Hi,

We observed a rather strange image viewing problem in KANBOARD Ticket systems for Images pasted directly to the board. More details see here: <kanboard.discourse.group/t/potential-issue-with-plugin-pitm/3188/12>, might be related to FF on Linux.

Some SeaMonkey expert here who is willing to help us with some SM diagnostics? With WIN SM I don't see any problem. And reporter did not see a problem with a WIN FF.

Rainer_Bielefeld, There is also #seamonkey on Matrix and Seamonkey mailing lists to ask and get maybe faster response . Also creating a bug report on Seamonkey itself might raise more people attention

I don't know matrix, but at least for the mailing list, IRC and the newsgroup are more likely to get faster or at least more responses

aaaaaargh [censored] discource

so supporting iDevices is more important than not breaking scrolling with overflow CSS rules

PITM means it was added using this? github.com/Chaosmeister/PITM

I've never heard of kanboard before, so can't really help much, I'd perhaps wonder if there are any JS console errors or messages when the feature fails, or if there are differences in network requests

Rainer_Bielefeld: oh and also, just to rule that out, using the UA of a working windows browser in the linux browser.

s/urce/urse/

njsg: Would be great if a Linux User / you simply could do a test on an existing KANBOARD. Simply create an account via newkanboard.bielefeldundbuss.de/?co…ewController&action=show&task_id=20, after I granted permission after your "Done" here I will grant permissions, and we will see whether Linux SM will suffer from the same problem like FF.

Here already the public link to the test KANBOARD Task:

The test can only be done with personal access, not with public access.

Rainer_Bielefeld: can you compare what address is the image served under for windows browsers? the address used gives 403 here (linux, no easy way to test windows right now)

or did you mean the image itself needs separate permissions?

(it doesn't load even inline)

(there *are* several JS errors, but that, at first, shouldn't cause issues on specific platforms, I'd say, although it's perfectly possible there is something going on)

(but those errors show on page load

)

ah and a "source map error" which I think was also mentioned in a github issue, either for PITM or MarkdownPlus?

njsg: Yes, of course!

Directly pasted into Text:

blob:http://newkanboard.bielefeldundbuss.de/6849a15e-6969-4b0a-ab4b-0542e22267e3

Attachment:

Both work fine for me with WIN SM and WIN FF

and: You need Personal access to the KANBOARD with an own account

but that error seems to be just for the source map, right?

Rainer_Bielefeld: same as irc nick, if you want to try to toggle permissions

source map?

Rainer_Bielefeld: only used for the devtools inspector I think, used because of the minified JS source

hello hello and hi

so is the goal to automate social networking so its just the bots that socially interact and just produce reports read by other bots that generate an infographic so the person can know how socially adept they are based on how well the bot generates content the other bots like and share?

CaptainTobin: yes. next question?

njsg: well I am obviously terrible at social communication so why don't *I* have an AI bot do this for me?

why? the AI bot won't excel at actual social communication

it will only be seen as excelling by its counterparts

which is why that content will be reinforced by algorithms ad eternum

njsg: Webrings

its the only way

we CAN'T REASONABLY run a search engine despite the tech existing

so.. webrings

with a ring of webrings?

yes

centralization might not be good, but in this case...

luckily Son_Goku surrendered binocnetwork.com back to me .. and know why I registered the domain for? To be a modern webring.

a ring to rule them all and in the google-unreachable bind them

well federation is a thing

webrings include other webrings because the webrings agree to be webrings

building a distributed federated index of the web.. and then we hook a searchbox to that

maybe we could build a system of menus with entries

a global index lookup of websites like what the search engine was supposed to be

.... DIRECTORY

THAT'S WHAT IT WAS CALLED

an INTERNET DIRECTORY

a searchable internet directory

but distributed and federated

maybe after 2030

njsg: this system of menus.. with entries did you say.. tell me more

ASCII-based for the menus

ASCII or ANSI

entries would have a type, and a description, along with some sort of pointer to the destination

ASCII as in what matters to the protocol would be 7-bit, but titles could well allow 8-bit chars

the format could be made extendable, by defining it as tab-separated and leaving it open to more columns on the right

types would be sort of generic, related to how it ought to be handled (downloadable binary, image, text, menu, ...), although it could easily be extended in several ways to include the MIME type somewhere

Anyway we should call it the Distrubuted Modern OptimiZed Directory or .. heh .. DMOZ Directory

I was thinking along the lines of Internet Gopher, but that could work as well

i know

the text specifics gave it away

what was that gopher replacement bullshit job did that go anywhere?

gopher replacement?

the only thing that comes to mind is gemini, the other thing being gopher+ but that wasn't a replacement, but an extension

someone was making a pseudo-modern gopher replacement to combat the web and its ads

gemini

If gemini I think it's getting some use, although I've never used it

that was it

last I heard of it, gemini: URLs were possibly being marked as malware by Google because Google probably wanted to use gemini: for something of their own

well, in theory, the comments didn't sort that out.

