Faster credit processing... but higher fees

[Perrie Juran]

---

Qie Niangao wrote:

---

Rya Nitely wrote:

Well, as Qie pointed out, there's $60 million per annum not cashed out, so it must be used for something. Maybe it is land costs.

---

I thought about this some more, and maybe some of the extra Lindex trades are by L$ currency speculators. It seems impossible to me given how stable the Lindex is, but apparently some folks have a way to beat the spread, sometimes. Well, there are people who

claim

to do that, though again, I've never understood

how

.

---

Gah, you're gonna make my head hurt cause I'm not in a math mood today.

Speculators?  I thought the math on the Lindex at the old rates made it almost impossible to make money speculating on L$.

[Rya Nitely]

---

Darrius Gothly wrote:

 

I do agree though, without being able to peek over LL's shoulder at the actual financials we will not be able to know for sure. But as it stands, based on the information I've gleaned from friends and cohorts, I'm pretty confident that I'm seeing a pretty reliable sample set and reaching very solid conclusions.

But .. y'know .. sometimes I do have my head up my ... (Hello? Anybody in here?!?)

---

Here I am again watching the LindeX at 4 PM SLT, which is a more sensible time, and now I see even larger orders going through than very early this morning - 400K, 600K, 700K within the past hour and these are all one order at a time, because I watch them complete, and again nothing pops up for a while until the next big order. These are individual sellers - not the accumulation of many small orders.

So, I at this point I conclude that if all the small sellers are squeezing themselves into 9 PM to 12 AM Sunday night, then the rest of the week is being taken by bigger sellers. The bigger orders are the majority. 

I think watching the LindeX is a much stronger and reliable source than asking your friends. We all have different circles or no circle at all. These big cash out people probably don't have a circle. They are here to make money and that's it.

 

Maybe I should get a 3rd life because obviously two are not doing it for me :matte-motes-tongue:

[Carlo Eriksen]

 

"Yet what is any ocean but a multitude of drops?"
D. Mitchell

[KarenMichelle Lane]

---

Phil Deakins wrote:

It was just a thought

---

“Sometimes the fluffy bunny of incredulity zooms around the bend so rapidly that the greyhound of language is left, agog, in the starting cage.”

― David Mitchell, Cloud Atlas    

[Alwin Alcott]

---

Rya Nitely wrote:

and now I see even larger orders going through than very early this morning - 400K, 600K, 700K within the past hour

and these are all one order at a time, because I watch them complete, and again nothing pops up for a while until the next big order.

These are individual sellers - not the accumulation of many small orders.:

---

just a question:  how / where can you see it are individual orders and completing?

the only buy and sell i can see are totals, not individuals, perhaps i look at the wrong place?

[Darrius Gothly]

I do it by refreshing the page every minute or so .. sometimes more often .. then mentally comparing the total available for sale to the number shown on the previous refresh.

This method presumes that only one (or maybe two) chunks of L$ have been put up for sale during the time between refreshes. But if you do it often enough, the chances of LOTS of little sales is minimized. It's more likely you will catch only one chunk placed for sale on each refresh.

[Darrius Gothly]

I'm not sure what to say .. other than "I swear I've seen it."

There may be lots of reasons neither you nor others are spotting this happening, but I'm at a loss to explain them all. I can only say that it's behavior I have witnessed many times. Backed up by conversations with my "circle", I am pretty confident in my estimation of the cashout distributions. Granted I don't have many "Big Fish" in my circle, but it only takes one or two to more than overmatch the total of all the little fish.

I think this is going to go down in the annals of "questions LL will never answer" .. which is really the biggest shame of all. They have their own reasons for completely hobbling the financial market, and the merchandise market, and most every other market in SL. They also have their reasons for obscuring or hiding every other datum that could be of benefit to anyone trying to improve their business. Those they don't obscure, they flat out destroy. (In-world search for example.)

To me the final bottom line is .. I will never understand why LL has now and has always done everything in their power to destroy any user-income from their platform. This is just another example of them shaking down and crippling one of the most obvious and vital sources of revenue growth stemming from Second Life.

[Phil Deakins]

---

Darrius Gothly wrote:

Those they don't obscure, they flat out destroy. (In-world search for example.)

---

They destroyed the inworld search? I never use it for objects and such. What's wrong with it?

