Reconsider the word "deed" #2
Conversation
|
Pinging: @freegold @mudgen @carloschida @inhumantsar @eordano @Nanolucas @Beskhue @jadhavajay @TCOA @dekz @7flash I noticed you had discussed the naming of the ERC-721 standard and thought you might be interested to join here. This is the official naming discussion. The result here will change or reaffirm the name of the ERC-721 standard by the end of this weekend. |
|
Title |
|
Just leave it at token please, everything else would be a huge burden on the developers who already work together using the current proposal using the name |
|
And, yes, for the name of the standard, please revert to |
|
Just throwing this in there: Unique Token Standard The term non-fungible means nothing to most people. The notion of a token and a unique token is much easier to understand. When I talk to people about these concepts, the phrase "unique token" helps people grasp the idea quickly. |
|
Yeah unique token is cool too |
|
NFT is OK |
Choice of word 'deed'Most unlinked references are to this Proposal. SUMMARY: I recommend employing 'crypto title-deed' whenever the term is firstly introduced and exhort the use of the following translations.
TODO: Request for comments on candidate translations to other languages.
I would say that the desired characteristics of the name to be chosen are:
It is correctly mentioned that the word 'deed' depicts that the specification tries to achieve:
Nevertheless, that is the second acceptation of the word in the Oxford English Dictionary, the first being
Fortunately, another entry exists in the same dictionary for a noun that refers exclusively to that sought second acceptation: title dead. With this we achieve so far Characteristics 1 and 2. Notaries and deedsDeeds are often related to a action/transaction performed in the eyes of a notary public or equivalent officer. In order to avoid this allusion, I also suggest the employment of the adjective 'crypto' to describe these title deeds. A historical guess on notaries and deedsConsider (economic) goods and categorise them by ease-to-transport as 'movable' and 'immovable' (likening how DE _Immobilie_ and ES _inmobiliario_ refer to EN 'real estate') and focus on the second. Now imagine a Roman citizen that acquired land. This citizen builds on it, inhabits it, and has thus so-called [factual possession](http://www.quinnlaw.co.uk/acts-that-constitute-factual-possession-of-the-land-in-claims-for-adverse-possession/) of this immovable good in question. This citizen decides then to travel away for an extended period of time. When they are back, they find a squatter inhabiting the land. Now we have a case of adverse possession. Nowadays, we have such problems solved by means of appending entries to an immutable ledger. But before this discovery, and even in the traditional systems of today, the claim was solved by enquiring a custodian of the centralised ledger, ie a notary public, in front of which transactions of immovable goods had to take place. Abruptly back to the naming question... I suggest, therefore, that these title deeds are also appended with the adjective 'crypto' as a way of stating that such a custodian is no longer needed and rather that the ledger is decentralised. In order to also comply with Characteristic 3 I propose that the full form of the object described in the specification be:
The word 'deed' complies with Characteristic 4 in, at least, German with the word Urkunde, and Spanish with the word escritura. I agree with @TCOA when they say that it is typical of real estate but must emphasise that 'typically' does not imply 'exclusively', and that Abstract firmly references such deeds implicitly. Without hesitation I could also say that some of the characteristics of this kind of deeds are desire by most of the stakeholders and that the semantic relation is clear. As a particular example, the word ES escritura (EN deed) is also used in Mexico as a proof of transfer settlement of the claim over an inheritance. 'Non-Fungible Tokens' (NFT)In the dictionary entry of the adjective 'fungible' we find:
Hence, the choice 'non-fungible token' is only partially correct as it does not specify anything about property or ownership of the good in question which is desired in Abstract. Other alternatives
|
The word 'deed' is not constrained neither by definition nor by employment to real estate I assume that with 'America' you mean the US. If so, my claim still holds: a deed, although usually related to real estate, is in definition, and employment a document that refers to the transfer of property. With the definition, this time I mean the US standard, Merriam-Webster (see also title deed by the same dictionary), and with employment I mean how the American Society of Notaries defines it. The line of thought of my comment was the following:
If you disagree with any point in particular or the execution thereof, please, do not hesitate to specify which one. I acknowledge that in the US the word is strongly related to real estate–and probably only to real estate–but this is not the case in the rest of the world:
