I feel as if I should preface this with a warning label: Potential Controversy Ahead. But then I realized I’d have to put that on almost every column, and then I’d have to make a graphic and then it all just starts to sound like a project and I then think “I have enough to do. Forget it.” But still. There might be. You’ve been warned.
The Reality of Virtuality
The number of times I’ve had a discussion revolving around the reality vs. unreality of virtual worlds are uncountable. You can have this discussion really about any topic. Any activity that you can do in both RL and in a Virtual World is fair game. Of course, everyone has their own lines as to what is “real” and what is alternatively, SL, VW, or “just a hyped up video game”. However, for me, the one thing that is undeniably real is the money. A virtual world economy using real money is for all intents and purposes a real economy. Since the money can be converted to real world cash, any distinction is trivial.
At a time where economic conditions are causing ever rising tension, potential instability, fear, anger and worry, many people seek non traditional routes for increasing their income. Say what you will about Second Life, but it has provided a platform for many people who might otherwise be completely screwed by current economic realities to add some money to their income each month.
Though it is true that most people in business in SL are (at best) are breaking even, or covering their inworld expenses, even if only several hundred people are making enough money each month to cash out again, that’s still several hundred people who would be otherwise out that income. For some it’s the difference between making the bills each month and not doing so. It provides artists, musicians, and designers (amongst others) with an additional means of gaining money and exposure by using their skills. In addition, it allows a flexibility that is invaluable for people who have other life cirumstances for whom a traditional schedule might be problematic.
One of the ongoing criticisms of Linden Lab is they are not really sure what Second Life is trying to be. Partially, this isn’t their fault. A virtual world is so much more complicated than an online store, or a video game, or a social network. It’s all of those things put together, and more besides. Trying to figure out where to put your resources to support it is very difficult. At times it seems the Lab hops from one project to another with no clearly defined goal in mind. But where the virtual world and real world collide inexplicably is within the realm of commerce and finance, and so when the Lab screws around with that realm and makes poor decisions, the consequences are both immediate and very, very real.
It would appear that after all these years, Linden Lab still doesn’t understand this basic truth- that to an important portion of the merchant community, this isn’t play money. It’s not a pretend job. It’s not roleplay. For some people it’s the difference between making the mortgage payment on time or not. It’s about keeping the lights on, or having enough for that last grocery run. Maybe in different economic circumstances that wouldn’t be true. But the circumstances aren’t different, and it is true. Because Linden Lab is still not taking commerce seriously (or, conversely, definitively stating they’re simply not going to), when they make poor decisions that affect the merchant community, they are screwing around with people’s very real lives.
Two weeks ago, Linden Lab CEO Rod Humble made an announcement covering various topics. Some of them sound potentially interesting/promising, such as the ability to introduce Non Player Characters (npcs)into Second Life. But what is missing other than a vague paragraph about customer service needing to be better, is any acknowledgement as to what’s happened within the commerce community for the past several weeks, particularly as it involves Second Life Marketplace.
This omission is telling to me. It’s showing me that even now, Linden Lab is not serious about the merchant community, nor are they taking the ongoing marketplace problem seriously. This is a grave error on their part, as content creators are an integral part of the Second Life economy. This is particularly true as Linden Lab has pushed people ever more toward SL Marketplace and away from inworld shopping. The more you do that, the more merchants go “marketplace only”, reducing their need for virtual land. Less land, means less tier (land fees), and the more support and efforts need to be directed toward Marketplace itself to keep the money moving. Whilst what he did have to say was interesting, the omission of addressing commerce issues is disappointing. So here I am, doing it here.
The Devil’s In The Details
It is no secret that I have long since been critical of how Linden Lab has handled Marketplace issues. From the outset of the transition from Xstreet to the new Marketplace system we have today, the entire business has been fraught with one serious problem after another, with the people most directly affected being told to just “be patient and file JIRAs.” But aside from the many ways Marketplace has been busted in the past year (and most spectacularly within the past month), I’m going to go back to illustrate some things that are still busted and have been that way since DAY ONE first. Why they haven’t been fixed yet, I have no idea, but the Lab has a long history of looking to add a bunch of new stuff that may or may not work whilst ignoring stuff we KNOW doesn’t work already. I’ll just cite two examples. I’m sure other people will have their own.
What you’re looking at is the back end of a marketplace listing, or at least part of one. Note the areas highlighted in purple. When Xstreet changed over to Marketplace, one of the new editing fields that appeared on the back side was the one at the top- SKU. SKU stands for Stock Keeping Unit. An SKU is a code, usually a series of letters and numbers, a business uses to keep track of various items in quick fashion. If you only are selling a few items, it’s probably not very important. But the more items you sell, the harder inventory is to track, and so they become more valuable. In theory, you should be able to look up an item by SKU and find it easily.
You notice in that photo, the lower two highlighted areas show the number 5 and the SKU at the top doesn’t? You see how at the end of the SKU is a “V”? The reason for that is, believe it or not, if you use numbers (digits) in your SKU the system will not return search results for them. I’m just going to let that sink in for a second. You cannot use a numerical integer in your SKU or the system will simply not find it when you try and search for it. That’s why that SKU shows a Roman numeral (V) instead of a 5. This problem has existed since day one of the new SL Marketplace. It’s a year later, and I can’t even tell you if it’s fixed or not because we were forced to switch all the SKUs over to roman numerals in order to avoid the problem. You can use numbers in the NAMES of things and it will search for them and find them by name just fine. But if you use a number in the *SKU* and try searching by SKU? Lost cause. We found this out the hard way and had to redo all the numerical SKUs with Roman numerals.
