-
Notifications
You must be signed in to change notification settings - Fork 11
Paid only features
While Querki is mildly idealistic in some respects, we are a for-profit company, and we do need to make enough money to pay good engineers and designers and such. We'd like to make that money in less underhanded ways than the modern Internet approach of invading users' privacy and selling their personal information.
So we're going to have paid memberships, and will be encouraging serious users to buy a membership. We're going to keep that as inexpensive as possible -- I'm hoping for a price point of $20/year, which I think is reasonable enough that many folks will go for it.
Some folks will buy memberships simply to support the service; I'm hoping that that gets us much of the way there. But there is also going to be some "upsell", of features that are for paid users only. This page discusses the current theory of what some of those features will be.
There will probably eventually be multiple levels of membership, once we start getting into business uses of the system. For now, though, I'm just going to worry about what I think of as the "Free" vs. "Standard" memberships, aimed at typical end users.
By and large, I'm trying to avoid the "crippleware" syndrome, of having a "free" version that isn't really usable and a paid version that you have to have in order to do anything meaningful.
To that end, I'm trying not to be arbitrary about requires a paid membership. Instead, we're going to try to follow some rules of what should require paid membership:
- Features that cost us money to run.
- Features that are limited resources.
- Features that are too easily abused.
- Features that prevent us from making income honestly from advertising.
Let's go through those separately.
So far, these are less about "what" and more about "how much". The service costs money in general to run, and how much stuff you have in Querki affects how much that costs.
So this area is mainly about quotas. The most important quota is simply how many Things you have -- since Things are in-memory (for the time being) while you are active in your Space, that's not a trivial cost. We'll be trying to set quotas for Free vs. Standard at a reasonable level to distinguish between relatively casual vs. more serious users. Even Free memberships will probably include at least a few thousand Things. For Standard memberships, I'm currently planning for a quota of at least 50k Things; if we can afford it, that will go higher.
(Important note: the quota for Free users is currently pretty ambiguous, and will depend on how effective advertising is there. See below for more discussion of advertising and Querki.)
The other quota I am concerned about is attachments, primarily photos. We're not going to be architected as a file-sharing service; for the time being, we may offload the attachments to a third-party service such as Amazon. That will cost us real money -- not a fortune, but not a trivial amount either. So there will probably be different quotas for Free vs. Paid users of how much photo storage you can have.
This one is all about economics, really, much more than cost per se. Certain things are in one way or another limited, by necessity. The most obvious example here is login handles -- this is a flat namespace, and is therefore horribly prone to squatting. We've all seen the syndrome where people swarm into a new service and begin grabbing up logins, with an eye towards selling them off later. It's against the rules of most services, but it's also hard to police and prevent.
The best way to prevent this is, frankly, economics. Login handles aren't going to be essential to using Querki -- your primary identifier is going to be an arbitrary alphanumeric ID; most people will log in via Facebook or Google+; and folks who log in directly will do so by email address. That said, a lot of folks like handles, and since we are not going to have a wallet-name policy, we want to provide a way to have them. They make your URLs more human-readable (since your ID is in the URL path for your Things), and I suspect they'll be a point of geek pride for some folks. (Certainly I know that I want to be using my usual online handle on Querki, same as everywhere else.)
So the current tentative plan is that you have to have a paid membership to get a handle. That'll be a one-off expense -- you won't have to maintain paid membership to retain the handle. But requiring some price upfront should at least significantly reduce the squatter problem. It won't eliminate it, but I would bet that it will cut it to a modest percent of what it would otherwise be.
Closely related to the limited-resources problems are some sorts of abuse problems. Here, the obvious example is email.
Mind, Querki is not intended as a serious email platform. Email is intended as a mechanism to invite members of an existing community to join a Space, and to communicate with that community; so far, there are no other uses envisioned for it. Large-scale blasts of email will not only be forbidden, they will be prevented by quota: the number of emails that even paid users can send will be modest -- in the hundreds or possibly the low thousands, with frequency monitored and throttled.