I just came back from my advisors' offices and I realised that we talked the whole time about the document with which my company was founded(notarised) and that ultimately determines the ownership as ES escritura (EN deed). |
If the final implementation of ERC721, say we decide to refer to it at the end as 'deed', is ERC20-compliant, then it is by all means an ERC20Token. This is not contradictory. Consider the base tutorials of an OOL where a |
|
Most arguments here conflate the definitions of "deed" and "title". They are not the same. This NFT is not accurately described (in English) as a deed. A deed is the record of ownership of a title. Since the standard refers to a unique NFT and the owner is just a fungible attribute, naming the standard according to its fungible ownership (i.e. deed) characteristics and not its un-fungible asset rights/title characteristics achieves the opposite of the standards' stated goals. If the accurate terminology of 'title' remains unpopular, please cast my vote for anything that isn't "deed". That this distinction is not understood does not bode well for adoption. @carloschida check that reference on the definition of a title again - there's nothing about title not being unfungible |
|
Neither choice is perfect so we need to take approximations, because we are creating something new. In that case, my choice would either be Deed or keep Non Fungible Token. I would vote against using title. YAY FOR TOKENMerriam Webster defines Token as: So the standard definition of token comprises of both fungible (a token for a bus fare) or non-fungible (souvenirs, challenge coins, etc). I would argue that if you built a trading card game with round metallic cards, they could be called "tokens" too. YAY FOR DEEDGoogle Defines it as: I understand you can argue that the Deed is the document, not the property itself, but for me it's a close enough approximation. If you have your name on the deed, you own it. NAY FOR TITLEOxford Dictionary: To me, definition 4 is the least used popularly when referring to the word title. Title is specially used in software, webpages, and other apps to mean the thing that appears on the top bar and I think it will generate confusion, as that's the first thing people will think when they see that word. Not a big fan of the other options. |
|
@d4v1dg I agree. That's what I meant when I said that 'title' violates '[making] allusion solely to the rightful claim over a non-fungible object' (Characteristic 1). And I also agree that title and deed are not the same, but there's another entry for title deed. |
|
Opinions, in order from most acceptable to least acceptable: Token: Fully support this 100%. It's the best option. There will not be confusion between ERC721 tokens and ERC20 tokens, because both are used with the qualifier (and will be used with the qualifier more, as other standards are also implemented). "Token" is what we are currently using in both our Nova Token implementations and our blockchain gaming implementations. And the name of the standard should be "ERC721 Non-Fungible Token", to ensure maximum clarity. |
|
I fully support deed. I think it's short, and represents well what we are dealing with. I do not agree with people saying "the ERC721 token is the actual object that is owned." In my opinion that is not representative of reality. It is true the ERC721 deed is owned by some Ethereum address, but I have not seen any implementation where people are excited about owning the ERC721 deed as-is. They are excited about the object the ERC721 deed represents. In most cases so far, this is some image (CryptoKitties), a collectible card (CryptoCelebrities), or piece of land with operatorship (ÐWorld). I have seen cases in the cryptogaming community where people were frustrated when the front-end application of an ERC721 implementation was taken offline: their ERC721 deeds now no longer represented anything. Besides ÐWorld, I have also used the "deed" naming in MetaGame (ABI). I am weakly against non-fungible token: too wordy. |
|
@paulbarclay You recommended "ERC721 Non-Fungible Token". Could you please clarify what your recommended name is for the smart contract that manages these?
Can you please clarify what is your recommended name for the actual things the tokens represent?
Thank you. |
|
We're already using Deed in our factory project, and considering the alternatives, Deed is what makes the most sense to me. FWIW though: When faced with describing it face-to-face, I tend to drop "deed" for "token". |
Sure - it would just be a "smart contract"; I don't see any need for a special name there (smart contracts can do more than one thing, and can implement more than one standard, so a special name seems unnecessary). Also, I don't believe my answers to any of these questions would change, no matter what we call the token.
This XXXXXXXX is "smart contract". Could be "An ERC-721 smart contract" or "An ERC-721 compliant smart contract" or "A smart contract that implements ERC-721". This one does illustrate the challenge of using some of the other words:
This YYYYYYYYY could be "asset". The XXXXXXXX is not the same as the XXXXXXX above. It would be "token". The last YYYYYYYY would be "assets".
