A possible solution for the dilemma?

Anchovis's Avatar

Anchovis

24 Feb, 2019 07:18 PM

Dear Cameron Kaiser, dear community,

I am an avid reader of the blog accompanying browser development, finding its content very informative and amazing to read. As a reader of the blog, this particular bit happens to make me sad:

" Like I say, I'm still concerned over the deficiencies accumulating in the browser that I don't know what to do with and don't have an easy means to patch into the browser core. That said, keep in mind that even if we did try to get a port of 52 off the ground to address these problems -- the functionality of which wouldn't guaranteed and has several major changes which would badly compromise TenFourFox's platform base -- we'd just have different deficiencies once Fx60 becomes the typical minimum, so it only delays the inevitable, and the Rust requirement for 54+ makes any later wholesale port impossible. Nevertheless, in the meantime these new features, although admittedly incomplete, at least give some additional functionality to the browser, and that's not worth nothing."

As far as I can tell, you are (somewhat) planning to use Firefox 52 as the new base. In my opinion, this is not a sound idea. The Firefox 52 base has been abandoned by Mozilla a long time ago and obviously lacks further development on Mozilla's part. Also, as pointed out in the entry above, Mozilla seems committed to go down the Rust route.

The team behind Pale Moon has taken a Gecko 52/53 base and has furthered development quite a bit, continually improving the code, thereby also improving web standard support, as evidenced here:

https://www.basilisk-browser.org/releasenotes.shtml

Furthermore, Basilisk uses neither Rust nor Electrolysis / e10s:

https://www.basilisk-browser.org/features.shtml

- "Does not use Rust or the Photon user interface. You can expect a familiar interface as-carried by Firefox between v29 and v56."
- "Does not use Electrolysis (e10s, multi-process browsing)."

This would get the Rust issue as well as e10s out of the way. Furthermore, Basilisk does not have Pale Moon's deficiencies, i.e. it supports both WebRTC and DRM.

I think it would be a good idea to use Basilisk as base, for the following reasons:

- The developers are not and will not be using Rust.
- Security updates are being implemented, so Mr. Kaiser doesn't have to do this anymore.
- Web standard support is for the most part taken care of by the Pale Moon team.
- Synergy effects between TenFourFox and Basilisk development, I think some patches can be useful for UXP.
- A great deal of development on the Gecko 52 / 53 base has already been done, contrary to the abandoned Mozilla development tree.

Please consider this proposal. The interface is the same as TenFourFox interface, so in terms of it being familiar to TenFourFox users, there should be no problem.

Regards

Anchovis

  1. Support Staff 1 Posted by Cameron Kaiser on 24 Feb, 2019 10:58 PM

    Cameron Kaiser's Avatar

    No, I think you misunderstand: the idea actually isn't to use Firefox 52 (or, for that matter, UXP/Basilisk). 52 made several platform changes that would make porting to 10.4 much harder and would cause us to lose much of the optimization work we have in 45. These aren't issues for Pale Moon, but they are for TenFourFox. The response was to make clear why I don't think such a port is a viable option anymore for maintaining TenFourFox, not to say it was being considered.

  2. 2 Posted by Anchovis on 25 Feb, 2019 02:28 AM

    Anchovis's Avatar

    I understand, but the question is still: With Mozilla implementing Rust code more and more, also getting rid of XUL in favor of HTML, how can you possibly see them as a good upstream source in the mid-term and long-term? I don‘t get it. Aren‘t they bound to become more and more incompatible with the FF45 codebase you are using?

    So, even if you don‘t create a full port of FF52 or Basilisk, wouldn‘t it make sense to at least use UXP as upstream in the future? After all, there is a team behind the browser developing it further, and they seem to be committed not to go down the Rust route...

    Other than ending TenFourFox, I am not seeing many alternatives aside from an UXP upstream source, either as a complete port or at least as a source for backporting code to FF45.

    Please elaborate, I‘d genuinely be interested in how you plan to use FF code in the future, with the divergance getting bigger and bigger. How would that be possible?

  3. 3 Posted by Anchovis on 02 Mar, 2019 10:08 PM

    Anchovis's Avatar

    Mr. Kaiser, is there a specific reason for you no longer replying? I just want to know (out of personal curiosity) how you plan to backport code from an increasingly divergent Firefox code base, one that comes to more and more rely on Rust, no less. I don't think this is viable in the long-term.

    Also, is there a specific reason why UXP isn't considered a future source of patches and fixes? Any kind of beef between you and the Pale Moon team? Is it technically infeasible, and if so, how is it infeasible to backport from closely related UXP as opposed to increasingly divergent Firefox?

    Please explain, I am looking forward to a possible reply.

Reply to this discussion

Internal reply

Formatting help / Preview (switch to plain text) No formatting (switch to Markdown)

Attaching KB article:

»

Attached Files

You can attach files up to 10MB

If you don't have an account yet, we need to confirm you're human and not a machine trying to post spam.

Keyboard shortcuts

Generic

? Show this help
ESC Blurs the current field

Comment Form

r Focus the comment reply box
^ + ↩ Submit the comment

You can use Command ⌘ instead of Control ^ on Mac