r3dfox, a modern Firefox based web browser for Windows Vista, 7, and 8!

[Mathwiz]

Re: Forcibly reinstalling uBO, I agree with NHTPG. Unfortunately the author appears to have missed the point:

Quote

I'd rather the average user have uBlock Origin by default rather than going without it. It's also convenient in testing because I don't have to go and grab it when I want to test something and not deal with 5000 ads.

Librewolf moved to the new policies.json layout, and if I were to revert back, the method to enable it in private mode was complex, and liable to break in newer browser versions. Librewolf doesn't use the old method anymore, so if it broke, I'd be on my own. So i just followed what they did.

The average user should always browse with an ad blocker, and if you wish to test without it, you can remove it from policies.json.

Re: the first point, I would too! On today's Internet, I'd never browse without an ad blocker, and uBO is the best IMO. But there's a difference between installing uBO by default, and reinstalling it if the user intentionally uninstalls it, effectively making it an unavoidable part of the browser!

I don't think the author fully appreciates the gravity of forcibly installing a third-party's add-on, no matter how good that third party may currently be. What if some hypothetical company (which I'll call "Poogle") makes gorhill an offer he can't refuse to exempt Poogle ads from being blocked by uBO? Or what if gorhill gets hit by a truck, and his heirs aren't software developers, have no interest in continuing uBO development, and just sell it to Poogle outright?

(As for having uBO always available when testing, the author could add it to his own policies.json; that doesn't mean it has to be in the distribution for everyone.)

As to the second point, editing a .json file is a bit beyond the ability of the average user. There needs to be a way for non-techies to do this.

But in any case, it appears the author has no intention of addressing this issue.

[NotHereToPlayGames]

I seem to recall that the developer has publicly asked for assistance and nobody stepped up to the plate.
Granted, at the time, that request was in reference to XP/Vista support.

The userbase (my opinion) has grown INCREASINGLY impatient with the "release schedule".
You can find Reddit posts asking if r3dfox is "abandoned" if it goes so much as ONE MONTH without a new release.
ONE MONTH !!!

There are also requests for NIGHTLY BUILDS !!!

And every time a Firefox ESR "security bulletin" is posted, the board jumps to life BEGGING for an update.
Begging, Begging, Begging!

Personally, in my humble opion, this is where A.I. should become everybody's "best friend".

The compile process takes roughly 1/2hr and while that may not sound like much, most everybody does have LIVES.
Not everybody is a "robot / machine" dedicatingly programmed into a "ritual" of posting software updates each and every week.
Let's be honest, not many of us are WILLING to dedicate the amount of TIME out of our LIVES that folks like @roytam1 does every week!

A.I. could do all of this for us!  And we'd have that "nightly" version of r3dfox.

If a young programmer wants to "make a name" for him/herself, the INSTRUCTIONS on how to build r3dfox is publicly posted.
No takers so far!  We can all read into that what we will.

Edited February 22 by NotHereToPlayGames

[user57]

was i really the only person that could do it ? from what i remember i wrote 150 functions or something but then rather going for XP that code was used to create a vista version

so i think the guys who can do this might should come out now 

[roytam1]

4 hours ago, NotHereToPlayGames said:

not many of us are WILLING to dedicate the amount of TIME out of our LIVES that folks like @roytam1 does every week!

not very "every week" now as upstream doesn't as active as before.

[user57]

i would ask samuel about this, he has a good base about this
it would be interesting to hear his meaning to this problem

[K4sum1]

So I have kept up with the messages in this thread via email, however I haven't taken the time to go here and reply until now. Since that was a while ago, I have a lot to reply to. Mostly on account of me either being busy at the time, not feeling like it, or forgetting that this thread exists here and that I needed to write a reply like this here.

On 2/15/2025 at 1:59 AM, Jody Thornton said:

I took neither from that.  It appears to me that r3dFox will use release v134 on the next build.

On 2/15/2025 at 11:31 AM, D.Draker said:

There would be no mystery then, @K4sum1 wants mystery. Those fellas (K4sum1and Compa), they love mystery, at least from what I know about them.

To be completely honest here, I had said there was going to be no r3dfox 133 in the past, and well there was no "r3dfox" 133. It also coincided with moving to a new codebase for the UI, so it also made sense to do it for that reason as well. I also wanted like a third branded browser separate from r3dfox and ESR, and I actually still use the Plasmafox build to this day as that third browser.

