Hello and sorry for the delay. Omise is a payments processing business for Asia-Pacific; something sort-of like Stripe in the West. Omise GO is a blockchain-powered, white-label ewallet platform for the region, which will use fiat-backed tokens (FBTs). Two separate businesses under one roof. We believe that the ewallet business supports that payments business and vise-versa; white-label ewallet customers will for example be able to immediately pay to all Omise merchants.
Thank you very much. :) We hope to offer more support to open Ethereum projects in the future, importantly including proof-of-stake research. If you have any questions or comments please feel free to join our Slack.
Hi! So sorry for the slow reply, it's been hectic, to say the least..
(1) It's not clear yet if there will be a requirement regarding a token minimum bond for validation, but there is definitely an advantage to the validator to have more tokens staked, in the same way that there is advantage to a miner (in a PoW network) to have more hashpower aimed at the network. And token staking can be delegated as well..
(2) Validation happens per block, not per transactions.
(3) It's essentially the same process as a PoW design (like Bitcoin and current Ethereum), or a PoS design (like Tendermint, or the future 'Casper-based' Ethereum). Beyond this, please wait for Joseph's upcoming paper this summer (mentioned in the whitepaper) which will discuss validation in detail.
Thank you very much for your questions!
The tokens created and allocated now, as laid out in the sale document, are it; we have no plans to create more tokens in the future, and doing so would require a fork (or 'upgrade' as some people might prefer to call it when there is some technical necessity).
The rate of return for processing transactions will depend on fees accepted by validating nodes and the tx throughput, similar to how it works in Bitcoin and Ethereum. Validating nodes that have more coins bonded (staked) will have an advantage in receiving more of the tx fees, yes.
The key word there is "pledged" (not "accepted") - that was the sum of rough figures Bitcoin Suisse gave us that people asked for (they can't be specific due to client confidentiality) and value from people that pledged to us directly over past months (who didn't register with BS in time before registration closed), factored by rise (and fall) in ETH value. To be clear, we are a bit overwhelmed by the level of interest, though we really appreciate the clear support. We will make an announcement on Monday about how we can proceed in the best interests of our community and project first and foremost. Please wait till then!
Since the video is recorded for posterity, he chose to speak Japanese because it's the language he is most fluent and expressive in. In person he speaks English (just with a heavy Japanese accent), and at the office, people go back and forth between Japanese, English, Thai, Chinese, and some others ... everyone can communicate just fine :)
Come to our slack, Jun also chats in English there!
Simply put, it's expensive tech to develop and build business on, both in terms of financial commitment and community commitment. We spent over a year doing dedicated research on it in our Ethereum lab, before we decided a token sale would be the perfect way to both kickstart this push to commercial development and to build community.
Ethereum is great for checking our validator proofs and maintaining economic security, among other things, but it doesn't have the scalability we need now, and we can't wait for Casper to be ready and advanced to the point where we can use it for our needs (or the needs of decentralised exchange in general).
From the sale document: 'The underlying OmiseGO network consists of cutting-edge crypto-economic constructions, including an on-chain decentralized exchange built into consensus that bonds to Ethereum for economic security, and a high-volume payments network capable of wider uses, with the ability to scale via off-chain channels.' (https://cdn.omise.co/omg/crowdsaledoc.pdf)
A short answer is that Ethereum doesn't have an on-chain exchange built into consensus (nor does anything else). The OMG chain will. If you read the whitepaper, you may find that what you are looking at is the successor to Lightning Network. What Lightning is to Bitcoin, the constructions underlying the OMG chain are to Ethereum... this is not merely a rapid, high-volume payment channel system, it's a scaling strategy. Unlike Lightning though, it's not something that can be thrown on top of another blockchain.
The OMG chain will offer a kind of Ethereum scaling that goes beyond just what we (Omise) plan to use it for, while remaining very linked to Ethereum, which it will rely on for economic security (and therefore, economically, OMG chain activity drives value to Ethereum and ETH).
Thank you for your questions!
The ICOAGE presale was meant to be China-oriented (because thousands of Chinese buyers were facing obstacles registering their docs with Bitcoin Suisse for weeks), but it was also meant to be open to a global audience. However, shortly before the sale started ICOAGE had problems with their email verification system necessary for non-Chinese users to pass the Sybil test (non-trivial KYC).
