#metabrainz

in fact, I'm actually not really ready yet.

Freso: ahh sorry

leave for next week, I'd say.

ruaok: We can skip to cookies instead?

Freso, perhaps a student introduction thread on the forums would be useful?

Leo_Verto: According to caniuse, Firefox 38 has 0.08% user base.

let's.

ruaok: EU & cookies!

so, we still have a pile of things to do before we can be in the clear wrt to cookies.

Leo_Verto: I was thinking to maybe have a "beginner's task" to have them come on the forum and say hi.

But again, talk after meeting/in 8 minutes. :)

if we read the documentation very strictly, we would need to have one of those stupid cookie messages now that we we are hosted in the EU.

(Note: Technically, hosting country is irrelevant – target audience is relevant. So not actually a change.)

we can't be free an clear just yet. Google Analytics as well as a couple of other aspects sink our being able to circumvent this rule altogether.

ah, interesting.

I'm certain a lot of web pages in the world are breaking that rule for their EU visitors then

do we know if anyone has ever been sued or otherwise gotten into trouble?

Yes, I believe Spain, Germany and (at times) the UK were trying to enforce it.

my feeling is that we should bring our cookie documentation up to date.

And France, never last to enforce stupid things

There were apparently some fines in 2014.

and add a message where people can find our cookies statement.

apparently some cookies are exempt from that law, notable cookies "used for the sole purpose of carrying out the transmission of a communication, and strictly necessary in order for the provider of an information society service explicitly required by the user to provide that service."

but not go as far as actually putting up a banner and requiring people to click ok.

Leo_Verto: Yes, we know.

that is another question: if a site does not meet the notification free cookie policy, what notification MUST be given?

is it imperative that a user click a button to acknowledge that?

chirlu: do you know?

That would mean that it had to be on every single page, and without click, no way to dismiss it.

I think we should do everything necessary to come clear on our cookies, but not add the stupid message.

But when you click, you store a cooking saying that you've clicked "OK"!

Piwik is recommended by CNIL (French authority) regarding cookies and privacy. More info on exemptions: http://piwik.org/blog/2014/10/cnil-recommends-p...

what if it was in the footer?

*cookie

Freso: yes, that means that you are ok with cookies, so that is ok.

Probably too inconspicuous.

yvanz: yep, that is already on the radar as a GA alternative.

well, the cookie message seems quite commonplace nowadays, I don't think it'd be too annoying

You can have implicit consent (“if you continue to use the site, we’ll store cookies”).

well, given that we're out of time, I would like to solicit feedback on how to do a cookie message that doesn't suck ass.

ruaok: More important, “first party analytics” are “exempt from the Cookie Consent rule”

But the user must actually take in the message.

yvanz: yes, understood.

we are not currently discussing the nuances of getting around this requirement. we understand that fairly well.

so, the challenge then is: how to we make a cookie message that isn't annoying as fuck?

Do we store any session cookies when people aren't logged in?

Freso: fin. we'll get back to this next week again.

Right, sure.

</BANG>

then :)

Freso: I forget.

Thanks for your time!

Freso: I’m not sure, actually, but we do store configuration cookies.

And GA of course.

Yeah.

(And permanent configuration cookies aren’t exempt.)

Is there any way that running install.sh could completely take down my computer or is it most likely unrelated?

I suggest the banner “We made a cookie to feed our flock of pink unicorns, thank you for your understanding.”

:-D

But we're maybe going to get rid of GA, I think(?). Not sure about the config. cookie though.)

most powerful of unicorns at that!!

Freso: yep, that is where we stand. we can't cleanly circumvent the EU rules.

anshuman73: Hi again! Did you have a question before?

ruaok: :'(

Anyway, as long as we can make it easy for cooking blocking rules to block the notice, I'm good.

*cookie

gcilou: And yes, go ahead and add whatever tasks. I trust you to be reasonable. :)

(And I need to look them over before they're published anyway.)