On 8/3/2025 at 12:41 PM, VistaLover said:

IMHO though, any r3dfox-specific change likely to interfere with user customisation, however small, IS worth mentioning in Release Notes...

Luckily for me, I found my way to 140's source code in GitHub and immediately noticed what looked to be the culprit commit: 

https://github.com/Eclipse-Community/r3dfox/commit/c69c8d07e9e421ef8b3f3ea6b730f8b3a4443fa8

With respect to the browser author , changing the "aboutAddons.ftl" file, part of the embedded en-US locale, isn't a smart thing to do, unless: 1) you restrict users exclusively to the embedded en-US locale, 2) you have the resources to produce yourself (i.e. translate from English-US) the rest of the locales; e.g., once one installs the Mozilla-provided 140.0.4_en-GB locale, the author changes are being reverted to what the Mozilla wording is:

I'm not sure why I didn't say this. Normally I check the commit history, and use that to make the release notes. Maybe I just didn't remember making any major changes and just skipped that process idk.

On 9/3/2025 at 12:13 PM, Jonathan-nighthawk said:

Til yesterday, all was good, today i found that all the passwords i had stored in the r3dfox (around 20 sites) are gone (deleted, disappeared) except one, and i don't know why.

I know the Firefox password system is weird, but that's not something I'd have touched here. I assume it was a browser bug, and being so long ago and no other report, I assume it was either a one-off thing or got resolved in upstream. I will say though in all my browsers that I use, I've never had this happen, Except in Chromium where it nukes your profile if you look at it wrong, but that's an entirely different engine.

On 9/21/2025 at 10:01 PM, Mathwiz said:

Not me. I've never understood that. I'm still using a 10-year old 4G phone. People like you describe act as if they actually have money trees, throwing away an over-$1000 piece of equipment every 2-3 years just because there's an even more expensive version out!

Same here. My PC is pretty good and modern because of pretty much just compiling. I got a 5950X when AM5 was out, and it was on discount for being last gen and old socket. GPU is nice to have for gaming, but like the realistic cap for me for a while will be a 3080 Ti/3090, and even then I had to downgrade to a 1080 from my 2080 Ti and that does fine for the games I want to play in the here and now. Outside of that it's just faster CPU for faster compile times and more RAM for more stuff open.

My main phone is going to be 9 years old this year, and my main laptop is going to be 13 years old this year. Around 8 or so months ago I was using my laptop as my main machine and remoting into build machine to make r3dfox, and albeit a little slow, it served me completely fine for that time. I only went back to main so I could play BG3 at a smooth framerate with decent graphics. Maybe with the exception of maybe soon upgrading to a phone that is a few years newer (6 years old this year), I don't see myself replacing either with something new before I upgrade my PC again. Likely won't be until AM6 comes out and I upgrade to whatever the last AM5 16 core part is that year, and probably 96GB or 192GB RAM, and repeat for next socket. (Assuming I'm alive then and I have the money, both of which I'm not sure of how likely they are.)

On 9/21/2025 at 10:01 PM, Mathwiz said:

Because that's how the World Wide Web was designed to work, that's why! The original idea behind HTML was that, no matter how many fancy bells and whistles were added later on, a Web page should still look the same to folks using a browser that didn't support the new bells and whistles. The page may be slow and look like one of those ugly pages from the '80's, but it's still supposed to work. (And for the most part, the WWW really did work that way for its first couple of decades.) I know we got away from that ideal long ago, but I still think it's an ideal worth striving for, rather than Discourse (or whoever) shutting you out of their sites completely because your browser/OS doesn't support all the HTML features they think they might want to use someday.

This is how the Eclipse website is designed, it looks nice in modern browsers, and it displays fine in even ancient browsers. Like the only major issues are old browsers don't grab the custom font, and there's not really a good fallback for the textbox background, so it's just overall lighter shaded. The only other downgrade I can name off the top of my head is the logo image requires a pretty recent browser to change on hover, not even Firefox 115 can do it. It uses the :has selector, and I don't know of any other way to implement it without using JS.

I do want to port the website to an even older layout that will display the same from IE3 up to latest browsers, and I have the base framework done for it, but I just haven't ported all of the text into it. I might even make it smart and serve the legacy one for http and the modern one for https.

