Here we describe, in abstract terms, concepts that must be communicated to the user during secure and insecure situations, so that they may distinguish them and be able to take appropriate action. We do not advocate specific concrete UX features here (though we give a few suggestions as examples), but we do intend for a designer to use these discussions to develop such proposals.
As explained elsewhere, we don't use a consensus mechanism for approving messages, and therefore the session transcript is necessarily a partial order and not a total (linear) order - even if there is (e.g.) a central server dictating a secondary linear order for display purposes.
This means that some messages may not be displayed directly below the one(s) that they are directly replying to. Most of the time this isn't a problem, and human users can naturally resolve any ambiguity. However, for cases where this is not possible, or if the session is security-critical, then there should be a secondary interface to indicate the real parents/ancestors of a message.
Some suggestions include:
annotations to indicate real parents, for example:
Alice: Who likes ice cream? Bob: Who likes SSLv3? Carol: Me! [2] # indicates replying to -2nd message relative to this one
highlight (gray out) a message's (non-)ancestors on hover / long-press
TODO(xl): generalise so these aren't specific to GKAs
It takes some time to run the GKA (group key agreement) to change the session key to include/exclude a user. During this time, messages may be sent (by anyone) to the older set of members. When the GKA succeeds, there might be a race condition like this:
Bob sends: (1) # final message of final round of GKA Carol sends: (2) # message encrypted to old members Alice recvs: (1) from Bob # Alices finishes GKA, and updates her view of the membership Alice recvs: (2) from Carol
In the above diagram, (1) and (2) are messages. Carol sends (2) before she has received (1) from Bob. When Alices receives (1), her UI will say something like "Dave has joined the room" and her next message will be encrypted to Dave - but later she will receive (2) from Carol that was not encrypted to Dave.
The UI must be able to show that (2) was sent to the old members rather than the new members. We know this information in the underlying system; the problem is how to indicate it in a non-confusing way.
Some membership operations may take some time to complete. If we exclude a member, and the greeting system does not immediately make this effective, then the UI for sending messages must clearly indicate that messages we send are still being encrypted to the old membership, that includes the member in the process of being excluded.
(The issue also exists for joining, but is less serious since any message intended for the newer group implicitly has permission to be seen by the older group.)
Some suggestions include:
- Disabling the send button whilst the exclusion operation is in progress. This degrades the user experience, and should be avoided.
- Have explicit notices that a member was included / excluded - in addition to the user list, which is typically not in the primary area of the screen. When the local user excludes a member, there must already have been an inclusion notice for that member, (hopefully) conveying to the local user that a temporary lack of an exclusion notice means they are not yet excluded.
We guarantee ordering - roughly, that messages are seen by recipients in the order in which they were sent - by buffering messages received out-of-order. Say Alice sends messages (1) and (2) to Bob. If the transport is unreliable and Bob receives (2) first, then we won't display (2) until we receive (1), so that we can display both in the correct order.
If we don't receive (1) after a reasonable time, we might try to communicate this to the user - e.g. a notice like "(2) received out-of-order; waiting for older (1)". Importantly, we must not display the contents of these messages in waiting, to avoid breaking our security assumptions. [1] We could also offer some advice on why this might be happening, but in a way that avoids giving a false impression on the exact cause: the symptom could either indicate an unreliable transport, or an attack by the transport, or even another member, but we don't know which.
On the other hand, the above may add unnecessary complexity to the UX, and burden the user with information that is hard to take action on. So, we might just simply ignore when messages are being held in the buffer - i.e. to not reveal this to the user at all, and pretend that we haven't received (2), to avoid frustrating the user. This may seem unnatural at first, but it is quite common for transport schemes that provide ordering guarantees - e.g. in TCP, if higher-sequence packets arrive earlier than lower-sequence ones, they are buffered completely invisibly to higher-layer applications and to the user.
As a failsafe to this simple option, we can trigger a warning if held messages remain in the buffer for too long. Optionally, we could automatically end the session and then display the contents of the buffered out-of-order messages - this is safe because there are no further messages to be sent, so there is no longer any chance of breaking the aforementioned security assumptions.
| [1] | Whenever we send a message, we also implicitly acknowledge all previous messages, in an efficient way. If we send a message after showing some messages but not their parents, then this acknowledgement is a lie that propogates and corrupts other members' views of the session and its consistency. See :ref:`consistency-without-reliability` for a more detailed exploration of this. |
After a message is displayed (authored by either us or others), we wait for everyone else to acknowledge it. During this time, the message is "not fully-acked", and we also know exactly who has not yet acknowledged it. Ideally, this set of people will eventually become empty, at which point we know that everyone has seen it ("fully-acked"). The question is how to display the "not fully-acked" status during the interim period. For example:
- We could gray out the message until it is fully-acked.
