Definition of Theft?

I agree that those places would be expected to be at or near the top when searching for couples animations, but this is where the square peg and the round hole come in again. The search engine is designed to index and rank pages, and it is normally used where there are pages to index and rank. But there were no pages in SL for it to work with, so LL contrived some, and that's why it's a square peg and a round hole. Pages are normally written by people, and they contain whatever the writers want to put in them. The GSA search engine works well with such pages. But SL's contrived pages are not written by people in the usual way. They are contrived by LL, and people can't normally write their content. We can't write a couple of paragraphs about our places, for instance, or describe our products sufficiently well. It's a completely different kettle of fish when you don't create the actual pages. Even now, it's probable that most place owners haven't much of a clue about search, or of their parcel's pages.

Nevertheless, the All search is still the best we have in SL, and it does provide very relevant results. To suggest that it doesn't, is to suggest that Google made a crap search engine, and I don't think many people would agree with that. Some people complain when a search engine doesn't produce the results that *they* prefer, bit that's different and unjustified. They still produce very relevant results. Where the All search fails is that people cannot write their own page's content, and the system is stuck with LL's contrived and very limited efforts.

LL probably doesn't know much about how the engine works, where ranking weightings are, and so on, but they could provide some sort of tutorial concerning the parts they do know about. As it is, it works on contrived and limited pages, which are not particularly good for it. Even so, it does produce very relevant results. A search for "couples animations" or "chandeliers" will certainly produce places that have them.

You are very welcome. I'm glad to have been helpful

As was proven before, even the ads were being gamed - and frankly, it puts the small business owner - the ones just starting out - at a complete disadvantage. I'm just not sure why LL can't use the same bots that another search engine inworld does to inventory for the "for sale" items and let the relevant results stand for themselves.

Oh, you've been more helpful in this post, than words can describe, Phil.

They do, that's not the issue, the issue is that items for sale have less weighting in search than parcel titles and descriptions.

There are a myriad of problems with any search engine here, the most glaringly obvious being that you might not be searching for the same term as the creator has put on the item.

It’s been awhile since I posted on the forums, mainly just stay away from the drama.

I have been in business for 5 years in secondlife and have grown from a small store on a 512 piece of land to 4 sims (soon to be six), and the one thing that I KNOW is that EVERY customer is important to your business and its success....... and any business owner who is arrogant enough to think otherwise is only fooling themselves.

And right there, is the most relevant statement in this entire thread. Every business owner should not only try to include every word they think is relevant to their business, they should ask their customer base as well. (The "second set of eyes" theory.)

I don't take chances. If I have shoes for sale, I also have shoe for sale, because I've seen myself that the Search doesn't pick it up, despite the fact that it's supposed to. If I have a house for rent, I have house houses for rent.

And in fact, I think as soon as time permits, I'll be going in and renaming all the products in my mall that I possibly can. A black gown will become a black dark floor-length formal ballgown gown, and so on. I can't afford $L100k a week in advertising, so being as precise as I can for purposes of the GSA is the best thing I can do.

Frankly, I have completely given up on the traffic metrics for my mall. I removed the bot that invited people to my group. I dissolved one group and put it on Hippo Groups. I'm still doing the picks contest, but only half-heartedly. Either my products will get picked up by the GSA or they won't. I'm truly tired of fighting it.

ETA: And the amazing thing is that since I "gave up"....I've had more traffic in the past week than I've had in the past two months.

What you should add then, is that picks soon after that did not have the same weight anymore then when I implemented the system. And I did not loose my place. Even more, since a week I do not have a picks incentive system anymore, after rebuilding my entire store, I decided to see what happens if I do not place the boards back. Up to now it did cost me a couple of picks, but my ranking stays.

Back to relevant: In the current Search All, results come up as relevant as they can be, provided that the parcel owners did some effort in being found. Of course, some of the results do not make sense at all because of parcel owners misleading people (advertise stuff they do not have) but those are minor. You can search in more ways then ever. Not perfect, but better then old search.

Micky, I do not agree with you in this thread, and I have nu clue in how far your influence would reach. But I do not use bots (not even a picks system anymore ) so I would be happy to have a place on that list of yours. You have been in my store so you must know what I sell, though I expanded a bit lately

That last paragraph is not true. If you can't find what you want by using search, it could as well mean that you do not use search in the best way. Search gives back the most relevant results to your query, but if the query is not exact enough, neither will your results be.

