Flea Yatsenko

well that's a good sign. I was pretty discouraged when I went to see this thread was made in 2020 and the problem still persists. But they must be aware of it and about to do something. Optimistically maybe they've been having problems because they don't have the hardware to properly sort, index, etc all the listings that exist.

All SL web properties throw random 500 errors.

The uploader is also not very useful for LODing. Depending on object size, some might never see medium LOD before it drops out of view distance and some might make lowest LOD at 5m. You can either 1. Just do the high LOD and set everything besides high to 0 in the uploader 2. Create LODs for everything when some won't even be used which can increase LI There really needs to be a full preview you can walk around and test different view distances on in the mesh preview. And something to indicate color wise when the LOD changes. SL doesn't incentivize making proper LODs at all. Granted there is some very obvious mis-use of LODs, but the only way to really do it right and not waste LI while not having your stuff turn into visible triangles is to upload in world, walk around and inspect everything, then do LODs for whatever degrades poorly. It's a huge time sink and given how the cost of products is dropping for a lot of goods, it's no wonder people don't take LODing seriously. Regardless it's something that should be fixed, especially because poorly done LODs aren't something you see in a marketplace listing. And if there's no in world demo, you can easily buy stuff that is useless depending on what you want to use it for.

The whole idea of deactivating stores that have items that don't sell and the store owner hasn't been online in two years is to get dead listings out of the way so people who are still active do not have to fight through so many dead listings. If SLMP is showing your products on a page filled with merchants who haven't been in here in 5 years. what works better for the customer? Buying from someone who is active or buying from someone who is gone forever? When search was Search is really messed up, the items showing up for keywords I am targeting are absolutely out dated, very high land impact, very old. When you list a new product, it ends up at the back of search with all the dead stuff with dead creators. It takes power users to find your product and start buying it for it to move anywhere in search rankings. The less irrelevant and old products SLMP has to sort and index the easier it is for SLMP to show you relevant products.

There are a lot of people on SL. The odds of something not selling once in two years and the creator not logging in in two years is awfully low if it's anything more than gestures, poorly done photoshop textures, etc. I did a random sampling of stores by changing the store ID at the end of the store URL. A lot of stores don't even have any items for sale. Some had things like Beethoven songs broken up into 10 second clips in a player. Some had some textures that look like they just put text on existing textures in MSPaint. There was a lot of very rough stuff. There are no barriers to entry to opening up a store except for upload costs. Meaning there's a ton of people who randomly decide to open up a store. I think you misunderstood the AND in there, meaning if someone makes a store and they don't sell anything for two years, but in the last two years they did log in, their store will still not be unlisted. OR if someone hasn't logged in in 5 years but they managed to sell something they won't be delisted.

We just went through months of SLMP search being completely borked and showing outdated and irrelevant results. The more bad listings SLMP team has to sift through to give you good search results, the more difficult it is for them to show you relevant products. I don't think people realize how many stores and products exist from people who tried making a store, tried making a few products, they never sold, and SLMP has to index and organize them all. I still sell sculpty stuff to this day. People are buying older products, whether by mistake or by choice. You can poke around random stores by changing the number at the end of a store URL. I spent time doing it, for every one legitimate store with something remotely worth buying there's 3 stores selling gestures, 3 selling songs (some of which are old and no longer have copyright), very poorly done textures done in photoshop, etc. I really don't think people understand how much old stuff and failed stores are on SLMP. People complain about search being bad and all this stuff, of course it's bad when SLMP has to index and organize stuff from 14 years ago that hasn't been bought ever. This hurts customers just as much as merchants. Someone new to SL who goes to search for something and sees outdated old stuff is going to give up. I used to think history was being deleted too until I started visiting random stores by the URL and store ID. I hate to sound rude but, well, this cleaning isn't as serious as you think it is.

