My Browser Builds (Part 5)

[roytam1]

New build of Serpent/UXP for XP!

Test binary:
Win32 https://o.rthost.win/basilisk/basilisk52-g4.8.win32-git-20260523-3219d2d-uxp-829418a939-xpmod.7z
Win64 https://o.rthost.win/basilisk/basilisk52-g4.8.win64-git-20260523-3219d2d-uxp-829418a939-xpmod.7z

source code that is comparable to my current working tree is available here: https://github.com/roytam1/UXP/commits/custom

IA32 Win32 https://o.rthost.win/basilisk/basilisk52-g4.8.win32-git-20260523-3219d2d-uxp-829418a939-xpmod-ia32.7z

source code that is comparable to my current working tree is available here: https://github.com/roytam1/UXP/commits/ia32

NM28XP build:
Win32 https://o.rthost.win/palemoon/palemoon-28.10.7a1.win32-git-20260523-d849524bd-uxp-829418a939-xpmod.7z
Win32 IA32 https://o.rthost.win/palemoon/palemoon-28.10.7a1.win32-git-20260523-d849524bd-uxp-829418a939-xpmod-ia32.7z
Win32 SSE https://o.rthost.win/palemoon/palemoon-28.10.7a1.win32-git-20260523-d849524bd-uxp-829418a939-xpmod-sse.7z
Win64 https://o.rthost.win/palemoon/palemoon-28.10.7a1.win64-git-20260523-d849524bd-uxp-829418a939-xpmod.7z
Win7+ x64 AVX2 https://o.rthost.win/palemoon/palemoon-28.10.7a1.win64-git-20260523-d849524bd-uxp-829418a939-w7plus-avx2.7z

Official UXP changes picked since my last build:
- Floor fractional border widths (4510bf92a2)
- Test fractional border width rounding (794f3fe2e7)
- Fix CSS border rounding and currentcolor clipping (c289641428)
- Flush layout for computed border-width shorthand (75136dff38)
- Fix app-unit rounding for border width edges (50a581840e)
- Issue #1826 - Implement broader CSS calc() parsing (29f5ff07d8)
- Issue #1826 - Support calc() in media queries (f9a90b9bb2)
- Issue #1826 - Canonicalize nested calc() serialization (f94a63864c)
- Issue #1826 - Add typed calc() arithmetic for media queries (12b120db24)
- Issue #1826 - Serialize special calc() number values (e85f778708)
- Issue #2982 - Follow-up: allow color-mix to work with oklab and oklch (20b2b3b9f5)
- Load mochitest modules without imp (ee00ac9826)
- Remove stale imagebitmap support file entry (4337565d3a)
- Issue #2404 - Enable CSS aspect-ratio sizing (0d684399b2)
- Support CSS sizing math functions (93899c0157)
- Revert "Cloudflare Image Resizing fix take 2" (98c9f7387a)
- Issue #3089 - Support logical border radius properties (27f2a0869c)
- Issue #1826 - Parse calc() weights in color-mix (4634a74b31)
- Issue #2506 - Support range media query syntax (6df85ff502)
- Support CSS shadow parts (51767db33a)
- Whitelist virtual GPUs (VirtualBox, VMware, VirtIO (QEMU and forks like UTM), Parallels) (235bcb010b)
- Issue #3092 - Refactor WASM compilation handling (a7a75b7851)
- Issue #3092 - Add new GC sweep tasks. (47746b476e)
- Issue #3092 - Implement BackgroundFinalizeTask for parallel garbage collection finalization (c06776336d)
- Issue #3092 - Implement parallel sweeping and compaction tasks for improved garbage collection performance (3433d538ed)
- Issue #3092 - Initial idle GC implementation (18ddd00afe)
- Issue #3092 - Perform a minor GC on tab close (1d3dad153b)
- Issue #3092 - Fix unsafe GC multithreading changes (f0cba41221)
- Issue #3092 - Safely parallelize GC background finalization (e9826f5559)
- Fix JS shell module hook build (6f47a2b0da)
- Merge remote-tracking branch 'origin/tracking' into custom (0933486bde)
- Implement ES2024 grouping and resolver builtins (e1b689d34e)
- Implement ES2024 ArrayBuffer transfer APIs (3be309faa7)
- Allow symbols as weak collection keys (e317bf10fc)
- Implement growable SharedArrayBuffer (2e51dc9f09)
- Implement Atomics.waitAsync (22cb023133)
- Implement resizable buffer view semantics (6f3f17ba86)
- Guard typed array JIT paths for resizable buffers (d97a2eb04f)
- Support DataView on shared array buffers (366476589f)
- Support BigInt Atomics waiters (8b09714bbc)
- Fix resizable DataView out-of-bounds semantics (aea80980ad)
- Validate typed array methods on resizable buffers (3cb76bb20e)
- Validate typed array set and constructors on resizable buffers (7613d2901f)
- Fix ArrayBuffer slice after resizable source shrink (61912d2318)
- Fix ArrayBuffer storage and error types (e21c4e2917)
- Fix incorrect variadic for `size_t` in `fprintf` statement in `hyphen.c`. (a02580dae4)
- Make WeakRef support always enabled (6861bedff6)
- Implement FinalizationRegistry (50c1419e75)
- Fix FinalizationRegistry constructor realm prototype (890fb3f399)
- Fix WeakRef constructor realm prototype (f3c6da5987)
- Issue #888 - Vendor dav1d 1.5.3 (ca94696239)
- Issue #888 - Use in-tree dav1d for AV1 decoding (4d1cefd2a1)
- Issue #888 - Enable dav1d LoongArch SIMD (d36f889688)
- Issue #888 - Remove bundled libaom (2d83afa321)
- Issue #888 - Enable dav1d SIMD on more architectures (d328cef713)
- Issue #888 - Enable AV1 playback by default (2bd46d4606)
- Issue #888 - Update AV1 configure comment for dav1d (3a2f21abfb)
- Issue #888 - Fix alphabetical order issue (c95062af66)
- Issue #888 - (potentially) fix MSVC builderr (83b19136b4)
- Issue #888 - Make symbols direct to dav1d, not AOM. (03a6729a9e)
- Issue #888 - Add dav1d_flush (7887ca767f)
- Follow-up Issue #888 - Enable dav1d SIMD on more architectures, add back missing nasm detection code in configure (c182d70c0f)
- Issue #2354 - Follow-up: Fix building WebRTC on 32-bit ARM without hardware float. (57cd574884)
- Issue #888 - Follow-up: Spot-fix; set PREFIX in Dav1d config on Apple. (3155b16f2c)
- Fix build on NetBSD/aarch64 (13339f6e55)

