PM x64 33.x crash with noscript

[andikay]

I think you might be reading a bit too much into it. People (not everyone of course, but possibly quite a few) come for the compatibility with old FF extensions and stay for a variety of reasons. I have said mine and I have also said that I am not against progress, I also evolve and grow. I have been using eMatrix for the past few days, I am still learning how to use it better and better and it's mostly fine (the font thing annoys me the most, that is a real oversight from the developer in my opinion). I am still on the fence about trying the forked uBO, especially in conjunction with eMatrix as I still don't see the point of having both.

[moonbat]

I suppose I have to try the uBO fork that got updated recently since eMatrix+ABL is not enough for me regarding security and blocking capabilities.

If you mean the one discussed a couple of weeks ago, it doesn't do anything besides adding a few extra filter lists that you could add yourself to the existing legacy version if you wanted to.

We've always known we were going to lose those users eventually... and yes, it does suck that we can't please them, and yes, losing them is a blow to the project

Not sure it will be a blow. These users aren't the kind to interact with the community except when they run into trouble, and then they show up here either with a huge sense of entitlement as though they're interacting with paid for premium customer support, or are such utter cretins that they can't provide basic troubleshooting information when there's a template in front of their face. I saw a rash of new account registrations back when firefox compatibility was removed in version 29, just showing up to complain and then vanishing once it was resolved - these are the invisible crowd who for all intents and purposes may not even exist let alone contribute financially or offer their time to help others. I doubt their leaving will be a big loss.

And then again - there's nothing as popular in the legacy extensions like NoScript, which could break the browser or behave unpredictably all of a sudden, and UXP isn't following the 'move fast and break things' anti-pattern that the whole software industry seems hellbent on these days. So the breakage of old extensions over time would be extremely gradual and not all at once.

I am still on the fence about trying the forked uBO, especially in conjunction with eMatrix as I still don't see the point of having both.

To reiterate, eMatrix is a host level content blocker and uBO is an adblocker focusing on content within individual URLs. Both have features that the other doesn't - eMatrix prevents previously set cookies from being read by the setter on other websites (Google/Facebook etc where you may be signed in but don't want to be tracked across other sites if you're not using their sign in features on those sites). It also can block XMLHTTPRequests specifically, uBlock can't.

uBO does cosmetic filtering on webpages with dynamic filters to get rid of ad placeholders and DOM popups, which eMatrix can't. Between the two of them you're totally covered when it comes to blocking ads and tracking.

[Mæstro]

…the only thing that is changing over time is that completely dead extensions that are not maintained at all might stop working unless they are forked…The moment we don't support every extension that ever worked on old Firefox, without modification, we will inevitably start losing people.

Because I suspect there has been some misunderstanding, I should say that I, personally, only use extensions targetting Pale Moon explicitly. There are no orange dots in my extension menu. NoScript 5·1·9 targets Pale Moon explicitly, so I have thought of it as unrelated to random extensions dredged from the CAA. Your point, that extensions need occasional maintenance, if this is rare because the browser is so stable, is sound. This is no change at all; it was true in the noughts.

Part of me thinks that, if web standards had solidified as I would have preferred, the notion of a new browser version each month would never have caught on, and the pace of browser change would have flowed like in the noughts, responding to security needs. I know Pale Moon as a browser is being goaded by capricious web standards; Pale Moon extensions are dragged in tow.

If we weren't striving for some kind of balance, extensions would break far more often.

This was indeed the result of the survey I had cited before, and we thank you for seeking this balance. I suspect that some NoScript users might resent the recent breakage because, given that nobody is maintaining NoScript still, the browser could be patched (with slight cost in speed) to restore v32’s behaviour. They neither care that it is bad practise nor that this could well signal further breakage yet to come. Perhaps, this apathy comes from our own technical inability to resolve this by patching NoScript. I like your idea of arranging a bounty to patch it for us whenever needed, and I have begun asking on this board about how to go about this.

The only hope is that we have enough users who weren't just here for unmaintained XUL extensions, and are happy to stay with us and appreciate the ecosystem that has been built around the browser, even though it isn't as large and complete as the one that previously existed around Firefox.

There are those of us here for the aesthetic and NPAPI support.
In all seriousness, I appreciate the garden you have tilled for us, and have adapted slightly over the years to fit better within it. As said, all my dots are blue, and I am trying to find somebody who could patch NoScript to meet the Pale Moon team’s standards (though I will need lots of advice in how to do so). We like the slower, saner pace that means that NoScript has only suffered occasional problems with the inbuilt film player despite a decade’s neglect.

We've always known we were going to lose those users eventually... and yes, it does suck that we can't please them, and yes, losing them is a blow to the project.

Where would they go if they leave us?

