Package Details: basilisk 2020.09.11-1

Git Clone URL: https://aur.archlinux.org/basilisk.git (read-only, click to copy)
Package Base: basilisk
Description: Standalone web browser forked from mozilla.org
Upstream URL: https://www.basilisk-browser.org/
Licenses: GPL, MPL, LGPL
Submitter: bm456
Maintainer: figue (jfigueras, figuepluto)
Last Packager: figue
Votes: 0
Popularity: 0.000000
First Submitted: 2017-12-25 20:34
Last Updated: 2020-09-16 20:41

Dependencies (12)

Required by (0)

Sources (3)

Latest Comments

« First ‹ Previous 1 2 3 4 Next › Last »

KlipperKyle commented on 2018-05-24 01:58

Well, it seems I was quoted out of context on IRC too, and apparently whatever I said went in one ear and out the other. So, here I am, signed into AUR for the first time in more than a year.

I tried to be nice, but frankly my patience on this issue is starting to run thin.

@eschwartz, there were at least 3-4 people who attempted to explain the basics of trademark law and how you may not use Moonchild's brand unless you are explicitly given permission. It doesn't matter what Mozilla told Debian or what Spotify told (or may not have told) Arch Linux. We explicitly told you that calling this package Basilisk is not OK.

https://lists.archlinux.org/pipermail/aur-general/2018-May/034102.html

What these jokers don't seem to get is that there is NO packages involved here. There is nothing here that violates the license as there is no redistribution at all. Moot point, move on and whine somewhere else.

@scimmia, that's incorrect. AUR is redistributing the package via build script, and the default configuration is insane, as pointed out repeatedly. We have a problem with this non-condoned configuration being called Basilisk, and we asked (no told) the package maintainer to stop.

@eschwartz, we gave you an option for peaceful resolution: renaming the package. I even personally contributed a suggestion.

So, why hasn't the package been renamed yet?

vonPalitroque commented on 2018-05-23 19:24

Greetings,

So it looks like things have been rather festive since I last looked at this PKGBUILD. First and foremost, thanks to frankspace and bm456 for their work on the PKGBUILD. There are a few things I do not agree with, such as hardcoding potentially lethal CFLAGS in the PKGBUILD for optimizations that may not be available in a given architecture (though I am pretty sure that all CPUs officially supported by Arch Linux have these features, but I do not like the idea of my CFLAGS being willy nilly ignored). If anything, leave it as "-march=native -mtune=native" which generates code optimized for the CPU it is being built on (dangerous if using distcc), or pull from the defaults in makepkg.conf and modify from there as most other packages do. Same with the LDFLAGS behavior.

IANAL: Now, regarding the issue with MoonChild Productions and their team. As it is, this PKGBUILD generates a browser that identifies itself as "Serpent", not "Basilisk". A while back, there used to be a package called "swiftfox" in the repositories, which existed as an alternative to "firefox". Furthermore, at some point, Firefox itself identified as a different browser. Pointing to a different distribution, Debian started shipping Iceweasel as a rebranded Firefox in response to what they perceived as non-free trademarks from Mozilla.

Given that this PKGBUILD does not generate a browser called Basilisk, the only instance of the name that remains is in the package name [and resulting binary], I propose that we just provide our own branding and rename the PKGBUILD to something else. The default unbranded "serpent" could work I guess. We also specify in the package description that this is a "Web browser based on the Basilisk source code with some modifications and optimizations for Arch Linux". As such, those looking for "Basilisk" will be able to find an alternative that is built from source, or the binary package that is already being distributed in the AUR. At this point, we are no longer claiming to be Basilisk. We are explicitly stating that this would be a derived product that does not use their branding, which, as far as I am aware, is valid under the MPL. This is what in effect Debian did with Iceweasel (maybe rename the binary that is outputted just to be sure).

Although somewhat understandable since by providing a potentially unstable product under a brand can tarnish the brand itself, I strongly believe the way this was approached from MoonChild Productions and their team was improper. Instead of demanding a very vocal, very public takedown, they should have privately contacted the maintainer voicing their concerns and offering assistance on how to resolve the issue. Their attitude and community response has left a rather sour taste in my mouth.

Cheers, Orlando.

[edit] Formatting.

Ashie_Princess commented on 2018-05-22 11:38

@eschwartz I would argue that theoretically, the most effective solution would be to go the way of the OpenBSD people and just outright remove the software from the AUR, not in compliance with the request, but from the fact that Moon "Acts like a child" Productions clearly don't understand what the AUR, or a trademark are.

From the INTA (https://www.inta.org/TrademarkBasics/FactSheets/Pages/Fair-Use-of-TrademarksNL.aspx) Nominative fair use permits use of another’s trademark to refer to the trademark owner’s actual goods and services associated with the mark. Nominative fair use generally is permissible as long as (1) the product or service in question is not readily identifiable without use of the trademark, (2) only so much of the mark as is reasonably necessary to identify the product or service is used and (3) use of the mark does not suggest sponsorship or endorsement by the trademark owner

To demonstrate this: (1) By naming this something else, such as "webbrowser" or "internet-page-loader" while describing what this product does, does not accurately describe which web browser it is. clearly, if you want to get the basilisk browser, then you would not be able to identify it without the use of the Pale Moon trademarked name "Basilisk". (2) Literally, the only use of the name "Basilisk" is in the PKGBUILD where absolutely necessary, and only to identify it. (3) At no point does this suggest endorsement nor does it suggest that you're agreeing with this. however, as many AUR repos have done in the past, if you wish to help out, by all means, contact the maintainer and do so.

