please tell me that's not in search results

at least one good thing from google is that they're still one of the engines that actually lets me see more than 10 results per page...

and where "more than 10" doesn't mean "20"

stopped using goggles long ago

switched to ddg or startpage

last time I tried switching to ddg didn't go so well, but I'll perhaps give it a try again

tbh i hate goggles

nothing is stopping them from supporting windoze 7/8.1 with the chrome browser

but they had to pull the plug now causing steam to do the same

they could at least handle the version check correctly

it's slightly laughable that they identify NT 6.1 as Windows 7.

SeaMonkey status meeting in 30 minutes

hello no1 WG9s

Hi tonymec

sily me hosul have sent that

status meeting in 20 minutes - wiki.mozilla.org/SeaMonkey/StatusMeetings/2023-06-18

status meeting in 15 minutes - wiki.mozilla.org/SeaMonkey/StatusMeetings/2023-06-18

status meeting in 10 minutes - wiki.mozilla.org/SeaMonkey/StatusMeetings/2023-06-18

hi .*

status meeting in 5 minutes - wiki.mozilla.org/SeaMonkey/StatusMeetings/2023-06-18

meeting time - wiki.mozilla.org/SeaMonkey/StatusMeetings/2023-06-18 - And now, Heeeeeeer's frg! ;-)

Lets wait a minute or two for rsx11m

IanN is busy today and traveling.

ok bet you did not wxpect an into

expect

In any case I do the meeting notes again late :)

and problably no one ouside the US understood the intro

for 30 years that is how Ed McMahon introduced Johnny Carson on the Tonight Show.

yeah just forgot which show.

Anyway lets start

Nominees for Friends of the Fish Tank?

Hi rsx11m

hi frg *

njsg for hi work supporing user issues on Mozillazine

rsx11m we just started. IanN is traveling today

sorry, few minutes late...

rsx11m: hi

second ald also for helping out with patches and reviewing.

^and also

rsx11m any additional nomiations from your?

^you

frg: no

oki

Action Items

Nothing here from me. bau

Status of the SeaMonkey Infrastructure

IanN and ewong set up more stuff for using BugSplat as crash reporting site but didn't make it into 2.53.17b1

not sure this applies here but I am no onger able to post anything to achive nightly

sent email to ewong still awaiting a reply

I created a substitute page on my website and changed the loink inthe topic to go there

Opps too fast on action items. Project areas have been updated in bug 1828328.

Mozilla moved from aws to google. Bug 1828540 is still open. Might delay the beta. Hope ewong sorts it out so that the nightly builds also will appear again.

Anyway have not heard any news about the new Rocky 8 builder and Server 2012 R2 will also go away in a few months. So there will be infra changes.

Status of the SeaMonkey Source Tree

All building I think. 2.53.17b1 no longer needs Python 2.

late, but hello world~

tomman hi

wow a refernce to the first program ever written in c

WG9s any problems with building central?

no but seems we have info in the meeting agenda about a no longer existing issue about how to get around the fact that mach bootstrap does not install a wasi-sysroot. well, it seems it now does so this whole section can go before the next meeting.

yes I did a build recently and seems now to import the world. Bloat**2

I think clang 16 under Windows with 2.53.18b1 pre now works too. Need to re-check

i am still working on the issue whcih you decided was distro dependnt on building under linux with clang 25 or later with elf-hack-enabled

Mozilla plans to abandon Windows 7 and 8.1 after ESR 115 starting with 116a1. No such plans here. If critical third party packages drop support it might come to it later but currently not even discussed.

I had hoped to figure this out yesterday but I did not feel well and am just now getting back to working on it.

Mozilla does not support their bowsers on any OS that the OS vendor has dropped support on.

if the os vendor no longer does scurity updates on the os they drop support of thir brwser onthe platform

WG9s well which is not true because Server 2008 R2 and 2012 are still supported and at least 2012 R2 will be for at least 3 years.

ooof, Mozilla, why!?

they feel if they support the browser on an OS that is not getting a security updates then they would be taking on the OS security vulnerabilies as their fault

tomman well you still have about a year and a half with ESR 115.

wonder what's their support policy with Linux builds then

I mean, plenty of EOL'd distros where the current binaries still run fine

WG9s I suspect it is just a good excuse. macOS up to 10.14 is also dropped.

but then, they don't even tell you which libc version they want

I am not trying to defend their position just trying to explain it.

yeah, understandable, too bad the average user doesn't care

anyway can't be helped and we keep the support as long as possible. I think SeaMonkey was one of the last browsers to support XP :)

but I think they do not need to drop support just need to not advertising as being supported to avoid any libility

And with 7 8.1 it is even easier because not much different from 10 when it comes to the api.

you now like we no longer support blah but it might still work but don;t blame us

Even MS needs to patch up Edge for now on 7 and the servers.

Release Train

2.53.17b1 is done. It is now in tmp archive.mozilla.org/pub/seamonkey/tmp/releases/2.53.17b1

Hope ewong manges to move it soon.

It will rpobably break a few ond add-ons as discussed here but I hope they can be patched up.

Otherwise the beta should be very stable (if you don't have mem errors) and please test

hey~ :D

tomman me? :D

so far no new crashes since detecting the problem yesterday

but that RAM is getting bought tomorrow!

I had some crashes 6 or 8 weeks ago but did go away. Some backports missing I suspect. Lots did go in then.

Anything else for release?

Extensions Tracking

I plan to do Bug 1083458 soon. Then hopefully no more add-on breakage for some time.

maybe there should be another note about Enigmail for SeaMonkey under the "Enigmail was discontinued [...]" item?

remind me what Mozillaism is that?