 

[Alwin Alcott]

---

Darrius Gothly wrote:

I do it by refreshing the page every minute or so .. sometimes more often .. then mentally comparing the total available for sale to the number shown on the previous refresh.

---

ehm... i don't think this shows the movements per selling order, it shows the buying

x adds 500, y adds 2000 and z 5000 = 7500 if one buys 7250 only 250 remains, and will show the total 7250 going down, but there are 2 orders filled and one partly.

Thats why i asked where to see the orders getting filled, because it's impossible to see on the buy/sell page

[Darrius Gothly]

---

Phil Deakins wrote:

---

Darrius Gothly wrote:

Those they don't obscure, they flat out destroy. (In-world search for example.)

---

They destroyed the inworld search? I never use it for objects and such. What's wrong with it?

 

---

For a time (up through about 2012-2013) it was possible to use keywording techniques to promote your listing in Search. I'm not talking spamming though, I'm talking proper use of keywords and object naming. They were using a Google Search Engine to index the listing pages, but asking it to do more than feasible using the incredibly sparse content on those pages. But then they chose to stop paying the service fees on the GSE and instead switched to an in-house developed search appliance that uses "secret techniques" to rank parcel pages.

In the process of trying to make the GSE work and then switching to the in-house system, they were assaulted by a LOT of pressure from big names to "get rid of the spammers". In essence what was requested was "get rid of my competition". The end result was that they tweaked indexing parameters enough to destroy the actual relevance of items for sale, overweighted words found in the Parcel Name and Description .. and then added a foot on the scale so that they can arbitrarily rank any parcel they wish much higher than any other. They also imposed some very arbitrary "rules" that totally eliminate smaller shops and overly weight those with massive amounts of entries.

The end result was that only those with a finger in LL's ear can easily rank; those just trying to make a living or that are part of the herd don't even show up ... and the whole process became yet another skewed game controlled by unseen hands for secret purposes. What was a useful tool, able to be used as a good advertising and marketing tool, was reduced to yet another worthless waste of time and computer cycles.

Like the scroller ads on Marketplace that show only the top names and sometimes sprinkle in ads bought and paid for by the many other Merchants, In-World Search is a showcase for the chosen few and has no useful purpose. As evidenced by how few use it to find anything any more.

As you well know Phil, Search can be a very useful tool to find relevant things and places. But since they've got it working the way they want, it really isn't useful for finding new things; it only works to aim you at things they want you to see.

[Darrius Gothly]

I don't pay attention to decreases, only increases. Decreases happen in blocks not related to anything we can measure. But increases happen when someone places a block for sale, so it represents one sale.

[MarmelaGramela Doesburg]

I do not see the logic in it. Paypal causes LL cost? Paypal at best charges the recipient fess for receiving money, but not the sender. So can anyone explain these "costs" to me?

Actually this was a genral post, not a reply.

[Phil Deakins]

Interesting. I knew the GSA, of course. I was the one who taught the forum how to rank well in it. That was back in days of yore. I even controlled 7 of the top 10 rankings for my main searchterm - almost all of them not even mine. And I knew when they ditched it favour of another one, but it wasn't an in-house one. It was a freebie, open source one, that was undoubtedly written to be pretty much the way that Google ranks pages - which they published a long time ago. All the main engines copied it after Google took off.

I haven't looked at search in many years now, so I just did a search on my main searchterm (low prim furniture) and I was surprised to see that I'm still ranking #1 - ahead of Loks, who tried and tried to get past me years ago, before eventually giving up. She IMed me to tell me that So, at first glance, I don't see how it has been destroyed, or how it favours those you described. The store (land) is just a fraction of the size it was, so it's nothing to do with that. And I'm certainly not favoured in any way by LL.

There's another little bit too. Yesterday, a strange thing happened in my workshop, which is about 3500 meters in the sky. The store is directly beneath it and not far off the ground. I was in the workshop but looking at an external programme, so I didn't notice a girl arriving. When I looked back at SL, I saw her on a bed trying the animations. We chatted a couple of minutes and it seems she really was looking for a bed with a particular type of animation in it. Of course, I didn't swallow that because the workshop is just too far off the beaten path. But it made me wonder if there was something there that was set to show in search. It turned out there was, and it had only been there a day, or 2 days at the most. It was just an armchair. But it may have been the cause of her arriving in my workshop.

