Addressing the "dead land" problem

I don’t really think dead land is that big of a problem yet…
Most people teleport around…
5% of lands being inactive is less than I thought…

My opinion is forget the problem until it’s actually a problem…

1 Like

@Canessa
The goal is not to solve the problem of how to incentivize large active holders to actually do something with their LAND - though that’s a very important problem and something that needs to be done too. But I’d tackle that separately

The goal here is what do we do with LAND that is literally not accessible by anyone because they are dead and their private key is lost.

Metakovan had transactions less than a year ago so he’d be considered active/“not dead”. The address for Festival Land has received the LAND over 3 years ago.

Wondering if we could add an extra oracle subsystem that knows of all LAND deployments and creates another merkle trees with all LAND that has been deployed in the last 5 years.

@BET
this might in fact be the case - trying to figure out how to pull numbers of how many LAND have not been deployed AND are owned by addresses that haven’t had any outgoing transactions in the last 5 years - in other words no outgoing transactions since genesis.

if that number is under 0.1% then I’d agree that tabling the discussion for now is probably best route. but given that we are talking about setting very long inactivity/grace periods (5 years) when we do want to tackle the problem, it’ll take at least 5 years before we can actually do something about it.

I don’t think this is a problem now… but at what point do we consider it a problem? 10%… 20%? It’s a good idea to have this conversation now and come up with a plan before it reaches a significant percentage. It’s probably safe to say that 5% of the land in DCL will never be developed for various reasons, one of them being we have to wait several years before we can auction off any inactive land. I think 5 years is a fair deal. I think the auction process should work similar to IRL… we should determine the market value of the land by appraising it based on comparables and auction off the land back to the community at fair market value. Thanks for bringing this up @seb…great stuff.

1 Like

hurray! - received a spreadsheet from hardworking volunteer (: the spreadsheet has address, number of land, number of estate, number of days since last Decentraland NFT transfer and number of days since last outgoing ethereum transaction.

Findings
About 0.5% of LAND and EST have not moved in the last 3 years - and the owner has not made any outgoing transactions in the last 3 years either.

A few examples of addresses that have not transferred (in or out) LAND/EST nor made outgoing Ethereum transactions for over 4 years

address numberLand numberEstate
0xe8641e43a2a34dadcb8fc5a0e9e244214ccd36e9 1 0
0x89eb01ff4fa87841cb3fbdfa85d13a88ed32ad55 0 3
0xa2a1df4bc4a4cae81bac66d94520af7b7e14c8d0 7 0
0xfca8a978ade10f699d9222b7cc41306980d24eea 32 3
0x31343bf81b7cc66988902f8eaa13e8953a76889e 16 2
0x1ed0acd4123f174e0d984b79b1ee0fca551c43d3 13 1
0x606123cd289bf2d9358c80b224364601c1d95a6e 13 1
0x0ed8889cb2d073efae7b0c82cb33bebff655e92e 6 0
0x5420f1137a4e80f3ad2b41130dc8554daf6eeac4 6 1
0x9d287dd272b9fc823b0f7c03f9f4ca07f87abc2f 6 1
0x74efcb25ef70802b81fedc099250c042c538c616 6 1
0x4aa9874e0d7898cec7ba879f584150fec226e368 16 2
0x1b38ddce13ab4116e09ca4d8435852bdb0985c01 4 4

Looking at aggregate, there are 169 LAND and 96 EST for which the address has been inactive for more than 3 years, and 142 LAND and 33 EST for over 4 years.

Num years since active Num land Num estate
0 16385 4590
1 1709 483
2 603 120
3 169 96
4 142 33
Total Result 19008 5322

can’t tell how many LAND are in these EST (could be 0), but with a conservative 1 LAND per EST, it means that about 0.5% of LAND would currently be in grace period - using 3 years as an example. If not activated within that grace period -f.ex. 2 years, then about 400 LAND would recycled.

we are probably under-counting a bit for two reasons

  1. some addresses have received NFTs recently yet are provably “dead”. F. ex. both 0x0000000000000000000000000000000000000000 and 0x000000000000000000000000000000000000dead have been receiving estates from people wanting to get rid of empty EST it appears. at least some of those EST are not empty… This nice 4x4 EST belongs to 0x000000 but would not be recycled because 0x00000 would be considered “active”: 乾坤宝地 | Decentraland | Etherscan

  2. there are contracts have received LAND or EST in the last 3 years - yet don’t have a way to transfer the NFT out, the MANA token contract being an example. It received this nice 5 LAND estate about 3 months ago Estate (EST) Token Tracker | Etherscan There’s a few more examples like the WhaleStreet Deployer (?) that deployed unverified contracts and got LAND transferred about 2 years ago Contract Address 0xb3c48ff6830ff7e9597882b207c7ac8191f7f208 | Etherscan

Conclusion
The issue is not as widespread as I was initially thinking… i guess most of the empty parcels really are just people not having time to deploy anything - yet still being active on Ethereum.

The true number of “dead land” is probably above 0.5% since more people have lost keys in the last 3 years and aren’t counted in the statistics above. it’ll be interesting to re-run the analysis in 1, 2, 3 years.

but even if that number is at 1% - it’s not clear to me anymore how much of a priority this should be compared to say, creating positive incentives for active users to build on their LAND.

yet - the problem still needs to be addressed sooner than later… it’s a bit sad for the neighbors of those LANDs that are clearly dead, to know that you’ll never have a neighbor and will forever sit to an empty LAND ):

This is still great info! We can use it to somehow encourage more building since we know the landowners are active and it will help us to keep track of this issue for the future. Muchas gracias secret volunteer person. Maybe this can be added to the library/repository thingie or made into a public data that everyone can access? :smiley: