Some operators started out with two pools in the first place, and are now consolidationg as none of them produces blocks. 
I hope daedalus will be able to suppot 1-to-n delegation soon…
That will be great. My only real concern for delegators / pools at the moment is when metrics shift and saturation threshold drops leaving pools with high delegations instantly saturated. I’m not sure how many delegators keep on top of that or will think to look in a few months. The pool operators don’t exactly have a way to communicate to current delegators that they should move some delegation elsewhere to maintain highest rewards. Delegators would need to monitor their pools or social media to a certain extent. Who would actively move their delegation? Who would assume others will move theirs first and do nothing? I’m so very interested to see how it all plays out!
What you call concern is my only real hope 
Bragging about being 100x more decentralized for quite some time now, but metrics support basically 50 to 60 pools producing the vast majority of blocks at the moment (apart from some lucky ones with single block here and there).
I guess, quite some delegators will review their options every few weeks, as all this is still quite new to them and d is decreasing. On the other hand: if k does not change during these times, there is no incentive for better spreading of delegated funds.
Yes! I too am a tiny fish in a small pond. I don’t see me getting a blocks any time soon. I would imagine saturation would have to be under 10m before I start getting any real consideration for delegation.
The new stake coming in is forced to decentralize as more pools fill up. We shall see some big days once some exchanges support staking and when Yoroi Ledger is activates.
What makes you thinks that people wont orientate towards the already big pools? I just randomly (!) picked three small to mid-sized pools which were even producing blocks before… they’re practically going extinct right now. 
So far we are consistently seeing a retiring of 20 or so pools per epoch. We are also seeing a new batch of pools every epoch in roughly the same number. I think we will see a regular cycle of new pools replacing retiring pool in this small to mid size range for a few more epochs. As long as new pools keep being built to replace them we will maintain the current number of pools. Maybe even see a few more. As soon as we run out of fresh fish the pool count will start to fall. I think we will settle in the 700 ~ 800 range by the end of the honeymoon phase.
I have no proof to back any of this but I think we will see a lot of short term pools coming and going in this manner as long as google and amazon offer their free introductory period. As soon as a majority of SPOs have to start paying bills on a non block producing pool and / or notice that they are actually losing money by not just delegating their own ADA to someone else’s pool instead of dumping it all into their own pool they will immediately shut themselves down. There will be a few in for the long haul. I hope to be one of them :).
After about 12 ~ 24 epochs (1 ~ 2 months) we will start to see some real numbers when all the new comers cloud services intro periods expire.
@mraybin: I wasn’t even aware of “free” hosting offered by the big players in the industry. I never considered using those. Thanks
I mean, what kind of decentralization is that supposed to be, if there is a big number of “logical” pools - which are hosted under (political) control of two or three big cloud hosting companies?
At the end of the day, this offers no advantage over a “classic” mining operation, where a few big players control the sake of the blockchain. Just scan the processes of VMs for cardano-node, shut them down, problem solved.
I hope you’re right, concering the “wash out” period. We’ll see who’s lipservice is actually backed by the will to pay operating costs out one’s own pocket. Just because
It’s a labor of love. :). I think of it like a hobby, golfing, fishing, running a Cardano stake pool… They can all cost as much or as little as you want really.
I don’t see a problem with this at all. These services are not going anywhere any time soon. These services are not “running” the pools. Individuals are. These services are all very good at what they do. It’s not like anyone could attack all three and bring them down at the same time. The world would most likely have bigger problems if that was happening. Each pool operator still has the control to manage their own pool with their own specs at their own price point. Lots of security pre built in. Even if it were to all fall at once there is still a good hand full of folks running bare metal or raspberry pie’s from their home. I think there is plenty of decentralized goodness going around considering we just started like two weeks ago. I think we will see even more of that in the future when the core software becomes more robust and can run on more devices.
I started on a cloud service. While that was running I tried bare metal at home. I didn’t realize how often my power went out :D, amongst other inconveniences and costs like maintaining multiple static IP addresses, multiple internet service providers, multiple local machines. Backup power sources. Cloud service has it’s place. In some cases it’s cheaper and more reliable than running from bare metal locally. If you have an office with multiple computers and business class internet all ready it is way easier to absorb the cost and infrastructure running your pool from there.
Again, just my thoughts, I have no fact to back any of this.
I’m not arguing about using a so called “cloud service” for hosting cardano nodes, at all.
My point is, that those huge datacenters being controlled by less than a handful of companies, often entangled with politics, which may become subject to economical and political… let’s say “ideas” - if something might threaten the company’s revenue, become a competitor to their own projects or against their political agenda. I mean, we’re not talking about sunny days like now, when nobody cares about obscure small projects like cardano… 
I’m talking about these, hopefully coming, other days when the project grows up.
We might be administering our machines, but most (myself included) do not actually run them. In terms of resilience, I personally would just prefer, less eggs in these big baskets.
Why not do the same with the actual physical hosting, as what we’re here for with our cardano-nodes? Why not support smaller companies running their services professionally and renting VPS to customers? You don’t have to care about power outages, public IPs, webinterfaces and all this stuff either (it’s standard anyway). Apart from that, if you cared about running a “bare metal” server, you would rent a physical machine, screwed into a rack in a datacenter for security and practicality reasons, anyway.
Thanks for the discussion, mraybin!
I mean, what kind of decentralization is that supposed to be, if there is a big number of “logical” pools - which are hosted under (political) control of two or three big cloud hosting companies?
That’s exactly my point! Over the years, I read through many masternode setup guides of different cryptocurrencies and most of them suggested to host a VM at one of the big worldwide companies. As you said, this is no decentralization.
That’s the reason why we encourage everyone to setup his/her own stake pool. We are here to help! Just a few days ago, I was contacted by a stake pool operator because he had troubles updating his newly created node. I published the information in our Knowledge Base - we appreciate it when it helps other SPOs of newly created stake pools as well.
It is important that a truly decentralized systems consists of many different machines - and not automated created virtual ones which all rely on the same set of (few) providers and same hypervisors (and thus the same security leaks).
You have an old machine in your basement and a stable internet connection? Go ahead and configure it as a node for cardano. Exactly these unique machines bring the diversity into the ecosystem of cardano which is essential for making cardano completely different and better than bitcoin!
Cheers,
Chris