[andikay]

Not sure it will be a blow. These users aren't the kind to interact with the community except when they run into trouble, and then they show up here either with a huge sense of entitlement as though they're interacting with paid for premium customer support, or are such utter cretins that they can't provide basic troubleshooting information when there's a template in front of their face. I saw a rash of new account registrations back when firefox compatibility was removed in version 29, just showing up to complain and then vanishing once it was resolved - these are the invisible crowd who for all intents and purposes may not even exist let alone contribute financially or offer their time to help others. I doubt their leaving will be a big loss.

That is an interesting take to say the least. So users are only relevant or worthy when they contribute? You do realize that the vast majority of users of any kind are simply users and nothing else, right? And making developers aware of issues is contributing in itself because the developers surely cannot account for every possible setup and eventuality.

[Moonchild]

In their eyes, "old extension no work," is the same and they won't take any excuse for it. The moment it happens, we are the same as Mozilla in their eyes.

It's a form of PTSD coupled with entitlement, a dangerous combination.

(I really like your free whiskey analogy, by the way - quite spot-on!)

The main point of argument seems to be though that we "shouldn't ever be losing users"? That's not how software publication works. People will always find one reason or another (whether out of their resistance to change, ignorance, external factors, or simply having a change of heart, etc.) to stop using a particular piece of software, and that's fine. It's the consumer approach (as opposed to the community/collaborative/FOSS approach), I'm well aware of it. But also realise we will (have to!) treat those users as exactly that: consumers of a free product. They have no rights to demand anything from us because there is no transaction, no balance, no reciprocal value. In short, we have to decide our direction and how we get there. As far as leadership goes, I've always aimed for a balance of features, user wishes as a whole, and the need for change.
So we'll lose users. It's inevitable. We'll also gain new users when we work on improving web compatibility (i.e. focus on the core task of what the browser should be doing, not added tools/extras). Long term, not maintaining sufficient compatibility with the web would make us lose all users. Which is the more sane approach? Big picture stuff, I know.

it does suck that we can't please them, and yes, losing them is a blow to the project.

I think we'd lose that particular group of users anyway if we were to stop developing altogether and froze the platform in time, because they would simply not be able to browse the web any more in the end. The only difference would be that they would be a lot less vocal about it. So... A blow to the project? I really don't think so.

Don't let a few loud voices persuade you otherwise.

This was indeed the result of the survey I had cited before, and we thank you for seeking this balance. I suspect that some NoScript users might resent the recent breakage because, given that nobody is maintaining NoScript still, the browser could be patched (with slight cost in speed) to restore v32’s behaviour.

Really? So what was the commit that broke NoScript then? Point me to it and I'll let you know if it could be "patched with a slight cost in speed"

[athenian200]

Not sure it will be a blow. These users aren't the kind to interact with the community except when they run into trouble, and then they show up here either with a huge sense of entitlement as though they're interacting with paid for premium customer support, or are such utter cretins that they can't provide basic troubleshooting information when there's a template in front of their face. I saw a rash of new account registrations back when firefox compatibility was removed in version 29, just showing up to complain and then vanishing once it was resolved - these are the invisible crowd who for all intents and purposes may not even exist let alone contribute financially or offer their time to help others. I doubt their leaving will be a big loss.

I guess my anxiety here is that, like I said earlier, XUL keeps the lights on. Some of the people donating to the project could be under the mistaken impression that we are offering them full support for all unmaintained Firefox extensions forever. And I can even understand why they would be confused... projects like Wine and various emulators target increased compatibility with whatever it is they are trying to simulate over time. As Wine gets older, it supports more old Windows programs. So it's perfectly understandable to me, that our users would be frustrated and unable to understand why it's the reverse for us... we supported almost all the old Firefox extensions when our project started, and instead our compatibility is getting WORSE, not better, over time. No one would tolerate this from an emulator or a compatibility layer like Wine, so all it would take is for users to mentally slot Pale Moon into the same category as projects like that, and boom, you have reasonable confusion and unrealistic expectations.

As hard as I am arguing against andikay's perspective, I really do "get it." The problem is that getting why users feel that way and why they are confused doesn't mean I can wave a magic wand and give them what they want. I think that is also why I am going out of my way to keep replying to this thread... I really want to make sure people, if they understand nothing else, understand why Pale Moon can't be like WINE or an emulator and target more compatibility with Firefox extensions over time rather than less.