Freso: haha thanks :)

Freso: nope, i was just mentioning that I can't claim tasks yet, but Im working on one. I hope thats not an issue

anshuman73|mob: Are you Tigran?

No, anshuman73 on his phone

Sorry for the nick shift, just put away my laptop and didnt want to turn it on again

updated browser stats https://gist.github.com/mwiencek/4546f854f7d944...

anshuman73|mob: Okay. Well, if someone claims the task, they'll get the credit. (Someone named Tigran has claimed one of the beets tasks, that's why I asked.)

anshuman73|mob: Also, I'm fairly sure you should be able to claim tasks - you just won't be able to get the task approved/closed until your paperwork is done.

Or maybe you need to submit before you can claim, and then it needs to be approved before tasks can be closed?

Ahh. Thats okay. I never really cared about " tasks " on GCI. I come here every year mostly to learn :)

I ignored firefox 12 and some chrome 22 versions

anshuman73|mob: ❤️

anshuman73: Hi! :)

ah yay the computer is working again

No, its just that it was too late to wake my parents up to even sign the thing, its 2:30 am atm in my place

Freso ^

bitmap: No Edge! 🙌

Or is that MSIE 11-or-newer?

chrome 16, 21, 25 have shares above 0.5%

(not 22)

anshuman73|mob: :)

Freso So i havent even submitted a request for parents consent thus the inability to claim tasks

oh, I forgot edge

updated to add Edge 13 or newer

glennren, Hey ! (Same person as anshuman73 here)

anshuman73|mob: Right, yeah.

bitmap: Aw. :(

I think it would be fine to... forget Edge.

(And MSIE.)

anshuman73|mob: Just get it done ASAP in the morning! ;)

well, IE11 has 4.72% which is the 4th highest share of any browser version

Freso, sure !

Edge mostly isn’t that bad, actually. It’s just a small bit behind the curve.

Edge 13 has 1.90% and 14 has 1.20%

And IE11, well, I gues we’ll have to live with it.

But hey, it does support SVG!

:)

My girlfriend uses IE.

(In contrast to some earlier versions. So some users of still-“supported browsers” can’t actually see the MB logo.)

I should make that my goal for the Dec/Jan visit to convince her to switch to something else.

I guess supporting opera 39 is ok if it uses chromium 52 http://www.opera.com/blogs/desktop/2016/09/oper...

bitmap: I’m comparing the versions to http://caniuse.com/usage-table and am a bit surprised regarding Chrome and Firefox. The versions seem to be off by one …

E.g., FF 44 has 0.08% on caniuse.

Chrome and Chromium's versioning run parallel, right? So Chrome 52 ~= Chromium 52?

FF 45 is the current ESR.

I see FF 44 with 1.00% in the stats, compared to 0.62% for 45

Is some kind of Ubuntu or Debian still on 44?

Strange. Perhaps there is a single high-profile user still at FF 44 for some reason.

I wonder if 44 is the one CatQuest is using maybe?

Pre-45 isn't supported upstream.

Yes, because 45 is the ESR. :)

16,538 New Users ("The number of first-time users during the selected date range.")

It’s the only non-current release supported by Mozilla.

(that's for 44 only)

Hm.

(our #1 browser, chrome 51, has 100,069 New Users)

Is there anything not in 44 but in everything else that we really want to use?

But it won’t matter much in practice. I looked up some newer HTML/CSS features recently, and it mostly was “not supported by MSIE before 10/11/not at all” “supported by FF/Chrome since eternities”.

*vs.

(Because if not, we might as well keep it another ~half a year.)

Yeah, I’m mostly just curious about the strange results.

chirlu: At least we get to bump MSIE to 11. :)

Yes, that’s really good news.

Although it *still* doesn’t support a lot of things that are standard everywhere else.

(I already said I'd be happy to drop support for it completely. :p (But I also understand why it might not be a good idea... yet.))

E.g. flexboxes are nominally supported, but there are so many bugs in the implementation that it is unusable in practice.