Oh no, the store itself would just have to release one item in the last year for all items to be in a higher status. I.E. if you have something that's 10 years old and it hasn't sold in the last year, if you released anything on SLMP in the last year all of your products maintain status. SLMP just needs checks to see who is active and who isn't. If Maitreya hasn't released anything in over a year and they were going to be penalized, they'd be active creators and they'd just post something to stay active. But mesh body stores don't usually sell their bodies on SLMP either. I really don't think that's unreasonable to ask for anything to be listed in one year if you're still an active creator and around in SL. Right now SLMP is 13+ years of products and SLMP team has to sort through all of that to show customers relevant stuff. Their job is gonna be a lot easier to get people good stuff that they're looking for if they can find it. 50k limit exists because there's simply so much stuff they're trying to sort. It needs to be filtered in some way to make it easier to sort. My point about 50k is that low quality avatar components will sell better than high quality niche things simply because there's more demand for avatar components. I don't think people understand how massive SLMP sorting and search is and how cutting out older stuff (and making it optional) would help active content creators and customers. They also need to implement the single listing for multiple variants thing, because having one search document for 15 different colors is a lot easier to search through than 15 different very similar Elasticsearch documents. I mean I'm usually a pretty humble guy but my other project is a web thing based on elasticsearch with 1.5m requests a month. No way you can ever have a reliable search engine with so much old and nearly duplicate stuff listed. That said I think deleting is a horrible idea. They do spring cleaning to reduce the overhead and stuff they have to sort through and try and make sense of when someone searches. I'm just proposing instead of deleting stuff, to make their job a little less impossible, is they move stuff to legacy status on SLMP and they do so more aggressively than they are currently deleting things in spring clean up. Dates and stuff I'm suggesting could obviously be changed. But I think there's a better way to reduce what they have to search through than deleting old stuff forever.

Basically, if they've logged in OR sold something in the last year, their store is untouched. I think they should go further, disable those stores AND then make a checkbox that's on by default to only search active stores where the user has logged on in the last year and created new listings. With the last marketplace update and only the top 50k items being given preferential treatment, people who make more niche stuff than avatar apparel are going to have a massive disadvantage. And FYI 50k items isn't all that much when you consider how many color variations and versions the huge avatar clothing/shape/skin/etc stores have. I don't think people realize the scale of marketplace. My store has been around for a long time, over 10 years, and my store ID is 28761. Meaning like 14 years ago when I started there were already 28,760 stores created on XStreetSL, this was before SLMP was even a thing. Actually, just typing in random IDs and the end of https://marketplace.secondlife.com/stores/, I made it a little more than 250,000 before I ran out of stores. Meaning there's over 250,000 stores ever created (not active). After SLMP going through a phase of "we're going to sell your 12 year old stuff and bury your new stuff at the back of search", I'm all for giving people filters to find newer stuff. I don't want to see it nuked though, SL has a huge history in the internet, they were doing metaverses before they were the trendy startup buzzword. That old content should be archived or kept around just for history's sake. SLMP has a serious problem with pushing outdated products over newer ones. It hurts creators, and it hurts customers because they're seeing outdated stuff.

I don't know if it's a long standing change or a fluke, but the last few days my sales have been better and it's been pushing higher quality products. I changed nothing in my store.

To be honest looking at your store I think you have keyword issues, I don't think most people search for appliers. They search for skins. According to some of the other members here, keywords can hurt you in your title now, so if applier isn't a popular keyword it's gonna hurt you very badly.

