While I think the idea is good, the more I look into it and discuss it with many members of the community, the more I start to see this as an flawed idea.
No one who worked on this system claims it is flawless. We know it has flaws, and probably many that we haven't even realized yet. That's why it says "draft for initial vendor submission and community feedback" in big red font at the top of cover page. =)
I feel that the entire idea of a rating system is stupid, because it treats the situation as a objective thing when it really isn't. there are so many moving variables when it comes to these situations that rating it objectively is kinda missing the point, there are far more moving variables than just if they delivered or not and if they are responsive for support, people can have a perfectly good track record and still just go away never to be seen. This feels like a way to minimize damage and not actively prevent it, which is better than nothing but doesn't fix the issue long term and will allow people to get scammed still.
Nothing in the system criteria is subjective. It provides clear definitions of requirements and how they will be tracked. Yes, there are many potential variables. The current ones were selected based on the collective experience of the vendors, mods, designers, veteran community members and others who dealt in some way with the ongoing vendor failures and have attempted to create a system that mitigates some of the risk of a zero-vetting GB free for all on certain platforms. There is no system that will "actively prevent" damage or scams - that's a fundamentally unrealistic expectation of GBs. If anyone is looking for that, it's called "buy in-stock items only".
I have concerns with how concentrated the amount of power is through this system, in the hands of a few people. It does not help that the system feels like it's designed to benefit the big players in the hobby and not the small ones. I am worried about the massive conflict of interest here and I think a lot of people wouldn't trust a bunch of discord and reddit mods to handle this in a way that is unbiased. What is stopping someone from excluding someone from this list, someone who has not done anything wrong and is a reputable business just simply because someone on the team doesn't like the person running the business, or a large company, secretly paying someone to get rid of competition of the list, limiting the ability to advertise to a lot of the community, the way the system is currently implemented seems exploitable.
The criteria of vendor ratings is transparent and objective measures, such as number of completed GBs, number of employees, response times, etc. There are no subjective criteria. If there are, please do let us know which ones, so we can correct it. There are multiple people working on the mechanics of taking vendor submissions and creating the profiles. It is in no one's interest to exclude a vendor based on personal feelings. If a vendor feels they have been unfairly treated, they can contact anyone on the team or make a public post, and we'll respond. But at the end of the day,
someone has to do the (thankless) work of actually entering data. Any affiliations of those people are disclosed in the document. Any volunteers are more than welcome. =)
Furthermore, this system is actually reliant on way more than just the limited number of people who are doing the mechanical work. It relies on vendors to submit info, community members to report issues, mods to verify info. platform mods to enforce rules, streamers to support rated vendors, etc. It will succeeded or fail purely on a collective effort by the community.
I also understand that this is very early in development but presentation needs a lot of work, I feel that the new people who need this info the most will not go and read a random spreadsheet about it, this information needs to be properly presented in a website in an appealing way.
Yes - we need to evolve this into a website eventually. Thankfully, we've had some people who have graciously volunteered their time, and will work with them once the requirements of the system are more stable and requirements for a web site can be clearly defined.
At the end of the day, a centralized data set about vendors is a good thing. But I feel that the whole thing seems rushed without any foresight into the long term effects and how this would affect smaller vendors and individuals. I hope things can improve and the system be fully fleshed out.
That's just how I feel though. Would like to hear other opinions.
Think of this as an minimum viable product (MVP) or prototype. There is no way to stress test all the possibilities of hundreds of vendors, proxies, and other situations than putting out a draft, collecting submissions, and refining the system based on those results. That's the process we have to go through. That process will result in some growing pains, but it's a process that needs to happen regardless. The entire GB model is an experiment, which in some cases has resulted in the failure of dozens of vendors and millions in lost consumer funds. All of these things are a symptom of the hobby maturing, painfully in some cases, but hopefully net positive in the end. Appreciate your feedback. =)