- A variant of this is to gray it only after a grace timeout period, if it is not already fully-acked by that time. The reasoning behind this is that every message will take some time for everyone to ack it, but graying every message immediately when displayed might be annoying or weird.
- We could add extra indicators to a message when it is fully-acked.
For comparison: At the time of writing, Facebook chat with 4 people looks something like this:
Alice : (1) Dave : (2) Carol : (3) [seen by Bob]
This means Bob has seen {1,2,3} - this is the equivalent of "acked" in our system. (Ours is more secure, but the intended semantics are the same.) With our strong ordering guarantees mentioned above, we can additionally deduce that Carol has seen {1,2,3}, and Dave has seen {1,2} - but what has Alice seen? She might have seen {1,2} or only {1}; it would be good to let the user know exactly which. Yes, this extra information is likely unnecessary in "most typical cases", but it is still good to make it accessible in a secondary interface, in case the user has critical needs for a particular case.
As with drop detection, if we reach a good state (here, consistency) after the timeout, we should cancel the warning or at least downgrade its severity. Also, here the warning is associated with a message that is already displayed to the user. Therefore, it is important that the warning is shown not as a point-event for the user to dismiss and forget, but as a persistent state shown together with the message, in the same space and time. TODO(xl): reword
We optionally send heartbeats to check that other people are alive. The purpose of this is to detect that "no messages received from Dave" really means "Dave didn't send any messages", as opposed to "the attacker dropped all of Dave's messages".
There could be some indication of this in the UI - for example, graying out a username when we don't see their heartbeat responses, etc. Extra information could be provided via a secondary interface, like "user not responsive since 19 seconds ago".
TODO
In most typical cases, the user may not care too much about the nuances of some security properties (e.g. group integrity and reliability), and therefore shouldn't be burdened by the extra security information that our system offers. However, in some cases, it may be a strong user concern, so we should provide a way for the user to access this information; it would be irresponsible not to. The existence of these features are also a good way of distinguishing our secure application from competing insecure applications; see next section.
By secondary interface, we mean UI features that exist away from the main view, so as to not overload the user with too much information. This is meant to achieve a smooth default UX for most users, but also make detailed security information accessible for users with high security needs. One approach is to provide extra information in pop-ups that are only activated by long-press or double-click on a related component in the primary interface; designers may evaluate a wide range of other possibilities too.
The properties above are fine-grained information about error conditions of specific messages or users. However, the desired state is for there to be no errors, and hopefully this should hold most of the time (i.e. with a non-malicious and semi-reliable transport). Rather than scanning our eye through an entire list of messages or users to check that all of them are OK, it would be more efficient to have a small number of "summary" indicators. Users can then tell at-a-glance if everything is OK - if so, there is no more work to do; if not, then they can check a more time-consuming secondary interface for specific information about errors.
For example, there could be one indicator for each of the properties mentioned above, that is quantified over all messages/users:
- no out-of-order messages being queued for display = "not missing any older messages"
- seen a heartbeat for all users recently = "not missing any newer messages"
- all messages acked by everyone else = "everyone else has seen what we've seen"
The first two could even be combined into the same indicator, to communicate "not missing any messages". However, it is probably good to separate the third item, since it takes longer than the other two to reach assurance about.
"Indicator" is used loosely; of course it is OK and maybe preferable if they are invisible for good states, and only show themselves when something is actually wrong.
Many messaging applications already have notions analogue to the ones described above, but these are not end-to-end secure. For example, with XMPP presence, the server tells the client when a contact is online/offline, or when they join/part a channel. Other applications have an idea of delivery receipts, but these are authenticated by the server rather than by the actual recipient.
Such information should be regarded as untrusted. We implement more secure (i.e. end-to-end) forms of these notions, so the UI should emphasise the properties from our system, and de-emphasise the less secure ones.
In some cases, it could be good to combine the information from both sources. For example, if we haven't received an (authenticated) acknowledgement, we are unsure if a message has been delivered or not. But, if the local internet is offline, or if the server is unreachable, then we know that a message has not been delivered. These differences in meaning could be communicated via different states (e.g. colours) of the same UI indicator.
In other cases, the untrusted source may be replaced completely by our more secure notions. For example, in the case of freshness, it is trivial for the server to report false presences, so our system (authenticated heartbeats) should be used where possible. For the case of joining/parting a room, this is probably more appropriate too - i.e. hiding XMPP join/part events, and only indicate GKA events in the UI.