Two items were on the agenda:
1 Continuing work on optimizing world start-up times
2 Discussion of a possible web-based marketplace
Continuing work on optimizing the asset and inventory systems
Ilan updated the meeting on progress.
The new asset system optimizations currently being worked on greatly reduces the load created by users logging in to a world and starting to download all the assets. It loads assets and users' inventory via HTTP instead of UDP (hence the recommendation to use a modern viewer such as Firestorm; legacy viewers will not be able to take advantage of these optimizations).
When the main Kitely system gets the HTTP requests, it has systems outside the OpenSim software handle them. This enables the load to be distributed elsewhere so that things that must run in OpenSim (e.g. inworld scripts) have more resources in which to do so.
Other things could be moved outside of OpenSim, such as chat, but inventory and assets were creating the most load -- hence the priority placed on them, and now the further optimization work.
A big part of world startup time is how assets are handled. This change will also make world starting times more consistent. They currently fluctuate depending on whether the server that is running the world had ran it previously.
Marketplace discussions
After asking who at the meeting had sold content on other grids, Ilan led a discussion on Kitely possibly setting up a web-based marketplace "with some twists".
*** IMPORTANT NOTE: This is still at the brainstorming stage and no final decisions have been made about features, functionality, behaviour, fee structure, etc. in addition some of the things that were discussed in the meeting have been removed at Ilan's request as he didn't want to reveal Kitely's hand to competitors before the marketplace is ready to be rolled out ***
Initial points:
- The marketplace would be limited to non-free items only.
- There would be a fee to list (or "publish") an item for sale on the marketplace.
- Items sold in the marketplace would be charged a fee of 10% in line with Kitely's other website-based monetization options. Items sold inworld don't have that charge now and won't have it later if/when the marketplace goes ahead. [This fee is separate from whatever costs there will eventually be to get real money transfered to people's PayPal accounts (using Kitely or a third-party exchange).]
- The marketplace will have a search system. Items will be ranked by statistics, "likes", etc. Users will be able to search for a particular creator.
- Items bought from the marketplace will be delivered directly to the avatar's inventory whether or not they are inworld at time of purchase.
Generally the meeting was positive towards this idea, thinking it might help to reduce clutter (at least at first). "It's worth trying, you can always change your mind later and allow freebies but not the other way around."
Freebies and promotional items could still be offered inworld, as well as full-price items, of course. Several in the meeting also commented that they liked to see items inworld before purchase, whether via a web-based marketplace or directly inworld. Marketplace listings could have a link to see the item inworld, or to direct people to an inworld shop.
Ilan pointed out that one problem in having freebies in a marketplace is that it reduces the amount of money people are willing to spend on non-free items, which hurts content creators. The aim would be to make the marketplace content "high quality".
This led to a question of who decides what "high quality" means, and whether products would have to go through an approval process in order to get on the market.
In the end, it is the purchaser that decides "quality". As one person put it: "In SL some of the freebies are high quality, and some of the expensive stuff is crap. Nothing you can do about that."
A pre-listing approval process for every item would be too labour-intensive, but listings could have a "Report this item" button to assist with post-hoc moderation. (See discussion under question 4 for more on item moderation.)
Question 2: What do you think the minimum price should be? What should the publish fee be in your opinion?
Neither of these questions were answered directly, but the subsequent discussion brought out a range of views and ideas, including comments on the 10% marketplace fee. But as Ilan commented later in the discussion: "It's important to hear possible objections as well. We can't build a good solution without hearing them."
Ilan started by explaining that the publish or listing fee would be charged "when you press the "publish" button. You can sell a million copies of that item and not be charged for the publish fee again. If you publish often then you will be charged the publish fee often. It's like paying for a classified ad."
A couple of people pointed out that Etsy and eBay use a similar model of charging a listing fee plus a fee when an item sells.
One person said she would prefer not to pay a listing fee unless the item makes money -- if it sold, taking a percentage was fine.
However, another confessed that she hated the 10% but wouldn't mind a higher publishing price. She compared Kitely's pricing options for inworld activities and those suggested for the marketplace: "The parts of Kitely I really like are where prices reflect costs logically [e.g. 10% for monetizing visitor charges -- Kitely's costs go up the more avatars are inworld]. That is not as true for a marketplace. That's why I don't like 10% for a Kitely marketplace. ... Long term, it's more sustainable if you can justify your charges based on your costs." She expressed further concern that this might make Kitely vulnerable to competition, in that if another site could do it cheaper, people would end up there eventually. However, "If you can justify and articulate your cost vs price strategy re Kitely Webshop, then potential competitors will be discouraged from undercutting you."
