VistaLover

... No, they don't ; Discourse are serving a browser-feature detection script (believe it or not, this is actually the recommended way to probe a browser engine, rather than blindly block it based on its UA string) and they're looking for these features: https://arkshine.github.io/discourse-features-test/ (NB: Their announcement was recently edited to also cover "import maps" (Cr89+, Fx108+)) In a nutshell, they're dropping support for Win7/8.1 , where the last Cr and Fx supported versions are 109 and 115esr, respectively... ... Please do ; the posted custom filter blocks Discourse's browser-checking script, so your browser (FxESR-115) will be served the full-blown variant of the Discourse-based forum of interest; please be aware that while this hack currently works, it's not future proof; in, say, 6 months' time, one of the browser features they're checking for might end up in production, hence one (or more) forum function might break, to some extent ... That's the way to go , though it doesn't offer any additional clue as to why all but one of your existing logins were mysteriously removed ... Glad to know and... you're welcome ... Saluti !

First, welcome to the MSFN forums ; I take it you are using FirefoxESR-115, now on v115.27.0esr, supposedly to be EoL'ed on Win7/8.1 with the next, last, release v115.28.0esr ... Do you, by any chance, have uBlock Origin installed in your FxESR-115 profile? If yes, there's a way you can continue accessing the full version of Discourse-based forums in FxESR-115; read this post ; basically, you'd have to create below custom filter: ! Discourse-based forums ||*/browser-detect-$script,important uBO -> dashboard -> My Filters -> add above code -> Apply changes (header of tab) -> exit dashboard then delete discourse-forum cookies (for good measure) and reload the forum page... Can you please be more specific? Which version is that exactly? Help -> About r3dfox -> ? r3dfox has recently moved into a new GitHub repo, with new Releases section ... I'll assume you currently are in v140.0.4 (not the latest, BTW) ... Did you just transplant/migrate your full FxESR-115 profile to r3dfox-140 ? Although r3dfox is Firefox-based, such big jumps between major versions are prone to profile corruption ... I would advise you start from a clean redfox profile and then, progressively, tailor it to your own needs (settings, extensions, importing bookmarks from FxESR-115, etc.) ... In FxESR-115, load about:logins (Password Manager); click the 3-dot-button (top-right) and you should see a context option to "Export Passwords"; this will save ALL your login credentials to a CSV file; via a similar, but now reverse, procedure, you can import this CSV file to r3dfox and have all your passwords restored! (Take good care of that CSV file, as it contains the passwords in unencrypted, human-readable, form). Best regards ...

Hello Reino (OT, but I'm really missing your Vista-compatible FFmpeg compiles ) While your OpenSSL binaries (archive "openssl-3.5.2-win32-shared-xpmod-sse.7z") may very well be XP+ compatible , they don't work out-of-the box as CPython dependency by swapping the already existing DLLs (of openssl v3.1.0-dev) inside cmalex's CPython assemblies with them; he's been using the same (quite deprecated and now rather insecure ) v3.1.0-dev DLLs in his 3.8.13, 3.9.13, 3.10.8 and 3.11.4 CPython offerings, but the thing is the libcrypto-3.dll and libssl-3.dll inside CPython work in conjunction with two Python DLLs, files "_ssl.pyd" and (possibly) "_hashlib.pyd"; the existing .pyd files are incompatible with your own openssl DLLs ... BTW, I'm on Vista SP2 32-bit, so I'm using cmalex's Vista-compatible 3.11.4 assembly; with the original v3.1.0-dev DLLs: python Python 3.11.4 (heads/My3.11.4-dirty:1c368b1ab42, Aug 24 2025, 19:29:33) [MSC v.1900 32 bit (Intel)] on win32 Type "help", "copyright", "credits" or "license" for more information. >>> import ssl >>> ssl.OPENSSL_VERSION 'OpenSSL 3.1.0-dev ' >>> ^Z <redacted>\3.11.4\cmalex\Python3114-32_Vista> Overwrite them with your own v3.5.2 DLLs and... python Python 3.11.4 (heads/My3.11.4-dirty:1c368b1ab42, Aug 24 2025, 19:29:33) [MSC v.1900 32 bit (Intel)] on win32 Type "help", "copyright", "credits" or "license" for more information. >>> import ssl Traceback (most recent call last): File "<stdin>", line 1, in <module> File "<redacted>\3.11.4\cmalex\Python3114-32_Vista>\Lib\ssl.py", line 100, in <module> import _ssl # if we can't import it, let the error propagate ^^^^^^^^^^^ ImportError: DLL load failed while importing _ssl: The specified procedure could not be found. >>> ssl.OPENSSL_VERSION Traceback (most recent call last): File "<stdin>", line 1, in <module> NameError: name 'ssl' is not defined >>> ^Z <redacted>\3.11.4\cmalex\Python3114-32_Vista> I guess newer versions of "_ssl.pyd" and "_hashlib.pyd" must be provided for openssl-3.5.2 compatibility ... @nicolaasjan openssl-3.1.0-dev inside cmalex's assemblies (both 3.10.18 and 3.11.4, both XP/Vista variants) can be updated to a slightly more recent version, openssl-3.1.2, by using the OpenSSL-3.1.2 DLLs cmalex had compiled in the past for another of his projects, ProxyMII; link below: https://mega.nz/folder/68dj2YTY#As2w31IO4Smr7gy6p1ciSg/file/Pl9yzapZ Proof: python Python 3.11.4 (heads/My3.11.4-dirty:1c368b1ab42, Aug 24 2025, 19:29:33) [MSC v.1900 32 bit (Intel)] on win32 Type "help", "copyright", "credits" or "license" for more information. >>> import ssl >>> ssl.OPENSSL_VERSION 'OpenSSL 3.1.2 1 Aug 2023' >>> ^Z <redacted>\3.11.4\cmalex\Python3114-32_Vista> NB: When using the Vista variants of py3.10.18/3.11.4 to compile yt-dlp_x86.exe for Vista, none of the wrapper DLLs (kernelxp.dll, ntext.dll, psapi.dll) are needed; because your resultant yt-dlp.exe binaries (for Vista) still contain file "psapi.dll", redundant on Vista ; come to think of it, I don't think file "ws2_xx.dll" is needed on Vista either, but I'm not dead certain; only cmalex can vouch for that ... Many thanks and praise to both cmalex and nicolaasjan for... keeping the torch lit (i.e. yt-dlp accessible on Win<8.1) ...