The forum is similar, but the user interface might actually be better for older browsers, however it requires the user to manually switch to the legacy UI, I don't think automatic switching is possible.

On 9/21/2025 at 10:01 PM, Mathwiz said:

You misunderstand the situation. We don't get to choose whether to use "old school" or "new school" methods! If the Web designer used the "old school" method of sniffing the UA (e.g., chase.com), then we have to use the corresponding "old school" method of spoofing the UA just to get in! Of course, we often have to use "new school" methods as well, like those built into a Web browser like R3dfox or Supermium, or the site will likely not work well (see above) but that doesn't mean we can ignore UA spoofing just because it's considered "old school."

I really hate this method of checking the browser. Like all they give is a non-descript website is not going to work on this browser f*** you, and there's no obvious way to force it to load. I had to block the discourse script and do other workarounds for other sites when I was using Mypal68 as my main.

On 9/22/2025 at 4:22 AM, NotHereToPlayGames said:

Not the folks that I know.

They live paycheck to paycheck, not a dime in savings, but can't keep a phone more than a mere six months or so without cracking it.

They upgrade OFTEN, *very* often, not because they can "afford to", but because they eventually get tired of the cracked screen that their NEGLIGENCE added to the phone they do have.

On 10/19/2025 at 9:41 PM, Mathwiz said:

Accidents happen, but cell phone insurance and repair shops exist, so I'd say that if you're (a) so chronically careless with your phones that you're repeatedly breaking their screens, and (b) using that as an excuse to buy a new phone rather than getting your old one repaired, methinks you really wanted a new phone all along. So evidently, they can afford a new phone after all (or at least, they think they can).

For me, I'm not the most gentle, but I try not to be careless. I've had multiple parts break over multiple phones, and unless it was such an old phone where getting another used was more economical than getting the required parts, I've repaired my phones every time. I assume the problem here is phones have gotten harder to repair and parts are more expensive. I also have no idea where I would even go if I wanted someone other than me to repair a phone. Back to parts, I've never been able to take an OLED screen off of a phone without obliterating the screen, while my main LCD phone I went at with 0 heat and a razor blade and it was completely fine. On top of that, OLED screens are much more expensive, like 5x or more the price of older LCD phones, and it gets to the point where it's almost more worth it to get another or a newer phone even if used it's like $200+ because the screen costs that much.

I would say the same for laptops, but that's a bit different and I've had like a bunch of sidegrades for xyz reasons. I had a MSI GS60-2QE with a 4820HQ and 970m, but I got tired of dealing with Optimus and two RAM slots, so I got HP 8770W with a 3630QM and upgraded it with 32GB and 980m, until the 980m died and that was a $250 part and I was tired of lugging the 8770W around and on top of that MSI decided f*** you you only get 4GB of available RAM, so I got a T430 and put the 3630QM in it, but then I got tired of Intel integrated graphics being barely able to run games, so I got a M4800 with a 4810MQ, and upgraded the GPU to a M2200 and display to a 1080p 120hz panel, and here I am now. I almost want to get a 8570W and transfer over the M2200 due to weird (I assume) BIOS related issues with the M4800, but the 8770W also had weird issues that I assume the 8570W would also have, and screen upgrade is rare to get the required parts or expensive, so eh. I'll just use this until the higher end P5x/P15 Gx Thinkpads become cheap and grab one of them. (P15G2 with A5000 is still $2k wtf)

On 9/26/2025 at 2:59 PM, Jonathan-nighthawk said:

I know that some peoples have a "psychosis" that make them change their devices each time one new pops out, but i just cannot understand them.

Meanwhile here, if I had the money, I literally don't know of any "new" phone or laptop that I would want to get. I'd just get what I've researched as myself wanting either because it's a better experience or just faster. The newest phone I'd want is 3 years old this year, and even then idk if I'd want to use it as main, and the P15G2 is 5 years old this year.

On 9/30/2025 at 7:28 PM, VistaLover said:

Also, I have many months myself to spot even a €100 banknote (green); the highest denomination in circulation for everyday use is currently the €50 banknote (brownish); besides, governments have made it mandatory to pay with "plastic" money (credit/debit cards, physical or digital forms of them, money transfers, etc.) for a sizeable portion of your expenses (40% here in Greece), else you'll be "fined" with an increased income tax; so, people tend to move away from cash; the youth of today don't normally have cash in their pockets, this includes coins, too... And you can't buy here something that is over €500 worth with just plain cash (in one go) - you can probably still pay for the first €500 with banknotes, the rest has to be paid with some other "money" form...

As an American, this sounds so absurd. A lot of places here don't take $100 bills, but like that is it. A lot of places, even if they don't take $100 bills, give you a discount for paying with cash, or charge extra for using a card. It literally costs you more to use a credit card, and it costs them more to take a card payment rather than cash.

On 11/12/2025 at 8:03 AM, ED_Sln said:

Does anyone know what happened? The repository on GitHub is archived, and the website is down.

On 11/12/2025 at 9:36 AM, NotHereToPlayGames said:

I haven't heard anything.

The recent Eclipse Community discussion seemed to be developmental proposals on the direction the project was headed.

There were some outspoken disapprovals of those proposals.  But it didn't come across as "hyper-critical" or anything like that.

But sometimes you never know, it is his project and maybe he just got tired of those "disapprovals".

I kind of doubt it, and hate to "speculate".  I'm sure we'll know more soon...

On 11/12/2025 at 12:05 PM, VistaLover said:

I believe so; he probably had a "hissy fit" and decided to "now I'll show you all", or I could be totally wrong and the GitHub repo archival was an inadvertent mishap ...

What had happened was I had gotten the feeling that the website and repo may have had the possibility of being attacked and/or spammed, and I was very tired and wanted to sleep, so I just locked everything down so I could sleep. I think I either forgot or was too busy to think about reopening them though until I was already awake for a while though, which is why they were locked for like ~20 hours if I remember correctly. 

On 11/12/2025 at 12:05 PM, VistaLover said:

That  ; and those "proposals" did materialise into a Librewolf-ification of r3dfox that many, myself included, never asked for  ...

As @Jody Thornton put it, 

Quote

If it were up to me, I would just want r3dFox to be stock Firefox ESR
140.x, with the ability to run it on Windows 7 and Windows 8. That's it.

... and I'll add Vista SP2 to the OS mix above  ; yes, close to "stock" Firefox but with the ability to launch on older WinOSes! This is what most site admins expect, this is what most extension authors expect and target...

I was never part of the "extreme web privacy" crowd to demand a change of route towards Librewolf (or similar forks); I understand a small portion of the LW code was needed to address a specific r3dfox technical issue, but that is different to incorporating large chunks of LW code "while we're at it" ...

As if it wasn't enough to deal with Mozilla "breaking" things (and locking down the browser) with each major version update, an "average" r3dfox user has to deal with "r3dfox-specific" changes, too (ones that not always meet with said user's "approval") ...

And my own words on DRM/EME:

I found r3dfox maintainer's "obsession" about DRM simply "blown out of proportion"; he goes to extreme lengths to disable EME at buildtime, but the browser itself provides an easy way to disable EME at runtime, if one objects to it for whatever ideological reason...

Let's face it; with Google practically owning the Web, they have leveraged the use of their own CDM (Widevine) in most media services, even the most obscure, but still free, ones... Yes, I totally understand the argument about "black-boxed code" etc., but DRM has become a necessary evil in the web era of 2025 and beyond...

A lot of focus has been put on the VMP (Verified Media Path) requirement associated with the majority of the prominent/commercial DRM'ed Video+Audio services (e.g. Netflix, Disney+, Paramount+, Spotify, Tidal, Apple Music, etc.) as a reason NOT to implement DRM on r3dfox (because VMP entails a very large sum of money, paid to Google, for certifying the browser for VMP purposes), but what about the rest of the lesser known services that don't impose VMP with DRM? Jody's example of 

https://www.cp24.com/now/

is such a case, there are many others...

On 11/12/2025 at 11:36 AM, Jody Thornton said:

I was worried that may be the case too, as I was a tad vocal on my thoughts of where r3dFox is headed, on the Eclipse Fourm.  However, that isn't my saying "You have to do this or that!".  I'm vehement on my opinions, and make no apologies for it.

In any case, I have now moved on the Firefox for Windows 7 and 8, since releases seem to be happening more quickly.  This is especially the case for ESR builds.  I'm glad I switched.  Even if my comments were not made, I'm sure there were a lot of vocal proponents of my mindset, including one being a moderator.

r3dfox originally started as mostly just Firefox but for 7/8, but then I wanted to make it so I had to change less to set the browser up for myself later, and then https://github.com/e3kskoy7wqk/Firefox-for-windows-7/ came to exist, and that fulfilled the idea better than I could have. So I started making the browser more in my image. I believe all the changes I make are beneficial for the normal user, and where realistic, I try to add options to change or revert new behavior or styling that I can see as being potentially unwanted in the settings. I would rather you use r3dfox since well it is my fork and I feel like I need a reason to exist ;-;, but like you don't have to. I've been kinda in the "extreme web privacy" crowd, but I run r3dfox announcing to every website that it is r3dfox which is a rare fork, so I definitely have a unique fingerprint, but I do all of this and run privacy extensions and tweaks idk why I'm like this.

DRM being removed is part of the same reasoning, I don't use anything that uses DRM, and although they do exist, not many services work with non-VMP or give you a good experience with non-VMP Widevine. Since I'm not using it, I don't need the DRM and I feel easier about it if my browser comes with no closed source DRM at all, so I disable it. On the Eclipse forum, they've devised methods to re-enable DRM even after my changes, and that's probably the way to go if you want DRM in r3dfox rather than me having to ship it to everyone.

On 11/15/2025 at 10:19 PM, Mathwiz said:

I just stumbled upon a curious issue with R3dfox ESR. I tried to update from R3dfox 139 to R3dfox ESR 140, and none of my profile settings were migrated! It was as if I was starting over with a clean profile.

A little investigation revealed the R3dfox ESR profiles were in a different directory: "R3dfox ESR" instead of merely "R3dfox." So my old profile was still there; R3dfox ESR just uses a different profile directory.

OK, I thought, I'll copy my old profile to the new "R3dfox ESR" directory and switch to it with the profile manager. But when I tried to "create profile" and browse to the copied profile directory, it acted as though it wasn't even there! All it would show me was the new, clean profile it had created for me. The copy of my old profile was definitely there, but R3dfox ESR wouldn't show it. I was forced to give up and install a regular "R3dfox" version (I picked 142 for now). This did work and all my bookmarks, passwords, add-ons, and the like came back. But WTF? Why can't I migrate a R3dfox profile to R3dfox ESR? Is this a new Mozilla thing or an Eclipse thing? Weird!

Another anomaly was that R3dfox ESR offered to migrate my settings, but not from MS Edge, as "regular" R3dfox had done; rather, from Internet Explorer! What good is that? I haven't used MSIE in years and years! I don't use Edge any more but it's still installed.

So, they use different profile directories, and profiles are weird. I think just renaming the r3dfox profile directory to r3dfox ESR would be the best way to do it, but I'm not sure if that works, or if that still requires manual intervention in about:profiles to set the old profile as default. I just use portable and that accepts only one profile, and won't try to make a new profile.

Also the browser should have the code to import settings from even IE6, but due to well not having an IE6 profile or the will to make one, I have no idea if it even works. It was basically only ever in the browser because it technically broke importing from more modern IE and it is technically related to 7 support. It's still in the browser because (I assume, I don't handle the base 7 support anymore) it has practically no maintenance cost, so it's like a why not.

On 2/8/2026 at 3:06 AM, NotHereToPlayGames said:

re: r3dfox and it being LibreWolf-ified, there is a way to PREVENT it from ramrodding most-recent uBO.

I personally love uBO, but I also prefer the user to be in FULL CONTROL of what extensions they opt to run.

You'll notice that the LibreWolf-ified r3dfox does not allow a uBO uninstall, and it reinstalls itself if you delete uBO from your profile.

Again, nothing against uBO, but I'm no fan of The Nanny State, and that includes r3dfox deciding for itself that I have to run uBO because that is "them looking out for me".

I don't know what to do here, I like the idea of including uBO by default, and if possible, I would like to make uninstalling it actually uninstall it and not have it come back. However as far as I know that's not possible. So when it comes to do I make both my experience in testing better (not dealing with ads in a fresh profile or take the time to manually install uBO), and make the experience of the layman who may not even know what uBO is better, vs not including uBO. I'm going to take the choice of including uBO. I don't know what to do here. I feel like a decent ad blocker is the bare minimum for a usable web experience.

On 2/21/2026 at 10:16 PM, Mathwiz said:

Re: the first point, I would too! On today's Internet, I'd never browse without an ad blocker, and uBO is the best IMO. But there's a difference between installing uBO by default, and reinstalling it if the user intentionally uninstalls it, effectively making it an unavoidable part of the browser!

