both -- aside from that, I am all onboard with the rest actually
also we had archive data on the picture polyfill before it was taken up
Who would you ask? In my workplace, once you get outside the UI team developers are back end developers that may be developing UI and THEY DO NOT REALIZE HTML IS A SEMANTIC LANGUAGE
I HATE that the order of [latitude, longitude] is not uniform (ex, taking goole map coords into leaflet)
q+ on this point @prushforth ... this is very hard in practice when it is not concrete, because people will do things if you have an element that you didn't expect, and that is often more interesting than the thing you even wanted
I think the argument that <picture> is an awkward markup API is true, though, but that's more because of the incremental development with a focus on fallback & reusing other attributes/elements, vs video & audio were more created all together.
Iván Sánchez Ortega
@edent_twitter If this were an in-person talk, I'd be nodding in agreement like 95% of the time
@AmeliaBR yeah - that is fair, and so is much of what the speaker said about the involvement of standardistas - I dont think picture was anyone's first take
one of the brians :)
I asked a q above!
@edent_twitter sure we may not know how to use it without looking at it, but if it supports formats we have all become comfortable with the transition into it is more seamless
I understand the <feature> element after looking at it once because I use geoJSON, but I didn't understand <layer> and <extent> after one look.\
q+ to talk about the chromium origin trials process
@zcorpan you next
Definitely, I will not have much to say until I make a few P.O.C's from user stories I understand
@zcorpan do you have a link to those origin trials?
Thanks for all the questions gang. I'll be sticking around for a bit - but it is quite late in the UK :-)
J Moules (via zoom chat): Code is write-once read many. "Latitude" is thus clearer than "lat".
Thanks Terence for your contribution tonight!
yeah like really far in that direction!
For developers it is becoming more important that APIs don't tighten themselves into a box. Extendability considerations are key. 2D should only be different from 3D by 1 dimension, and that dimension should be possible with a small extension and vocabulary.
The time element additions should be standard, and work in all dimensions it affects
@prushforth Sure, but it means doing things to a lesser degree of freedom than being able to do whatever you want. Implementing without a standard is just that, be as free as I want to accomplish my task. Standards seem an extra layer of complication for what seems unbeneficial as an overall part of finishing the task.
Of course now I know this is not the case, but I had to filter though some standards to understand why. A contract developer often does not have the time to go through the gauntlet.
But if you say "you must use this standard", then they have their rails
A standard doesn't stop you going your own way if you want. You can still use Flash video, or build your own component. But a standard makes it easier for people who don't want to roll their own.
Yeah I guess people in the framework domain complain about the DOM, and wish they could re-invent the Web because they aren't document-oriented, they're app-oriented
Also, a standard should enable "progressive enhancement" by providing a reasonable simple api that can be built upon
No, it doesn't stop me. But then all my labels become strings, and we know that doesn't work for internationalization or accessibility technological concerns. Or it means I use <a> instead of a <button> because I like it more.
Thankfully there is not native carousel element.
Iván Sánchez Ortega
I agree with @zcorpan here - a <slippy> 2D pannable/zoomable thing is sensible, and some better way to build over WebGL/OpenGL (in a way that could fit with the D3 data model with less friction) would be awesome for dataviz in general