Not sure if this was clear here, but when you create an account on this site, the docs do sync to a server automatically. The encryption key is derived from users' passwords (Userbase handles that part). This way if users lose their locally persisted data, so long as they still have their password they're good to go (or vice versa). It's only if they lose both that they lose their data.
But still, given your experience, it does sound like it would be challenging to sell an exclusively offline product using browser storage. Appreciate the insight.
There's also a challenge here where browsers will literally just start deleting data from IndexedDB if a user starts running out of space on their hard drive.
I added this statement: "Keep in mind this is a demo app subject to change in the future. You can download your docs at any time." as a warning to anyone who might be using this site with expectations that the data will be persistent.
Ultimately I figure the only solution for a production app is to either do a better job educating users on these issues, or scrap the offline persistence selling point entirely (and just rely on it as a cache).
For starters, I think the number 1 most important thing is open sourcing code. Everyone should be able to read the entirety of the software's code themselves, verify the claim, build the app from source, and run it locally if desired. Unfortunately not everyone will have the skill to do this, so many will likely have to resort to trust in others to do this verification for them.
Given that this code also runs on a website, you have to trust that the code loaded from the website is the same as the source. Notice this statement I also added in there: "Please be aware: if someone evil... takes control of hushdocs.com, they can access your docs!"
There are some ways to mitigate this:
1. I could offer a native downloadable version of this site, and you can verify the downloaded file's integrity using a hash and signature.
2. I could add a service worker to the site such that after the website loads the first time, you have to manually allow a new version of the site (that would also be signed by me) to load and run in your browser. Check out this blog post for ideas on how to go about this: https://blog.airborn.io/post/2017/08/03/Transparent-Web-Apps...
3. I could host the static site on IPFS and you could rely on a browser extension to ensure the correct site loads, described here: https://blog.cloudflare.com/e2e-integrity/
But at the end of the day, to really verify the claim yourself, you have to be able to read and build the software from source yourself. Otherwise, yep, you are trusting that I'm giving you code that's doing what I say it's doing.
For one, conflict merging sounds like it could be a nightmare with that approach. Like imagine you have 10 people working on a doc at the same time, and they each have different versions. Hush Docs takes care of keeping a doc in sync out of the box.
For two, real-time collaboration isn't possible with that approach.
For three, it's end-to-end encrypted if you care about that sorta thing.
Also just a note, comments and change tracking could be implemented pretty easily if there's interest. I spent like a week and a half on this, it's pretty barebones.
Hush Docs is an offline-first, private Google Docs alternative. You can turn off your internet and the site works totally fine. Your docs are stored in your browser (using IndexedDB/Dexie.js), so you can create and edit your docs whenever you want, no internet or account needed. If you decide you want to sync across devices - or collaborate with others on a doc - you can create an account (which you do need internet for). Your docs are then end-to-end encrypted, so no server ever has access to your plaintext docs! Docs stay in sync using CRDTs (Automerge) - collaboration should be smooth even if someone works on a doc while offline for a while (like on an airplane with no wifi). Code is 100% open source. [1]
I decided to create this because someone asked for a tutorial on how to create offline-first apps using Userbase, which is a developer tool I’ve helped work on to make it easy to create secure and private apps [2][3][4]. I started out with a super basic tutorial, and then one thing led to the next and ya know: Hush Docs. A basic tutorial on how to build something like Hush Docs with Userbase is coming! [5]
Hush Docs is super barebones, so I’m calling it a "demo." If there's serious interest, I’d happily put more time into it and get it into something more production ready.
For reference, here are a couple other sweet E2E encrypted Google Docs alternatives I came across while scoping this out:
Airborn did something especially cool worth pointing out: they used a service worker to guarantee that code loaded into the browser came from the latest GitHub commit. [6]
We use scrypt for password hashing. From the scrypt paper (which keep in mind is assuming hardware from 2002, and isn't assuming an attacker is using ASICs which have been developed since then), the estimated cost of hardware to brute force guess an 8 char password in 1 year is $4.8 million with our chosen parameters. [1]
Ultimately we strongly recommend that developers using the end-to-end encryption mode of Userbase recommend their users use a password manager, since losing their password means losing their data (and we try to make this extremely clear to any developers using Userbase via the admin panel and docs). A password manager randomly generating passwords makes this a non-issue.