I dunno I think we need to split the World Wide Web and the OpenWeb Service Network into two protocols

I say we start by versioning javascript

I say we go back to EMCA 5 and rethink how we wanna do JS

ES5

ES6 could be entirely done with ES5 poly fills and is only justified by the later living standards

I say we bring back VBS on the web, if we're going to have to deal with scripts, let it be with Links :-P

there are perhaps nice things on the newer syntax, although it's really bad how it then ends up breaking a lot of things...

Let's burn an effigy of Eich for inventing Javascript.. nothing burns like a good effigy

njsg: let's fork the c language

maybe I will just learn objective-c and gnustep my way to victory

njsg: how does that sound?

(loop (print (eval (read))))

gitlab wip updated mostly python build stuff from 2.53 and from Myckel

what are the new python requirements frg_Away

CaptainTobin Nothing new just Python 3.

3.11 right?

3.6 should be fine. 3.7 for sure.

3.12 currently broken

if it wasn't for all the python codegen (or if I ever got a handle on it) other build system options wouldn't be so out of the question like i dunno str8 autohell

I defy anyone to explain just why Mozilla couldn't have updated to a newer autotools build system or cmake

"Unlike Google, XScreenSaver will never try to make WebP a thing."

Unlike Google, Mozilla, Moonchild..

I was against WebM and WebP support

h264 and png work just dandy

I think we need to stop with build backports soon. Under Windows central is total cr*p pulling in gigabytes in ms tools for bootstrap

despite the fact I was the only one who ever bothered to update libvpx

And they are not even using the compiler...

CaptainTobin: is there good documentation on current and past mozilla build systems?

njsg: .. the changelog and .. well.. me unless you can get jcrammer to go along with you

the current build system is documented on readthedicks but its all based on former MDN content and is sparse in places formerly documented

more about documenting the bs like fluent

or how to build and contribute to firefox cause that is the only mozilla application ever made right?

.. ahem anyway

njsg: the question isn't is there any good documentation its what do ya wanna know lol

and I will try and find out

you know.. typical mozdevelopment

Ok next year I will document the build system

which build system?

don't bother it changes so fast....

I could prepare some form of document that would AT LEAST point to more specifics by going over the basics from UXP, SM, and Current ESR

time to do something I've not done in years: a new SeaMonkey install from scratch, but moving stuff like passwords and bookmarks and extensions...

tomman if same user zip profile and be done with it.

yes, it's me, but it's on a whole new metal

maybe his profile is fucky

and I want to remember how I did things years ago

right now my new laptop FINALLY arrived

also setting up a new routerbox too

so it's time to relearn things for fun and profit

zip it together and be happy

CaptainTobin: I was just wondering if you knew of anything

could at some point be helpful to understand changes or what was pulling what

dir %userprofile%\AppData\Roaming\mozilla

njsg: besides Firefox Source Docs and whatever is still on UDN no .. a lot is in my head tho

Build System was an area i worked hard to be able to mangle into some sort of order

I really should tho do a series of Mozilla Build System Classes.. The Pure Autotools build system (1.8 and 1.9-21.0?) Early MozBuild, Late MozBuild and Current.

esr45 is where I seperate early mozbuild from later mozbuild cause .. mozconfigure hit the tree

njsg: so whatcha wanna know LOL

CaptainTobin: btw I think this page sounds like it'd fit the idea of webring/directory: jwz.org/xscreensaver/google.html :-P

ALL HAIL JWZ

xscreensaver is amazing

I have mine set on flying toasters (because I got distracted from setting it to the one I was wanting)

we should all adopt that privacy policy

it's sincere, it's real, and it goes straight to the point: "we are NOT Google!"

heh

And Google knows better than trying to act against this in any way. Or so I hope.

Do any of you know about that big house by the coast in Malibu?

I think I am about over this Github is your website hosting crap.. Keep it for thereisonlyxul because why not.. but I think BinOC is best with active server side code

njsg: is that a reference?

CaptainTobin: Barbra Streisand's house

nop

knowledge of it was amplified in what's now known as "Streisand Effect". I suppose if Google tries to backpedal on accepting jwz's policy, a similar outcome might happen :-P

The Beast is still with JWZ

certainly isn't at Mozilla

or not as much as it should be

njsg: OH i thought you meant something specific not just the streisand effect

i mean hell all those old and insecure redditors kept Pale Moon's name alive

when no media was covering it

no don't use Pale Moon it's old and insecure and the developers are assholes.. and it gets more users.

re: build system, I don't have any specific question at the moment.

it's like the meme that says that "Debian is old and stale"

(ironically JWZ believes that)

the issue with debian is not its age or claims of staleness vs stability its that debian is very opinionated in ways that are kinda nuts

tomman: binaryoutcast.com/updates