I still wasn't convinced, but I checked the visitor log later, and another girl had come to the workshop. It just doesn't happen that anyone arrives there, let alone 2 in quick succession, so I'm inclined to think that the chair in search was the cause. It did have 'low prim furniture' in its description. And if that's so, I can't really see that the inworld search has been "destroyed".

 

[Darrius Gothly]

GSA! Thank you. Appliance not Engine. (duh!)

I'm not sure where you're searching, but I don't see your store listed. I do see LOK's as the top ranked place though. Your store must be listed under a different owner name. (btw: I see her store ranked #1 as a Region too) But the situation you are describing doesn't begin to reveal anything about Search in total. All that happened seems to be one item drew the attention of two people.

Spend some time plugging in different search terms that casual shoppers or searchers might use, then look at the results returned. Study the places and their listings to see if you can understand why they have been ranked highly for those search terms. I'm fairly confident that in short order you'll start to see how irrational the results seem to be.

There are terms that makes sense, but a vast number of others that don't. It's that focus on the high end and total disregard for the full picture that makes the tool mostly useless. If you want shoes, land or low prim furniture (in your case) then it will send you to places with top traffic. But anything else?

(The Places tab of Search is purely Traffic ranked and has no bearing on anything. Just make sure you have the search term once then load your land with Bots and you win there.)

However, I gave up on search a few years back. I stopped caring when they added their own internal weighting and destroyed its usefulness. It may have improved some over the years but .. I don't see it being as useful as it was or could be.

[Rya Nitely]

---

Alwin Alcott wrote:

---

Thats why i asked where to see the orders getting filled, because it's impossible to see on the buy/sell page

---

It's not impossible. It just takes patience and many refreshes. I only look at sell orders at 249. Very often there is nothing, so it's easy to see a new order come in, and then with each refresh you see it get sold in small portions. I always watch my order go through this way, so that I can process credit as soon as it completes.

Buy orders appear to be many small amounts and are a constant flow, but the sell orders are usually large blocks. Now, if sell orders were made up of many small amounts then I would expect the same constant flow.

[Rya Nitely]

---

Darrius Gothly wrote:

I'm not sure what to say .. other than "I swear I've seen it."

There may be lots of reasons neither you nor others are spotting this happening, but I'm at a loss to explain them all. I can only say that it's behavior I have witnessed many times.

---

I will concede that small sell orders below 15K may get placed and go through so quickly that I miss it between refreshes.

[Phil Deakins]

I simply did an inworld search for 'low prim furniture' (without quotes). I had it set to Places and no restriction on Adult. I'm ranked #1. Lok's is ranked #2. Lok's listing that is above mine isn't a ranking. It's a placement, so it doesn't count as a ranking and there's nothing we can do about it. Btw, it's not a placement for any of the reasons you mentioned. They placed exact name regions at the top when they used the GSA. It's the equivalent of search engines like Google placing advertisements above the rankings. In the past, you couldn't tell the difference between them with some engines. GoTo (it became Overture later) actually placed paid ads within the rankings, and there was no way to spot them. That caused a change in  U.S. law, and it became illegal to do it.

The point I was making with those 2 visitors was that, in a very short space of time, an item was listed so that it caused people to come to the workshop. I don't know the mechanism behind it but it almost certainly happened because of inadvertantly leaving a chair set to show in search for a very short space of time. (I haven't totally ruled out the possibility of it being someone who knows me, but the chair is by far the most likely reason, imo).

---

Darrius Gothly wrote:

There are terms that makes sense, but a vast number of others that don't. It's that focus on the high end and total disregard for the full picture that makes the tool mostly useless.

If you want shoes, land or low prim furniture (in your case) then it will send you to places with top traffic. But anything else?

(The

Places tab

of Search is purely Traffic ranked and has no bearing on anything. Just make sure you have the search term once then load your land with Bots and you win there.)

However, I gave up on search a few years back. I stopped caring when they added their own internal weighting and destroyed its usefulness. It may have improved some over the years but .. I don't see it being as useful as it was or could be.

---

It doesn't do that. My top ranking is above others that have much higher traffic than I get. One place, ranked 2 down from me, has over 3 times the traffic that I have. And there are more higher traffic places below that.

Incidentally, there is no Places tab in search. There is in the old search that can still be used with some 3rd party viewers, and that's always been listed solely on traffic. Could it be that your mind is getting things a bit mixed up?