No official Pale-Moon changes picked since my last build.

No official Basilisk changes picked since my last build.

My changes picked since my last build:
- dav1d: port WinXP threading code from mypal68, guard AVX usages with GetVersion() > 5, exclude AVX* assembly from non-x64 build, add ipred16 smooth_weights in sse for non-x64 (808697c1df)
- Revert "Issue #3092 - Perform a minor GC on tab close" (3d04e32447)
- Revert "Implement FinalizationRegistry" and related commits. (87776c74a9)
- Follow-up rev 87776c74, restore Helper prototype changes (829418a939)

Update Notice:
- You may delete file named icudt*.dat and icu63.dll inside program folder when updating from old releases.

* Notice: From now on, UXP rev will point to `custom` branch of my UXP repo instead of MCP UXP repo, while "official UXP changes" shows only `tracking` branch changes.

Edited May 22 by roytam1

[roytam1]

New build of BOC/UXP for XP!

Test binary:
MailNews Win32 https://o.rthost.win/boc-uxp/mailnews.win32-20260523-40a79c75-uxp-829418a939-xpmod.7z
BNavigator Win32 https://o.rthost.win/boc-uxp/bnavigator.win32-20260523-40a79c75-uxp-829418a939-xpmod.7z

source repo (excluding UXP): https://github.com/roytam1/boc-uxp/tree/custom

* Notice: the profile prefix (i.e. parent folder names) are also changed since 2020-08-15 build, you may rename their names before using new binaries when updating from builds before 2020-08-15.

--

New build of HBL-UXP for XP!

Test binary:
IceDove-UXP(mail) https://o.rthost.win/hbl-uxp/icedove.win32-20260523-id-656ea98-uxp-829418a939-xpmod.7z
IceApe-UXP(suite) https://o.rthost.win/hbl-uxp/iceape.win32-20260523-id-656ea98-ia-c642e3c-uxp-829418a939-xpmod.7z

source repo (excluding UXP):
https://github.com/roytam1/icedove-uxp/tree/winbuild
https://github.com/roytam1/iceape-uxp/tree/winbuild

[roytam1]

New build of post-deprecated Serpent/moebius for XP!
* Notice: This repo will not be built on regular schedule, and changes are experimental as usual.
** Current moebius patch level should be on par with 52.9, but some security patches can not be applied/ported due to source milestone differences between versions.