... Initial post deleted; apologies, but I misunderstood ; I later realised what was asked for was the CPython (3.11.4) assembly itself used to compile the yt-dlp.exe binary linked in nicolaasjan's forum signature...

I've made the uBO-legacy custom filter even more generic, because I've come across Discourse-based forums where the browser detecting script wasn't served from within an "assets" subfolder ; so, the version I'm currently using is: ! Discourse-based forums ||*/browser-detect-$script,important The now broken userscript just told Discourse's browser-feature-detection script that "CSS aspect-ratio (Fx-89+)" IS supported in UXP-based browsers; maybe Discourse are now looking for something new entirely or, besides CSS aspect-ratio, for additional (recent) features that the userscript (and, alas, UXP) doesn't support... A very savvy person would have to dissect the new version of Discourse's script, identify what new things they search for now and author a new userscript... Discourse, in March, issued the following announcement: https://meta.discourse.org/t/dropping-ios-15-other-old-browsers-in-july-2025/358131 so it's probably one or more of the new requirements there that broke things; FWIW, relative color syntax & subgrid aren't supported in UXP, probably the same applies to import-maps ... Recent upstream forum thread about Discourse's shenanigans : https://forum.palemoon.org/viewtopic.php?f=70&t=32600 Last post there is from Aug 13th, though...

Different builds exist based on the capacities/age of the processor (CPU) on the machine the browser build is to be launched; if your CPU is archaic and doesn't support the SSE instructions set, use the -ia32 variant (sometimes referred to as "-nosse"); if your CPU is somewhat newer and does support SSE (but not SSE2), use the -sse variant; if your CPU does support SSE2 (and later), then you should use the "normal" build, whose filename ends in just "-xpmod.7z"... PS: Not relevant for XP users, but if you're on Win7+ x64 OS and want to use roytam1's NM28_x64 fork (you may have your reasons ), there's a fourth (64-bit, only) build variant that's meant for CPUs supporting AVX2 ; filename ending in "-w7plus-avx2.7z"... So, now you know ...

Slight correction: Vista+ only, for quite some time now ... https://github.com/Eclipse-Community/r3dfox

... My local AV solution, after a Custom Scan of the executable (yt-dlp.exe), found NO threats (red rectangle in Greek, haven't had time to consult VT) :

; is that py3.9-based (so will get deprecated in October ) ? Can you at least upload manually to GitHub, where it'll be more visible and with a permanent link? Thanks ... (OT: I have serious issues with my landline since Aug 19th, my tel. number isn't working at all and my ADSL+ internet access is quite fickle; my ISP don't really know what's goin' on, they said I have to give them time until Sep 3rd (!) for them to identify the true nature of the problems ... ) Sometime later today, hopefully, I'll further comment on your previous post ...