I might suggest that that is the reason for your pessimistic views. Perhaps your interest faded too quickly, before they got the new thing working as they wanted it to work. I think you may have become a bit out of touch, Darrius. I certainly have. It's been years since I looked at search in the detailed way that I used to, but I see nothing wrong with the results that I saw yesterday and today. For that reason, and because I have no real interest in it, I'm not going to study it in the depth you suggested. I was simply curious about your statement that LL had destroyed it, and I wondered in what way.

[Darrius Gothly]

---

Phil Deakins wrote:

I simply did an inworld search for 'low prim furniture' (without quotes). I had it set to Places and no restriction on Adult. I'm ranked #1. Lok's is ranked #2. Lok's listing that is above mine isn't a ranking. It's a placement, so it doesn't count as a ranking and there's nothing we can do about it. Btw, it's not a placement for any of the reasons you mentioned. They placed exact name regions at the top when they used the GSA. It's the equivalent of search engines like Google placing advertisements above the rankings. In the past, you couldn't tell the difference between them with some engines. GoTo (it became Overture later) actually placed paid ads

within

the rankings, and there was no way to spot them. That caused a change in  U.S. law, and it became illegal to do it.

The point I was making with those 2 visitors was that, in a

very

short space of time, an item was listed so that it caused people to come to the workshop. I don't know the mechanism behind it but it almost certainly happened because of inadvertantly leaving a chair set to show in search for a very short space of time. (I haven't totally ruled out the possibility of it being someone who knows me, but the chair is by far the most likely reason, imo).

---

Darrius Gothly wrote:

There are terms that makes sense, but a vast number of others that don't. It's that focus on the high end and total disregard for the full picture that makes the tool mostly useless.

If you want shoes, land or low prim furniture (in your case) then it will send you to places with top traffic. But anything else?

(The

Places tab

of Search is purely Traffic ranked and has no bearing on anything. Just make sure you have the search term once then load your land with Bots and you win there.)

However, I gave up on search a few years back. I stopped caring when they added their own internal weighting and destroyed its usefulness. It may have improved some over the years but .. I don't see it being as useful as it was or could be.

---

It doesn't do that. My top ranking is above others that have much higher traffic than I get. One place, ranked 2 down from me, has over 3 times the traffic that I have. And there are more higher traffic places below that.

Incidentally, there is no

Places tab

in search. There is in the old search that can still be used with some 3rd party viewers, and that's always been listed solely on traffic. Could it be that your mind is getting things a bit mixed up?

I might suggest that that is the reason for your pessimistic views. Perhaps your interest faded too quickly, before they got the new thing working as they wanted it to work. I think you may have become a bit out of touch, Darrius. I certainly have. It's been years since I looked at search in the detailed way that I used to, but I see nothing wrong with the results that I saw yesterday and today. For that reason, and because I have no real interest in it, I'm not going to study it in the depth you suggested. I was simply curious about your statement that LL had destroyed it, and I wondered in what way.

---

I'll start with your last paragraph. After you gave up and quit watching the SEO activity, I was still in the thick of things. I was active in it for quite some time and saw everything going on. I was involved daily with many clients working around the clock to keep them ranked well. I gave up after I saw the writing on the wall. I'd spent months working hard to make good stores stay in search, only to watch "Big Names" with near zero keywords or traffic fly to the top of Search and stay there.

I was the one that attended Community Meetings with Jack Linden, and I was the one that confronted him in one meeting with hard evidence of their "tinkering" .. only to have him lie to my face about it. I didn't quit too early, I stayed through the whole process until LL made it abundantly clear they had no interest in being fair, only in helping out specific people.

Now about the "Traffic Only" ranking in places. You are correct, I was looking at the old style Places search as shown here:

Hardly anyone uses that .. unless they see the "Places" tab up top before they see the one along the side. I use Firestorm exclusively and never run the LL Viewer, so I'm biased in that respect. But nevertheless, that's the tab I was referring to when I made my earlier comment.

Now the "New" (or "Web") Search: What you refer to as a "Placement" is to the untrained eye the same as a regular listing. Here's what I see and I believe you see from the Everything search:



The first three entries are a Region and two Groups. Untrained eyes will see those as #1 through #3 in results. They will be able to TP to the Region listing but won't get far with the Groups .. at least not for going somewhere. I still don't understand why they place Groups at the top, but I'm sure most people that click on those entries either get fed up and close Search or understand what's going on so it probably doesn't do too much harm. (Although it would seem more logical that they should be ranked within Everything appropriately and not pushed to the top as they are now.)

