Web engines are so insanely complex that you can't just create a new viable competitor without millions on fundings. They're practically as complex as operating systems themselves.
this post was submitted on 16 Nov 2023
72 points (100.0% liked)
Free and Open Source Software
19581 readers
91 users here now
If it's free and open source and it's also software, it can be discussed here. Subcommunity of Technology.
This community's icon was made by Aaron Schneider, under the CC-BY-NC-SA 4.0 license.
founded 2 years ago
MODERATORS
In the past, all complexity was delegated to NPAPI plugins that worked with all browsers.
Now it's all Web APIs that every browser engine has to implement.
Yeah, Opera tried but eventually just gave up and now uses Blink, Microsoft tried with Edge (Spartan) but gave up and now also uses Blink.
Blink is the render engine made for Chrome
load more comments
(3 replies)
it often gets caught pulling anti-user shenanigans
I'm not aware. Can you list a few?
Receiving funding doesn't necessarily mean serving. Google is interested in funding to keep it's position. Mozilla still provides alternatives and regularly criticizes Google.
I’ve not been tracking them because I tend to only collect dirt on the greatest of evils. What comes to mind:
- default search engine: Google (this is what that Google money is for officially)
- Mozilla gave the boot to a lot of plugins and imposed some kind of control-freakish trust mechanism. Plugins/extensions were evicted from the plugin repository and they made it hard for plugin creators to distribute their plugins. I lost several very useful plugins when Mozilla took this controlling protectionist stance.
- MAFF ditched. Mozilla abandoned a good format for archiving websites. I had a lot of content saved in *.maff files which Mozilla dropped direct support for and at the same time they blocked MAFF plugins.
- Without Firefox, Google would be easily targeted with anti-trust actions. Google props up Mozilla just enough to be able to claim they have “competition”. Google can be most dominant when it has a crippled competitor under its influence.
- Google killed the free world JPEG XL format. When a browser as dominant as Chrome withholds support JPEG XL, there is then no reason for web devs to use that format. Google did this because JPEG XL competes with a proprietary Google format. Firefox does not support it out of the box either, likely because of Google’s influence. Firefox users can enable it by going through some config hoops, so if Chrome alone did not kill it, that certainly would.
I vaguely recall a slew of Mozilla actions that were anti-thetical to privacy and user interests which caused me to move them from “a decent browser” to a “lesser of evils”. Hopefully others have better records of Mozilla’s history.
update May 2024
- Mozilla uses data abuser Cloudflare for their exclusive access-restricted blog
- Mozilla has decided to add more tracking to their browser to collect people’s search activity.
Most of this is self referencing. Like the default search engine is not an example of Google's control, it's Mozilla's revenue model.
The remainder sounds like personal gripes that you're misconstruing as evidence of nefarious intent.
There's also plenty of evidence to the contrary, total cookie protection to name but one.
Additionally, beurocratic processes produce terrible software. Log in to any govt website as a refresher.
Finally, browsers are incredibly complex, if this model worked you'd use it for much simpler projects first.
Like the default search engine is not an example of Google’s control, it’s Mozilla’s revenue model.
It’s both, of course. Mozilla’s revenue enables Google control. If Mozilla changes the default search to one that is not in Google’s interest, they will lose their revenue.
The remainder sounds like personal gripes that you’re misconstruing as evidence of nefarious intent.
It’s both. I’m a user so I notice when Mozilla makes an anti-user move. Businesses serve their customers. Mozilla’s customer is Google, not me. So Mozilla serves Google, not the users. W.r.t evidence, I gave no evidence. I did not say “this is evidence”. If you want to challenge a claim because you can’t find the evidence on your own, you can ask for the evidence.
And as I said, I did not keep track of all Mozilla’s anti-user shenanigans over the years. So you’re not looking at a complete list of issues. It’s disingenuous to treat it as if it were.
There’s also plenty of evidence to the contrary, total cookie protection to name but one.
I did not mention anything about cookies, so which of my points do you think cookie protection counters what I’ve said?
Additionally, beurocratic processes produce terrible software.
Nonsense.
First of all, capitalism produces terrible software when you’re the product rather than the customer. It’s often shit even when you are a paying customer. The best quality software is produced outside of capitalistic structures.
I’ve worked on both gov and commercial environments. The gov process was superior for quality. On a commercial gig I was actually told not to fix bugs as they were spotted because it was important for the customer to discover the bug & report it so the supplier could charge them extra for the bug fix. The whole commercial work environment was rife with chasing profit (of course) which means cutting corners to cut expenses. If a developer produces something high quality in a fortune 500 company, they get back-roomed for “gold plating” (which means they’ve invested more in quality than necessary for the consumers). That doesn’t happen on gov projects.