Test binary:
Win32 https://o.rthost.win/basilisk/basilisk55-win32-git-20260523-287867cac-xpmod.7z
Win64 https://o.rthost.win/basilisk/basilisk55-win64-git-20260523-287867cac-xpmod.7z

repo: https://github.com/roytam1/basilisk55

Repo changes:
- ported from UXP:
 - Issue #888 - Vendor dav1d 1.5.3 (d1b46d7347)
 - Issue #888 - Use in-tree dav1d for AV1 decoding (a468f81a5e)
 - Issue #888 - Enable dav1d LoongArch SIMD (5c0a35a358)
 - Issue #888 - Remove bundled libaom (eb487c2ca1)
 - Issue #888 - Enable dav1d SIMD on more architectures (9a8cb2e22e)
 - Issue #888 - Enable AV1 playback by default (e61621692c)
 - Issue #888 - Update AV1 configure comment for dav1d (0ccaab55ca)
 - Issue #888 - Fix alphabetical order issue (a30f439a9c)
 - Issue #888 - (potentially) fix MSVC builderr (6e4bb257a4)
 - Issue #888 - Make symbols direct to dav1d, not AOM. (aa69d5ec99)
 - Issue #888 - Add dav1d_flush (b7b1918a36) (874efea8e)
- import from `custom` branch of UXP: dav1d: port WinXP threading code from mypal68, guard AVX usages with GetVersion() > 5, exclude AVX* assembly from non-x64 build, add ipred16 smooth_weights in sse for non-x64 (808697c1) (77a233ced)
- import from UXP:
 - Floor fractional border widths (4510bf92)
 - Test fractional border width rounding (794f3fe2) (02b2cac86)
- ported from UXP: Fix CSS border rounding and currentcolor clipping (c2896414) (edee89bdf)
- import from UXP: Flush layout for computed border-width shorthand (75136dff) (597bf8b86)
- ported from UXP: Fix app-unit rounding for border width edges (50a58184) (d9dc061b2)
- ported from UXP: Issue #1826 - Implement broader CSS calc() parsing (29f5ff07) (c8457d302)
- ported from UXP: Issue #1826 - Support calc() in media queries (f9a90b9b) (9873ecca1)
- ported from UXP: Issue #1826 - Canonicalize nested calc() serialization (f94a6386) (b2f897bcb)
- ported from UXP: Issue #1826 - Add typed calc() arithmetic for media queries (12b120db) (256258afd)
- ported from UXP: Issue #1826 - Serialize special calc() number values (e85f7787) (0160a02ff)
- ported from UXP: Issue #2982 - Follow-up: allow color-mix to work with oklab and oklch (20b2b3b9) (306ec5e3e)
- ported from UXP: Load mochitest modules without imp (ee00ac98) (21a0bd5f5)
- import from UXP: Remove stale imagebitmap support file entry (4337565d) (abb2284a4)
- ported from UXP: Issue #2404 - Enable CSS aspect-ratio sizing (0d684399) (71c1eca45)
- ported from UXP: Support CSS sizing math functions (93899c01) (309cc07f6)
- import from UXP: Revert "Cloudflare Image Resizing fix take 2" (98c9f738) (db50dc617)
- ported from UXP: Issue #3089 - Support logical border radius properties (27f2a086) (23873f6c0)
- ported from UXP: Issue #1826 - Parse calc() weights in color-mix (4634a74b) (e84b76843)
- ported from UXP: Issue #2506 - Support range media query syntax (6df85ff5) (d7e253b9a)
- ported from UXP: Support CSS shadow parts (51767db3) (c84685d78)