But alas, we do recognize not everyone will, which is where scrypt comes in to play.
Adding to this, properly defending against enumeration also sacrifices a level of security in addition to privacy, since the average user would likely need to store some additional identifiable data (such as an email) in our database in plaintext that would be compromised in a breach.
>You built a safe with no guard... and you're telling attackers where you put the safe. You are almost guaranteeing someone will compromise it eventually.
Userbase is built on the assumption our entire database and server will be compromised, and the attacker would still not be able to access protected user data. Validating that we protect user data in that scenario was the goal of our security review. [1]
On top of this, requiring users to provide an email or some other identifiable means to sign up, which is the practical way to defend against enumeration, compromises a level of privacy AND security in the average user (since this data would be leaked in the event of a breach). So this is a significant tradeoff, not as simple as one way is secure and the other is not.
Finally, we recognize the impact of allowing user enumeration. We will offer protection from user enumeration for those who are comfortable with the tradeoffs in user experience, and with sacrificing a level of privacy and security for their users.
>Does it download the entire history of all transactions and replays them into the local database?
This is what clients do initially, until the database grows in size. Every time the transaction log increases 50 KB, the client takes a snapshot of the database's state at a particular point in time, compresses and encrypts it, and uploads this state to the server. We call this a "bundle". This way when clients reopen a database, they load from the bundle first, and then apply any new transactions that come after it. Rather than needing to query for the history of all transactions and decrypting them individually and reapplying.
>Hot do concurrent modification get resolved (several clients try to modify the shared stage at the same time)?
The server assigns each transaction a distinct sequence number via an atomic operation. So clients always apply transactions with the same distinct sequence number, in sequential order. The client relies on this to enforce uniqueness and versioning. Only the lowest sequence number itemId gets applied to a database if 2 clients insert with the same itemId at the same time, and similarly, only the lowest sequence number version of an item gets updated or deleted if 2 clients update or delete the same item at the same time.
With regards to bundling, it's a bit more complicated and there are layers to our approach in safely handling it under high concurrency. When a client uploads a bundle, the database records what sequence number the bundling took place at so clients can use it to retrieve the latest bundle. And the server retains copies of bundles at prior sequence numbers. This way if two clients attempt to open a database right around the moment a bundling process completes (client 1 receives a bundle at lower sequence number, and client 2 receives a bundle at a higher sequence number), both clients receive the same set of transactions regardless. The server sends all transactions in the log after the bundle sequence number, so client 1 just needs to decrypt and apply more individual transactions to rebuild the state compared to client 2.
Some may find this interesting too -- we specifically test for safe concurrent behavior across 2 clients using a makeshift testing framework that opens 2 browsers at the same time and does some neat tests: https://github.com/smallbets/userbase/tree/master/test
If you clone the repo and run `npm run test:concurrency`, it will run those tests and output test results to the consoles of the 2 browsers.
Userbase is built on the assumption that in the event an attacker compromises the Userbase server and database, the attacker would not be able to access protected user data. We chose this assumption to build on because we figure that users and developers alike should assume that data stored at rest in cloud-based databases will eventually be leaked, as we've seen countless examples at almost every major company. Thus, we figured the default assumption is that usernames would not be expected to be private (and so yes, to answer your question, user enumeration is currently possible).
Additionally, practically defending against user enumeration beyond rate limiting sacrifices a level of security and privacy (for example, by requiring users provide an email to sign up to your service, or through some other means that likely ties the user to an identity and storing this in our database in plaintext), rather than allowing them to use pseudonymous usernames alone.