That said, even modest numbers of emails will be horribly abused if that's possible. So once again, we will resort to economics and quotas: free users will only be allowed to send a very small number of emails, likely on the order of ten on any given day. Paid users will be allowed more, but the general objective is, frankly, to make it uneconomic to use Querki for spam. By and large, the spammers are pretty rational, so the issue isn't to make spam impossible through Querki; rather, we need to make it slow and expensive enough that the spammers will go elsewhere.
Here's the complicated one.
While Querki would prefer to make its money from paid memberships, we need to be realistic: at best, maybe 5% of the target market will actually do that. Folks on the Internet are used to getting things for "free", and either don't understand or don't care about the privacy implications of how advertising works online. Much though I might wish otherwise, those are the facts on the ground, and it's not worth trying to deny them.
That implies that we will eventually be selling advertising space on Querki pages. I'm hoping to keep that to a minimum -- frankly, I have found that a small number of well-targeted ads are far more effective than scads of them -- but it's likely to happen, and we need to be thinking about it.
That said, there are some principles that I want to stick to, and the most important is that Private Means Private. If I declare a Thing or a Space to be Private, then it should be Private -- it shouldn't be subject to secretly having my information resold or data-mined or anything like that.
(There will be limits to privacy, of course: for example, if Querki is presented with a properly-executed court warrant, we will comply with it. I'm not talking about that sort of outside-our-control problem. The point here is that Querki should not be profiting from exposing user's private data.)
The trick here, of course, is that online advertising is strongly based on contextual information. You can't get any real money from placing completely random advertisements -- the game is all about knowing the page's contents. (And they prefer knowing who is looking at that page.) The implication is that we cannot in good conscience put advertisements on non-Public pages.
Along with that, I would really like to provide an "out" from the advertising game. Google and Facebook have encouraged people to just give in and put up with ads as a fact of life, but I find that annoying at best. I often wish I could just pay a bit and make the ads go away; some services offer that, but sadly too few.
So the current plan is:
- Paid users will not see advertisements, and their information will not be resold. This also implies that other users will not see advertisements on their pages -- basically, if you are a paid user, your Spaces become ad-free zones. There might be opt-in exceptions to this, but only likely if we have, eg, the ability to voluntarily have ads on your pages and you get a cut of the revenue.
- Free members will see some advertisements -- but only on Public pages.
- As a consequence of this, Free users will have a quota on how many Private pages they can have. I don't know what that quota will be, but it won't be huge.
Again, this fits in with the general plan, that Free membership is intended for more "casual" users of Querki, with more-serious users encouraged to get a Standard membership. People who are just using Querki to keep their CD collection organized likely don't care so much, but people who have a lot of Spaces -- especially those who care more deeply about the information in those Spaces -- will likely be more willing to pay a modest amount to do it well.
There are some corollaries here:
- We will need a way to give gift memberships.
- The Private-Thing quota probably must be assessed when adding new Things to your Spaces. Private Things should never become Public unless specifically requested, so if my paid membership lapses, that doesn't affect my existing Things. (There will be potential abuses where people buy a one-year membership, create lots of Private Things, and then let them just sit there. I expect that to be rare, and I'm not concerned about it.)
Similarly, CSS and "chromeless" mode will probably be paid-user features. "Chromeless" is basically intended to let you partially "white-label" Querki -- displaying just your content, with no Querki "chrome" such as menus. (Full white-labeling, including custom domains, will probably be added down the line, for a bit extra.) In conjunction with this, Querki allows you to use most of CSS to customize the look and feel of your pages to a much greater degree than most cloud-based sites. But both of those features are likely to sit poorly with advertising, and could probably be used to circumvent it entirely. Neither is in any way a "casual" feature, though, so I suspect it will be reasonable and appropriate to make them paid-only.
I expect to get some pushback on this point, especially from people who believe that online services should never have advertising. But I believe it's an appropriate compromise, recognizing the reality of the modern Internet and the desire to run a profitable company, while making it reasonably easy and affordable to opt out of the advertising game.