How would you like a "Gamers Seeking Gamers" system to work?

[Mark]

All you can do is make it work best, and work best for those who want to take full advantage of its functionality. Trying to make it schemer proof will get you nowhere and you'll wind up with a less useful tool. The tech to have a Lat/Long system exists and it seems like it would be the best. So maybe you have a requirement for a zip code, leave the other info up to the user, and have a report function so all users can help spot the most egreious of misusers to help get those cleared out as needs be.

 

[Janx]

Or the user could enter a fake - but close enough - address.

but once again, it's close enough. If the user is entering truly bogus info, it won't help them find gamers.

And odds are good, the city, state (which is nearly the same as the zip, from a functional standpoint) is good enough.

In nearly all design cases, the programmer is going to snap whatever data he has to a set of lat/long coords and then do the math to find nearby players.

Doing a match on city just isn't sufficient with more complex regions.

For instance, I live in Houston, but the address is really Spring. The city limits isn't that far from me. And "The Woodlands" is just north of me. However, "The Woodlands" doesn't really exist as a postal code, it's really Spring, or Ponderosa, or Conroe. In any event, looking for Spring, TX gamers by matching on the city name would fail to find very viable candidates for me.

Thus, converting the address data to lat/long internally lets you do a "who's within 15 miles of me" search.

The DB stores the user's address (or at least the major parts like city and state or zip or country, depending on region). The code converts that, maybe by another table that translates, or a web-service call to Google Maps even to a lat/long. Problem solved.

Don't over analyze location mapping. Just say you want it, and what you don't want exposed to other users (your privacy concerns). Let the developer figure out the internals.

Why I say this, is Morrus is effectively doing Requirements Gathering. Finding out what the tool should do. You don't determine HOW it does it, just make a list of the features. From there, that starts turning into a feature list and the developer has something to work from.

 

[Thornir Alekeg]

The lat/long coordinates provided from Google Maps go out six decimal places. If the EN World locate function were to automatically truncate the coordinates to two decimal places, that would narrow the members' location to about 1 mile. Truncating to one decimal place drops it to about 10 miles, which would likely be enough obfuscation for most people seeking a game, while still providing enough information for a meaningful search radius.

 

[Scott_Rouse]

I have been using Meetup.com and second the vote for that sites functionality. If you are creating a system from scratch I would look at that meetup for direction or invest your money into other areas of the site that don't have 3rd party solutions and create enworld meetups. my 2cp

 

[Janx]

another thing to consider is partnering up and outsourcing that functionality.

Much as enworld partnered up with others for e-publishing and e-booksales, don't write a tool if somebody already makes it.

At the simplest, just link to meetup.com. Or send them some mail, and talk to them about a little partnering.

 

[maddman75]

Would that be accomplished by essentially tying the existing Groups functionality in with the Gamer Search functionality? So that a search gives your results both in terms of matching user profiles and matching groups? (Assuming someone gives their group a "location")?

Yes, I would say so. I'd picture it as a user enters their location. If there's a group for that location, the server informs the user. If there's not, it will look for the closest one or ask the user to create one.

One reason I think Meetup worked for us was the emphasis on real life meeting for networking. That is, this isn't a tool to find someone to fill a spot in my D&D game. This is a tool to meet with other gamers in my community. I would advise against having it create a group for each gaming group, but rather one for each community. The gamers can meet each other and network from there - you meet the group at the game store, and invite the cool kids back to your place to fill in that weekly spot.

 

[Lord Xtheth]

I would like to see an activity log perhaps. Somthing to tell me I'm not trying to talk to/invite someone that hasn't been on for a bazillion years or so.

Having been stopped by postal code verification because I'm Canadian, Id like to make sure that it can accept any postal code.

 

[Festivus]

I really like how PenandPaperGames is setup. I get email notifications of new players and games in my area (within a radius I establish), and allows me to reach out and contact those people and let them know about the monthly D&D meetup we have.

 

[Morrus]

I would like to see an activity log perhaps. Somthing to tell me I'm not trying to talk to/invite someone that hasn't been on for a bazillion years or so.

Way ahead of you there! Think Facebook!

Having been stopped by postal code verification because I'm Canadian, Id like to make sure that it can accept any postal code.

Don't worry, as a UK gamer running a primarily US-populated website I'm very aware of such issues!

 

[Morrus]

I really like how PenandPaperGames is setup. I get email notifications of new players and games in my area (within a radius I establish), and allows me to reach out and contact those people and let them know about the monthly D&D meetup we have.

Other than the radius (and a lower membercount), it's the same as what we already have here - the vBuleltin search engine searching the member database for profile fields. Not sure how they've done the radius bit, though.

We can definitely drastically improve on that. I get over a couple dozen results when I use EN World's current system to search for "Southampton UK", but zero on PnP's.