Directly below those is what is called a Placement as you pointed out. That is for LOK's. Even though it has a different icon, it provides a TP function and is every bit as useful as a normal Parcel listing. Below the Placement is your listing. So as it stands now, you and I both see the same thing. And what I see now is not what I saw earlier today. So as it stands, my earlier observations are no longer correct.

When I looked earlier (but did not screencap darnit) you were not in the listings. LOK's Placement was immediately followed by two other listings and then her Parcel listing. That is what I saw and why I made my comments. But it has changed.

The New Places tab shows something a whole lot different .. and here is where the power of the Placement works well:

 



LOK's Placement comes ahead of your normal Listing. But they are identical in appearance save for some very tiny differences. Those differences won't matter to most people .. if they even notice them. The upshot is the Placement puts her shop "ahead" of yours for anyone using the New Places search.

Now .. back to why you are ranked first in New Everything:

We can only get real gut-level details by doing a manual "Client Search". The previous ability to see what items were marked to "Show In Search" is no longer available from the Web Search. So how those two visitors came to be in your shop instead of your store is beyond me. However the Client Search does allow you to open the Parcel's web page. You and LOK have entries that link to your Listing pages as follows: (and btw her listing comes before yours in Client Search)

 

 

Furniture at LOK'S LOW PRIM FURNITURE

 

Low Prim Furniture

These two links open the Parcel Listing Pages for you both. They allow you to see what items are marked to Show in Search. You have 21 items listed, she has one. Directly below you is a Parcel with four items listed. So clearly the content of your Parcel, the items marked to Show in Search, are not very heavily weighted. You've also pointed out that Traffic doesn't really count much either. So they've eliminated (for the most part) Items for Sale and Traffic. That leaves only Parcel Name and Description for ranking.

Here is where the "Foot on the scale" comes into play. During my final days of caring about Search, I conducted a number of tests using the exact same Name and Description on multiple Parcels. I found that my results would wind up on pages 10+ or further down. Even duplicating the Items listed and besting Traffic only resulted in moving up a few slots. But the slots on Page 1? Always stayed the same small group and never varied. In short there was no way to advance to the first page unless you were already there.

Final words: I didn't become cynical too soon. I stayed at it and flogged at it for months after they stopped listening and stopped communicating. I finally came to the conclusion that they had "frozen" rankings as they were and thus were not allowing the system to work. In other words, they killed it as a Search Tool and turned it into a Listings Service for the chosen few. Who those few are and why they were picked .. and more to the point are they ever switched up, I cannot say. But in my months of testing I never saw things vary. No one new had a chance .. and that just wasn't fair in the least.

And THAT is when I quit, stopped taking consulting work, pulled my SEO devices off the market and stopped blogging about Search. It wasn't fair to anyone to spend money when it was all too obvious that nothing could be done to realistically get them good placement.

Now to bring this full circle and back on topic for this thread: Search was a useful tool and needed only some careful engineering to make it really work properly. Instead LL fired the GSA team, went to an in-house system and completely buried any information that made it valuable to the entire platform. They have done the same thing with Resident to Resident transaction information, Marketplace Advertisements, page views on Marketplace and now .. they are "adjusting" Cashout volumes and fees.

It all comes back to the fact that LL does more to destroy income and customer participation than any Griefer or Hacker could. Rather than look at the big picture and make decisions based on the bigger benefit, they almost without exception take a path that results in alienating a majority, hiding basic needed information .. and ultimately harming more than should be harmed. So yeah, I'm cynical .. and remain that way any time I hear "News" from the Lab.

[Phil Deakins]

Phew! That's some post to reply to, Darrius. I'll do it in chronological order, as it'll be easier on my brain

 

1st and 2nd paragrpahs:

Jack had gone before they stopped using the GSA. I was the one who announced in the forum that he'd been sacked before it became public knowledge. I was very quick off the mark with the information (on the very day he got the news) because of a friend in the right place. It was a mutual friend of ours So those meetings with Jack were during the GSA period and not during the period we are discussing. At least I'm not .  I didn't see anything of what you described during the GSA period, and I did take a great interest in how to achieve top rankings. I even played with it to the extent of controlling 7 of the top 10 rankings - as I mentioned earlier.