As to the second point, editing a .json file is a bit beyond the ability of the average user. There needs to be a way for non-techies to do this.

But in any case, it appears the author has no intention of addressing this issue.

I don't know of any good way to add uBO on first run, but let the user uninstall it without editing policies.json. Unless there is some other way I just don't know what to do here.

On 2/21/2026 at 10:16 PM, Mathwiz said:

(As for having uBO always available when testing, the author could add it to his own policies.json; that doesn't mean it has to be in the distribution for everyone.)

I feel like that would be just as time consuming as manually grabbing it, because I would have to manually replace the policies.json every time, and I'd have to keep a separate version of it up to date which I know I will forget to do. Also I would go about testing r3dfox, not grab uBO, see ads, get annoyed, then get uBO. I'd need to preemptively apply the uBO installing policies.json, which I also know I will forget to do.

On 2/21/2026 at 10:16 PM, Mathwiz said:

I don't think the author fully appreciates the gravity of forcibly installing a third-party's add-on, no matter how good that third party may currently be. What if some hypothetical company (which I'll call "Poogle") makes gorhill an offer he can't refuse to exempt Poogle ads from being blocked by uBO? Or what if gorhill gets hit by a truck, and his heirs aren't software developers, have no interest in continuing uBO development, and just sell it to Poogle outright?

In that case, I'm not going to ship something I won't use. If uBO gets enshittified, I'll replace it with something else. The only way that scenario could play out is if "Poogle" or someone gave me millions to sell them my browser, something which nobody will ever pay millions for. If that somehow happened I'd see that as a no-brainer easy money, and I would still need a decent browser for myself, so I would just make a new fork, or even sit back and pay someone to develop it for me and continue there. (Or I guess it could happen if I die, r3dfox gets discontinued, and then enshittifed uBO, but in that case r3dfox will be outdated and something better will exist and you would probably just want to use that instead anyways.)

On 2/22/2026 at 4:32 AM, NotHereToPlayGames said:

I seem to recall that the developer has publicly asked for assistance and nobody stepped up to the plate.
Granted, at the time, that request was in reference to XP/Vista support.

The userbase (my opinion) has grown INCREASINGLY impatient with the "release schedule".
You can find Reddit posts asking if r3dfox is "abandoned" if it goes so much as ONE MONTH without a new release.
ONE MONTH !!!

There are also requests for NIGHTLY BUILDS !!!

And every time a Firefox ESR "security bulletin" is posted, the board jumps to life BEGGING for an update.
Begging, Begging, Begging!

Personally, in my humble opion, this is where A.I. should become everybody's "best friend".

The compile process takes roughly 1/2hr and while that may not sound like much, most everybody does have LIVES.
Not everybody is a "robot / machine" dedicatingly programmed into a "ritual" of posting software updates each and every week.
Let's be honest, not many of us are WILLING to dedicate the amount of TIME out of our LIVES that folks like @roytam1 does every week!

A.I. could do all of this for us!  And we'd have that "nightly" version of r3dfox.

If a young programmer wants to "make a name" for him/herself, the INSTRUCTIONS on how to build r3dfox is publicly posted.
No takers so far!  We can all read into that what we will.

I want the AI to replace me. When it gets to the point where AI can handle the process of making the browser and do a better job, maybe this could happen. I have no idea how it could work though, or if I will see hardware capable of hosting a local AI able to do all these steps within my lifetime.

How I make r3dfox, is I import e3k's patches as commits into the next version branch, then I cherry-pick over the r3dfox changes. If a conflict is encountered, I check the commit to see what it's changing and what commit to the browser resulted in the conflict. If it's simple, I just adapt the old commit to the new code, or if it's complex I have to revert the bad commit. Which since I'm cherry-picking code over is a pain. I can't just cancel the cherry-pick there, cancelling it will revert all the applied commits, which may also include previously resolved conflicts. So I have to upload the changes, cancel the cherry-pick, which will go back to the previous branch I went to cherry pick from. (Also multiple times I've had the cherry-pick just silently fail mid-way through undoing everything I've done.) So I have to switch to the new branch, and then pull all those changes from the uploaded branch, then revert given commit because if I start by reverting the commit that will f*** up the tree and require merge, then go back to the previous branch, select the rest of the commits to cherry-pick over and continue. Then I have to go and test everything, and if I have an error, I have to dive into that and that's it's whole own thing depending on the component. Mozilla has been f****** with Brotli lately, and I was able to fix it before, but why the seemingly simple changes in 147 broke it to the point where I was unable to compile with it hurt my head so much that I just dropped the Brotli compressed omni.ja idea.