One of the other features of the new marketplace system (and frankly, a good one) was the ability to “associate items”. So, for example, if you have an outfit in a bunch of colors you can link up to 8 of those outfits on every listing page. Someone finds it in red, they can also see it comes in up to 8 other colors as well. Or in the case of furniture, if someone sees a sofa, they can see that there’s other pieces to the living room set they might be interested in. It’s a great idea, really- one of the true improvements of Marketplace. Except that it’s stupidly implemented on the back end.
In this photo you see the problem. When you associate items, they show up as links, which you click on either one at a time (which makes that popup window disappear, forcing you to hit another button to recall it and re-enter your search term) or frantically trying to hit as many as you can before the popup window disappears (I can usually get two, sometimes three if I’m very lucky) and you have to click again to recall it. This seems like only a couple of extra seconds. But the larger your marketplace inventory, the more of a giant time waster this becomes. To have to recall the popup window and search for the right items to associate for each one of 8 items per listing takes more time than you think. For reasons I can’t possibly tell you, there are no checkboxes. The simple addition of checkboxes when associating items would move this process along enormously, as you would only have to use ONE associated items popup window in order to hit 8 boxes(once you hit 8 it would hit a hard limit) and then you could move on to the next one. But it’s a year into the new Marketplace, and this awful association procedure is still here. The only thing that makes sense about this is if the Lab completely operated on the principle that everyone only lists fewer than 20 items in total. In short, focusing on the end of the market for whom screwups really don’t matter, as they are not actually relying on marketplace at all. The logic of this rather escapes me, to be honest, but there you are.
Oops, They Did It Again
But the two examples above, while (really, deeply) annoying, are not showstoppers. However, four weeks ago the whole show stopped entirely for about a week, and the situation still hasn’t been completely fixed. This particular cascade of failure shows an enormous lack of understanding and basic respect for the merchant and content creation community and near as I can tell it’s still not really being addressed.
About four weeks ago, the Lab did a deploy of new code on Marketplace. It was only supposed to be down a short while, and people were told that the Lab would give an all cler when things were back to normal again (or as normal as they get, anyway). It was not known before the deploy what the code changes would be. After the fact we learned that this was a major code change. It was doing away with the last vestiges of the old Xstreet system and making some large changes in advance off a much touted delivery system for Marketplace called Direct Delivery. Using this system, items would be sold directly out of a person’s inventory rather than out of the “magic box” system we use now, eliminating (in theory) the need for magic boxes at all.
As you may have guessed by now the code deploy did not go as planned. In fact, it went so not as planned that it become clear rather quickly that at best this entire deploy was “poorly tested for QA” at best, and “Not tested at all” with a side order of “Use the live marketplace as a lab rat” at worst. A number of critical marketplace functions became damaged, such as: names of items and their permissions becoming wiped out completely if the item was edited, a massive number of delivery failures and “operator exception” errors, a complete breakdown of any real way of tracking split payments, duplicate entires appearing every time an item was edited or listed, listed items showing up in searches for unlisted items, a complete disconnect between a marketplace order number and a transaction number, and a failure of sale notifications being sent to merchants. Here, let me sum this up: They really, really broke it. This deploy spawned no less than five separate JIRAS. At least two are still live (I think at least three are, actually).
But setting the side the details of what they broke, there’s a larger issue at hand- which is that the Lab seems to think the merchants view the money they make like monopoly money, and that the fact that (again, at best) they deployed code to the live marketplace that had been inadequately tested was really not a problem.
That, more than the details of the (numerous) things that went wrong is really the crux of the matter, because time and again they are showing that they really aren’t grasping that when they screw around with content creators and their stores, they are also screwing around with very real money. What’s more they are not really interacting with the commerce community, many of whom are extremely knowledgeable and would gladly offer any and all assistance asked for in order to create systems that would actually benefit the community as a whole. The larger the merchant, the more goes wrong when things break, and yet the Lab seems focused on the other end of the pool, because they are more numerous as people- but again, if things go pear shaped like they did four weeks ago, they are the ones least likely to feel a real impact. Although virtual goods make hundreds of millions of dollars a year worldwide, it’s as though those micropayments don’t count if the money belongs to the people actually creating the products.
That has to stop, because aside from it being unfair and unreasonable, it simply isn’t sustainable as a business strategy. The first place to truly support content creators and treat them like real businesspeople will grab the brass ring.
So far, the Lab has done a fine job of jerking them around and one of two things will happen:
- a) Linden Lab will change its strategy.
- b) someone else will use the correct one and draw creators from SL in large numbers, because people will go where they can make the most money for the least hassle.
It is now four weeks later. At least two of the problems (the lack of sales notifications and the listed items appearing in an unlisted items search) are still ongoing. Split payment notifications are also still screwed up (but in fairness they were screwed up before this too- the payments themselves are happening but there’s almost no way to effectively track them unless you’re the one holding the inventory). Unfortunately there is also no sign that the Lab has figured out that an overall change in how they interact with the commerce community is needed, or that maybe deploying untested or barely tested code to the live marketplace is not, perhaps the best plan of attack.
So I’m throwing this out there to the Lab directly – the commerce community is a valuable resource. It contains lots of smart people who have a vested interest in making things work. They have managed to make them work very often not because of the assistance of Linden Lab but entirely in spite of them. Stop ignoring them and acting like not only their skills, but the money they make isn’t real. It’s great to put in lots of new features, but if they’re not tested, needed, or cause more trouble than they solve, then there’s a meta problem that needs to be addressed.
Frankly, I’d like to see THAT in an announcement coming from Battery Street.