Proposal: Changes to Popular Places

Cross-posted from http://slhomepage.com/2006/09/02/pro...opular-places/
and http://forums.slhomepage.com/showthread.php?t=50

What to do about Popular Places? I’ve seen this question asked in-world and on the forums. The debate normally centers on how the list only shows the places that normally grab people through camping devices such as chairs and dance pads tht pay residents to park their avatars in a location for a few L$. The following proposal contains a major part (revision to the Popular Places tab of the “Search” dialog) and a minor part (discussion about how traffic is computed).




You can view all of the images created for this proposal by visiting this page. This new design is cenetered around breaking out the current list which can be a maximum of 40 entries (though usually less as there is a big overlap between PG and Mature locations).

The first change is the removal of the “Include mature places” checkbox and replacing it with the same combo-box used on the “Land Sales” tab. This allows a “Mature” only list to be displayed which is currently unavailable. In place of this checkbox in one for “Include places not listed in ‘Find Places’” to include locations where the parcel owner may not have the L$ or wont to include standard advertising.

The second change is the addition of the “Category” combo-box, taken from the “Places” tab. This list is currently broken, but is supposed to be fixed in the next 1.12 release. Thedefault choice would render the list as it is now, but selecting an option would filter the list based on the parcel category choice (for all but the “None” category). This change would allow many other locations to become easily visible to new residents in particular, as well as providing long-standing residents with locations they may have missed or are just curious about. The only issue with this addition is the chance of “gaming” the list, such as lower-rung casinos listing themselves as “Stages” since they can’t crack the major list at first. Beyond a new category of “Abuse Report” to handle mis-labelled parcels, I’m looking for suggestions on how to handl this issue if other residents believe it may be a problem in the long run.

The third change is an integrated map. It’s true, the “Show on Map” button can be used for this functionality, but people looking for “hot spots” could scroll the list with the map tab active to see the current population in the location rather than opening, closing and scrolling mechanism that is required today.

I apologize if there are any rough spots on the images, I did my best in trying to duplicate the look of the current UI elements. Hopefully they give a good overall design indicator. I believe these changes would be helpful for residents in finding locations and other residents and enhance the Second Life experience.

Now for the minor point: traffic computation. I still don’t know the formula, I tried to figure it out, but without sitting on a parcel 24 hours and recording the comings and goings I haven’t been able to figure out even a rough formula. Whatever the current formula is, I believe now isa good time to revisit it. The big change I think is needed is to create a greater weight on unique resident visits rather than time spent. With the ability to run multiple instances of the Second Life client and proliferation of alternate accounts some people artificially increase their traffic using these “exploits”. It would be great if the hardware hash mechanism could be used to filter alts out of traffic entirely.

As for a basic computation, I’m using round numbers: 500 traffic points for the first 5 minutes a resident is on a parcel during the 24 hour period; 1 point for each additional 5 minutes spent at any point during any part of that 24 hour period. This would make a maximum of 788 points awarded for a single resident managing to be on a parcel for 24 straight hours. With more points awarded for unique residents, this would up the traffic of places that are great to visit but don’t hold the “linger” power of current in-world games.

I am hoping to get feedback on any part of this proposal and am leaving it open for Linden Lab to determine if these changes are warranted. If someone wishes to make a voting proposal out of this, feel free, but I am not a fan of the voting system and don’t believe software development houses should wait for a vote when something could be implemented to help all users (as with the recent change in the “Sell Land” dialog).
Thank you very much for reading this post and adding your comments!

I've thought about this myself...

camping chairs no longer serve the purpose they used to: with no dwell, chairs can't pay for themselves, so they must exist entirely for the purpose of being a loss-leader: you sit on the chair and hopefully buy something from the vendor.

Of course, that simply means that you're giving them back money that they gave you, or money you earned from another camping chair... which again defeats the purpose.