Best selling has been broken for a while. My sales have been very poor the last few months as well. In fact I've been selling a lot of older builders kits that are around 10 years old more than anything. Lots of people have been pretty upset, customers for wondering why things look outdated and merchants for wondering why their new stuff isn't selling when it's superior in every way. Regardless my year over year compared to last year is absolutely abysmal and I'm not alone with this problem. They have been playing around with search result rankings a lot lately. The new algorithm, from my experience, doesn't like short keyword searches and prefers you search for multiple keywords. It happens to work very well if you're searching for avatar accessories because you can search for body and item, i.e. "jake tank top shirt" gives good results but "shirt" gives horrible results. If you're selling stuff that doesn't have multiple keywords in a typical search, you're doomed. LL has gone silent. And I understand why, the search results have been so skewed and weird the last few months that they have no data to calculate what's been selling well and what hasn't and what deserves to be selling well and what doesn't. For me it's created a weird situation where popular items people like aren't selling yet older, outdated stuff is selling. And I don't want to unlist my older stuff because there's a small amount of users who still prefer older stuff, and I'm afraid if I unlist it I just won't sell anything anymore because it's not picking up on newer stuff. Personally I'm demoralized, why spend 3 times as much on texture uploads when all I'm gonna do is sell old baked non-material stuff, let alone even bother uploading PBR materials to SL. Hope they fix it soon, you're not alone. But thankfully they are switching to a more industry standard workflow so if SLMP isn't working right it'll be very easy to sell stuff on other marketplaces instead of having to try and build specifically for SL, which is some sort of unholy custom baking workflow that's stuck between modern games and 2nd and 3rd gen 3d consoles where you have to bake certain maps and lighting yet let some of the engine do some of the work. Anyways I'm on a tangent but my point is that I hope LL gets a little more transparent and starts to communicate a little better because the stuff you sell on SL for $10USD can easily be sold on other game marketplaces for 5 times as much.

Nintendo is crazy about DMCA an stuff. Never use their stuff in SL and you'll save yourself a lot of headaches. Sony is very strict too. Luckily "Vita" means "life" in italian so they are kind of limited, but my experience outside of SL, Sony DMCA will hit you quickly.

A lot of my recent transactions are for products made between 2011 and 2015. In fact it's by far the majority.

Why is anyone gonna build PBR when most of their stuff that's selling is already horribly outdated and search doesn't push newer stuff?

There is a problem with it showing outdated products. It really struggles with simple keyword searches and shows lots of sculpts and prim builds even though there's much more modern stuff out there. It also feels like it does better with more mainstream products, like clothing, than niche products. Then again that's still not great, take a look at at the Linden example of searching for "blue shirt" and see. It really needs more keywords to do well, like if you search for clothing and add the name of the mesh body you're looking for. But I think what's happening is search has gotten very bad in some situations with people searching for short keywords or a few keywords and it's obliterating sales for people who don't have multiple keywords to find their product (i.e. (body type) shirt). It's like it just searches by keywords now and if ignores everything else including product age, ratings, etc. For example if you're selling products that people would find by searching for "cave", you're buried in outdated stuff.

I don't know how long this has been going on for everyone, but for a few months I've been way below average year over year. It's just finally reached a point where things are way too bad. I've been pretty depressed over it. I have a bunch of new stuff to upload and I'm not even bothering spending the time or upload fees.

SL doesn't have to compete with AAA games, there are gamers who only play indie games or at best AA Japanese games. I haven't touched a triple A game in years. And I don't plan on doing it anytime soon. I play a lot of games, that quite frankly, look terrible, and I have a lot of fun. I can't even remember the last time I played a pretty AAA game, I think it was when my 7970 was new. Even if you're the type of gamer who doesn't care about graphics, SL's latency issues mean it's relegated to only playing turn based games like Greedy Greedy instead of anything remotely real time. They can start adding lag filters to their viewer to mitigate it like I mentioned earlier in another of my posts. And it's so easy to start making money in SL as compared to trying to launch a game on Steam that has low odds of ever being purchased. The framework of creating games in SL with microtransations (i.e. buy this gun in the shooting game for more power or a cool skin or whatever) is all there. But the game play part of SL feels terrible because of the lag/latency. Sure graphics or whatever are a problem, but SL will always be held back because it takes so long for your avatar to respond to movement or for you to interact with the world. SL needs a mode that makes the net code operate like a legitimate online shooting game. It should be part of an experience in which movement method it uses. Experience should also be able to change movement parameters, like speed, etc and it shouldn't have to rely on scripts. It needs to happen on screen as fast as possible. And give scripts more keybinds. I think SL could be amazing for making small games like VRChat has. I played a game that was a cross between an FPS and Among Us in VRChat and it was a ton of fun and it wasn't horribly laggy. But the latency/lag in SL would make it worse. LL has better resources to make game levels and stuff than VRChat, people could make way better games. But they need the tools to do it. There are a lot of laid off game devs who are trying to free lance and SL could be a great place for them to do so.