- import from UXP: Whitelist virtual GPUs (VirtualBox, VMware, VirtIO (QEMU and forks like UTM), Parallels) (235bcb01) (1f850cf78)
- ported from UXP: Issue #3092 - Refactor WASM compilation handling (a7a75b78) (995ff8e86)
- import from UXP: Issue #3092 - Add new GC sweep tasks. (47746b47) (cada0ccb8)
- ported from UXP: Issue #3092 - Implement BackgroundFinalizeTask for parallel garbage collection finalization (c0677633) (fd2399f50)
- ported from UXP: Issue #3092 - Implement parallel sweeping and compaction tasks for improved garbage collection performance (3433d538) (9a90ec0ba)
- import from UXP: Issue #3092 - Initial idle GC implementation (18ddd00a) (5a6d30bb9)
- import from UXP: Issue #3092 - Perform a minor GC on tab close (1d3dad15) (54a4438d6)
- ported from UXP: Issue #3092 - Fix unsafe GC multithreading changes (f0cba412) (a208479a1)
- ported from UXP: Issue #3092 - Safely parallelize GC background finalization (e9826f55) (8ff6df57a)
- import from UXP: Fix JS shell module hook build (6f47a2b0) (db6ada440)
- ported from UXP: Implement ES2024 grouping and resolver builtins (e1b689d3) (5ea63a633)
- ported from UXP: Implement ES2024 ArrayBuffer transfer APIs (3be309fa) (25d353174)
- ported from UXP: Make WeakRef support always enabled (6861bedf) (f2854c5a5)
- import from UXP:
 - Implement FinalizationRegistry (50c1419e)
 - Fix FinalizationRegistry constructor realm prototype (890fb3f3)
 - Fix WeakRef constructor realm prototype (f3c6da59) (d0757ae7b)
- ported from UXP: Allow symbols as weak collection keys (e317bf10) (e7c381f1b)
- ported from UXP: Implement growable SharedArrayBuffer (2e51dc9f) (a05676bc4)
- import from UXP: Implement Atomics.waitAsync (22cb0231) (ef078cc66)
- ported from UXP:
 - Implement resizable buffer view semantics (6f3f17ba)
 - Guard typed array JIT paths for resizable buffers (d97a2eb0)
 - Support DataView on shared array buffers (36647658) (84e30bc61)
- ported from UXP: Support BigInt Atomics waiters (8b09714b) (0e7642199)
- ported from UXP: Fix resizable DataView out-of-bounds semantics (aea80980) (7f0c3f82c)
- import from UXP: Validate typed array methods on resizable buffers (3cb76bb2) (210966fa1)
- import from UXP: Validate typed array set and constructors on resizable buffers (7613d290) (26f6831f1)
- import from UXP: Fix ArrayBuffer slice after resizable source shrink (61912d23) (a086de655)
- import from UXP: Fix ArrayBuffer storage and error types (e21c4e29) (b85992b66)
- import from UXP: Fix incorrect variadic for `size_t` in `fprintf` statement in `hyphen.c`. (a02580da) (ab8034125)
- Revert "import from UXP: Issue #3092 - Perform a minor GC on tab close (1d3dad15)" (19d4d490b)
- import from UXP: Issue #888 - Follow-up: Spot-fix; set PREFIX in Dav1d config on Apple. (3155b16f) (a0dd0b793)
- import from UXP: Fix build on NetBSD/aarch64 (13339f6e) (c3fc55652)
- Revert "Implement FinalizationRegistry" and related commits. (287867cac)