Allocating to Bitcoin Suisse and ICOAGE isn't a perfect system but it is unquestionably superior to the ways BAT, Bancor, or Status did it. We are able to avoid having a whale-dominated sale, and we are doing it without recklessly raising more money than we actually need and/or clogging the Ethereum network for days. Sorry you missed the contribution period but many people did; we did zero advertising, it has been pure word of mouth, and people who have been following us for a while (since we announced at EDCON Paris) were able to get in. Hope you can pick up some OMG later and come help run the network when it's time to use the token.
Yes just to clarify, that $45M figure in crunchbase includes the $25M token sale that OmiseGO - the blockchain unit of Omise - just did. Otherwise, Omise has previously raised around $20M.
The $25M raised in the token sale will be used exclusively to build and maintain development for the OMG network, which will be a totally open and public (permissionless) blockchain that we won't control. The funds won't be used to fund other Omise operations outside of OmiseGO.
Yes, we are the first Japanese startup to do an Ethereum project. We've been involved with Ethereum since 2015, and actually we were one of the first 3 companies in any country to support Ethereum (the other 2 were Microsoft and Wanxiang Group in China).
Unlike in China and Korea, Ethereum hasn't really taken off yet in Japan.. however, we hope to be able to do our small part to help.
Though much of our business is also abroad, in other parts of Asia-Pacific, we had a Tokyo meetup in June and are trying to do a 2nd meetup sometime by the end of summer or beginning of autumn. Might have some special guests too (hint: see who our advisors are). Please stay in touch!
Article has some good points but one thing that should be pointed out is that, unlike several other projects, our product isn't for 'the Unbanked' ... it's for Unbanking the Banked!
From the beginning of our sale document (found at https://omg.omise.co):
'At OmiseGO, “Unbank the Banked” means disrupting banking services: providing alternative financial and digital commerce tools for everyone without having to go through traditional banks, institutions and card networks. OmiseGO is a next-generation Ethereum-based financial platform enabling real-time, peer-to-peer value exchange and payment services agnostically across national jurisdictions and organizational silos, and across both fiat money and decentralized currencies; enabling true financial inclusion as well as freedom from monetary limits that are not aligned with the people's interests.
'Our target customers are both the 73% of the Southeast Asian population who do not currently use or have access to formal financial services (the so-called “unbanked”) and the 27% of the population currently using formal financial services (“banked”) because it's been their best option until now.'
A large number of Western banks will flag any activity abroad for fraud unless their card-holder notifies them in advance travel plans. Unfortunately we can't stop this practice, nor is it necessarily a good idea, as it does catch a lot of fraud that happens in high-risk places. All the more reason for us to develop better payment solutions in Asia-Pacific!
Almost! Joseph Poon is the guy who invented Bitcoin's Lightning Network (original paper here: https://lightning.network/lightning-network-paper.pdf), and Raiden Network is the Lightning concept re-written for Ethereum. Both concern enabling a way to do secure, cheap, and super-fast transactions using 'payment channels', which allow network actors to transact do a lot of transactions off-chain then settle on-chain.
However, what Joseph is working on with us is different, and goes beyond payment channels: it's a decentralized exchange that happens in a blockchain itself (in the consensus rules). You could say that it's the successor to Lightning Network - a scaling solution that makes full use of Ethereum's scripting capabilities. See more here: http://www.coindesk.com/striking-twice-lightnings-joseph-poon-takes-on-ethereum-exchange-project/
Beyond that we can't say too much yet, more details later this summer!
We are definitely going hard.
We are also racing to make this the very first real, completely mainstream, product using Ethereum.
Paysbuy was also the first digital wallet service to open in Thailand. It's historic, as the takeover marks the moment that the first digital wallet service will also become the first digital wallet service to decentralize onto Ethereum.
From a community angle, we're launching OmiseGO to support Ethereum scaling and to provide a decentralised exchange, that is not only totally decentralised but also has correct incentive alignment with Ethereum (because it uses the Ethereum mainnet for economic security, OMG will not siphon off value from ETH like most other tokens have the potential to, but will rather support it).
But from a business angle, we're launching OmiseGO to support Omise Payments and all existing stakeholders, including our existing and future merchants, such as McDonald's in other countries in Asia - McDonald's Thailand is just the start. Through OmiseGO, all Omise merchants will be able to seamlessly accept payments in multiple currencies, including ETH or BTC or other cryptos, without needing to know what their client is paying in.