[As an amendment, Ilan adds that Kitely will in fact incur a cost for managing the maketplace, both for moderating item listings, for handling inevitable customer complaints about merchants, for handling more attempted credit card fraud by people who will try to launder money via the marketplace and for ongoing development and operations costs related to the marketplace.]
Ilan explained some of Kitely's thinking behind the 10% marketplace fee: "Application stores (iOS, Android, and others) are charging 30% from sales revenue. We intend to charge 10% for the revenues which we enable using our tools but not charge for those that are created inworld without our tools. Our long term goal is to make inworld access cheaper but we need to have a sustainable business to do so. One of the most commonly accepted ways of doing so in virtual worlds and online games is via monetizing content sales."
Others commented that one pays for the value of using a service, and that a strong, viable marketplace would be a good draw to bring users to Kitely.
The discussion moved on to whether there would be rules about pricing an item that is for sale both inworld and on the marketplace. Ilan confirmed that the price rules would be for marketplace sales only.
A further example was put forward: "Let's say someone offers a house for sale in the marketplace for 5000 KC. They mention in the ad that if you buy it in-world, you can get it for 4800 KC. That would circumvent the 10% fee and still garner a sale from the marketplace ad." Ilan said this would be fine -- "The goal is to offer a tool that people can choose to use. We're not forcing anyone to buy from the marketplace -- we're just adding that option." [As an amendment, Ilan adds that while selling an item cheaper inworld will be possible, directing people from an item's marketplace listing to purchase inworld would probably be against the marketplace's terms of service.]
Question 3: What is better in your opinion: having item entries in the marketplace be editable cheaper or minimising item changes so that it's easier to track new items that have just become available?
A range of views was expressed, but the meeting seemed to move to consensus that a fee would be chargeable only on initial listing and there would be no additional fee on editing (much as in the eBay model). The "What's new" tracking would be based on the date the item was first listed and not take any editing time into account.
Question 4: How about having multiple entries for the same item with different color?
The meeting quickly agreed that this would be A Bad Thing -- that it would add clutter, was a form of spamming.
It would be better to have a single entry for a group of similar items with an option to see/select from those variations when viewing that listing.
At this point the discussion moved to the issue of moderation: preventing duplicate items could potentially require items to be actively moderated before they are listed. It was agreed that including a "Report this item" button on each entry would probably be the most practical solution, so that only reported items are moderated. It also means that content creators could bring items to the marketplace without having to wait.
The question of converting KC back to real money
This was seen as a key issue by many attendees. Wouldn't it need to be in place before the marketplace is set up?
Ilan thought not. The marketplace can be rolled out before being able to turn KC into real money.
One person commented: "I have a list of Content Creator SL friends that are very interested in Kitely but monetization is THE issue holding them back." Another agreed: "For them, until they can make money, they won't want to invest"
Ilan suggested that they might like to get a head start on others by listing their items sooner and hoard KC (which don't expire) until they can monetize them.
Someone asked whether converting KCs to Linden dollars might be a simpler option. However, the issue is (as with monetization to real currencies): who does the conversion?
Rolling out an exchange has legal issues, not just technical ones. It can't be rushed. Kitely could possibly go for a Virwox exchange but that might not be the best option for them long term.
Ilan continued: "I wish we could do everything we want yesterday. Unfortunately we can't so we have to prioritize for what should help us grow optimally at each stage. For example, getting a lot of people from SL without the inventory system we recently rolled out would have been painful. We're now optimizing assets which should give us a lot of growth room before we need to do additional scalability stuff.
"Right now you can come over from SL and get great performance inworld but finding content is hard. With the marketplace, those content creators that understand that KC don't expire will start selling thus making it easier for non-creators to move over to Kitely. The more people we get the more attraction for more content creators.
"Eventually monetization becomes real then people can start selling all those KC. In the meantime they can use them to offset Kitely subscriptions. That is a feature we rolled out a long time ago because we wanted to make it so that the majority of SL content creators (who don't even make tier) will be able to at least make tier on Kitely."