Does Pale Moon have more rendering trouble than other browsers?

[andyprough]

Critics on tech media sites sometimes say Pale Moon has an especially hard time rendering modern web sites. But I just went to the Chrome and Firefox and Vivaldi help forums yesterday to see what people complained about, and not only do they get a lot of complaints about websites not rendering properly, they even get complaints about some of the same sites that are complained about on this forum (Godaddy and Facebook and banking websites were well represented).

I've also had difficulty with other "major" browsers rendering certain sites, some of which work much better with Pale Moon. So after my fact check investigation, I'm giving 4 Pinnochio's to the Tech Media website Critics. Does Pale Moon sometimes have trouble rendering some unusual sites (especially improperly coded sites)? Yes. Does it have more trouble with more sites than other "major" browsers? A brief review of the evidence does not support it. The overwhelming sense I got is that nearly everyone in 2022 has a "main" browser, and a secondary browser for all the sites their main browser can't handle.

[Sob__]

Yes, it does.

One problem is speed. No matter how much I like Pale Moon for its sane UI that's not breaking every few months "just because" (and so I still have it as primary browser), it doesn't hide the fact that the engine has trouble keeping up with "modern" overscripted web. Find few "modern" sites with heavy scripting and do side to side comparion between Pale Moon and some Chrome clone, the difference is very noticeable.

Another worsening problem is support (or lack of) for new features. Whether they are proper standards or not, the fact is that their use increases and too many websites don't bother with any backward-compatible alternative, so they either don't work in Pale Moon at all, or are severely broken. You can complain to them, sometimes you may even succeed, and they will make backward-compatible version (for now, but it doesn't mean that they won't break it again few months later). Trouble is, for every such site there are hundereds of others that won't care, so it's futile.

[andyprough]

One problem is speed. No matter how much I like Pale Moon for its sane UI that's not breaking every few months "just because" (and so I still have it as primary browser), it doesn't hide the fact that the engine has trouble keeping up with "modern" overscripted web.

Another worsening problem is support (or lack of) for new features. Whether they are proper standards or not, the fact is that their use increases and too many websites don't bother with any backward-compatible alternative, so they either don't work in Pale Moon at all, or are severely broken.

But you'll find the same complaints about speed and broken "modern" sites (often the same ones) on Chrome's help forum and Firefox's help forum. So some of those things are going to be browser specific, some are going to vary by the individual system setup, and some are going to be problematic for some users on all "major" browsers. You may be able to get around that with the use of a secondary browser, but as I said, it would appear that today nearly everyone needs a 2nd browser for all the stuff their first browser fails on.

An interesting thing I noticed is that Edge is the most common 2nd browser I saw mentioned. I've never tried the new Edge, although I've heard there's a version available for Linux distros. I doubt I would want to, since like with Chrome/Google I'd just end up sending all browsing and system data to Microsoft. But for whatever it's worth, Edge may be the one that is actually doing the most contortions to render the most sites.

[Sob__]

It depends. People may complain about speed of every browser, but major part of it is the hardware they run it on. If they got slow computer, then even fastest browser will be slow for them. Speed only makes sense as side to side comparison between different browsers on same hardware (and with same website of course). Also, some "modern" overscripted websites are so bad that they feel slow almost regardless of used hardware.

As for things breaking, it can happen everywhere, websites are basically programming like any other, bugs happen. And having hundered times more code than required for the job doesn't help either. Pale Moon has problem with different kind of breakage, when websites use code that it simply doesn't support.

[Moonchild]

Actual rendering is generally not the problem. but for an average user it may appear that way. in almost all cases it is because scripting on a website breaks, and "modern websites" are more often than not entirely script-generated (IMHO still a big fallacy of the modern web, as it should build on the triplet html-css-js for structure-styling-scripting respectively. Using js for structure and styling as well creates the obvious pitfall that a very non-forgiving scripting language like js (especially in the over-used "strict" mode!) will abort entire content building routines when it throws an error, and that results in the appearance that Pale Moon can't render something -- however, it could, if only the scripting to create the content had not aborted before the content was actually fed to the browser's renderer.