The XXXXXXXX is smart contract, and YYYYYYYY is asset. |
I think these three illustrate the spectrum of owned object to property deed very nicely. On the owned object side of things, you have CryptoKitties - their tokens have functionality embedded into them, and actually describe what you own directly, without outside help. Here, the object and the ownership are one and the same. "Deed" does not make sense here (and, given that the contract code is publicly reviewed, will cause actual confusion among users). In the middle is CryptoCelebrities. The thing people care about is the visual representation, and that's mostly divorced from the contract itself. However, the only underlying object is the object in the smart contract itself, and the UI layer that translates this into what you see. And that makes it closer to CryptoKitties than to ÐWorld. "Token" makes more sense than "Deed" here. On the other end of the spectrum is your ÐWorld. Here, the smart contract is a true deed - it's recording the title of a piece of property, and is only enforceable in a legal sense. "Deed" makes more sense than "token" here. While property management is a valid use case for ERC-721, it will be the minority of use cases - the vast majority of uses in 2018 will be similar to CryptoKitties/Celebrities. And in 2019 and beyond, the applications will tend to be more like CryptoKitties and less like CryptoCelebrities |
|
I don't really think we have this discussion just because of the name of smart contract. What if we are faced with the more underlying issue? I mean we consider only few use cases right now. Most of them are not really Deeds. But in the next year we will find strong use cases where tokens would reflect real deeds instead of pictures. It would be good deeds, or bad deeds, talents, and decisions. What is the fundamental difference between Deed Token and Asset Token? |
|
@7flash We have delayed the discussion of naming substantially until very late in the process. At this time we have a very good understanding (and good documentation) of potential use cases. Of course more will always be dreamed up. But you can be sure that MANY people have reviewed and discussed the ERC-721 standard and how it will be implemented. |
|
To put some bearing on the matter, so that the choices can be extended. Distinguishable = Non-fungible = Unique If the metadata can specify divisibility within the unique asset -- as it's own subclass or method -- then isn't that the antithesis of what the proposal set out to accomplish? There's an obvious paradox at play. If you enforce structure onto the metadata, then you imply a value method, which again brings one into the question of ubiquitous growth versus finite divisibility. |
|
Beyond that, why not Merit or Part, as each can be made particular in english? |
|
By the same token: |
|
@ravachol70 To clarify, the standard metadata now has one distinct URI for each deed/token/asset. There is no concept in divisibility of any deed/token/asset. |
|
I don't like the implied agency (or the idea of an agent acting in absentia) to accept the decision to call this construct a 'deed'. Why not drop the legacy flourish and go with something very dry, logical, and completely basic? |
|
Seems the consensus is heading towards some usage of 'token'. I personally prefer 'Unique Token' over 'Non Fungible Token' simply because it'll be easier for most people to grok. |
|
I am against token and asset because in my opinion they are too broad of a description (All ERC20 are tokens, bitcoin is an asset...). |
|
Token is to ERC20 as WETH is to ETH. Take your ERC20 tokens and leave the ERC721 standard to its own devices. I propose Ticket, which I believe to be neutral and flexible enough to cover it and Ciwb, from the Welsh for cube -- just because it's a natural 2nd-order primitive and, at this point, why not? Everybody holding out for an ERC721 token is simply trying to give weight to a weightless matter. Everybody calling for a deed is trying to put my metadata into a pre-existing legal container for the sake of a status-quo and frankly mundane narrative. The value that anything deserves should remain free of as much, especially considering all of the constraints in the default understanding. The money is not in the distribution of weighty matter. If it is, then this is the hard-line between fungibility and non-fungibility. Give me a ticket for that kitty, please. Do you have a token? Ciwb it and make it unique. |
|
Let's use Deed for tokens that anyone can mint and Asset for tokens with fixed supply? |
|
Hello everyone. Thank you so much for participating in the discussion, seriously. It was important to me to test the status quo and see what options are out there. Here is what I learned:
End result:
I hope you can accept this result. And let's move forward with solidifying the standard. Onward ➡️ |
|
"Consensus was found to use an existing word with a recognized meaning." Source? This has been a decision by single-vote proxy. |
|
Nous as late suggestion. No commentary available. |
|
@ravachol70 Clarification: some people set forth novel words or new names that do not yet have an established meaning (so that we could give them meaning with our new concept!), however the majority favored word that were already well known. Sorry, this is what I meant to say! Brevity sucks! |
|
Sorry for delayed input. Good to see it is being concluded.
I am for Non-fungible Token.