It’s also wrong to attribute bureaucratic processes strictly to government projects. You may have a shit-ton of bureaucracy in the governance outside of the project which leads to: “build a Mars rover”. How bureaucratic the processes are within the organization is independent of whether it’s a commercial project or not. Fortune 500 corps are inefficient due to their bureaucratic structures. I could not reuse code from one project to another within the same company because there were rules about one project benefiting from another internal pot of money. So a piece of code had to be rewritten from scratch on the other project which means more bugs than you would have if the audited code could have been reused.
Finally, browsers are incredibly complex
Precisely why lack of competition is problematic.
LOL. Sure mate. Keep smelling your own facts and I'll eat a bag of dicks when ... checks notes ... the Italian government produces a FOSS browser to compete with Chrome & FF 🤣
For example as I recall some people are complaining about Pocket because it's not privacy friendly in some way. Idk about specifics of this, my only complaint about Firefox is that the CEO absorbs huge amounts of money to herself despite the shrinling userbase of Firefox YoY.
The reason why firefox and chrome work so well, is that they literally have been in development for over a decade. In Firefox's case, it's actually over two decades now.
Instead of trying to reinvent the wheel, why not support some currently existing alternative browsers that look promising? You have servo, you have webkit, and you even have a ladybird now. That's three potential browsers.
All three are under somewhat active development. Servo, in my opinion, looking the most promising, that shares a lot of dependencies with Firefox still, which means maintenance cost is not super high. It's easy to hack on, and of course it's rust. ~~who doesnt love rust~~
Sounds like a great idea, so long as Servo has not sold out to Google in any way. If Servo is really an independent browser govs would do right by the public to make that browser officially supported by all web services by the gov and do the necessary to ensure the Servo project is funded.
load more comments
(3 replies)
Finally: Pasta Browser
Logo: 🤌
Finally, a browser that can play the bass
For a time, konqueror was a thing. Khtml was even used as the basis for safari and WebKit, and later Chrome. Could always reinvigorate that project.
I say we either resurrect KHTML, or try and turn NetSurf, Servo, or Ladybird into something that you can realistically use as your main browser.
That's where Google succeeded. They bloated up the web standards so much that developing any of the alternatives to the required level is extremely hard. I doubt that even Google can create an alternative to chrome from scratch.
At this point, the only way for any of these to succeed is for the vast majority of people to actively avoid chrome.
In principle, if a government is going to distribute content to the public, they also have a duty to equip the public to be able to consume the content. Telling people to come up with their own private sector tools to reach the public sector is a bit off.
This statement is a rearrangement of events. The governments of the world didn't create an online presence and then tell the private sector to create browsers. Governments joined in an already existing method of communication because it was convenient, popular, and browsers already existed to view the content.
My comment does not imply when the first browsers were developed. Nor is it relevant. The problematic status quo sequence:
- offer web-based gov services
- leave people to their own devices.. to fend for themselves and pawn themselves to the private sector as needed to reach public resources
.
The sequence should have been:
- ensure sovereignty-respecting public tools exist
- offer web-based gov services that officially support the tools distributed in step 1
.
The internet began as a military project (government). The graphical web later emerged in the 1990s. So all governments have had 25+ years to become sovereign and ensure that the gov itself is not subjecting people to a US surveillance capitalist.
It was only in the past ~2—3 years that my local government closed its doors and decided to force everyone to do public administration tasks online. Indeed things are happening in a reckless sequence of events. Sovereignty from US tech giants should be sorted out before a government forces people to interact with their web-based services. So w.r.t my local gov, the status quo (first sequence) now has a third step:
- force people to use the web-based gov services without equipping them.
.
Do you see the problem? Step 3 is the most abusive, and that’s quite recent.
That sounds like your government has an issue. That isn't the same as governments as a whole using the web.
In the US, we still have the option to do things in person. The online presence is a convenience. That's how it should work everywhere.
Even in the US people are forced to use the web for public service even if it’s not officially announced.
E.g., suppose you want to see the state secretary’s records for a corporation. A lot of SoS websites try to force you to solve a #CAPTCHA. Fuck CAPTCHAs. I don’t do CAPTCHAs. So there’s an offline option, right? Ha. Try it. Send a snail mail letter to a state secretary requesting the registration records for an arbitrary business you know they should have records on. They just ignore it now. They don’t even have the courtesy to respond to say why they will not treat your request. Offline services have been quietly taken away without people even noticing.
I can walk in to the library of Congress and make a face to face request.
The web is a convenience for any public need in the US.
SoS records are state records, not federal. Are you saying every state shares their corporate database with LoC?
I would not be as fast as you to call the web a mere “convenience” to 99.9% of the country who are not a walking distance from Washington DC. If the analog way of doing something requires thousands of miles of travel, the online way is not a mere convenience. It’s a requirement, in effect.