[andyprough]

Actual rendering is generally not the problem. but for an average user it may appear that way. in almost all cases it is because scripting on a website breaks, and "modern websites" are more often than not entirely script-generated (IMHO still a big fallacy of the modern web, as it should build on the triplet html-css-js for structure-styling-scripting respectively. Using js for structure and styling as well creates the obvious pitfall that a very non-forgiving scripting language like js (especially in the over-used "strict" mode!) will abort entire content building routines when it throws an error, and that results in the appearance that Pale Moon can't render something -- however, it could, if only the scripting to create the content had not aborted before the content was actually fed to the browser's renderer.

Thanks for correcting me on that, and for the bit of education, that context is intriguing given that in my brief review there's hardly any complaint about websites on this forum that won't be mirrored on a Chrome or Firefox help forum. So certain builds of those browsers running on certain systems under certain conditions are going to run into the same (or similar) scripting breakage. Whichever browser contorts itself enough to allow the most scripting to run without aborting is then "the winner" in the eyes of us un-knowing users I guess. But it would appear that no browser gets a free pass.

[somdcomputerguy]

IMHO still a big fallacy of the modern web, as it should build on the triplet html-css-js for structure-styling-scripting respectively.

My site is mostly HTML and PHP for structure and scripting, and CSS for styling. I have ONE line of JS to change the name of the current page on the title bar. I also use a JS based 'local service' to generate the non JS code that displays email addresses and phone #'s or any other text I want to be visible (and clickable if need be) on the page but not displayed in the source code.

[moonbat]

All thanks to Google making standards irrelevant. Time was there used to be a standardized HTML 4 test suite and a browser had to pass it to be declared compliant. Websites didn't have to care what browser you used if their HTML syntax was correct (Except for IE of course, with its quirks). Chrome is the new IE times ten, thanks to being the default on mobile phones, which dominate internet use today, and Google pushing their javascript frameworks that work only with it.

[Moonchild]

thanks to being the default on mobile phones, which dominate internet use today

I still don't understand this actually. Aside from casual browsing standard display content, it's not like you can actually do any work on the web on a phone, and the sites impacted most by the Googlification are complex, interactive sites one would use for work, not entertainment. Do you see businessmen doing their corporate work on their phones? I don't. If mobile they have a laptop and are on desktop applications and browsers if on the web, not phones.
So i do think that this metric of "internet use" is very skewed and measures casual requests for information from storefronts and wikipedia as a heavy factor, and maybe even includes all the automatic background data/metrics requests that flow to Google... but the -type- of use matters too, IMHO, in how websites should approach their code.

[moonbat]

I think they just don't care anymore - if it's corporate work, most of it is cloud based now and a company can simply mandate Chrome for its employees. For regular folk, the internet has become what TV used be before it - just passive consumption with slightly more choice. Notice how the majority of sites that people complain don't work with Pale Moon also fit this category - Facebook, Youtube, Instagram, Reddit. I don't use Facebook or Insta here, Youtube works fine for me with the userscript I keep recommending and with adblocking to the max, and Reddit still has its old desktop site available. So if you're just going to look at videos or scroll photos, a dumbed down mobile-ified site experience won't matter.

One trend I've seen in SaaS is the proliferation of dashboards for everything, from network load to internet traffic to sales figures, and they use Chrome to show some bling to impress the corner office folks.

[Sob__]

It's not difficult to understand, it's the combination of things, people are lazy, time is money, etc. If they make website only for few latest versions of major browsers, it's less work than making one with broad compatibility. And since almost everyone either uses major browsers, or at least can use them as backup, ...

[damjang]

All thanks to Google making standards irrelevant.

For me this is actually the first problem. It backfires to all users and also the same browsers (read crome etc.) ...

[vannilla]

If they make website only for few latest versions of major browsers, it's less work than making one with broad compatibility.

Surprisingly this reasoning is actually worse: Pale Moon is closer to the actual standard (even if the standard is Google) than the rest of the browsers, so targeting Pale Moon means doing less, as making a site (or "webapp") work there means it works everywhere else.
Less features, especially less experimental features, is a good thing* as you automatically target the common set of features that are sure to work and is already "battle-tested" on the actual web.
It's not even the Internet Explorer situation: there you had to explicitly write markup specifically for that browser inside some <!--if--> garbage and doing it for almost everything adds up fast; with Pale Moon it's the same HTML as other browsers, with the added benefit of having scripting code that is guaranteed to work everywhere.
The only real problem would be actual quirks in this set of common code, like the infamous "interactive content inside <button>", but if you target what Pale Moon does in these situations, these quirks either disappear or are so few that taking care of them doesn't take that much time, especially when behind a framework.
Basically, relying on the "caniuse" method of checking browser compatibility, as is done today, is more time-consuming and work-intensive than straight up targeting Pale Moon since the beginning.