When I test the quality of search, the last thing I search for is low prim. Because I know who are in the top 10 there, so I am kind of biased. But I search for things as much as everyone, and I find them most of the time. It did take me some time to understand how to use search though.

And finally: you do not have to play games with search, you have to optimize. Your parcel name and description, your item names and descriptions. No bag of tricks needed.

How the heck did you learn this? Where is all this information about the details of LL's implementation disseminated?

I teach a free class at least once a month on how to increase traffic. While I get new biz owners all the time, there are definitely a great deal more who attend who have been in biz for quite awhile and need to learn this info. I do not pay for picks, use bots, camping, lucky chairs or really any of the stuff in question and i do not advise others to simply because it isnt necessary in my experience.

I completely agree that the best source of information is your customer base. I often sit in my store creating and talk to those who come in to understand how they found my store, what they are looking for, what terms they used to search on, etc. I get the absolute BEST information from them and have tweaked my land descriptions etc based on their input and constant trial and error. I also change my offerings based on their input.

I have many loyal customers who come in over and over, bring others and that have become friends. But that is due to my efforts to interact with them personally and make them feel special and welcome. And also my efforts to make their visit to my store productive.

So while optimizing google search (all search) is effective, the MOST effective way to learn the terms to use when optimizing is to KNOW your customers...literally. I found out recently that folks could not find the sims i manage when searching for eco village. We have the words Eco-village in all our info instead. If I had not taken the time to stop a visitor, welcome them and ask "how did you find us here?" i would never have learned that and been able to make adjustments to make searching more effective.

I guess what I am saying is that none of these methods will be successful...and no store owner will be truly successful...unless they offer outstanding customer service AND exhibit true interest in their customers above and beyond any search methods, using bots, offering camping etc.

And just what are those sphere things you sit on to have your avie get animated called?

Hint: it's "*pose* ball".

Poses and animations in inventory both fall under the singular label "animation" as well and since I don't want inventory animations but rather prims with animations in them "couple poses" fits the description.

That last sentence really just sums up the entire problem. Your store isn't relevant in and by itself, you *make* it "relevant".

Once again: it wasn't about "I don't like it so it's 'crap'", it just was junk, period. Noone in their right mind would buy from those stores unless they were under the mistaken impression that that's really the best there is which search certainly seems to do.

If I could name and shame the stores I ended up in from search and contrast those with stores offering the same thing but that have actual quality then I would.

And you're really not that dense that you can't figure out that if someone searches for "chandelier" they're not really interested in a list of all the stores that sell them as much as they mean "give me a list of stores that are most likely to have a kind of chandelier I want".

Search doesn't exist as a tool for you to rub your store into people's faces, it exists to let people find what they want in an efficient and productive fashion. I could care less about whether this or that store ranks high or low, I care about getting results that are most likely to offer what I want.

Please feel free to hand out your psychic ability potion How the heck does search know what sort of chandelier you want? Come on Kitty, play fair. Search just isn't capable of doing that, you might be able to narrow down by the type you want but if you search for chandeliers, you're going to get results of places that have chandeliers.

we need to be able to set our own search filters why would i want to find a high traffic place if im just searching for clothes or shoes or hair? maybe im looking for something diffrent then what everyone else has... just let us have our own control of what to search for and stop Lindens forcing us to find what they want us to find or who pays them the most

It isn't the search engine's job to determine whether a product is crap. If the store owner optimized their products and parcels in such a way as to get highly placed, then they have done the appropriate marketing for their business. You can visit 5,000 websites through Google, and find the same problem. The products you were searching for are there - you just don't like the version of the product being offered.



Horse of a different color, and another thread entirely - Quality of workmanship.



There isn't a search engine in existence that can determine exactly what it is you're looking for, unless you specifically type in "brass antique chandelier with offwhite globes"...and then the search engine can only find it if the manufacturer has listed it that way.

Of course, even if he has listed it that way - it's still no guarantee that you're going to like the workmanship.

How about a search comparison then?

I'll tell you what I want and you can use search to find it and see how productive search is for something I want to use it for.

Then you can tell me what you want and I'll use search to find and see how productive that turns out to be.

You and I are very likely going to have a wildly different range of things we search for and until you actually compare experiences for the same thing and keyword it's just bickering back and forth about whether your/my experience is representative or not.