Not sure about deed word.
|
|
Neither deed nor token done right, please. "this isomorphism is not natural" |
|
Please consider that ERC721 could be the functor at the heart of the Yodeda lemma : the representable. With the use of the word 'token', the object is demeaned as codified representation, the debt of its appropriation to the blockchain pre-supposes the unseen settlement by society. All that to say: there's a paradox here that needs to be broken down into something with proper resemblance to logic. In 2018, I can tokenize my left foot independently from my right. The burn rate might be similar on each leg of the market but still the two will only communicate through some prescient singularity. Where is that contained in the notion of token ownership? As beings whose livelihoods are likely to be wholly placed into the blockchain by 2028, why not remove these extra chains (eg. of language) as and when they do pop up? The open-source craft of divvying out logic willy-nilly has led us to us debating whether or not two things are the same when both (and all defining parameters for that matter) are generally misappropriated. That is: they've seemed to become the same while, in reality, we've only just begun to discuss how different they are from each other. What if the distinction is an actual non-representable figure? You can draw a circle around it all you want; that doesn't make it real. The token is always, already involved and that's cribbing Wittegenstein, in case anybody is wondering. |
|
We're trying to open-source the natural language indicator to a difference between something that is defined and something that isn't yet developed. It's the same whether it's in stacks or in heaps: so why must it be so dimensionally limited, i.e. by the limited connotation itself of the work, which calls everything a token? ERC20 can thereby be left with its own common parlance and ERC721 left to its own devices -- esp. as that concerns overall readability of the resultant code. I know from personal experience that to debug/audit code that includes entities that are named so similarly as be be mistaken, then such a thing will occur and causes panic and confusion when up for debate. The goal is not to embed the nature of tokenization into everything. |
|
Why not Chattel? |
|
|
I also suggest two more words. The first is Prop, which I found as a nice translation of the chinese 抵. (I believe this is close in understanding to deed: something worthwhile, offset.) The second is Breve which is a very basic mark, suitable for accumulation. (I had mentioned that 'Note' was on the list; this reinforces it.) |
|
@ravachol70 "Chattel", strictly speaking, excludes Real Estate but it's a great word, not in use anymore - it could be hijacked. This standard can not, without extension, be applied to either Chattel or Real Estate, however as 1) it does not accommodate ceding the jurisdictional nexus of title and 2) multi-dimensional rights types that both Chattel and Real Estate tokens require are not in scope. For now, I personally think that "NFT" is a good starting point for naming this foundational standard and for building more tangibly useful extensions to deal with the vagaries of collectible Chattel and liquid Real Estate in further extensions. |
|
@d4v1dg I appreciate the recognition that this is a foundational standard. I find it very tasteful that you also mention building and tangibly useful extensions. What seems clear from your understanding is that this is indeed an architectural standard. We should go with that spirit rather than "making do and moving on" as programmers generally need to do (by the very nature of producing code). Let's try to consider or include non-fungibility as precluding the existence of an existing phenomenon that needs to rolled into the codebase -- not as something that needs to be labeled simply to carry on with the success of something else (e.g. via tokenization, even of the name). To my ear and understanding, such continual cross-referencing to this idealised economic sub-class or sister category which is supposed to match up and be overlaid with the legacy types and business models: it seems such a wordy hassle. There is no need for a continued use of the concept of token, let alone the word or label. On that note, there was an appropriate suggestion from a few days ago:
Today, left-over from other research, Der Ersatz, and not much deliberation much at all, I found this page full of sturdy references: https://www.idixa.net/Pixa/pagixa-0508281130.html "Difference is neither a word nor a concept: it is a clean beam to think the most irreducible of our time" - J. Derrida The term pre-supposes the other working notions (Gas, Ether, Token) through the 5W1H -- the five W's and one H. Furthermore, the Mantle must use two opposing supports of equal bearing. |
|
|
Yeah.. "Chattel" is not the greatest word association you are all looking for.. |
|
I think quite a lot of these suggestions are pretty obscure. The concept of a token is just starting to go mainstream so I think it could help if that word stays.
Right now, that is being prepended with "Non-fungible" and shortened to NFT. This is starting to spread throughout the community.
After talking to dozens of people about tokens and testing "Unique" and "Non-fungible", I can safely say that people find the concept of a "unique token" much easier to grasp. We are all familiar with terms like fiat and fungible because we think about the nature of money and value all the time. Normal people do not think about it and are certainly not aware of the concept of fungibility.
Just some thoughts based on user testing :) Enjoying this discussion a lot.
Sent via Superhuman ( https://sprh.mn/?vip=richard@balance.io )
…On Thu, Mar 01, 2018 at 13:39:48, Alex Van de Sande < ***@***.*** > wrote:
screen shot 2018-03-01 at 10 39 01 am (
https://user-images.githubusercontent.com/112898/36847351-cd0dbac2-1d3c-11e8-9097-f2b8e5071cd2.png
)
Yeah.. "Chattel" is not the greatest word association you are all looking
for..