 

3rd paragraph (Places):

Since we're discussing the current search, the old search system doesn't come into it.

 

Placement vs Rankings:

Yes, I know that the top placement looks (almost but not quite) the same as the rankings, but that's not the point I was making. We can't do anything to get above the placements, so I don't count those as rankings when I say that I am ranked at #1. I didn't say I was listed at #1. I said, and meant, ranked.

 

The rest:

If you do a search in the website it's the inworld search. The difference between doing it in the website and in the viewer is that you can get to the parcel pages from the website results but not from the viewer results. I've no idea how you got to the page you linked to with "Client Search" as the link text, but when I do a website search I get the results in the normal format, just like the search that's done in the viewer. The one difference between them is that the "More info" button in the website search takes you to the full parcel page, whereas in the viewer it doesn't.

Your statement and 'evidence' that page content is pretty much eliminated doesn't hold up. At first glance, it may appear that way but search engines are not like you suggested. In the 90s, each major search engine had what it thought of as an 'ideal page', and the closer a webpage got to the ideal page, the higher it ranked. An ideal page might have had nnn number of words on it, with nn number of the searchterm in it, and the searchterm appearing in particular sections of it, and so on. A simpler engine might rank simply according to the percentage of the words on the page that match the words in the searchterm. What I'm trying to say is that quantity in a page never did decide rankings in any reasonable search engine. So I'm saying that your statement that "that leaves only Parcel Name and Description for ranking" is highly unlikely to be correct. If your tests were only about adjusting the parcel names and descriptions, and didn't also adjust the quantity of body text, and searchterms in it, not to mention the H tag text, and so on, then you missed a lot, and your results were skewed by it.

Remember that my RL job (self-employed) before coming to SL was seo. I was well-known enough in that world at that time that Microsoft invited me, along with 11 others from around the world, on an all expenses paid trip to their headquarters in Seattle, to help them with our expertise when they were creating their own search engine. I'm not just guessing about search engines here. It was my profession, and you garner a lot of knowledge on a subject when it's your profession.

Now I'm not insisting that LL doesn't favour certain people or places. I can't say that. What I am saying is that I have seen no evidence of it. I am also saying that, from my limited view of it yesterday and today, it looks to me as though the engine is working well, and is not "destroyed".

 

About those 2 visitors:

I had a longish chat with Qie today, who took an interest in that particular 'mystery', as he put it. The only realistic conclusion that we came to is that they came to my workship from a GO link in the parcel's webpage that they reached via the website search, and the GO link could only have been there because I had rezzed an armchair that was set to show in search the day before they came - 2 days at the most. At first I didn't quite see the mechanism behind them doing it, but now it seems to be clear. They did an inworld search in the website, looking for something that I sell, and got to my parcel's page, which has plenty of GO links in it. As chance would have it, or maybe they were interested in arnchairs, they clicked the only GO link that brought them up to my workshop.

There are other possibilites, but none that fit everything. If it was a person using 2 avatars, then they said an did nothing to make their venture worth doing. If it was a spying competitor, the second one only stayed a few seconds, which doesn't match spying. The competitor idea is outlandish. The individual person is more acceptable, but to what end? Nobody has teased me about it, and nobody has got on my back. I honestly believe they got there via search, as the first one suggested when she said she was looking for something in particular. I wasn't there when the second one came, so she couldn't say anything to me. And, of course, they only came during the very brief period when the chair was rezzed. It all fits with the search method. However, it doesn't provide any evidence that either search works well or is destroyed, because the chair would only be listed on an already high ranking page. It only provides evidence that the crawling and indexing is working fine, so me mentioning it was a bit of a red herring. Sorry.

 

[Darrius Gothly]

1&2) My memory was faulty, My confrontation with Jack Linden was October 21st, 2010 .. long before they dumped the GSA. I stand corrected.

3) Old search is moot. I said that too.

PvR) I don't count Placements as Rankings either. My point was that to the USER they are Listings and just as useful. So trying to beat them is moot but they still figure into the user experience/response.

Client Search is the automated search API used by devices. AFAIK It uses the same index and data as Web Search (which is also Viewer Search) but it presents it in an older, more "edible" format. As I said, it's been a few years since I studied it or SEO. If they have disconnected it from the real search results then they did so only to hobble use by responsible Merchants and to cripple useful devices, not to benefit anyone. But near as I can tell, even today, it still returns useful and valid results.

