've become a firm believer in forcing features to prove their worth before implementing them, and the rating moderation system seems like a good candidate for the chopping block. In the many years of running the site, it simply hasn't shown itself to be of significant value.

Am I wrong? Perhaps. I welcome any thoughts on the subject.

Comments ported from original blog:

• At 3:12 PM, Doug said…
  From 3 moderated ten-point ratings to 5 unmoderated stars. There won't be any ones and fives, so that gives us maps rated 2-4. What's the point?
  
  
• At 3:14 PM, Doug said…
  Well, ok, now that I think about it there probably will be some ones. Those will belong to mappers without friends.
  
  
• At 11:20 PM, John said…
  Hmmm... I'm not sure I agree with that assumption. Members give ratings in whole numbers of 1-5, but the tallys will be decimal values -- an average taken across x ratings isn't always going to be a whole number.

Categories: