Popular Places tab goes, but the Places tab remains

...with the horrible, mind-numbing, bone-chilling implication that camper bots actually *work* as a viable marketing device; an aspect which is usually left untouched when the talk falls on bots, their unfairness and their resource drain.
Driving the business *is* actually the original intent of camper bots; not merely bogging down the system to spite people.
(Not that I condone the practice. Buying better visibility - one way or other - is part of running a business, but I find it ridiculous that the best way to do it is currently a technical brute-force approach).

Yes, it really does work. And, they don't bog down the system as some people like to think.

The sim obviously has more information to track. And if they happen to come into "consideration distance" of a client, they can cause quite a bit of a hit, as the client starts to download and calculate the detailed information about them.
They do log in, they do count in various metrics and transactions. So it would be naive to think they are invisible to performance.
But not moving, not using scripts and attachments, not making asset transactions and causing sim handoffs, they *are* a lot more lightweight than a typical "true" user, *especially* on the asset server, which is what is really stressed these days. Once logged in, they don't touch "the grid" much, staying largely confined to the sim.

-That still doesn't change my opinion that it's a pretty ridiculous, technical round-about and bizarre way to buy visibility; something which could be done a lot easier with some other metric the advertisers could battle over.

While it can be argued that 20 camping bots can be less-load than one person with a load of attachments and scripts, the problem is that those 20 camping bots are still using resources that legitimate visitors or other land-owners in that sim could be using. And even if they do nothing except keep themselves active, they use network bandwidth which can become scarce during peak hours. When logged-in users hits 50,000, if 1,000 of those are bots then that's a good chunk of resources that actual users could be using to log-in.

I see these changes as positive ones, but as usual I'm going to have to remain skeptical since a lot of good ideas LL have never live up to the hype. I hope there's going to more categories for these recommended places though, and maybe even ways to subscribe to them via RSS feed(s) or a newsletter or something so we can find out when it changes.

As for the places tab, it's still better than the main search for finding some things, I suspect it will remain for a while longer while search continues to be refined.

I agree Tali. I didn't say or mean that they are completely invisible to the system resources - just that they don't bog the system down as some people like to think. Campers are the worst resource users because they do come into range of people but, if a place is laggy, and there are campers in it, some people simply blame the campers for the lag without even considering that the myriad of objects and textures around might be the reason. They are responsible for some of it, of course.

I also agree that ranking on traffic is a very poor way of ranking, and I'm all in favour of getting rid of it altogether.

Yes, now that you mention it, that would be an interesting thing to see. especially for no copy items

I agree that camping and bots artifically inflates the traffic numbers. I don't agree it makes traffic completely useless. If you ignore the people with 20k traffic and above, you get some pretty realistic numbers. It's also pretty clear to see when someone uses camping/bots or not. You can't hide those green dots.

That's utter and complete bullshit and you know it. Nearly all mainland sims are limited to 40 AVs, and island sim limits range from 60 to 100 AVs. Unless you are running a camper bot on an island you own and do not share, you are burning up limited sim resources at the expense of others. It's especially harmful on the mainland, with its tighter avater limits.

Even random order can be gamed with places. Whoever chops land into the most listable parcels wins.

The problem isn't "gaming" the search system. The problem is that the search system doesn't produce relevant results.

By responding to the problem as one of "gaming," Linden Labs spends its time not making a rich search system, but making a mysterious search system.

There are two problems with the mysterious search system. First, honest merchants who want to place ads that will show appropriately in searches are unable to figure out how to do so. Thus, the honest merchant is punished. The reward goes to the blindly lucky, or the merchant who invests time into "breaking" the system.

The search system instead needs to be richer, to allow the searcher more control over the results. The function of the search does not need to be mysterious or secret if the search variables are actually relevant to the search queries. Design a search system in which "gaming" the system coincides with making a helpful, search-relevant ad.

Second Life search is the crudest search system, with the most limited search capabilities, of any such system I have seen. In the past fifteen years.

I wouldn't have said it if I thought it was bullshit. They really don't bog down the system as some people think. People who have a thing against them often jump to incorrect conclusions. You know what happens: person arrives at a place and finds lag, then spots some campers and assumes that the lag is caused by the campers, but they have no evidence of that, except that there is lag and they can see campers. I can demonsrate that they don't bog down the system - they can't demonstrate that they do.

What I continue to not understand is why we can't have a "products" search like Google does with shopping. Why can't I type in shoes*, click a box that says "products for sale" and see a list with pictures and prices -- and be able to sort by price, or sort by color.

*not that I shop for shoes, being a barefoot hippie and all, but you get my drift.*

Well there's currently no way to associate images with items. It would be nice, that's why places like slexchange and onrez are good because you can browse in a more structured and useful way when shopping.

I suppose it's what we're moving towards, but until we have the ability to provide proper listings for items then search is a bit backwards. Currently it works by telling you where to go, then you have to find what you want in that place. While this method can be interesting, a lot of stores get layout wrong, or load slowly, or place you miles away from what you want.

I know, but how hard can that be when you an already do it for your places search, your profile, your group, etc...
All it does is add a function to the Object Properties.

Check "Show in Search"
Check "For Sale"
Hit the "Item Image" button and upload

The set up your Google search tool to produce the results. Still make people go to the store to purchase, but what is the point in listing items that a store has for sale without a picture? Not to mention, a significant number of creators don't list their stuff at SLX or OnRez.

Google's "Shopping" system, which is what you mean, is a totally different system to their search engine. The results are not pulled from the engine.

Oh, I realize that. But the reality is that when we discuss traffic and everyone gets their undies in a wad, what we are really discussing is business traffic. I honestly do not think there is any system that cannot be gamed - so the way to solve the problem is to allow customers to bypass the whole search results ordering entirely and let them view the products and decide for themselves if they are intrigued enough to teleport on over for a look-see. Give them a picture, a price point, and permissions. Let them decide if it's worth a teleport or not.

It's an interesting idea, and I'm sure it would be very good, but i can't see LL writing such a system in the near future.

It would take a hell of a lot of resources. Google's system is for companies to put their catalogues into it. Stores here would want their catalogues in it. I can't see it happening for a long time, if at all.

Yes. leave "places" alone and implement a "Product" search. This would do more to help inworld businesses IMHO. I do most of my shopping on SLX because I can easily just search for a product AND also see the top sold products. This would have more relevence for retail businesses than killing traffic.

Certainly.

I think it would be good if those of us who are unhappy with the idea of a Spotlight search tab of Linden picks would figure out a way to let the Lindens know we are.

coco

Firstly i use new ALL search almost exclusively for things i need to find. I particularly like the listing of items for any given search result (i use that a lot)...you'll also notice x,y z co-ordinates for those listed items. You can map those co-ordinates either beforehand...or once you arrived at location to almost take you to that product.

I think it's great that Popular Places is being scrapped....it was really a contest who had the most camper bots or camping spots.

I think the Showcase tab is a good idea if it only listed places of interest (Apollo, Greenies etc) or non-profitable organisations. There are already enough search options for businesses... they don't need this extra one.

If they must include businesses in "Show case" tab, then maybe LL should allow Creators to file their company's details (mainstore location,products etc etc )....to enable "Showcase" business listings to change on a daily or weekly basis. This would be a fairer system and include businesses of all shapes and sizes from all the business info files submitted. Obviously LL will have to inspect these locations first and pass off as suitable.

To be honest, I've never paid much attention to Land traffic as i prefer to use invisible low lag visitor counters at all my main shops /mall locations. Unique visitor numbers provide
far more useful information than say multiples of 1400 traffic units =1 camping bot during 24 hrs. With that in mind...i would be happier to see PLACES SEARCH switch to a unique visitors count as opposed to current metric. Sure it will be gamed...but say if you had 2 companies and 1 company decided to add 60 bots to login into that location during the course of the day...you might for example have company X show (e.g) 673 visitors (of which 60 were bots) and company Y show 450 visitors (no bots or campers used)......its not a great deal of difference relatively speaking and they would not be too far apart in the listings. Under the current system however those 60 bots camped for 24 hrs translates into 84000 traffic units on top its their genuine visitors......whilst company Y with 450 visitors averaging a 5 minute stay might only show traffic of around 1900 units. Company X will no doubt be on top of page 1 for that Keyword, whilst company Y will buried deep in page 1, possibly even flipping onto page 2. That is were the problem lies with the current value of 1 traffic unit.

If the Places tab has to stay, then I don't think that unique visitors will work.

Take a typical store that has an average of maybe 1 or 2 people in it at any one time - sometimes no people for periods and sometimes several people for periods. How many uniqure visitors would it have in a day? Let's say it has 100 (most stores probably don't get anywhere near 100 different people in them in a day). It won't take long for bot runners to have 100 bots, so their unique visitors start at a number that non-runners can't reach. Add to that the fact that people will visit the runners' stores from search because they rank highly, and the situation will be made worse. It will be no different than it is now - places with bots top the rankings.

The problem with unique visitors is that the daily numbers are very low, even in the more popular stores, and it wont take many bots, plus real visitors, to top the rankings, and nothing will have changed.

The Places tab need to go.

Sure the BOT runners would be top, thats why they employ bots in the first place!....but the differential in placings won't be as great. Instead of seeing results of say 86000 vs. 1900......you'll see maybe 673 vs. 450 as in above example, so placings differential will be that much narrower and more helpful to guy not using bots or camping who generates genuine traffic. Not everyone selects the 1st placed location.

I have 2 main shops that regularly top 200 unique visitors a day....and another 3 that are in the 100-150 per day bracket. I'm sure that the large main (BRAND) stores nearly top a 1000 a day. (places like MM skins, Celesrial. Last call, Simone, Nicky Ree etc etc)

I'll have to count my uniques

I doesn't really matter what the differentials are - it's the ranking positions that matter. The top few of any search results get the most click-throughs. Even if there was only 1 between each of the top 10 (e.g. 450, 449, 448, 447, ....) then the top few would get the most click-throughs. Bot runners would still top the rankings, and the top ones would prosper more than the lower ones - assuming the product qualities are equal, of course. Nothing would have changed.

i'm not convinced of your argument. Even if both systems are flawed,..which they are, as Bot logins are being used in both cases.....Unique visitors is a little bit fairer than the 1400 traffic units per 24hr Bot being used right now.

*PS For the record i'm in agreement with you...they should scrap Places search as you can get adequate results in new ALL search with a couple of clicks. LL won't be ditching Places tab, so we have to think of a fairer metric.....anything to neutralise the use of camping bots imo would be welcomed.

I don't really disagree with that. Neither of them would produce fair results in the way that you mean "fair". Imo, the only change would be that the numbers are smaller and, depending on the field that a business is in, it would be a little more difficult for bots to take control of the results. E.g. it would be hard for bots to top Xcite's uniques, but it would be easy enough for them to top most results. Neither the current method, nor uniques, is a good solution, imo, and the Places tab needs to be removed.