Nowhere does that violate your trademark

eschwartz commented on 2018-05-22 11:31

Fixed some markdown formatting, cf. https://lists.archlinux.org/pipermail/aur-general/2018-May/034100.html

mattatobin commented on 2018-05-22 11:17

Since what Eschwartz pasted is completely unreadable here is the complete, readable, and unbiased log from the conversation taken by globbot.

https://freenode.logbot.info/palemoon/20180522#c1555880

Starts at 01:50 and ends 03:10 globbot time. Enjoy.

eschwartz commented on 2018-05-22 03:17

1) Pale Moon and everything that comes from there is gross.

2) There's no rule against gross things being in the AUR. If someone wants to maintain it, and users want to use it, that's their choice.

3) On the topic of license violations, I joined #palemoon on Freenode to try to clarify things... @mattatobin responded to me there.

[10:11:52 PM] <eschwartz> so if I understand correctly, it is utterly forbidden to use the basilisk name in any way, shape or form, without exception, under any circumstances whatsoever, bar none? Then what is the purpose of having a trademark...
[10:12:49 PM] <NewTobinParadigm> eschwartz: Your rights are clearly stated in the Mozilla Public License
[10:13:03 PM] <eschwartz> so just to be clear, this is your objection? ^^ you don't allow any sort of use whatsoever, ever, no matter what, ever?
[10:13:37 PM] <NewTobinParadigm> Not for Basilisk.. If you call it something else and don't use the word "Basilisk" but otherwise comply with the MPL.. well that is your business

I'm completely unsure what to do with this information.

I'm inclined to say that since we don't host source code, binaries, branding files, or indeed anything other than download URLs and compilation command lines, we're covered under the same rules that let us host PKGBUILDs for completely proprietary software, rather than basilisk where only the branding is proprietary. But it seems they disagree:

[10:08:23 PM] <eschwartz> But I still don't see what it is doing wrong
[10:08:29 PM] <NewTobinParadigm> because of the Pale Moon redist license
[10:08:34 PM] <NewTobinParadigm> eschwartz: You don't have to
[10:08:55 PM] <NewTobinParadigm> eschwartz: You merely have to comply with the Mozilla Public License
[10:09:12 PM] <NewTobinParadigm> which does not grant any rights to the name "Basilisk"
[10:09:27 PM] <NewTobinParadigm> You do NOT have the right to call it Basilisk in any form
[10:09:29 PM] <NewTobinParadigm> simple as that

[10:24:08 PM] <NewTobinParadigm> You are misrepresenting us is what you are doing..

[11:04:09 PM] <eschwartz> MoonchildPM|Away: I asked an hour ago, what is wrong with the build configuration and what you would prefer. I got flatly stonewalled, and told that NewTobinParadigm as a representative of the Pale Moon team was flatly asserting the trademark rights to unconditionally forbid its use, with the presumed logical conclusion that there is no build configuration other than -when-hell-freezes-over which would be acceptable.
[11:04:37 PM] <eschwartz> then I got the output of `yes MPL`
[11:04:56 PM] <NewTobinParadigm> Colorful
[11:05:32 PM] <NewTobinParadigm> essentually true but not so dramatic
[11:05:34 PM] <eschwartz> are you saying there is something this package could be doing, which would meet your approval? this is news to me

Despite my repeated requests to know what, exactly, we are misrepresenting via this build, no one was inclined to say anything other than "you're in violation of the license, please delete everything".

So as far as I can tell, this really and truly is entirely down to their complete unwillingness to see their name used at all, and not for any reason specific to this package.

In which case I'm unsure why this is worse than distributing Spotify via an AUR package too. We'd need to establish a rule that proprietary software is completely forbidden from the AUR.

... which I pointed out, and was then told:

[10:43:49 PM] <KlipKyle> eschwartz: you are distributing build scripts, like Gentoo ebuilds except less automation.  The same rules apply.

But Gentoo explicitly contains USE flags for proprietary-non-redistributable software, on the grounds that the user can choose whether they want to include non-redistributable code as a general thing on their built system (perhaps they want to redistribute the system as an ISO image or something).

And anyway, we've got no rule against this AFAICT.

...

All this being said:

[11:07:23 PM] <MoonchildPM|Away> eschwartz: if you plan to go that route, that's fine and someone can have a look over your build configuration (which I could do as well if it is was not 5 in the morning) and can tell you what's wrong with it. In the interim, until permission is granted, you are NOT allowed to keep these packages up since you're in violation. You ask permission first, get it granted first, THEN are allowed to use it if OK, not in any other order

So if @bm456 would like to work with the Pale Moon team to establish a mutually satisfactory outcome, that would certainly be the most... effective solution.

FredBezies commented on 2018-05-20 07:33

Reminds me something about the behaviour of Moonchild productions... https://github.com/jasperla/openbsd-wip/issues/86

This commit sums it up: https://github.com/jasperla/openbsd-wip/commit/59812ef678fa40b306cee4ef3f3a7278c12dd7ae

"Remove. clearly, they don't want users, and we're not gonna give them…"

Well, what do you think now? Who needs this browser, after all? Just asking, of course !

jonathon commented on 2018-05-20 00:14

@mattatobin If you feel this package should be taken down file a deletion request with all the necessary details, e.g. where/how your license and trademarks have been infringed.

On the other hand, if you wanted to work with the community instead of alienating them, post a "compliant" example package file so the changes could be considered and adopted.

Ashie_Princess commented on 2018-05-19 18:05

@mattatobin could you also clarify what is meant by the term "squatting"?

phw commented on 2018-05-19 17:50

@mattatobin: You should clarify what parts of https://aur.archlinux.org/cgit/aur.git/tree/?h=basilisk you claim violates any branding rules.