I think it's about time to share some plans for our near future.
As you know, some time ago we've launched the point system on read.cash along with the read.cash fund. This was a really cool development for us, because suddenly there was a big influx of new users. You can easily guess where the point system went live:
Though it solved one problem (stagnation of read.cash), it brought a new problem: the spam....
Obviously, when you tell people "Hey, we're going to pay you for articles", all that some people can hear is: "Hey, they're paying for letters, let's just find a good source of letters to copy from". So, they start producing trash content at a break-neck pace.
Well, we've tried building a system that detects if an article (or rather its parts) appears somewhere on the Web. That definitely solved the problem with copied content, but also that led to unending string of fights with authors (public here on read.cash, on Reddit and in our inbox, Twitter, everywhere) who wanted to publish their content in many places at the same time. (I don't really see the point in that, but OK)
One proposed solution was to leave only the "OC" (Original Content) mark if the content is detected as exclusive and do nothing if content is not original. We thought about it, but it didn't seem like a solution at all. I mean that won't stop any spammers.
You have to understand that these people aren't bad per se, they just want to earn money and don't know of any better way, so they copy articles or comments in hopes of getting paid.
Our big goal of course is to channel their energy into something more productive, but so far the only thing we can try to do is to stop them, because spam prevents read.cash fund from growing (no, seriously, that's not a guess or something, we know for sure that it can be bigger, but not with that amount of spam).
So, that got us thinking.
Removing the point system didn't seem right at all, since we strongly believe that authors should be rewarded and the previous system where only those who write about Bitcoin Cash get tips from other users of Bitcoin Cash didn't seem to do it (which is what all our predecessors like yours.org and honest.cash did... and they aren't going anywhere, frankly).
The point system and the fund solve that well, but they also have an unintended consequence of financing the spammers.
We've thought long and hard. We threw away hundreds of ideas on how to stop it. We can't of course personally moderate all of read.cash ourselves - we're a very small team (just a few guys) and we can't possibly read 1,000+ comments and 250+ articles each day.
Sometimes it's not even possible even if we had unlimited time - I don't speak Bengali and I can't verify the claims about some Bengali articles being copy-pasted from somewhere...
Sure, we can use Google Translate, but then there are claims about content being of poor quality and how would you judge that?
Then we remembered that we do have something called "Communities", which many of you haven't even noticed up until now.
Communities have admins and moderators and actually some of them are keeping pretty clean pages right now, without any pay. (More on the pay for moderators later)
We should use their power for good and we should stimulate them, we decided. Then came a lot of days when we went back and forth thinking about the way communities can help us in reaching our goals.
(Just in case you missed it - our goal is to make as many people use Bitcoin Cash as possible)
Finally, about a week ago a plan crystallized that should become "the read.cash of the future".
Our plan
Step 1: we'll upgrade communities to first class citizens on read.cash, meaning that you'll see us actively promoting communities on the main page.
As you can see, there's "Crypto and tech (c20c)" community staring you right in the face, between two articles that weren't as active in communities.
During this first step if an article receives an upvote or a tip on a community page (as opposed to via the article itself) it would get a blue community link on the main page (hint to moderators: upvote good stuff on your community page). We might even add some direct links to interesting communities on the homepage too.
Step 2: we'll turn off points for non-community-submitted articles. So, only if you have submitted your article to a community and your submission was approved you'll get the points.
That should really stimulate people to submit articles to communities and use them more.
Comments will also bring points only if you write them to articles inside of communities.
Cherry on top: moderators of communities will be getting 20% of points accumulated by authors/commenters via posts in their community. Authors will get the same amount of points, 20% goes on top.
Note though that only first approved submission will be bringing the points to users and moderators (to avoid the situation where a user submits to one hundred of his own communities and gets 100 times the points).
Step 3: we'll develop the guidelines for what we consider a good community. It won't be anything too complex, mostly it'll be about the fact that the community moderators should keep their communities clean from illegally copy-pasted useless articles and from useless comments (especially copied from other comments), also keeping the approval times reasonable.
At this stage we'll also redirect spam reports to community admins and moderators.
By the way there will be an option to make community pre-moderated (like now), post-moderated (everything is allowed, but something might be removed later) and pre-moderated, but only for first few posts of a new author or commenter.
Step 4: we'll start assigning a "featured" status to communities that uphold the high standards from the previous step.
Only postings to these featured communities will be getting points.
In other words only first approved submission to a featured community will be given points.
(Remember though, that you can always create your own community easily! Even if its just a community for you only)
The future
Hopefully, by this step we'll see spam in total decline, because it will be in community moderators best interests to keep spam at bay.
Any featured communities where the quality will decline will be stripped of the "featured" status with ability to re-apply later, of course.
We will, of course, give moderators a whole arsenal to help them (such as full feeds of everything that happens in community, results of our own internal "copy-paste" detection, abilities to mute users for certain periods for rules violation or even block users from participating in communities, detailed history of users' rules violations and previous punishments).
But all of these tools need to be developed, because they don't exist yet.
We really hope that it will solve the problem about the fact that we can't moderate 1,000+ comments per day (and growing daily), but an army of moderators with good tools can.
Commercial appeal
There will probably be one more slightly unpleasant thing. Some communities will be giving more points to authors than others, despite being equal in quality.
It really depends on commercial appeal of the community.
The majority of our fund will come from advertising on read.cash and of course there are much more companies willing to advertise on some specific communities close to their business than on any generic "My life" community.
It doesn't mean that "My life" is bad... it's just the law of supply and demand. If there's demand for advertising on certain communities, it's in everybody's interest to promote that community (because that money will also finance other, less commercially appealing communities).
Why is it so slow?
So, these are our plans for some upcoming weeks. Maybe even months. It's really hard to change read.cash right now, because we need to test everything rigorously to make sure we don't break something and lose some precious data in the process. That's why we can't move faster - because even at this pace, something like this happens. We should note thought, that despite the fact that communities looked messed up in the process - nothing was lost.
In the early days of read.cash we could just scrap something and build it from scratch. That's why we could move much faster, but right now we need to be really careful.
Topic explosion
We really think that this change would bring read.cash to a new level, because right now we have two big problems: one is about the spam and people hunting for easy money without any regard to how much trash they produce along the way, and the second one is about topic explosion.
When you visit read.cash homepage right now, you'll get overwhelmed with information. There are articles about everything. That's good. But it also creates overload. Maybe you aren't that interested in cryptocurrencies or sports or gaming, yet you are very interested in startups. There should be a way to make read.cash much more personal, much more about you and your interests.
Personalization
That's where subscriptions to communities come into play. Our home page will feature your subscribed communities much more than others and you'll have an easy way to hide content from communities that you aren't that much interested in.
Along with our new quick registration system (it's less than 20 seconds now probably, and we're not done optimizing the process yet), it should become pretty easy to personalize the homepage to your tastes.
Though, as with everything, we have our plans, but life has something else up its sleeves. We are always surprised by how our plans are destroyed by reality after we roll something out.
Let's see how this one goes.
In the meantime, we have crossed 3,500 users on read.cash. Thanks to Bitcoin Cash low fees, almost everyone gets to try Bitcoin Cash, even if its just a few cents.
SpaceX might be launching some rockets, but we seem to be heading into space too.
Thanks to all of our users, writers and commenters!
Thank you read.cash; this seems a very positive and useful development. Looking forward to it......