[AstroSkipper]

Here’s an alternative take on the situation:

The drama behind the scenes: Why Google lost its soul

There are three fundamental reasons why Google Search has become so abysmally bad and all our observations are absolutely spot on:

Surrendering to SEO spam (the ‘Dead Internet’ theory): The web is inundated with automated, AI-generated pages written solely for the Google algorithm, with the sole aim of selling advertising space. Google has, in effect, lost the arms race against this flood of spam. They’re trying to patch it up with filters (like the latest Core Update), which then leads to total failures where forums like MSFN end up wrongly in the digital gulag.

The displacement of the ‘old web’ (Digital Oblivion): The permanent deletion and de-indexing of old content, which has been going on for a long time, is a tragedy. Google is optimising its server costs. Old forum threads, static HTML pages from the 2000s and small private tech blogs are mercilessly kicked out of the index because they don’t generate enough traffic or aren’t ‘mobile-optimised’. This systematically wipes out the collective knowledge of the internet.

The compulsion towards ‘synthetic answers’: Google no longer wants to send people to external sites like MSFN. They want them to stay on their site so that you see their adverts. That is why they are trying to summarise everything in AI overviews (SGE). This leads to the actual search index withering away, because it now serves only as fodder for the AI engine, rather than as a guide to real websites.

Edited May 23 by AstroSkipper

[roytam1]

On 5/13/2026 at 6:58 PM, roytam1 said:

Update: electricity company postponed this to early-June, exact date will be notified later.

Notice: a 5-yearly electricity equipment check is re-scheduled in 03 June 2026 (10:00-16:00 UTC+8). Download server is unavailable during the period. And just in case machine can't automatically power-up after this outage it may be needed waiting me to go to the site to troubleshoot so the outage ending time may be postponed due to this.

[modnar]

19 hours ago, AstroSkipper said:

Ok. Memory Fox 1.0 is a webextension and not a legacy one. That explains a lot. Therefore, it won't work in New Moon 28. I had a look inside the XPI file, and the popup.js is a script for discarding tabs. No more, no less. In case you are running Serpent 52 in single-process mode, it can't really help. Real tab discarding is only possible if running the browser in multi-process mode. 

You're quite correct, I already deleted this extension from my installation.

Sad too, because this latest version (2026-05-22) is super leaky, runs out of memory so fast (on imgur) and even if you just go to majorgeeks.com and then come here writing the reply is all choppy. What is being put in the water upstream?

Edited May 23 by modnar

[roytam1]

1 hour ago, modnar said:

You're quite correct, I already deleted this extension from my installation.

Sad too, because this latest version (2026-05-22) is super leaky, runs out of memory so fast (on imgur) and even if you just go to majorgeeks.com and then come here writing the reply is all choppy. What is being put in the water upstream?

most suspected code that may cause this is GC related one. https://repo.palemoon.org/MoonchildProductions/UXP/pulls/3093

[reboot12]

@roytam1

Make patch mailnews and icedove for properly handle icon default system browser. Now icon is not readed and fallback to CMD icon:

I noticed problem on WinXP 32/64-bit and also on Win7
Patch you need apply probably in omni.ja\chrome\toolkit\content\mozapps\handling
 

Yea, this maybe fixed in this way - tested on WinXP 64-bit:
Replace line 154 in dialog.js:

elm.setAttribute("image", "moz-icon://.exe?size=32");

to this code:

try {
  let wrk = Cc["@mozilla.org/windows-registry-key;1"]
              .createInstance(Ci.nsIWindowsRegKey);

  wrk.open(wrk.ROOT_KEY_CLASSES_ROOT,
           "http\\shell\\open\\command",
           wrk.ACCESS_READ);

  let cmd = wrk.readStringValue("");

  wrk.close();

  // extract path
  let match = cmd.match(/"([^"]+\.exe)"/i);

  if (!match)
    match = cmd.match(/^([^\s]+\.exe)/i);

  if (match) {
    let exe = match[1].replace(/\\/g, "/");

    elm.setAttribute(
      "image",
      "moz-icon://file:///" + exe + "?size=32"
    );
  }
  else {
    elm.setAttribute("image", "moz-icon://.exe?size=32");
  }
}
catch (e) {
  elm.setAttribute("image", "moz-icon://.exe?size=32");
}

Edited May 23 by reboot12

[johk]

10 hours ago, roytam1 said:

most suspected code that may cause this is GC related one. https://repo.palemoon.org/MoonchildProductions/UXP/pulls/3093

I hope that is a fix for the regular and constant CPU spikes happening after visit sites with loads of JavaScript (example comes to mind: Twitter). After leave the site, is a constant rate of CPU spikes wherever you browse then.

For now I leave the update for tomorrow or later, but I really hope it fixes it. With the summer heat, having spikes constantly rising the CPU temperature is insane, even when you are browsing plain text sites.

 

OFF-topic, isn't the Google chatty a bit off-topic here? I don't use Google since about 10 years, preferring, by far, Bing/Yahoo (same engine), and since Bing hardcoded redirects as links to sites, duckduckgo now. Google is not a trustable search engine since long ago. I have a personal example that... I prefer not to share... but if you find my content, won't be because Google, so make a favour to yourself and use any other search engine that actually index Internet.

Edited May 24 by johk

[Guest raddy]

On 5/23/2026 at 1:59 AM, roytam1 said:

Win32 https://o.rthost.win/palemoon/palemoon-28.10.7a1.win32-git-20260523-d849524bd-uxp-829418a939-xpmod.7z

Failed for me, high memory consumption and cf captcha tends to loop..
 