You could be searching for a kangaroo (random example where I'm guessing only half a dozen stores actually sell one) while I'd looking for a stuffed teddy to put on a chair where I'll have 500 different results to muddle through. Both are equally specific but your search is likely to be far more productive than mine.

To search more effectively then it is necessary to add in the "kind of" chandelier you want into the words you search on. However, it is just as true that if the creators calls the exact chandelier you want a name like "crystal perfection" that you will not ever find that particular item in search. I coach all of the folks I help with biz to make sure they dont call a photo just a name like "enlightenment" since in search that wont direct anyone to the photo. i tell them to call it Photo - "enlightenment" so that the search will pick it up.

Ultimately effective search comes about through both the search-er AND the creator using the the same words to describe the same item. That is just the nature of the beast.

Well... imagine we really were given the job of coming up with an in-world search engine that could do such a thing. I think that would be very possible given the wealth of information that SL has about buying histories. I mean, Amazon does it--and with spooky precision.

It really wouldn't be all that difficult to use statistical buying patterns to create user-tailored search results. I don't know that this would be a good idea, but the apparatus wouldn't need to include a crystal ball.

People don't want to pay for Second Life, yet LL is supposed to cough up for this kind of technology??

Good point smartarse However Amazon relies on a hell of a lot of number crunching and I'm not sure the infrastructure here is up to it at this time but I certainly wouldn't rule out such a feature.

However Amazon is a website purchasing utility, the closest similarities around these parts are SLExchange..I mean XStreetSL and Onrez. Certainly SLX has the customers who bought X also bought Y feature, but that's a different beast to a google type search we're discussing here.

If you want to use the term "play fair" you might want to start by not twisting what I actually said.

I never said search should be psychic and return THE store where I'm going to find exactly what I want. I said search should produce results "that are *most likely* to have it".

If a large portion of people find "chandeliers" in a select group of say 20 stores then guess what? I'm most likely to find something I like by visiting those 20 stores as well.

Maybe there are 480 other stores that sell them too, and maybe one of those has THE chandelier I'm truly after but the 20 that a large portion of people are happy buying theirs from would be the best place to start looking.

What I expect from search is to yield results that are most likely to have what I want and going by popularity works perfectly fine for that just as long as manipulation doesn't get out of control the way it has with both the old and new search.

Simply by reading the forum, and learning from my own experience. Indeed when I started the Picks system, it helped me a lot to get in the top 10 for my keywords. After that, I started optimizing the other parameters. I already saw what was happening, stores were outbidding each other and I did not want to participate in that.
After that I saw stores with much more picks then I have, lower in search. It looked like inbound links from picks were getting less important. About that time, other people starting posting the same experience. There is, however, no information about what parameter counts for how much. That is something you can only learn from testing and tuning. And, reading Google's own information.

I think the overriding point is that Google isn't, and wasn't, designed to be the sole search engine for a closed content delivery platform. That's why other content platform sites, like Flickr and Youtube, don't typically use GSA based search.

Google was designed for an environment of many linked sites which connect to each other. Even that is inappropriate to Second Life, because land parcels can't link to each other. Picks of individuals are a bit of a fudge, since they can never be anything more than personal lists with no references to them - in other words, the Page Ranking algorithm doesn't work on SL because all links are in a fixed structure. Moreover, Google itself has been criticised for causing stagnation of web results.

Any argument along the lines of "anyone can be on the front page if they work hard enough" is kind of bogus because if everyone works hard, there are still only so many slots on the front page, and it isn't desirable for the number of content providers for any given field in SL to be restricted to the number of slots on a page of Search. I believe there is some research underway on non-textual (or at least non-paginated) ways to display search results in order to address this problem for the web as a whole (also, for example, to overcome the problem that it's impossible to represent two sites of equal relevance - if you use a list, one must be listed first).

Normally I would say yes to this challenge, but from your other postings I learned that is is not of much use.

Somehow you think that the search engine should anticipate on what your tastes are, and in that case you are right: search sucks. Search can never know what you taste is, it can only give you back from what you put in. And there is no input field for taste.

But it is not fair to use this as an argument why Search is not working right and only being gamed. Again: every business owner has to optimize his business for search. Nothing to do with gaming. And you will have to work yourself through the results. Some of which you might like, some of which you may not like.