... Probably not, so let me summarise the gist of what @j7n was talking about in his recent posts: 1. The H.264 codec is probably the best choice when you're about to playback video on a system with low H/W resources (weak CPU, low-end GPU) and a legacy OS like Win7 SP1; the decoder is natively supported by the OS (not the case in XP SP3) and via implementations like WMF/MSE, an MP4 video can be played back within a compatible browser with hardware acceleration and hardware decoding (not the same things); most Win7-era graphics cards have H.264 decoding support built-in, so even when the MP4 video is being played back in a standalone media player, video decoding is performed by the GPU, not CPU... 2. Google have recently started producing their h.264 (aka avc1) youtube encodes at very low video bitrates, even for "high" resolutions, such as 720p/1080p; thus, while the h.264 YT encodes should be preferred for downloading (because they need less computer resources to be played back once downloaded), the stingy bitrate results in sub-optimal visual experiences (blurry moving images, video artifacts of various sorts, etc.). 3. OTOH, Google continue to produce their VP9 YT encodes at exceptional bitrates, especially for the FHD/UHD resolutions (>=720p); these are large files to download, to begin with, but, once downloaded, playback on low-end machines puts a heavy toll on the CPU, because VP9 is being software-decoded; I've not researched this properly, but graphics cards with integrated, native, VP9 decoding support (aka VP9 hardware decoding) only came in the later 2010s, as part of mostly Win10 machines... TL;DR: To have the best visual experience, j7n has to fetch the VP9 YT encodes, but these are detrimental to his under-resourced machine (mostly the CPU is being utilised for video playback, resulting in excess heat and energy consumption, etc.).

So did I now: https://developer.paypal.com/tools/sandbox/ so probably NOT what you're after ; if you're limited to using UXP-based browsers (on Windows XP SP3?), then am afraid they're now dead in the water with regards to PayPal; the issue needs to be escalated upstream by users of the "official" UXP browsers (Pale Moon and Basilisk, on Win7SP1+), in the hope upstream can come up with a solution, but the relevant PMForum thread hasn't seen any further activity since July 30th... OT for this thread, but you may also appeal for help to Feodor2 in MyPal's GitHub issue tracker ; if still on XP, you might try how the Supermium browser fares on PayPal these days; might prove your only recourse...

@kuja killer , follow up: Some 2 hours after my previous post, my modem-router had to be restarted (details that necessitated that are OT here), so my ISP assigned a new IP address to my internet connection; whether this is somehow relevant or a sheer coincidence, I have no idea ... Today, after reading your latest reply, I tried to re-access https://www.paypal.com/signin (as I had done yesterday) with the same NM28 copy as before; and guess what? I was hit by the same breakage you reported: I additionally have a page header instructing me to: "Please enable JS and disable any ad blocker" but JS is enabled and uBlock Origin (legacy) completely disabled; as you detailed in your post, sliding the blue button to the right (to "prove you're a human") turns it for half a second into green (with a white tick), but then the "you've been blocked" (in Greek, in my case) page appears... It is then impossible to access the proper sign-in page ... I opened Dev Console (which PayPal detected and did NOT like ) : but nothing stands out there for me... On the same machine (with the same IP address), browsers like r3dfoxESR-115.13.0 (or even the most current, r3dfox-140.0.4) will also be presented with the human verification challenge, but is then successfully passed, after which a redirection takes place to the sought-after sign-in page... So, PayPal have started requiring "a feature" (JS, CSS, both) that is to be found on relatively (more) recent Firefox incarnations, but isn't found currently in UXP browsers (and MyPal); this is based on feature detection, so a SSUAO won't work, as noted... And yes, PayPal are using their "own" captcha: https://geo.ddc.paypal.com/captcha?..... Further researching this, Google have provided an alternative sign-in page, https://www.sandbox.paypal.com/us/home => https://www.sandbox.paypal.com/signin which loads for me straight away in both NM28 and St52; as said already, I don't have valid credentials to test whether one can successfully log-in, hopefully it'll work for you (assuming that signing-in to "www.sandbox.paypal,com" logs you, too, to "www.paypal,com") ... Regards ...

I've also observed this myself; the other day I DLed a BBC-made 30min documentary off their official YT account (videoID=Pxozxj03l18) and though I initially wanted the 720p25 variant (in h264), I went, in the end, for the 1080p25 one (h264, ca. 1600kbps), because 720p25 was a measly 500kbps ... ... They're probably targeting ChromeOS laptops and/or Android-based UHD SmartTVs; little do they care for "legacy" OS/HW setups; and h264 still costs them patent fees, whereas their own VP9 codec (and av01) is royalty-free...

@kuja killer Paypal-related issues have also appeared in the upstream forum: https://forum.palemoon.org/viewtopic.php?f=61&t=32581 https://forum.palemoon.org/viewtopic.php?p=264391#p264391 Mixed reports from various users there, probably depending on their physical location; PayPal are conducting various A/B tests, it seems (read e.g. this PMForum thread ); user at the second link I posted thinks that it's NOT CloudFlare, but a PayPal-specific recaptcha (security check) that is NOT compatible with UXP-based browsers ... Can you try again with a fresh NM28 profile, where you block nothing? FWIW, with my ISP's Greek IP, I can access https://www.paypal.com/signin in NM28 at this very time without issue, but, as I don't have an account with them, can't proceed any further...

...It wasn't working for him, but now it is : PS: The forum software doesn't send a second (e-mail) notification to thread-subscribed members after one has edited one's original post (for which an initial notification has already been sent) ... Browser profile corruption is always a possibility, so, as others have noted, regular profile back-up and back-up of important info on physical copies is always strongly recommended ... ... Probably NOT ...