On Find: I think that there should be a find option that looks at every parcel on SL, whether or not the owner wants to pay for advertising. I don't sell anything in SL, but would like to be findable based on keywords. So forcing me to pay for advertising doesn't work - since I'm not making any SL money to begin with.

Why do people throw the word "exploit" into the wind when this clearly isn't an exploit? Linden allows people to create upto 5 alts per payment source. They do not limit how you use the alts except for the fact you can only buy first land once.

You honestly believe that Linden Lab planned a "traffic" system with the sole thought that it was entirely designed to reward people with the least exciting, least enjoyable, least interactive content imagineable?

You really believe that Second Life was meant to be used entirely by zombies logged in but not at the computer?

Camping chairs and dance pads are dishonest ways of getting traffic, totally removing the entire functionality of the whole system, and as a result, anyone who is here trying to make it honestly, is screwed.

Most of the popular places are full islands - which pay more money per month than the average user and majority of us. Therefore Linden Lab do not want to upset them by making them actually have to work for their popularity.

A very simple change that I have mentioned previously is that traffic should be divided by the amount of land that the parcel contains - so the big sprawling islands that are full of campers have to work as hard to appear popular as the little club on 2048 sq m that really is quite good.

Lewis

Can we then agree that it's not an exploit?

An exploit is defined, within Second Life, as "something happening in a way that is not intended to, giving someone an unfair advantage over others".

I would say that camping chairs to boost traffic is definitely an exploit, given that definition.

Lewis

Nice work, Clubside, it's really good seeing user created interface ideas with pictures to make it clear. I wish LL would give us the means to whip up displays like this in an easier manner than cutting and pasting stuff in a bitmap editor.

I like the idea of seeing the most popular places in the different categories.

From the Linden Blog...

So, just what do we consider an exploit? Exploits are any bug or viewer-side security hole that…

* gives you unauthorized access to scripts, objects and other assets from within Second Life. Note that this does not include programs that pull textures from the datastream. Unauthorized access includes copying, modifying or transferring assets you don’t have permission to.
* allows you to destroy inworld content. This also covers the ability to return objects by staring at them cross-eyed.
* allows you to steal L$ or create L$ out of thin air. Bugs like this include LindeX bugs and viewer bugs.
* can result in a Denial of Service (DoS) attack on the grid, effectively causing the grid to crash.

It's very much an exploit.

I wonder is theiving folks land from under them an exploit? huh HULKA

Buying land on the public market, even at below market prices is not an exploit. But let's keep the personal digs out of the thread and focus on the topic, k?

Put simply, any system will be duly gamed.

Originally, voting on popular places was the norm. This was quickly abused I am told, much like ratings were when they still had a point. Simply, people less interested in the point of these systems are very easy to sway -- or dare I say, control -- making the system fall prey to mobocracy.

The same thing happened with dwell, and with its "successor," traffic. These systems were put in place quite similar to feature voting: quite unmindful of their implications. Unlike feature voting (which is largely ignored by the Lindens), popular places is at the same time a benchmark and an eyesore. Quite a difficult role to keep.

The Linden picks are notable in passing, but also missed the point.


Your suggestions, Clubside, are good ones. The selection of "Mature" and "PG," as well as by category, is a step in the right direction. More choice should be present in the current system. Including the creation of entirely new categories.

Such is what search engines currently do. But I've already talked at end of those.


The second thing these lists should provide for is genuine user feedback. For papers like M2. Sites with information and viewpoints, as with the Second Life Herald and countless blogs. And for other community sites, like alternate forums.

What we're grappling with here, quite simply, is a fundamental lack of understanding and conflict of interest. Linden Lab, for all their achievements and faults, is in the business of placing their interests first. As such, the mechanisms are limited and skewed grossly in their favor, despite it being in their best interest as a company for them not to be.

It's hard to say what will happen, but the discussion helps. If not Second Life, then for whatever comes after it. And it's a pretty good idea to start with. Keep at it.