Visitors) The indexer does run more often and seems to run on changes to Traffic or Parcel pages. That change came as they also changed the format of the human-readable Parcel pages to remove a lot of information, including the page generation time. That killed any chance of automatically knowing when search results were updated.

Because it runs not on a schedule but "as needed", it is abundantly clear why that chair brought two visitors. People like going to "new" places. An item location that is not where everything else is would draw people right away. "Chair in shop? New! Lemme see!"

Removal of the Parcel listings page from the Viewer was another dumb move making it necessary to go to the web to see what's REALLY on a Parcel. That's akin to taking the excerpts out of Google; it makes each headline listing in the Viewer essentially worthless and does NOT serve the end-user in the least. Another useless hobble meant to obscure and not reveal or help.

Finally: You've looked at the windshield wipers and decided the car works just fine. When you spend some quality time studying the results deeper then I'll be more willing to accept your opinion.

Where you really stepped on my toes was your assertion that I gave up too soon. Long after you quit and stopped posting on the Forums or anywhere else, I was still out there daily helping folks try and get decent placement. I spent many many hours every day, uncompensated in any way, doing my best to do what I could. And at every turn, LL played hob behind the scenes and undid anything accomplished.

When they finally just stopped allowing the indexer to properly rank a page or a Parcel and just locked things into place, that was the last straw. You had long since stopped watching or helping anyone, so you have no place to speak with authority on how things wound up. You left, I stayed.

Perhaps LL has changed things since then, I don't know either. But you telling me I gave up too soon? No, that's an insult to me, my professionalism and character .. and I won't take that. Once you go talk to the people that I spent endless hours working with and for .. then you can pass judgement on me. Until then .. talk to the hand.

 

[Phil Deakins]

The odd things is that I remember writing that you may have given up to soon, but I can't find it. Wherever it is, I only meant that maybe you didn't wait long enough after LL got the new engine to see how it all worked out in time. It would have taken them long enough to get everything how they wanted it to be. I certainly didn't imagine that it would step on your toes, and I'm sorry that it did. I apologise.

Yes, I agree that I've only looked at the windshield wipers - but they do work very well, and they give me some confidence in the whole car. If I find that the headlights also work well, then it's a slam dunk, of course

Yes, to the user, a placement is top of the 'rankings'. But I only said that I rank at #1

It's interesting that the index is updated only as needed. It does make good sense and I'm starting to think that I have a very vague recollection of it from back in the day. The idea might have been around while the GSA was running, and mentioned in private. It does make a lot of sense though as far as spidering and indexing time is concerned.

---

Darrius Gothly wrote:

Because it runs not on a schedule but "as needed", it is abundantly clear why that chair brought two visitors. People like going to "new" places. An item location that is not where everything else is would draw people right away. "Chair in shop? New! Lemme see!" 

---

You seem to be saying that places only stay in the results if they have visitors to them. I wonder. However, it's not the case with the 2 visitors, because I know that people come to the store from search, and my store has visitors every day so, if places are dropped because they don't have visitors, my store isn't one of them.

 

---

When they finally just stopped allowing the indexer to properly rank a page or a Parcel and just locked things into place, that was the last straw.

You had long since stopped watching or helping anyone, so you have no place to speak with authority on how things wound up.

You left, I stayed.

Perhaps LL has changed things since then, I don't know either. 

---

Maybe, just maybe, they took that as an interim measure to ensure that relevant places were seen high in the rankings, while they sorted the new engine out. I didn't see it, as you pointed out, and, also as you pointed out, you haven't looked for some time either.

I haven't spoken with authority here. I simply asked a question, and, apart from clearing up a few facts, such as Jack leaving while the GSA was still the engine, my replies haven't been in any way definitive. They have been discussive. For instance, I said that I can't know whether or not LL fixed favoured results, so I haven't stated that you have been wrong about that. I've discussed it, using my windscreen wiper example

ETA: I've found it. I did not assert that you gave up too soon. I actually said, "Perhaps your interest faded too quickly, before they got the new thing working as they wanted it to work". Big difference. I didn't assert anything and 'perhaps' has a very significant meaning

[Phil Deakins]

---

Darrius Gothly wrote:

1&2)

My memory was faulty, My confrontation with

Jack Linden was October 21st, 2010

.. long before they dumped the GSA. I stand corrected. 