When you press the key to move your avatar, the movement is sent to the server/simulator, the simulator processes what your AV would do, then it sends how your AV moved to the viewer where it updates. This is going to be an absolute nightmare when the mobile client is released, because mobile data latency can be horrendous compared to landline. It's a structural problem with SL, but they could mitigate it by doing things like having the viewer making a guess, I.E. "Flea pressed walk forward so I'm going to start him walking forward and when I hear back from the sim I'll update if it's wrong". The latency between telling your AV to do something or interacting with the world is a show stopper for a lot of people. I don't know if you play Switch, but SL's netcode is set up like Smash Bros Ultimate where everything is synced and all clients have to agree with each other what's happening on screen. Meanwhile, Splatoon 3 makes guesses. Both provide different types of lag, SL/Smash lag is more like, "I pressed a button and it took 300ms for me to see what happened", and Splatoon lag is more like "I didn't see what happened on the screen, I just died suddenly."

Checking out some of my keywords, some of these results are really questionable. Like #1 rank going to a very outdated full perm object with a single 1 star review saying physics are messed up. Cheap FP outdated products aren't really that popular, are they?

No offense but if you're doing SEO, app store optimization, etc, you're always chasing secret changes they make, redoing listings, etc. This stuff is difficult and a lot of work. You are going to have to make changes to your listings if you want to be competitive. You will have to do a/b testing. Search before really sucked, it can be a lot better now. But it's going to take work from merchants and buyers to get things where they belong. Ranking in something like Google search is endless changes and optimization and all sorts of work. In fact if you have a blog you're probably spending more time with SEO than actually making content. I never expect slmp to get like that, but I think a lot of people were actually selling stuff because relevance was messed up and it's all they could find. I know I've always sold stuff and wondered why people are buying it because it's so outdated. And honestly I think it's because it's all they can find.

Even if LL managed to somehow completely fix marketplace overnight it would still take days or weeks for the proper changes to show up. I.E. if they are weighing results with sales in the slightest, and the last several months have been boosting sales of lower quality or older items because they've been given preferential treatment in search results. It takes time for users to correct it by buying the right stuff. Mainly the power users who can search for something, skip past all the prim/sculpty/etc stuff and find the good stuff, then buy it. There really needs to be an opt in thing for each store, where the merchant does an action that shows they are still active, and then users can filter active stores or not. I.E. a merchant just clicks a button on their dashboard to say they're active, then when people search they can filter by stores that have clicked the button, have not, or ignore that filter. I think no matter how hard LL tries they will never get SLMP marketplace search and sort working that well, because they listings they have to index are not very high quality a lot of the time and there's so much legacy stuff. There are a lot of stores older than a lot of very popular websites that exist right now. Even just excluding stores that haven't created a new listing in the last year and removing them from default search unless someone goes out of their way to enable it. LL needs more filters to sift through all the listings. And FYI no matter what LL does to the search there are guaranteed to be winners and losers. My BFF also has an SLMP store and with the constant changes to search, we've always been at odds, when my sales are good and his aren't, and then they change the algos and my sales are bad and his are good. My sales are still well below normal for year over year (this April compared to April 2022), but they've finally started trending in the right direction. I hope this is a sign things are slowly starting to improve because if they kept trending they way they were I wouldn't really have much reason to create any content for SL anymore. Anyways, whatever LL has been doing, even if they fixed it completely (which they won't without adding some sort of extra filters to limit what they have to search and sort), it could take weeks for it to show up properly in your sales. I don't want to sound like a suck up, LL's shenanigans have cost me a lot of money. But it's going to take time for whatever they've done to show up as a positive for us. I have been checking lighthouse and other website performance tools on marketplace and the site is getting pretty quick, honestly. They are putting in some good effort.