Sorry if I messed up the proposal, I only just discovered the bad link. If you check the screen shots you'll see I have replaced the "Show Mature Places" checkbox with a "Include locations not advertised in 'Find Places'" for just that reason. I also moved the "Mature" option to a combo-box the same as on the "Find Land" tab.



I apologize if exploit was the wrong word, I put it in quotes because I wasn't sure how I should label the problem I was trying to address. The problem I mentioned was people using the multiple login option of the client to park multiple alts on a parcel and manipulate traffic. I just thought it was fair to t least bring up the topic in terms of both Popular Places and traffic computation.



While this could be a factor, it's simple enough to parcel off just that section and achieve nearly the same effect. Ultimately some non-camp places could suffer as well like larger wilderness/park areas like Bonnydoon. Perhaps there could ne a valid multiplier rather than a divider with a logarithimc scale. I honestly hadn't thought about the effect of size. I also didn't take into account the possibility of islands set for more than 40 avatars (some of the Popular Places currently are) and was just trying to treat traffic as traffic.



If someone can point me to a quality XUI COM component I'd gladly whip up a prototyper for people to use. Unfortunately the one included with the Second Life Windows installation doesn't even have a company name on it. I did some searching to no avail. I feel your pain, though, as the mock-ups I'm doing regarding my communications proposals are a lot more difficult.



Thanks for the comments. You are correct, gaming the list will certainly remain an issue, though it will require more work on the categories. As for expanded categories, I'm trying to address the category issues at SLHandbook, and will have a preview next weekend. I've sought input for devloping a shopping site style category/sub-category trunk system to no avail.

Ultimately this list is more of a browse structure which is often better for newer residents not sure what to "search" for. That is the goal of SLHandbook as well. Ultimately the content in Second Life is too diverse, and I hope we can have more services for residents to see the breadth of options.

Just bumping this one last time since I planned on adding my next proposal (regarding in-world communications) tomorrow. If you don't have any particular comment on this, I'd at least appreciate any input regarding the formal of the proposal as the next one has far more imagery.

Clubside this is great! I'm all for anything that will make the "Popular Places" list a true representation of what is popular in SL. My one question is about cagegories. What about the places that represent themselves as a mall, casino, club, etc? If they truly are all of that, would they be stuck having to pick just one facet of their business?

Ultimately we're stuck with the current categories Trent, as they would be the easiest to hook in both from a owner tagging and developer perspectives. Since the "All" category would remain, the most to benefit from this change would be those using the categories like "Parks & Recreation", "Stage" and a few others.

It's true a better system is needed, and I am working on that at SLHandbook. I haven't been able to get much direct feedback so I am doing some basic building before opening it up to parcel owners. After a round of blanket input I will try to consolidate the entries.

If that wasn't vague enough I will just switch to the more direct analogy: the shopping site. Search engines have their virtues, I use them plenty as well. I also enjoy browsing using a drill-down structure, and in the case of Second Life I would use the drill-downs of shopping sites where major sub-categories can exist under multiple categories as my preferred method. If you follow sites like amazon.com or compusa.com through their categories you will find a "trunk" system where you start at a specific top and then filter through the sub-category choices.

In the case of your example, something like "Gathering Place" would be more appropriate than a top-level casino, mall or club choices. Once choosing "Gathering Place" you'd find Casino, Mall and Club as sub0categories, of which a location could belong to all. In the case of a location that was only a mall/casino, it would be filtered out if the user at any point chose "club" for filtering. And below these choices would be more direct sub-category trees, the extension of the root trunks. So "Club" could lead to "Live Music", "Adult" and many more branches that if needed could be branched further.

If this system could later come in-world I think that would be great, and my intention in developing SLHandbook was for just that purpose. In the meantime, my hope with this proposal was to give more exposure to great resident creations beyond a global traffic number, choice of "rating" and the bizarre "only include locations with images" options.