---

I just read that blog and it seems you were right about believing that LL purposely favoured certain places in the rankings while the GSA was still the engine. Specifically you mentioned sim owners. I don't know what change triggered the outburst. Maybe it was when they first placed exact match regions at the top.

What I can say is that, in my opinion, you were mistaken when you called Jack a liar. You quoted him twice and judged each quote to be a lie. Imo, they were not. He said that he could not discuss the finer details of how search ranks. There's nothing wrong with that. No public search engine does it.

You then quoted him as saying that "No search engine provider would talk abut the details of how relevance is calulated", and you judged that to be a lie too. In my experience, it wasn't. You went on to say that Google does it. You mentioned pages and pages of it, but they don't do it. You may have been thinking of the papers that Google's founders published when they created their search engine at Stanford University. The engine was called Backrub, not Google. Those papers did indeed divulge some of what you said, especially the way that PageRank worked, together with its actual formula, but it wasn't a public search engine that website owners could submit their sites to, or want to be ranked in, so it doesn't count. It was just university course work. Since Google became a public engine, they have offered us some information to help us rank our pages better, but they have never told us how the Google engine calculates relevance. No search engine would do that.

In all probability, Jack didn't even know the details you were asking for. He was the search team's manager, among his other jobs, and how the programme ranked pages isn't something he would necessarily know. He couldn't know because it was the GSA, and LL didn't have access to it at that level.

[Prokofy Neva]

If they're already charging to *convert* the Linden dollars to real US dollars, then why clip us *again* on the cashout?

The reason is: because they can, and they need to raise revenue.

Keeping in mind that you will pay Paypal a fee as well to get this money - er license to view virtual content - into your bank account, it starts to be really eroded.

Of course, if you cash out in order to pay tier, mainly, then you don't have to pay those fees, after paying the exchange fees.

The Lindens always mention these figures of $45 million or is it now even $60 million that residents "earn" in Second Life. But I'd love to know what percentage of that is going to tier payments, and what's left out. They never tell.

[Rya Nitely]

---

Prokofy Neva wrote:

 

The Lindens always mention these figures of $45 million or is it now even $60 million that residents "earn" in Second Life. But I'd love to know what percentage of that is going to tier payments, and what's left out. They never tell.

---

To quote Qie in previous post

'Appears to me from the Lindex data that volumes now average about L$85,000,000 a day. That would be about US$ 124 million a year, or about twice the $60m / yr cash-out volume Darrius cited. So it's not "the vast majority" -- it's about half and half.]'

When LL use the word 'redeemed' they must mean taking money out of SL. Otherwise it's useless information if it includes money going back into SL.

 LL didn't need to tell us about the $60 million. It's quite surprising that they did. It's an Ebbe thing as he mentioned it at that virtual world meeting as well. I think he's very proud of it. So proud that he wants some.

[Rya Nitely]

---

Darrius Gothly wrote:

Being someone that analyzes processes for optimal outcome, I learned that the very best time to cash out was just before midnight SLT on Sunday night. That way I would almost always receive my money on or before Friday of that week. Cashing out Monday would often result in receiving my money Monday of the following week .. a full seven calendar days. But beating the midnight deadline Sunday meant that I generally waited no more than 5 calendar days.

My monitoring of the LindeX would thus occur starting around 9PM SLT Sunday night and continue for several hours. I would choose a time near midnight when I saw the L$ amount for sale one point above LL's "wall" was low enough that I could ensure a full sale before the cutoff time.

Watching the LindeX right now, I see no activity similar to what I have witnessed in the 9PM to Midnight window. So based on this time frame, your observations are correct. However the absence of the patterns I have observed late Sunday night does not mean they are not there; it simply means they are not there now.

 

---

You're right. The LindeX has been very active, and it does look like lots of small orders totalling 900,000 to 1 million at 249. There is buying activity but the amount is not coming down.  It means that many people are selling small amounts late on a Sunday night to get it in 5 days.

Note to self: Never place a sell order late on a Sunday night SLT time if you don't want to wait for it get filled.

Although this behaviour should change somewhat in April.

Edit: There are orders at 250 going in now to beat the queue. I mean really, all to get the money in 5 days?