Doing nightly means doing this process more often, and I just don't have that in me. Also even for like updating to point releases or updating the ESR which normally is easy just rebase current branch or pick the changes to new branch and build, except Mozilla has made breaking changes in 140 ESR twice in a row now, I just don't feel like doing it or running that on my PC at the time, so the builds get delayed. The long gaps between versions at times are because I just don't feel like I want to make a new build and I don't feel like it's worth it to force myself to update it yet.

On 2/22/2026 at 5:50 AM, user57 said:

was i really the only person that could do it ? from what i remember i wrote 150 functions or something but then rather going for XP that code was used to create a vista version

so i think the guys who can do this might should come out now 

When was this? I did the Vista compatibility mostly on my own because I finally felt decent and I wanted to get closer to XP. Vista is an obvious target to get on the way, and it was pretty close, so I went for it. 

Also I don't think you can just write your way around the compiler. You would also need to replace the current compilers, something Mypal68 did. I never figured out how to set things up for Mypal68 to compile, so I have no idea how to make a similar setup for r3dfox.

I feel like compiler would be the main starting point, and/or get one step closer to XP by attaining Vista RTM compatibility by either bringing back layers and/or removing the WebRender dependency on DWrite, and then getting GDI and/or Cairo mode to work and not be super unstable (may also be WebRender related).

Now these are outside my current skillset and are open issues. Maybe I could figure something out finally out of sheer will sorta like how I got Vista support in the first place, but currently I feel like s*** and at this rate there's a higher chance of me blowing my own head off than things getting better enough for me to feel not like s***.

So that's all the replies I have for you all. I'll try to reply to the thread more often rather than making one big post after a number of months in the future, but I have no idea how that will go.

[user57]

the LLVM (v17) i made would be an option, it would be newer then VS 2019 
the idea was to connection the functions as usual with the compiler itself

there are certainly different ways to solve that specific problem, one solution is samuels one core api solution 
a other solution would be to rebuild the import table and connection them via assembly
dibya was trying a programmical solution, he tryed to lead the calling functions to his function and then to our functions - but it are over 300 functions - and maybe at some point the .obj trick apears (description below)
so the approach i did was to connect these via the compiler (see below why this dont just work - but it is doable due some approaches anyway)
to do so i wrote a file that compiles up with the firefox code, the missing part was for now the connections

the problem mainly came from the common problem, in VS2019 microsoft made a hidden .obj file 
(for those who read this and dont know what this is - a .obj file is a file with functions - the compile generate these and connect these)
and here is microsoft´s trick
for many functions (like c-runtime) and maybe some SDT functions - microsoft force this .obj file to be used 
after that you get a error "already defined in (that microsoft .obj file)
i wrote about in a other thread, because you can edit the code - but when i changed the code nothing happend
this is because the code is not readed - it just use that .obj file - it never read out that code (the code is already pre-compiled in that .obj file)

to start with version 140 or something would probaly be to much, from chrome i think we started one at v80 or something - then going upwards (mypal for example is still on v68)
i dont exactly remember what compiler was used - but it had that .obj force problem - its not coming so much from the compiler itself, rather that .obj trick

since the firefox community is getting annoyed - just to roll over this one is not that simple - its work that has to be investigated - if it would be simple someone would already have done it
for me i have a lot on the list, having severial chrome´s raise questions to solve this one

from what i remember you dont want to talk much you just want to have it done in like 1 min otherwise you where quickly gone - not having it solved shows that there is a huge problem in just solving it up

i was not making smalltalk i tryed to figure out the things that are needed, but going like that it can be done quickly, then you might should proof that
firefox isnt doing well for them just to kick in new function at the cost of users - one of the reasons it became less popular 

the discussion about the HEVC codec also gone like this - the discussion rather gone like "just build it in there" - but it are things to dig in - the chrome question for example took 2 weeks to compile up - just first to type something

r3dfox is a nice project 

so why firefox with a newer version is not happening ? at least v80 or something 

[K4sum1]

15 minutes ago, user57 said:

the LLVM (v17) i made would be an option, it would be newer then VS 2019 

That might work, Mypal68 is built with LLVM, I think the same LLVM that the Rust compiler generates as it's being built. LLVM 10 and later appear to not work, "version above 9 fails on winxp, something about "sesuretu" protection"

I feel like I'd need to figure out building Mypal68, get a baseline of something that works, then try extrapolating that to modern Firefox. If I can get a build without needing mach bootstrap, I could build up, 78, 91, 102, 115, and so on, fixing issues with the build system as they come.