* I'm not advocating that Pale Moon should not strive for compatibility with modern websites, just that not going crazy on adding new features is good.

[Sob__]

Let's say I don't disagree with you, but can't completely agree either. I used to do web development for living as my main job (I stopped around ten years ago), so I do have some idea what it involves.

The part I agree with is that using cutting edge stuff may not be the best choice, because if too many browsers don't support it yet, backward-compatible version still has to be made, which may be quite different, so it's extra work. And if what the new stuff brings is good, but not exactly something it wouldn't be possible to live without, it's not worth it. It's better to wait a while and adopt it gradually, as browsers catch up and old versions die out.

But it's not possible to wait forever. Well, now I probably could, because when I do some hobby stuff now and then, I'm perfectly happy with what Pale Moon can do. It's paradise compared to what I had to work with in the past, I don't need anything more. I'm lagging behing latest trends, but I'm sure that this new stuff like Web Components does have some advantages, major browsers support it for few years already, other browsers that are not clones of major ones are nearly extinct, ... you see where I'm going, too many people will feel that it's not needed to wait anymore.

[Moonchild]

It's not a matter of waiting.

The observation that "what works in Pale Moon will also work in Chrome/Firefox" holds true in practically all cases. And in the extremely rare cases it doesn't, it will be features that are useful but that the big players (usually for some marketing reason) have phased out in favour of something almost-but-not-quite-the-same that was touted as a new innovation. Scoped styling comes to mind.

Unless you actually want to exclusively target the quirks of Chrome and increasingly obscene moving targets of ever-more-complex specifications that are shoved to web devs with "Please use this! Don't you want it? It's new and shiny!" and the eager new bloods dutifully implementing it because it means job security (or so they think), then there is little reason to actually use much beyond ES6. And even that can be argued as ES5 was already pretty complete.
The arsenal at your disposal for web programming with that already monstrous specification (which we fully implement, by the way, and then some) is absolutely formidable, and can even be used to implement the quirky extensions if you really happen to need them. But what we're running into is that web developers and especially framework developers will only target the exact and specified feature set of one engine, even if more general implementations are available and in fact have been in use for years before.
If your approach is to "forever wait until Pale Moon becomes exactly like Chrome and Goanna becomes exactly like Gecko/SpiderMonkey" then there simply is no future at all in which Pale Moon will ever fit your bill.

[Sob__]

As I wrote, I'm not following latest trends closely, so I'm not able to argue why new feature X is so wonderful and so much more efficient than what was there before, in a way that it's almost impossible to live without it.

But even if it's just little things, for example nullish coalescing or optional chaining. It's entirely possible to live without them, after all, everyone did in the past. But also, they do have some appeal, otherwise they would hardly get in. I started to use the former in PHP and it's kind of nice. And seeing that JS now also has it, it would be nice to be able to use it there too.

But it's not about me, I can live without it, no problem. But that's not true for the rest of the world. They'll think that it's officially in specification, all major browsers support it for two years already, so why not use it? And that's exactly what they do. Because if not now, then when? Next year, the one after that, never?

[Moonchild]

In a fair world your argument would be valid, but many people use new stuff not because it's wonderful, but because it's new, and only because it's new.
The issue really is that a lot of it is syntactic sugar that makes things available to save on keystrokes for developers but often are very difficult to implement in an engine that already has established ways of doing things, especially if the spec introduces ambiguity. And in many cases when frameworks are concerned it even boils down to what the framework devs decide is the "default configuration for code generation" -- hardly anyone actually programs in JS directly anymore and stuff tends to all be generated code; so in what way does that syntactic sugar matter then? Save a few bytes on the minified blob that is spit out?...

If you want us to support ?? or .? then we can use your help making it happen; maybe not personally if you're not a programmer yourself, but see if you can find someone else who can help in that case. I certainly understand the appeal of those features for programmers.
I've tried (several times already) to get it implemented and thus far was not able to make the parser be happy with my changes, so maybe it really does need someone with more Mozilla knowledge of that part of the code base. Parser changes are actually rare and I personally simply don't have much experience with it. Of course something like that would be bounty-eligible as it's one of the main hangups for web compatibility at the moment.