And without the significantly distributed data redundancy that the Storj network was built upon. It’s difficult to understand why else Storj would even openly propose these significant pricing changes on their SNOs if the goal is not to explicitly drive their SNO base away from the network and move the data to a new model.
Lol. That’s all I have to say to that.
So much for Exabyte storage expansion…
I understand that stopping test traffic is a decrease in SNO money.
I understand that deleting test data is a decrease in SNO money.
BUT everything you are talking about here is based on the data, taking into account these test data, for which Storj is 100%, then even if you agree on a price of $0.05 and it turns out that 97% of the test data is online, it will be a loss.
Even according to the table you presented, + -50% is test data = then it makes sense to talk about something?..???
Based on the above, do not do calculations based on what you have now, because at least 50% of what we receive for is test data that drags Storj to the bottom and which they will be the first to refuse without expanding SNO payments. + 50%-75% reduction in the tariffs for SNO = calculate and understand that SNO loses anyway.
Understand the first thing to do is to stop synthetic data, after we all see the payments, in 1-2 months you can return to discussing prices for real useful traffic and volume. It’s stupid to talk about it now!
If you are so afraid of a complete rejection of test data and synthetic tests, justifying this by the need to reserve disk space, then leave the data already stored for a certain period (at the same time, implement for SNO the display of the volume of this data and the amount of traffic in case the download of test data resumes in the future ), but loading and, most importantly, unloading of test data must be stopped immediately! And only based on the data that we receive, we will be able to talk about the amount of remuneration for SNO and nothing else.
+ deleting test data volumes will clean up old nodes and they will start receiving new data and will give outgoing traffic, thus their payments will be balanced.
I suggest the following:
The 1st wave of decrease is a stop and removal of test data
2nd wave of reduction or increase in tariffs for SNO 1-2 months after the complete elimination of test traffic and data
Protecting test data and synthetic traffic, you follow the fear of lower payouts, stalling for time, but it is inevitable and it is better to lose test traffic and volume first, and then talk about prices for real data.
Uploading test data hasn’t happened for a long time already. And downloading it has slowed to a grind already. So it’s basically already that way.
That said, all the payout calculations in this thread have just mentioned what customers pay for storage and bandwidth and compared that with what storagenode operators would receive. These calculations have never included test data and traffic to begin with. They’ve ran on the assumption that test traffic would eventually be gone or negligible. We don’t need to wait for that data to be gone to know what the economic model will look like.
What technical options are there to reduce those costs? I have not much idea about internet backbones and routings and all this stuff, but the idea or question of direct peering comes into my mind. Something like this:
I have no idea but could it be possible and help to ‘peer’ directly with other networks to reduce network costs?
Isn’t there also even revenue sharing in place for referred customers? I don’t know how significant that is but it reduces Storj earnings even further but never gets accounted for in the calculations for payout.
Yeah, there is a partner program with revenue share. Not sure what the numbers for that are though.
Just checked, outgoing traffic to test satellites is about 20-25% today.
Even 20% is significant for calculating payments.
+ for storage of about 50% of the amount of payments
If you don’t understand it, all these people here who believe and try, but at the very beginning they are mistaken.
Good luck to all 
That sounds like a bug. Can you share the numbers? According to the public dashboard not a single of the test satellites has a significant amount of download traffic.
I see only 2 ways to end up with high download traffic. One of the test satellites has a lot of repair traffic for some reason. If that is the case I would like to investigate. We have deployed a few changes that could cause that. Or your node is running graceful exit. That would also transfer a lot of pieces.
Don’t speak for all of us. I have installed the lovely payout script that is available here in the community. It gives me a very good overview about my income per satellite. I copy paste that into excel, wipe out the test data, calculate my payout with the new payout rate. Yes it is that easy. No guessing is required. I could even update the script to just show me that result directly but I think that is a task for later once we know how the transition will look like.
Yes, this is an old and complete node, but why should I expand now if I know prices will go down?
Why should I expand if I understand that the test data will be removed and then the space will be filled with new data?
In fairness, small and free nodes really have less outgoing traffic to test satellites. BUT it exists and Store Lab pays for it on its own, and most importantly, it pays 50% for storage (out of 8.5TB, 3.4 + 2T = 5.4TB) of disk space is taken up by synthetic data only to reserve space.
once again, test data needs to be deleted, stop paying for it, stop sending test traffic.
And in order for Store Lab to understand I want to continue on mutually beneficial terms, I want to expand and install new equipment BUT the calculations must be predictable with the correct starting point. We all know that it is impossible to guarantee traffic and volumes from clients, but only stupid people can make forecasts on test traffic and make investment forecasts and calculations based on this. Once again, based on this, the first unpleasant wave of payout reduction is a complete rejection of test traffic and volume, and then we will meet and discuss prices and the future…
Starting with “contains strong language” and ending with “I will not reply to this post” pretty much invalidates everything in the middle.
Join the discussion, exchange ideas, be prepared to be challenged.
Is it this?
Maybe membership here could also be rewarding (Backblaze is there):
Yes, it’s very simple. Do not forget that STORJ requires constant growth of storage on disks, that is, you need to constantly buy new disks, spend money. The first thing I did after seeing the proposal was to limit the size of the space for the nodes at the current level, so as not to buy hard drives in a week and not increase costs. Most likely, large operators do the same.
increase prices? just a bit… 1 or 2 $
i know this might require a lot of work and wouldn’t apply to all SNOs.
but some of SNOs would be able to take over the gateway / edge type services which StorjLabs pays a lot of money for…
it would also partially fall in line with the emerging project of custom satellites.
ofc the storage side of things still needs to be profitable, but adding more possible paths of revenue for SNOs would never be a bad thing…
obviously one would first have to determine, if that is a viable path…
but it wouldn’t surprise me if it could be…
and i for one have been looking for an excuse to upgrade to 10gbit internet 
That was disappointing, I was expecting as least one “so you can fuck off” in your post. You have a strong opinion but nothing ban worthy
Im questioning why you needed a burner account for this, This is a place to speak your mind I always do because I dont really care that much cause im pretty out spoken and I dont use a burner account, people need to know your opinion about how Storj is operating.
Money Maker: Create a simple backup app that everyday users can install (Windows, Mac, Linux, Docker) that allows them to backup to Storj. Either monthly unlimited fee per device or pay-per-TB.
This, I think, is the big hole that would bring in revenue and fill up all that empty space on the network.
As it stands, the average user doesn’t want to figure out how to set up an account, generate compatible (S3) credentials, etc.