Like with Rust, there is no sympathy in that design for someone who wants to target an older version of the JS and DOM APIs. They just expect you to keep up forever with changes, hop to the latest version and forget previous ones existed, and won't let you target specific versions of the API so as to ensure backwards compatibility. If XUL had been designed with that kind of foresight in mind, we likely wouldn't have a lot of the problems we do. Don't get me wrong, XUL is great, it got a lot of things right, like using native widgets and being very powerful. But that sure doesn't mean I think it's perfect or don't get frustrated with it being... well, a Mozilla technology, and having a lot of annoying issues that I've come to associate with the Mozilla mentality in general.

[andyprough]

Some of the people donating to the project could be under the mistaken impression that we are offering them full support for all unmaintained Firefox extensions forever.

What will frustrate me as a frequent donor will be any of the devs spending any time worrying about what a handful of users of one extension want rather than just making this the best browser they can within the limits of their skill and creativity. At the end of the day our financial contributions are because of our faith in you and your vision, not because of a project that Giorgio Maone coded like "ass" (apparently) and then abandoned back when gasoline was $1.80 a gallon and no one had ever worn a paper mask on their face to shop at Walmart.

I don't speak for all the donors of course.

[Mæstro]

Thank you, Athenian, for understanding our perspective. I hope that I have been polite to you, Moonchild and others. (I know you are also autistic; I trust you know how one can never be sure without being told.)

Some of the people donating to the project could be under the mistaken impression that we are offering them full support for all unmaintained Firefox extensions forever.

I am under no such delusion myself, but I believe your following explanation is sound diagnosis for many. Certainly, I would have preferred a static, standard web that would have enabled Wine-like targetting, but as you have said below and I have suggested before, organising this is beyond your scope.

As hard as I am arguing against andikay's perspective, I really do "get it." The problem is that getting why users feel that way and why they are confused doesn't mean I can wave a magic wand and give them what they want. I really want to make sure people, if they understand nothing else, understand why Pale Moon can't be like WINE or an emulator and target more compatibility with Firefox extensions over time rather than less.

You do get it. I notice you have not been replying to rebut me, and I am not sure whether to take this as a good sign (that there is less disagreement than I feared) or not. You have helped me understand. Likewise, I have been replying so much to offer all the introspection I can, so the future mood surrounding NoScript is more civilised than it has been. I hope that i have been able to show that there is less conflict than it might seem: that you meet our needs quite well without compromising your mission. You have also inspired me that, with the right guidance, I can facilitate a patched NoScript which agrees with Pale Moon and can put this whole matter to rest, at least respecting this eristic extension.

Like with Rust, there is no sympathy in that [XUL’s] design for someone who wants to target an older version of the JS and DOM APIs. They just expect you to keep up forever with changes, hop to the latest version and forget previous ones existed, and won't let you target specific versions of the API so as to ensure backwards compatibility. If XUL had been designed with that kind of foresight in mind, we likely wouldn't have a lot of the problems we do.

Thank you for saying this. I would not have known this otherwise, and this indicates the ultimate root of any conflict as having been from outside forces which we have all been powerless to affect. Hearing this comforts me. I wonder how the Otter and Konqueror users fare.

Really? So what was the commit that broke NoScript then? Point me to it and I'll let you know if it could be "patched with a slight cost in speed"

I wish I knew. DBsoft was the one to diagnose the breakage last time. I have been speculating as a layman by analogy from the last time NoScript broke Pale Moon upon opening MP4 files. I am sorry that I can do no better than guess and gesture.

[moonbat]

So users are only relevant or worthy when they contribute?

We cannot roll back the browser to what it was 6 years ago (essentially what it would take to get this sainted extension to work) just because of a subset of entitled users who have nothing to do otherwise with the project. This is neither paid software nor subsidised by other lines of business; so there's even less moral ground for such behavior. The least one can do to get an extension to work is contact the original author (I've done that for several extensions I liked and some like Master Password + have been forked for Pale Moon by their original authors) or get to work fixing it themselves.

Just goes to show the general profile of NoScript users that in all these six years and counting not one of them has offered to fix or patch something that they themselves use; going by Maestro's appeal. And if everyone who uses NoScript lacks the technical expertise to do so, why not then look at the alternatives?

[suzyne]

I think it is a positive step to see that another topic with a different approach has been started to address the issue of this thread. viewtopic.php?f=46&t=30943

But here it feels like there has been a comprehensive overview of the situation both historically and technically, and an unmovable line has been clearly articulated.

Is there anything new to add that would constructively add to this topic? Is it the right time to lock this thread? 

[moonbat]

Is it the right time to lock this thread?

I'd say it is. The one thing that comes out here is - use eMatrix or get cracking on fixing NoScript, the latter being focused on in the other thread.

[Moonchild]

Alternatively, keep using an old version of Pale Moon forever until either the last of web compatibility and sec updates make it blow up in your face, you can finally pry your fingers off of the relic you're clinging to, or the heat death of the universe, whichever comes first.