recently thought about this topic again. there is no clear date for Ceres, but I would still like to have the topic covered somehow.
what is the general sentiment, should the auction system be changed according to the aspects mentioned above?
generally lower the time for the auction end and have a fixed frame of e.g. 480 keyblocks until the auction would end
bump of the timer on new bids only in a certain frame of keyblocks (e.g. 3 keyblocks)
I think it would be super cool to be able to have enforced voting on those parameters in the future so that no hard fork would be required if there is a general consensus that it should be changed. not sure if this would be a viable option for Ceres though (cc @hanssv.chain)
shall I open a poll? or do we want to track the sentinment on the thread only?
I voted and made my points already clear above. I think there is definitely an adjustment needed.
In regards to the governance mechanism to change params without having to wait for a hardfork. I think this would be cool to have, but I don’t think it’s urgently needed right now. Probably more important topics to be covered by the core devs
I voted, and my ideas should also already be represented above, but TLDR: Yes, shorter auctions is a good idea, I would probably divide the length in three groups short (2-6 characters), medium (7-12 characters), and long (13- characters). No auction for long names, like 2 days for medium names and maybe 5 days for short names. Bump the auction time up to about an hour (20 generations?) if there is a bid closer to the end of the auction.
I am also fine with that. just shouldn’t be a month or even more. few days are ok I think.
I think I remember we discussed a short period for the bumping already somehow. I just have a fear that people get distracted and thus very frustrated if they have to wait for a long time period for the auction to finish as they might miss the opportunity to place a new bid if they do not have a bidding service running for them.
what’s the risk / downside with a small bump time of e.g. 3 keyblocks again?
The opposite of getting distracted, i.e. that you don’t have time to react (can be mitigated by tools to a large extent). Generations don’t have fixed length, and 3 generations could happen in a minute… Also the discussion on finality - but forks of length 3 are not common, agreed.
Also you don’t have to wait until the last generation to overbid
that’s not the point. I am more worried about people not noticing outbids as they move away (from) the phone / tablet / desktop. you need to wait for the auction to finally end. you want to know if you won the auction or if somebody outbids you.
forks could certainly happen but that’s also sth. for tools to support the user in such cases. in worst case the auction ended without noticing and some other party winning it
I don’t see any problem with the timeframes nor with the pricing. If people want to get a short name quick they can put a very high initial bid so that nobody wants to start a bidding war with them because it feels already overpriced. If people want to get started quick they can select a bit longer name like alexanderthegreat and get it instantly. Short names should be difficult to get by design to prevent squatting. I also don’t see a problem with the price being driven up to the max amount people are willing to pay for a name since the fee gets burned and this way the total AE supply gets reduced effectively acting as a donation to the aeternity community.
As an alternative we could create another namespace like “.space” to have very cheap short instant domain names. Could also be solved with an establishment of a new naming prefix and as well introduction of syntax for this, like “shortnames—alex.chain” or “a___l___e___x.chain” could be translated to alex.space so that even no hard fork is required for this, just some UX work. More “.TLDs” could be introduced this way. And it is backwards compatible.
A marketplace for the dot chain names would great as well! There people could buy instantly short dot chain names. But introduction of a new name space and addition to the existing UX should be the quickest way to solve the problem perceived by some of the users. Which .namespace do you suggest?
Let’s not change anything about the auction mechanism without that it is not absolutely needed. Better create another namespace with whatever rules you want. Open source rocks.
it is not a year. but a last minute bid could currently cause an extension of the auction for ~ 3 months. this sounds really crazy to me and makes the auction mechanism for short names quite unnattractive.
everybody watches the end of an auction or uses an automated tool to bid until their personal max-price is reached.
having async auctions in terms of time is nonsense to me. this is why I think it should be as short as possible. having to look at the same auction every now and then to finally see it settled kind of sucks. I am really wondering how you come to the conclusion that nobody can react to this
note: I hope it is clear enough that the initial auction time will never be less than ~24hours (according to the current params set). I am referring to the actual expected end of an auction.
Yes @marco.chain your description looks on point to me. I am with @philipp.chain on this one, 12h which I set is an absolute minimum in my eyes. Automatic tools are great, but we need to give humans a fair chance to react WHEN the tools break
this is exactly what I want to achieve. nobody wants to check the end of an auction every 12 hours. humans get distracted if it takes too long. auction-ends are typically happening in a very short time period. I am really wondering where your point of views are coming from
I think we should more discuss which events you expect to happen that would hinder “humans” to react within 5 keyblocks. I would say in 99,99% of the cases 3 keyblocks should even be enough time to react.
but anyway, I am fine with everything. I just want to see the auction system improved in general what we (almost) achieved now
agree to disagree. I think core developers should be working on reducing the rolling checkpoint time and launching hyperchains and decentralization instead of changing the economics of the naming system. I find it a big blocker for adoption of aeternity mainnet to wait for 100 blocks so that transactions are considered confirmed. with bitcoin its usually 30 minutes wait time. what is planned to improve this? other ceres improvements?
aens is only a small number of domain name holders such as Dashu in China have a strong demand; They want to sell their domain names. I don’t know why the team took the time to improve this? Aren’t core functions and hyperlinks more important? How long has the hyperchain been delayed; The hyperchain release promise has been delayed。