Mobile client isn't for building. The benefits of it will be more users to fill up empty sims and places. Mobile users will be strictly content consumers, which SL desperately needs. The fact there are so many empty sims and parcels means there's too many content creators for the amount of content consumers there are. Which is an extremely rare situation for a social platform as usually there's not enough creators. And I wouldn't just limit creators to just people who make mesh, it includes people who create their own social spaces, whatever you want to call them. Mobile SL isn't meant to replace desktop, you won't be building on it. But it has the possibility to bring in a lot of users whose only objective is to explore, be social, work on their AV, etc. If successful, SL mobile could easily increase the population and breathe life into a lot of places in SL that are fantastic and really deserve the traffic but are rarely populated as much as they deserve. Desktop will be for building and power users, mobile will be for people to explore the world and be social. And honestly SL really needs more users who just want to explore, be social, and do things in SL. Mobile client isn't going to replace anything. Even if they make the Unity viewer work on Windows, OSX, and Linux they'd still keep the legacy viewer around because it has all the features to build. But honestly I'd be happy to see a simplified Unity viewer and for the official viewer to be treated like the power user editor for the world, and for LL to be able to add advanced features for building and creating worlds while not having to worry if it'll confuse new users.

Marketplace has been doing this kind of thing for a very long time. It's only recently gotten so bad that people finally started to notice. It has some serious problems and they need to be fixed. I've always had patterns where I sell very poorly, sell old stuff, then sell new stuff. It's never performed the way it should for an ecommerce site and it's always left a lot of sales off the table. I hope this upgrades are a part of what it needs.

The internet shifted from people having fun and doing cool stuff online, exploring technology, to becoming excessively egotistical and individualistic. The internet now is an ego boost. Post pictures, get positive attention. Post stuff on forums/reddit/etc, get likes. The wholesome internet of people just doing things for fun is mostly gone. SL was a bunch of people using some "prim"itive tools to let their imaginations run wild. Yes, a lot of it didn't look good, but it was like creating your own lego creations. Now it's more like just buying stuff and throwing it together. I.E. you used to build your own picnic table out of prims, now you just go buy one. You can see how the old internet was "never post any real life information on the internet" and it was like another life. The new internet is "post as much of your real life as possible to get internet points and attention." Big example of the "new internet" is how YouTube switched from "don't use your real information on the internet, keep your real name safe" to "Google+ update is here! Your real life name will be on your channel unless you jump through hoops and cancel it!" Definitely not rose tinted glasses, the internet has changed. Second Life has changed far less than the rest of the internet, and I think between Metaverse being ran by Facebook and people getting tired of social media, a "Second Life" on the internet away from your first might appeal to a lot of people soon. I hope SL doesn't lose that, it's one of the few places on the internet where you can be someone else still.

I can't imagine it not being successful unless LL really drops the ball somewhere. Mobile users are so valuable when it comes to this stuff because people pull out their phones when they're bored doing something else, i.e. waiting in a line, waiting for food, on a long car ride, etc. They don't actively seek out and say "I'm going to hop on my phone and play games and browse the internet for an hour", they usually do it when they're doing something else. Desktop/laptop users are more like, "alright, I'm going to sit in front of my computer and drop everything to play a game/go on SL/etc." Big difference in how often people want to get on. VR/metaverse is an even larger commitment to get into. If you have a wired unit, you have to clear your play area, set up the software on your PC, then strap the whole thing on your head. It's a huge pain. Oculus tried to make it more convenient but every time want to play you end up moving most of the room around unless you just want to sit and play. I have a Quest, some of the games are really cool, VR Chat can be fun. But VR is a 100% commitment. I can log into SL to be social and work on unrelated programming projects, blender, etc. VR is 100% commitment to playing VR. Which is what I think is a major reason why metaverse failed. I'm logged in to SL now, if I was in VR I couldn't also be on this forum reading and writing. I'd be 100% in SL and unable to multitask. People like to show off, as long as the AV editor is great and people can make a half way decent AV without spending $50 (possible with NUX) they will stay, and if they stay and enjoy themselves they'll eventually want to stand out more by not having the "noob avatar" (no matter how good it is, just because it's so easy to get people will probably view it that way). I'd even argue the AV creator is one of, if not the most important part of making mobile SL a success. That and making the AVs look good and functional, which is what they are focusing on right now, which is a really good sign they're doing what needs to be done to make mobile SL successful.