BTW, it’s worth noting that the LoC has an access restricted Cloudflare website. So their exclusivity makes an offline option essential. If that means face to face in DC, that’s fucked up indeed. You should be able to use the postal service.
I think you're struggling with the difference of convenience and difficulty. Doing things without the web implies you are going to do them in the same way you'd have to pre-web. That makes the web more convenient.
Pre-web, postal correspondence was treated. Now it’s not. Convenience and difficulty are inversely proportional measures of the same thing. When you take away one out of two options, the other option is not a convenience. It’s a requirement.
The idea that you think people nationwide traveling to DC to get a business record is mere inconvenience is absurd. Are you drunk? You’re making a lot of bizarre assumptions, starting with assuming the travel is even possible for everyone nationwide who needs the service. If someone needs to sue a company for $200 and travel costs to DC to get the registered agent of the company is $400, you’ve effectively killed their access public service by nixing correspondence.
Your perverse understanding of convenience is ultimately just a language game that changes the language but not the problem. So let’s say traveling from California to DC to get an address is a mere “inconvenience” and using the web is “convenient”. That so-called “convenience” is essential in countless scenarios. And because what you refer to as “inconvenient” is actually not plausible in a scenario, the need for convenience in your language becomes essential.
load more comments
(9 replies)
It is very unlikely that anyone can develop a new browser from scratch. Just too hard both initially and maintenance in terms of the rate of new web specs. This is why most everyone except Firefox and Safari is a clone of chromium.
It's called WebKit and it's used by Safari.
That's your third option.
I just had a look at Debian’s official repos. No Safari browser. Did a search… found “how to install Safari on linux… start by installing WINE…” (yikes)
So in terms of a government offering public services that need to serve all people, Safari is not an answer unless the gov finances porting it to linux.
Gnome Web. Literally one Google search to find a browser using WebKit on Linux.
But sure, pretend Italy of all people will build a new browser just for you.
This misses the point. Governments are designing web services for Chrome. So you have two choices:
- pawn yourself to Google and use Chrome; or
- experiment with unsupported browsers, which even if they work you’re still limited to the window of standards Google decided was good for their business
It’s a lousy idea. The gov should be supplying services that are wholly free of Google’s influence.
Why does a government need to make their own browser to combat this?
Just make sites to existing web standards that work in Chrome, Firefox and Safari. Even the UK government is capable of this.
First of all choosing the subset of standards that Google chooses is a sovereignty problem. Gov services should not be constrained to what Google in the US decides to implement. Of the 3 browsers you mention, Chrome is subject to google snooping. Firefox is limited in Google’s influence as well. And Safari only serves Microsoft and Apple users officially.
The gov need not produce a browser from scratch, but they need to officially support a non-controversial browser that is not tied to US tech giants.
Opera has been great since the revamp.
But that's chromium anyway
Opera is chromium based
Isn't Opera Chinese-owned though?
Developed in Norway funded by a Chinese multinational.
Long-time Opera user here (I paid for Opera 5 - yes, that long ago). Opera was a truly visionary browser up to version 7 or so based on its own Presto engine, but nowadays it's just another Chrome clone with dubious features IMHO.
Opera's successor in spirit is probably Vivaldi. I use it as my standard browser on Windows and Android and am very very happy.
I remember purchasing Opera mobile. I can't remember if it was for Windows Mobile 5/6 or if it was the super early Android. It was the only browser that really worked on that phone, whatever it was. Well worth the $5 or whatever.
That app was revolutionary for its time and a true life-saver. (For the young ones: at a time where mobile networks were still really, really slow, Opera's technology would load the requested web page on its server farm, parse it, strip out the unnecessary stuff and send it to your phone in a highly compressed form. It basically enabled you to surf 'normal' web pages without overloading your phone and with load times measured in seconds instead of minutes.)
ISTR news/discussions where several shady companies were said to want to buy Opera for that technology, for the simple fact that everything you did on Opera Mobile was at some point processed on Opera's servers in unencrypted form. I dread the day where Google or Meta start offering such a service.
I think I used to know that. Thanks for the reminder.
Regarding your 2nd paragraph, that’s indeed what Cloudflare has started offering. Your browser is moved to the cloud and you effectively run a dumb terminal and get remote desktop of sorts. I think it’s pitched as a security benefit. Cloudflare has a tendency to always assume everyone fully trusts them with everything. Indeed the technology is great for snoops who want to see everything you see and do.
I wasn't aware of the CloudFlare thing, thanks. Though I'd trust them more than Google or Meta, I share your reservations. Looking at how big a part of the web they serve, there's already way too much power in too few hands anyway.
Pretty sure Opera was in the list of backdoored software, revealed by Ed. Snowden
I've been saying this for years, no one took me serously 😒.
Servo is a nice start, but it needs a lot of work.
And regarding the gov financed browser, we're way past that option IMO.
view more: next ›