(hope I don't need to patch Redirector once more!)

njsg noted 3.14 is out.

tomman awful syntax It will not be missed. Check the bug.

uuugh, now that I can see it, yeah

it looks real ugly

good riddance then

need to watch out for that one

frg: huh? bug 1083458 (Core/JS Engine) was closed RESOLVED FIXED 5 years ago. Typo?

I hope I can get the legacy getters and setter in.

tonymec no. Still not done in 2.53.

KI think we have lst control of the meeting i have no diea what tomman is even taking about if it is enigmail this was not a mozilla decision this was a TB decision.

this? web.archive.org/web/20141019225043/…/Guide/Closures#Expression_closures

WG9s talking about bug 1083458

WG9s: I was talking about that deprecated syntax for expression closures

yeah, my bad, should have waited a bit before mentioning enigmail

enigmail now has a new 3.14 release version which is fine.

njsg just a bit of multitasking

re: enigmail, possibly useful to link in the meeting notes: enigmail.net/index.php/en/download/changelog#enig3.1.4

Yes. Whish we had the resources to set up our own add-on page. Not sure if SeaMonkey ones even get reviewed any longer.

anything else on extensions?

Next, Feature List, Planning

bau for me

pdf.js

would be nice to see a new SM-compatible release from somewhere

but the current one STILL works fine

aside of that, nothing new here, business as usual~

we have updated the internal one in browser but needs to be adapted and packaged. No time as usual.

(what was the sidebar for facebook error? maybe I'm mixing things, but I wonder if that kind of message could be the initpump change?)

s/initpump/InputStreamPump/

gotta go, blackout :/

njsg usually it was the pref rename lately.

Anything else or should we call it a meeting?

nothing here

Then I jsut say thanks for your time and contributions and see you in 3 weeks with our regular SeaMoneky show host IanN again.

:-)

nothing here either

nor here

needed to change the nick to block an idiot using it.

and we're back

...for now

it's raining, and this is the season where you get what you pay for

it... crashed again :/

back to memtest!

OK, turns out I was actually isolating the wrong address

because memtest86 and the kernel see different address maps

because... x86 MMUs are hard?

memtest86 was saying my bad RAM was at 0x00000000a00f2888

so I had isolated the entire megabyte at 0x00000000a00f2000 to be safe

but... I was still tripping the crashes

turns out the kernel now also has a memtest tool: boot with memtest=<num-of-passes>, and it will EVENTUALLY catch the bad area and isolate it

not as effective as memtest86, but it actually found the proper address: 0x00000001e00f2888

a 20-bit difference on the offset!

tomman main reason I only use ecc ram on my desktops. Life is too short to track these errors.

no1: sadly this is a laptop, no such luxury :/

yes, I know, the answer is "buy new RAM lolololo", but IT'S SUNDAY, IT'S DOWNPOURING RAIN OUTSIDE, THERE ARE NO COMPUTER STORES IN MY CITY ANYMORE, and we already had one blackout!

this stuff is in general poorly documented, especially memtest86's reliance on badram patterns, a set of kernel patches that not many distros include, and the kernel has better tools for that these days, but those are lacking good end-user documentation too

tomman yeah thanks to Intel no ecc support in these cpus and chipsets.

oh, the last time not only SeaMonkey tripped the crash for the umpteenth time, even STEAM of all things also managed to trip it!

the kernel gotta love those high memory addresses :D

lovely one-bit errors :/

tomman: perhaps later in the week run it again (memtest) just to be sure it's constant

njsg: at least today and yesterday, the addresses seem to be constant

but I do intend to get this thing sorted out ASAP, as soon as I can buy the new RAM

frg - Windows Server 2012 R2 will go out of extended support Oct 10 2023 and after that date updates will be delivered by Microsoft

frg_Away: just checked and boostrap.js:533 in Sidebar for Facebook version 52 has Components.interfaces.nsIPrefBranch2, so it's the prefbranch rename indeed

no updates will be delivered by Microsoft

guest__ there will be esu updates for about 3 years.

no one - except large organisations - will ever buy ESU for Windows Server

frg - how is your progress with updated regex support?

more and more sites stop working in SM due to fucking Google Chromism

guest__ I know but slow go. Python 3 was another priority so that we are not kicked out of distros. Now halfway done but the current web is like wack-a-mole

cerebralab.com/Imaginary_Problems_Are_the_Root_of_Bad_Software related

the main reason of why the modern web sucks and it's overengineered to hell and back: because it's solving problems noone had

AKA the "shiny stuff"/"I am bored" syndrome

I'd go with: some people actually think backwards compatibility is bad, others don't care about it; when people thinking like that are in charge of frameworks, all it takes is using incompatible syntax once and it might break many sites

it's perhaps a bit like offering content without scripting

starting with some version, the mastodon web UI apparently doesn't want to do that

now that can be frameworks all over again too, some are designed to be so, so dynamic that such a thing is perhaps inconceivable

animenewsnetwork.com oh god, they HAD to deploy this regex garbage too!

> SyntaxError: invalid regular expression flag s

also wonder what's this: TypeError: event.path is undefined

Anyway, Anime News Network is now broken: images won't load anymore, and it completely ignores my layout settings

OF COURSE, ANOTHER CHROMEISM

...wait, event.path no longer exists on Chrome as of v109

...uuuuugh

wait, Event.composedPath is the standards-compliant one

and SeaMonkey indeed supports it (also Firefox since FF59)

why ANN went out of their way to implement a nonstandard Chromeism for which the standard implementation is old as dirt, AND the original proprietary one was removed a few months ago anyway?!

do they even test their junk!?

animenewsnetwork.com/bbs/phpBB2/viewtopic.php?t=3187693 fixed for you, sweetheart... not for me

is it possible there are two versions of some script and that they fixed whatever is served to chrome/firefox UAs but left the other uncorrected?