24 minutes ago, user57 said:

rom what i remember you dont want to talk much you just want to have it done in like 1 min otherwise you where quickly gone - not having it solved shows that there is a huge problem in just solving it up

i was not making smalltalk i tryed to figure out the things that are needed, but going like that it can be done quickly, then you might should proof that
firefox isnt doing well for them just to kick in new function at the cost of users - one of the reasons it became less popular 

I'm not sure I remember this or who you are. I know this wasn't going to be fast, or likely to be done with the latest version. So I just set some basic goals that are maybe doable before doing something like this, hence Vista support, and the issue with DWrite and WebRender, and GDI being broken.

I might try seeing if I can get Mypal68 to compile sometime again and see what I can do from there.

[grey_rat]

In Firefox,  can enable Webrender via OpenGL 3+ previously, this was possible in MyPal 68 on XP.

[feodor2]

4 hours ago, K4sum1 said:

Also I don't think you can just write your way around the compiler. You would also need to replace the current compilers, something Mypal68 did. I never figured out how to set things up for Mypal68 to compile, so I have no idea how to make a similar setup for r3dfox.

 I may could help you just ask.

I already passed by 100 firefox and its fine

rust is greater problem: once I wanted to move to 57 but failed, sure I shall overcome later.

 

[feodor2]

1 hour ago, grey_rat said:

In Firefox,  can enable Webrender via OpenGL 3+ previously, this was possible in MyPal 68 on XP

I never did it myself you please elaborate, suppose webrender on direct9 cards is good feasible.

[K4sum1]

1 hour ago, feodor2 said:

I may could help you just ask.

I asked before about cbindgen, but I still got stuck there and I didn't know what to do or ask. You say to compile it without cargo, but I don't know how you compile something that needs cargo to compile without cargo.

It would really help if you were to set up a VM from scratch to compile Mypal68 and make a step by step guide similar to this. https://board.eclipse.cx/viewtopic.php?t=812

1 hour ago, feodor2 said:

I already passed by 100 firefox and its fine

I love the 68 style UI, I'll have to give it a try as main again sometime.

1 hour ago, feodor2 said:

rust is greater problem: once I wanted to move to 57 but failed, sure I shall overcome later.

I feel like Rust would be the easy part given the existence of Rust9x. Unless you mean other parts of the browser tree, which that I could understand.

1 hour ago, feodor2 said:

I never did it myself you please elaborate, suppose webrender on direct9 cards is good feasible.

I'd need to see this to believe it. I'm pretty sure I tried testing Mypal68 here and I wasn't able to get WebRender to work, or get it to enable here.

[feodor2]

29 minutes ago, K4sum1 said:

I feel like Rust would be the easy part given the existence of Rust9x. Unless you mean

I think not, it has a very poor condvar stuff, even more than mizilla, it was about condvar from the beginning, I think best condvar for xp is chromes, I moved most on it on mypal cpp, except rust which is still on mozilla version.

condvar is major core stuff at all

38 minutes ago, K4sum1 said:

I asked before about cbindgen, but I still got stuck there and I didn't know what to do or ask. You say to compile it without cargo

Okey I shall make detailed instruction about this

[feodor2]

47 minutes ago, K4sum1 said:

It would really help if you were to set up a VM from scratch to compile Mypal68 and make a step by step guide

II is there https://github.com/Feodor2/Mypal68/wiki/Building-from-the-source  still valid all, certainly some pats needs update

 

[K4sum1]

1 minute ago, feodor2 said:

II is there https://github.com/Feodor2/Mypal68/wiki/Building-from-the-source  still valid all, certainly some pats needs update

I mean like I want a guide with exact commands that I can blindly follow that has been recently tested so the chance of me doing anything wrong is practically 0. That's what I made the r3dfox build guide to be.