And any other payment gateways, including Omise's biggest competitors, as well as global giants like Stripe or Cybersource, will benefit from using OmiseGO/OMG, and are welcome, no partnership or permission needed. The OMG chain will be an entirely public and permissionless network, connected to Ethereum.
OMG is a token that secures a public permissionless POS network, whose primary use case is to host a decentralised exchange for digital fiat and crypto alike. It is backed by significant mainstream financial firms in Asia who will use it as the backend for existing digital wallets - think Venmo, Google Wallet, WeChat Pay (examples of digital wallets only) - and for hosting digital fiat tokens for their existing merchant payment network.
OMG holders can use OMG to validate the network. If they validate a block correctly, they get paid the tx fees from that block (but if they validate incorrectly, they get punished). Like mining but without the need for hardware (and if you do bad stuff the network doesn't accept, you don't burn expensive electricity, you lose money in other ways).
Because you may want help validate the OMG chain, and process transactions going across the network and through the decentralised exchange that's baked into its consensus layer. It's unlikely OMG will be used as a currency the way ETH has started to be (though that wasn't the original intent for ETH either).
Roger Ver got into crypto to help bring about the rise of fair money, not to advance any particular clique or faction. He threw his support behind OmiseGO because the project isn't about Bitcoin vs. Ethereum vs. others, it's about fair finance for everyone. Quoting from https://omg.omise.co:
"The OmiseGO network is intrinsically agnostic between fiat and decentralized money: as far as adoption and usage go, the system is constructed so that the best currencies will win."
That's why Roger joined us.
Bitfinex listing came as a real surprise to us but a very welcome one. Judging by their announcement, they are in this for the long. They clearly understand the benefits of the tech and financial relationships we're developing, so we really welcome their support.
I think Bitfinex have far better things to do than try to pump OMG, especially since we refused to raise more than 25M.
Ps. we sold in 'presale' for the anti-Sybil properties, and the sale was totally open and public.
Wow, you really did post it here. Ok, so for redditor context, that would be like if a foreign country's Stripe, or Square, bought a Venmo connected to the userbase/user data of AT&T or T-Mobile, and said, "Hey we're going to make this an Ethereum-based thing now."
Staking token, used for bonded validation of the (decentralised exchange happening in the) network.
It depends on the amount (in tx fees) that it will take for validators ('miners') to include transactions in a block in a reasonable time, and is dynamic, which is to say that it depends on the validators, like bitcoin or Ethereum mining depends on the miners, the network congestion, the transaction size, etc, and will probably constantly adjust. More specific than that is near impossible to model at this point, and won't be in OmiseGO's control.
In short, it's too early to say.
No, a validator would have to stake their coins by putting them into bonded deposits (as a kind of insurance against validator misbehaviour). You lose money if you try to cheat the network and get caught.
Most users won't know their everyday mainstream wallet backend is running on the OMG network.
National regulations aren't really an obstacle for OmiseGO, as they're for prospective wallet providers to handle themselves, and more than enough already have (that's actually part of the problem!). From a wallet provider's perspective (especially a pre-existing one), the product is mostly a backend, facilitating interoperability and removing the limitations around their systems. The wallet framework is a nice-to-have; the real gem is the blockchain and the decentralized exchange inside it, finally tying together crypto and mainstream online consumer money, and allowing for mass adoption of both.
We have already encouraged several other major sales to do this. The response was very positive. Keep watching :)
In short, and not an exhaustive list:
Problems being solved on the crypto side:
Scalable decentralized exchanges don't exist yet. Preliminary ELI5 on our unique/brand new approach here (full paper coming very soon): https://www.coindesk.com/striking-twice-lightnings-joseph-poon-takes-on-ethereum-exchange-project
Fiat<>crypto exchange is hard to decentralize because secure, decentralized fiat is hard to make and distribute.
Problems being solved on the mainstream side:
Online money is hard to access, credit-based money is not currently suitable for most communities (high fraud/expensive), coordination problems between online money issuers prevent network effect and mass adoption.
Transactions are more difficult and expensive than they should be, especially for low value transactions.
Full reserve banking is not available to most depositors.
We expect that exchanges have a strong incentive to distribute OMG to their ETH holders, they make money in trading fees! So if it's technically feasible for them (might depend on how they have their ETH deposit system set up), they will probably do it. But this is up to them and their users, not the project owner.
We expect the pressure is unlikely to be but barely noticeable, if even that. At today's prices, the majority of people will get airdropped an amount of OMG worth cents. Very few people will get more than ten dollars, and the only people who will are likely to be people for whom ten dollars is not worth selling.
