Meeting minutes
<jgraham> RRSAgent: make minutes
Data collection for cached reponses
github: w3c/
jdescottes: data collection for cached responses. the current spec is has that we get a stream from fetch but we don't have anything coming from the cached version
… my feeling is that we should consistently return data even if it doesnt go over the network
… or should we have it handle the cached version consistently?
sadym: could we start with WPT tests ? to verify that we have the same behaviour
jdescottes: we can, we have mozilla specific wpt test that does that
… in chromium it does it for all where firefox doesnt do that for css or images
… we can have the spec to have it say to return a response text
jgraham: if the spec doesn't say that we should have the right response... i'm not sure what gecko is doing here
… we need to make it consistent as authors don't expect to have inconsistency here
Data collection for data scheme
github: w3c/
jdescottes: this is the same topic as the previous... we don't have a stream we can clone
… we've updated Firefox to be similar to chrome. Are people happy for me to return a response body for data urls?
blaze_apple: In general this is fine. What do we do if the page is navigated away... e.g. if the process dies or ...
jdescottes: we don't clean collected data. We currently do this with a queue
… and we can handle the situation when collectors are cleaned up
… but we don't do anything for navigation
sadym: I wonder what selenium thinks about this?
jimevans: how does selenium feel about cached responses or when the navigation has happened?
sadym: about data schema
jimevans: I am of the opinion that we are doing network interception or of data
… whether it comes over any scheme we should be able to retrieve that data
… and I expect selenium users to think the same
Autofill trigger
github: w3c/
sadym: this PR has been sitting there and we need to have other vendors review so that we can unblock and get this landed
jgraham: I havent had time to look but I think the last time this was discussed and it needed changes but there doesn't appear to have been updated. I will need to re-read the PR
Pending Fetch spec changes
github: whatwg/
sadym: we have a couple of pending PR changes. I wonder if we can get these unblocked and landed?
… how can we make prgress on this?
<jgraham> github: w3c/
jgraham: I have linked the original issue that is blocked on fetch changes
… I think we reuse events and we can't tell the order properly
… when it was reviewed Anne he wasn't happy and we have made changes
… I am not sure if the new model is sufficient here and we need to make sure that it is before we can get it updated/landed
jdescottes: nothing much to add. I will try settle on what we wanted and I will take jgraham's PR and update to get it landed
Preparing horizontal reviews on the way to Candidate Recommendation
github: w3c/
tidoust: A meta issue has been raised to get this to CR
… I have started to get the a11y feedback
… for privacy and security, we don't have that section yet and we should get that added
… the classic webdriver spec has it and we might be able to copy and paste it
… so how should we proceed?
jgraham: for privacy and security, we use the classic spec as a starting point but we need to have it updated
… the other thing from before we didn't have a11y consideration but we might have i18n issues. Is that going block review?
tidoust: I will ask the i18n group and we need to know how to handle error messages for example
… I don't think we should block on this. I think should go for review and just highlight it and it's not a priority
jgraham: I think in practise that people are hard coding error messages in english but if people arent then that would be great
… the assumption I have is no one is doing it makes how we handle it simpler
tidoust: I will draft the initial PR for the privacy and sec
… and I will get moving on the other reviews in the meantime
WPT for extra headers
sadym: this is an area that i will start working on this after I finish the offline work unless someone else wants to pick this up?
jdescottes: sounds good from our side. it's not something we will be working on any time soon?
sadym: do we want just custom headers or do we want standard headers
jgraham: I think we should do standard headers too incase there are interop issues there that we might be unaware of
<jgraham> RRSAgent: make minutes