[roytam1]

34 minutes ago, raddy said:

Failed for me, high memory consumption and cf captcha tends to loop..
 

make sure your browser sends referer.

[AstroSkipper]

12 hours ago, johk said:

Bing/Yahoo (same engine), and since Bing hardcoded redirects as links to sites, duckduckgo now.

Microsoft's Bing is just as bad as Google, and DuckDuckGo is dependent on Google because it uses its data and therefore no longer usable.  You don’t seem to have fully grasped the implications. Google, Bing, Startpage, Ecosia, DuckDuckGo, etc. have thus become useless. This is also about search engines under New Moon 28 and Serpent 52. If that isn’t important, then I don’t know what is important at all. As far as I’m concerned, all these search engines are dead and are now just history. 
 
You're worried about your CPU spikes when using the most recent version of New Moon 28.  But just think about what browsers were created for! The main function of browsers is to access websites, which must first be found by search engines if you don’t know their address. That’s what I do most of the time with @roytam1’s browsers, such as New Moon or Serpent. And I do it on a real installation of Windows XP.

Edited May 24 by AstroSkipper
Update of content

[NotHereToPlayGames]

There is the other way to look at it, and it is just as valid.

The OWNER of MSFN has not (or can not!) "fix" the 'hack redirects' being discussed in that thread I directed you to, that those have been getting WORSE and WORSE since they were first brought to our attention, that the OWNER of MSFN has just ACCEPTED that as "it is what it is", and all of those search engines removed MSFN from their results AS A RESULT of the OWNER of MSFN not being able to "fix" those 'hack redirects'.

CAUSE AND EFFECT.

[NotHereToPlayGames]

Those "spam redirects" were brought to our attention in FEBRUARY.

If you ask me, we should consider ourselves "lucky" that all of those search engines didn't remove MSFN results WAY BACK IN FEBRUARY.

We can blame Google and Microsoft all we want (that is an "easy" blame-game that we have all played over the years).

But if you ask me, the "first domino" was LACK OF SECURITY PROTECTIONS, the second domino was those "spam redirects", the third domino was the OWNER of MSFN "unable" to patch the security hole that ENABLED those "spam redirects", the last and final domino is the search engines left with no other choice than having to remove *ALL* search results because the MAJORITY of the results were now SPAM REDIRECTS and not REAL "url results".

We can blame any single one of those dominoes.

But the "first" seems more important "blame" than the "last", in my opinion.

[AstroSkipper]

The issue at hand goes far beyond MSFN. In my article

On 5/23/2026 at 1:10 AM, AstroSkipper said:

Here’s an alternative take on the situation:

The drama behind the scenes: Why Google lost its soul

There are three fundamental reasons why Google Search has become so abysmally bad and all our observations are absolutely spot on:

Surrendering to SEO spam (the ‘Dead Internet’ theory): The web is inundated with automated, AI-generated pages written solely for the Google algorithm, with the sole aim of selling advertising space. Google has, in effect, lost the arms race against this flood of spam. They’re trying to patch it up with filters (like the latest Core Update), which then leads to total failures where forums like MSFN end up wrongly in the digital gulag.

The displacement of the ‘old web’ (Digital Oblivion): The permanent deletion and de-indexing of old content, which has been going on for a long time, is a tragedy. Google is optimising its server costs. Old forum threads, static HTML pages from the 2000s and small private tech blogs are mercilessly kicked out of the index because they don’t generate enough traffic or aren’t ‘mobile-optimised’. This systematically wipes out the collective knowledge of the internet.

The compulsion towards ‘synthetic answers’: Google no longer wants to send people to external sites like MSFN. They want them to stay on their site so that you see their adverts. That is why they are trying to summarise everything in AI overviews (SGE). This leads to the actual search index withering away, because it now serves only as fodder for the AI engine, rather than as a guide to real websites.

I have shared some in-depth insider knowledge and outlined the reasons behind Google’s total decline. And Google is dragging most search engines down with it. From now on, we can remove the term ‘to google’ and, in my native language existing for 30 years, 'googeln'  from our vocabulary, unless you’re into adverts and misinformation. MSFN is just one of many to suffer the consequences.  If it were just a question of spam, there would be hardly any websites left to visit. It's all about money and power.  That's something you should know well from your boss. 

Edited May 24 by AstroSkipper