But don't read too much into it. Of course he supports us, we're the Plasma decentralized exchange, he's co-author of our paper. But please judge OmiseGO by its total merits, not just by who's got our sticker on their laptop. We guarantee there's a lot more going on! The tech is real, the biz dev is real. We don't do aggressive marketing (or any marketing at all, to tell the truth). We don't need to; OmiseGO isn't hype, or built on derivative tech. We are fiat-and-crypto decentralized exchange backed by a multinational mainstream payment company's business development and financial integration, pioneering the use of the flagship Ethereum 2nd layer scaling solution.
We will deliver something first class: a public Plasma network that is open and scalable and useful not just to OMG token holders but to the entire ETH and crypto community, fully justifying the purpose of an ETH token sale. Further, the work we will do on top of it will be useful to the entire populations of the countries we can reach, serving millions in their everyday lives, bringing crypto to the masses and bringing the masses to crypto.
And we're not leveraging legacy centralized tech like VISA/MasterCard, we're using Ethereum to replace it. With a decentralized solution that is almost ready to go to market, and run by you.
Forget the sticker!
Just for posterity, don't forget OmiseGO (OMG) - the Plasma decentralized exchange. https://twitter.com/omise_go/status/897003999403982848
(And we're supportive of all 3 other projects to the extent that they can provide useful services to people. Especially interested to see how Kyber's on-chain matching system will work out!)
Plasma works without Casper. Does not need PoS on the root chain.
Oops... The article missed the fact that the OMG network is a scalable on-chain decentralised exchange :-)
He skates, just mostly between the car and office these days.
Don't make too much of the stickers guys. Like someone else here said, we do zero marketing, so this isn't any different. Same as Vitalik's laptop, it's just taste, enthusiasm, and sentimental value.
Hi! Thanks for the great video talk. It's really inspiring to see community members engaged and supportive. Just in case it's helpful, we would like to offer a few clarifications on some topics discussed.
We moved from doing a simple capped 'first through the gate' on-chain sale to a 'presale-only' sale because of unacceptable whale attacks on simple capped sales. Because our project goal is to launch a public PoS network (that works on top of the Ethereum mainnet), getting a strong distribution of token holders is crucial; it would have been terrible if what happened with BAT and other simple capped sales happened to OMG, and so we thought about it very seriously and decided to run everything through 'presale' platforms for their property of non-trivial Sybil resistance. Other models were considered in long discussions with our advisors but ultimately we decided this would be the way to go. It is also important to please note that all 'presales' were 100% public (according to the terms of service of the platforms used); there was no private sale whatsoever.
The OMG network is the Plasma decentralized exchange, and contributes directly to Ethereum scaling via Plasma development and to Ethereum's economic security by driving value to ETH (on which hashrate OMG will depend for its own security). OmiseGO will use the network for our own business purposes but ultimately once it is built, it is out of our hands and we will have no more claim to its power than anyone else. Community service has been the chief spirit of our engagement in Ethereum since 2015, and OMG will be a 100% open and decentralized network, no less so than Ethereum or Bitcoin.
Our advisors have been strongly engaged in OMG development because, simply put, OMG is good not just for the Ethereum community and network, but for entire crypto space as a whole. Two of them participated in writing the whitepapers that describe what we are building: OMG network & Plasma. However, ultimately we take full responsibility for what we are building under their stewardship.
Thank you so much for all your support. We will work tirelessly to accomplish our goals. 🙇
The OMG network will be a Plasma chain that uses Ethereum as its root chain. Vitalik has called it an Ethereum sidechain several times but Joseph is adamant that that's not the right way to look at it. While we get the vocabulary sorted out, see https://plasma.io to understand what the vocabulary refers to. Alternately, this reddit comment provides a decent summary: https://www.reddit.com/r/ethereum/comments/6sqca5/plasma_scalable_autonomous_smart_contracts/dleskw5/
Hope that helps!
Sure, and even more accurate still to describe the relationship as root <> trunk, with branches (and various orders of twigs and leaves) coming off of the trunk. The only thing that connects to the root is the trunk, otherwise known as the Plasma parent chain (and the sub-chains are known as child chains.
Root, parent, children.
Possibly, or possibly it would get forked and switch over to a different root chain. Ultimately Plasma is root chain agnostic but it would be easiest if the root chain ran the EVM.
So an immediate difference between RSK (Rootstock) and Plasma is that Plasma doesn't depend on merge-mining, nor is it a federated sidechain (so it has better decentralization and security properties than a sidechain would permit). See section 4 of the plasma whitepaper draft (and relevant citations) for more details: http://plasma.io/plasma.pdf
That said, something like Plasma could run on Rootstock, if the goal were to use Bitcoin as the root chain.
However, OMG - the Plasma decentralized exchange - will use Ethereum as its root chain.
The OMG network is a Plasma chain (see plasma.io) with a decentralized exchange built directly into its Proof-of-Stake consensus layer. It is a totally open, public, permissionless blockchain, like Bitcoin or Ethereum, except it has the unique feature of being directly linked to, and ultimately depending on, the Ethereum mainnet for its economic security. In this way, OMG has a special incentive alignment with Ethereum that no other ERC20 token shares.
OmiseGO is the business building this network, so that we can use it for our business applications. Anyone else can use it, for any reason they see fit, too.
OMG holders can use OMG to validate the PoS network. If they validate a block correctly, they get paid the tx fees from that block (but if they validate incorrectly, they get punished). Like PoW mining but without the need for hardware - and if you do bad stuff the network doesn't like, you don't burn expensive electricity without reward like in PoW, instead you lose some of your OMG.
Hope that helps!
The token described in these articles is for sharing a form of company revenue. That is a totally different use from what OMG is for. The utility-value of OMG is as a network validation token, on a public decentralized exchange (that will serve as a single point of execution for trading liquidity which will be mostly unrelated to OmiseGO's own business uses).
Both the business use cases of the tokens, as well as the network use cases of the tokens, are completely unrelated. Please don't mistake them!
Yes. OMG functions as a security deposit which solves the "nothing-at-stake" problem, comparable to Casper (the planned Ethereum PoS upgrade).
It's the same as any staking token (or, ultimately, currency, for that matter). Ultimately, the expectation of value, which is a security assumption for the network, is predicated on the expectation of value. Relatively, if we are to ignore this 'turtles all the way down' aspect of the nature of monetary reality, we can say that the expectation that people will use and continue to use an OMG-based decentralized exchange with high transaction volume, is what gives OMG value. Hope that helps.
Yes, Plasma enables OMG to perform mass transactions that the Ethereum network isn't suited for. The OMG network isn't a system where the network token is used to propose trades (for which ETH could well be the token used), it's one where the token is used to secure the network consensus. Mechanically it works out better if it's a separate token from the root chain's token.
It's a protocol token for a protocol that really does need a token, yes, and this protocol is really far down the stack. Can't comment on Kik Kin but as for other projects... well, I hesitate to write it because it's becoming a bit of a meme for us lately, but "news coming soon" (sorry! It really is) [insert waiting skeleton here]
Good points in this response, especially concerning the business use cases. One point to add: the OMG network is actually a single-point-of-execution decentralized exchange for all fiat and crypto that runs across it. Unlike the projects you mention, which run mechanisms either off-chain or as applications in Ethereum, the OMG decentralized exchange uses Plasma to run on-chain, and directly in consensus, alongside Ethereum. So we're fully decentralized, and also fully on-chain, without either the security loss or performance loss. As a further note, the OMG network is also cross-chain compatible (it can also handle decentralized trades sent from other blockchains).
Tl;dr: OMG network is the Plasma decentralized exchange
OMG network does not require Casper. But Casper and sharding on the root (Ethereum) blockchain would help even more!
There is no Ethereum Foundation involvement with OmiseGO, but we have been supportive of the entire Ethereum community since 2015, and we genuinely believe correct and careful application of Ethereum can responsibly address many of society's financial needs. Plasma will be a big part of that, which is why we have been backing scaling and proof-of-stake research intensively, and why the OMG network has ended up the Plasma decentralized exchange running on a proof-of-stake network alongside Ethereum.
We hope that our efforts will lead to benefit for the entire Ethereum community, and the general crypto community as a whole. Further, we hope this work will help bring the technology and economic benefits accessible to people in the crypto community to mass adoption for the benefit of all people, going beyond financial inclusion and into harmonious financial equity.
The OMG network is a Plasma chain (see plasma.io) with a decentralized exchange built directly into its Proof-of-Stake (PoS) consensus layer. It is a totally open, public, permissionless blockchain, like Bitcoin or Ethereum, except it has the unique feature of being directly linked to, and ultimately depending on, the Ethereum mainnet for its economic security. In this way, OMG has a special incentive alignment with Ethereum that no other ERC20 token shares.
OmiseGO is the business building this network, so that we can use it for our business applications. Anyone else can use it, for any reason they see fit, too.
OMG holders can use OMG as a security deposit to validate the PoS network. If validators validate a block correctly, they get paid the tx fees from that block, but if they validate incorrectly, they get punished. It's like Proof-of-Work (PoW) mining but without the need for hardware - and if you do bad stuff the network doesn't like, you don't burn expensive electricity without reward like in PoW, instead you lose some of your OMG.
Hope that helps!
Nakamoto consensus was "too good to be true" too but here we are.
Potential threats: lack of network effect due to unpredictable coordination dynamics, short-sighted state-level interference early on, critical flaws in Plasma that can't be fixed.
Negative things that people complain about, as far as we can tell from reddit: perception that Jun (CEO) hypes the project by posting misleading photos or comments on his personal Twitter; no public repo available means we aren't committed to open source; redditors (presumably token holders) excitedly posting OmiseGO/OMG related content where people are tired of seeing it. All these things are genuine misunderstandings and we'll address them as soon as we have time (coming soon! sorry couldn't help the joke).
Thanks for the support!
We have a dedicated team working with potential implementers to explore possibilities with the OMG tech stack - we'll DM with contact info.
Dateless roadmaps are actually pretty common. To name a few (randomly chosen but well known) examples:
Stellar put out a single post at the beginning of 2018 with broad goals for the year, then periodic (postmortem) progress reports and announcements of completed major developments throughout the year: https://www.stellar.org/blog/2018-Stellar-Roadmap/
Cardano doesn't give dates but shares progress trackers similar to our own: https://cardanoroadmap.com/
Zcash shares product releases and upgrades at the time of release: https://z.cash/upgrade/
EOS shares past and present phases by season, with a one-sentence description of what they're working on in summer/fall 2018, no target dates for any specific milestone, and no indication of dates beyond the current phase
Presumably all of these projects have internal roadmaps with target timelines, as do we. Early on we chose to share those internal targets in the interest of complete transparency; but despite disclaimers that all targets were estimates and subject to change, it became clear that people were making investment decisions based on far-future milestone estimates. The cryptosphere is particularly susceptible to impulsive speculation, and we don't want to play a part in that if we can help it.
Between Jun's well-meant but admittedly overenthusiastic comments in 2017, to the widespread interpretation of best guesses as guarantees, we've learned our lesson. From here on out we will be as transparent as possible about progress on our task list; continue to provide detailed weekly updates; share our broad goals and the work to be done to accomplish them; and of course never excite. But we're going cold turkey on dates.
If your concern is limbo: we aren't in limbo and we haven't hit any major unexpected roadblocks. We're not hiding anything. But the thing about unexpected roadblocks is that they're, well, unexpected. On the off chance that progress slows because of something we didn't anticipate, we do not want to have made promises we can't keep. One promise we can and will keep is to continue to work hard every day to keep checking tasks off the list, and share that progress with you as we go.
The one thing that is out of our hands to be fully transparent about is announcements about specific companies that have decided to implement OMG. Dates of those announcements are entirely up to the implementers and many of them want to wait until they have a working proof of concept. All we can say is that we haven't had anyone pull out of PoC development and our efforts to bring volume to the network through partnerships and implementation support are ongoing.
Full interview here.
This is all leading up to Ethereum integration; the features we're working on are required to provide a smooth integration and save us from making assumptions about how providers want to run their eWallet.
Everything we’ll do to interact with plasma will require prior interactions with Ethereum - Ethereum wallet creation, creation and minting of tokens, ENS registration, deposit and exit of funds to/from Plasma all require direct connection with Ethereum.
The reset password feature introduced for end-users is only enabled when running the eWallet in standalone mode (which means there are no other web application running in front, users interact directly with the eWallet). By default, this mode is disabled, and that’s the way it should stay if you have a website where your users sign up, and only your website actually interacts with the eWallet. When standalone is disabled, end users don’t have passwords in the eWallet since they already have one in your system.
It’s up to the eWallet providers to take the appropriate precautions in terms of their own security, creating sufficient redundancies and ensuring coins aren’t lost - the same way that both private and commercial users of crypto do now when dealing with their Ethereum or Bitcoin accounts.