While we do recognize username enumeration is an issue (because users tend to reuse passwords from other sites, or don’t want to be found out using a site), we concluded that properly defending from user enumeration by default would have too material of a negative impact on user experience for little gain on top of what we already provide in way of protecting user data, and instead focused on defending against potential follow-up attacks by limiting brute force login attempts, and recommending that you tell your users to use a password manager at sign up.
The most significant place defending against enumeration affects is during sign up. When a user’s account already exists, we say the username is already taken, which isn't possible when properly defending against enumeration.
We're planning to allow you to enable email verification in your app if you want to, so users will need an email to successfully create an account. Once that's in place, we'll defend against enumeration more concretely. There are other places in addition to the salt retrieval that would be modified in similar fashion. For example, password reset will need to always successfully return even if a user provided the wrong username, and sharing a database with another user will always successfully return even if the other user doesn't exist (e.g. from a typo).
openDatabase loads the database's state into memory from the server, and then keeps it in sync with the server using the Web Socket. When you insert a transaction via one of the database operations, the server assigns the transaction a monotonically increasing sequence number, and broadcasts the transaction along with its sequence number out to the clients connected to a database. Clients then apply transactions in sequential order to the local state, and keep track of the latest sequence number applied. When a client goes offline and comes back on, it automatically reconnects the Web Socket and re-requests any transactions that it may have missed above its currently applied sequence number. We handle reconnection logic automatically under the hood, retrying on failure with backup delays.
I'm also working on an offline-first Google docs alternative that will write to IndexedDB, and stay in sync with Userbase using CRDTs. The tutorial on how to do it will be here: https://userbase.com/docs/
Userbase is a tool for developers to build secure and private apps. We launched 1 year ago [1], and have worked hard to widen its use cases. Userbase offers built-in user accounts and authentication, an end-to-end encrypted zero-management database, file storage, streaming and sharing, and logic to process and manage subscriptions (via Stripe). All Userbase features are accessible through a simple JavaScript SDK, directly from the client. 100% open source, and all platforms are supported (browser, iOS, Android, desktop).
Today, Userbase can be used in apps that don’t require end-to-end encryption, for those who like that Userbase can handle authentication, data storage and real-time syncing in a few lines of code.
We also completed a security review by an independent team [2], and wrote up a comprehensive specification of our architecture [3].
Personally I joined in working on Userbase to store end-to-end encrypted data in a performant way for an accounting app. Under the hood, each write to a Userbase database is an append-only transaction to a log stored in DynamoDB (therefore constant time), which is then pushed to connected clients over a Web Socket. Each client then decrypts and applies this transaction to its local state of the database in memory (real-time syncing is provided out of the box). In this process, the server ensures each client receives transactions in a consistent order, 100% of the time. This is unlike some of the (very awesome) decentralized alternatives that exist today (OrbitDB, GunDB, Scuttlebot), which generally rely on CRDTs to stay in sync, and CRDTs can be pushed in any order. For certain applications, the consistent ordering guarantee a central server provides may be extremely useful (such as in an accounting app), on top of the added reliability and performance.
Invoices are encrypted locally using your password, before being sent to the server for storage (thanks to Userbase, a super simple tool to add user accounts and persistence to a static site https://userbase.com).
The goal with Prinvoice is two-fold:
1) Get a feeler for how much interest there is in end-to-end encrypted financial management software.
2) Demonstrate a sample app that uses Userbase for auth and storage (I'm also a Userbase dev). A kicker, if you check the dev console in the app, you'll see SQL statements in action.
It's still a relatively fat client that needs to get loaded at page load (handling end-to-end encryption + client-side SQL with sql.js), so there isn't too much I can do. But I think the site loads at a much more reasonable speed now.
I also noticed the site isn't working in webview and is exceptionally slow there, looking into that.
Not sure if this was clear here, but when you create an account on this site, the docs do sync to a server automatically. The encryption key is derived from users' passwords (Userbase handles that part). This way if users lose their locally persisted data, so long as they still have their password they're good to go (or vice versa). It's only if they lose both that they lose their data.
But still, given your experience, it does sound like it would be challenging to sell an exclusively offline product using browser storage. Appreciate the insight.