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub (
#2 (comment) ) , or mute
the thread (
https://github.com/notifications/unsubscribe-auth/AAavkFRLsZkonHkzFjrWyy59TInGLxCgks5tZ_ojgaJpZM4SQjHs
).
|
|
I hear ya @alexvandesande. Don't worry, it wasn't my first choice. Regardless, all of this doesn't exonerate 'token'; I remain against it as much as 'deed'.
English, by the way, is a slave patois used in very many countries. Maybe we can get some sanskritization happening in here? I have tried to avoid arbitrary translations. |
|
My only reservation with Mantle is that it's somewhat awkward to collect. In this case, I'm leaning more towards Prop, especially as it's synonymous with beam (mantle) and can also be construed as a short-form way of saying "some Propitious article", which is next to Auspicious, which is next to Token, which half of the coders have already voted for in advance. Lastly, it also holds space as the nickname for the term "theatrical property" and so provides a physical analogue in the realm of the virtual.
|
|
I suggest using the term "nungible" for a non fungible token. It is shorter and people will get used to it. It turns a negative description of something into a single short name. |
|
I've taken a look at the implementation. 'Token' is everywhere in the code and 'Deed' is plugged quite readily alongside in the commentary -- so the election has been rigged. Not only has the standard been implemented in the wrong way, it has already rolled out, copy/paste, search/replace. The bounty is set to expire in less than a 48hrs: the first deployed ERC721 contract with the moderator as the author and deployer. How many experts out there are really debugging it? How many even know about it? I'm in the Lobby, looking around, and I see zero interaction. I'm not particularly new to programming and have thankfully little experience with the runaway Open Source mentality, so please consider this to be blank-space criticism. Where is the author's prerogative other than to have the very standardisation goals met quickly, first and foremost? This makes no sense to me as the copyrights have already been appropriated, even before the bug bounty is effectively offered. There are people trying to read this code and reconcile the implications with the legal jargon. I mean no offence in this analogy: if the task was food preparation, ERC721 would be a fast and cheap french fry of an order, a flabby cudgel of empty caloric value. This is not for deployment. The bounty program is open for only three days. People are struggling to keep up with the education that arrives with the explication. Technically minded people are becoming confused by the sub-classing and super-classing -- in the language usage -- and by the containment of the things that they value; they are having trouble reconciling the features of a new abstraction with the barely retainable notion of the original ubiquitous token. Yes, we all know what NFT stands for. It's a construction and we have yet to name what it's become in regards to Ether directly: a signifier of final cause. I would like to pragmatically avoid the long-term code fatigue aspect of this nft- prefixing and token centricism, especially as ERC721 has, in my opinion, a much broader wet space. With regards to the vastness of the aether, the token is a sunken ship.
|
|
Here are the normalised choices as presented in order of appearance so far.
*I've set these according to the suggestion's context. For example, Escritura is a common translation of 'deed' and so should not be considered new even though it was a new term amongst the others when originally mentioned. ** I vote for these. Yes, they all happen to be of my own inspiration but the work was done so intentionally; therefore, it's not surprising. Anyway, it's not to hold the acknowledgements captive. This is volunteer work and I'm as nameless as the rest of the hive. Lastly, note that I would withdraw the other eleven suggestions that I made and/or vote against them if they were presented to me at this time (as is) but that's neither to regret, discourage nor preclude further contemplation -- only to lend better credence to the suggestions left standing. |
|
I would also submit Policy as a very generalisable term. The neutral policy would naturally encapsulate the underlying contract and would go half-way to appeasing all of the deed mongers. |
|
@ravachol70 The discussion was concluded 8 days ago, I don't think you're going to get anywhere by continually introducing additional options. |
|
Hello @buhrmi. I see you had supported the change to token naming. If you are now satisfied with the result, would you please consider to remove 👎 on ethereum#841 |
* Proposed EIP for address and ERC20 transfer rules * Update eip-X.md Updating creation date * Update eip-X.md (#1) * Update eip-X.md * Update eip-X.md * Update eip-X.md Rule -> IRule consistently fix missing links improve abstract * Update eip-X.md typos small improvements adds implementation section
Discuss / vote on the name for ERC-721
This pull request changes the naming of ethereum#721 / ethereum#841 from what it currently is to something else. I will use the consensus achieved here to update and merge this pull request into ethereum#841. I hope you will respect the transparency and fairness of this process and also support the final choice. My role in this process is limited to bringing intelligent and hardworking people into this discussion. I will not advocate for ANY of these choices.
Choices (alphabetical order, a random contract listed for each):
Ground rules: