Closed flavio closed 2 years ago
In terms of icons, looking here, I think these ones are nice:
Context aware policy | NOT context aware policy |
---|---|
CloudDownload |
CloudOff |
Sync |
SyncDisabled |
Wifi |
WifiOff |
I sorted them by preference, the cloud one if my favorite
In terms of icons, looking here, I think these ones are nice: Context aware policy NOT context aware policy
CloudDownload
CloudOff
Sync
SyncDisabled
Wifi
WifiOff
I sorted them by preference, the cloud one if my favorite
What does context-aware
flag represent/what is the meaning in this context? By the name itself and the common known meaning I would propose alternatives:
Visibility
|VisibilityOff
(it's an eye)BlurOn
|BlurOff
(Business
|DomainDisabled
CenterFocusStrong
|CenterFocusWeak
something like the policy has view/knowledge to take decisions depending on the context . What do you think? Otherwise, the cloud one is my favorite too.
Btw, the idea would be to place the new icon closed to the mutation
one, would it be ok? In case the flag it's true I'd go for a green or blue color, in case it's false I'd go with dark-gray instead.
Everything is 'cloud' these days.
I wonder what 'context-aware' tells the policy user ? Why would a user need to know ? How does a user's behavior change with that knowledge ?
context-aware
is a term that was coined by OPA, which features also on different k8s books.
The concept is: a context-aware
policy has awareness of other resources already defined inside of the cluster. This is opposed to regular policies which have 0 knowledge of what is going on inside of the cluster, they can only see the incoming object to be evaluated.
Proceeding with my stream of thoughts: regular policies can be seen like in "airplane mode", while context-aware ones are "connected" to the network. Thinking more about that, maybe the sync
icons are the best fit :man_shrugging:
I'm aware of the meaning of "context-aware". :wink: But I'm not sure if a regular user of a policy needs to know if a policy is context-aware or not.
As a user, I'm interested in the semantics of a policy. What does it check ? Which user-settable parameters does it accept ? etc. I'm (probably) not interested in how the policy implements its checks and how the policy collects its data.
As a security-focused user, I would say it is good to show this information clearly. Despite Wasm is already a very safe environment, a user might be interested in only installing policies that are not able to fetch any kind of external information, and only take decisions based on the resource it is evaluating.
JFYI I forgot to update the status here: the UI part is already implemented and published since June. I'm not sure about the template though, but maybe this can be closed or?
Thank you @ncounter. I think we can close this issue. We still have to do the last stretch with context-aware policies.
Let's close this and reopen if necessary.
Some policies can be "context-aware", these policies have
context-aware
key inside of their JSON definition:To be done
hub.json
files inside of our policies + the repo templates - @ereslibre is working on that