Hacker News new | past | comments | ask | show | jobs | submit login
Browser diversity starts with us (zeldman.com)
172 points by ccnafr on Feb 4, 2019 | hide | past | favorite | 91 comments



We lost some diversity when Microsoft decided to adopt chromium[1].

I assume, though, that Google will eventually make some change that Microsoft doesn't want, and they'll be forced to fork chromium.

The recent proposals from Google[2] that hobble ad blockers and extensions like tampermonkey might even be enough.

[1] https://www.theverge.com/2018/12/4/18125238/microsoft-chrome...

[2] https://www.bleepingcomputer.com/news/security/chrome-extens...


> I assume, though, that Google will eventually make some change that Microsoft doesn't want, and they'll be forced to fork chromium.

This assumes that Microsoft still has a browser engine division by then. I suspect that adopting Chromium was largely a cost-cutting measure and this division is on the chopping block already.


I think they they had to kill their own rendering engine to keep Edge alive.

Sticking to their own engine would have eventually killed both. Hard to get people to adopt your browser if webdevs are not testing sites with it. And webdevs don’t care if your market share is marginal.

I believe the alternative would have been most of the enterprises deploying Chrome on Windows to access their apps. With this move Microsoft might still retain control of the app level. This is valuable because then they can for example provide integrated windows authentication for the apps.


Yes...you have to make the leap that they would see some opportunity worth reincarnating a browser group for.

That set of proposed changes from Google looks like one to me, but it's possible they don't to MS.


I think it was a cost cutting measure too. But more about removing duplicated effort based on the premise that Microsoft was already spending a lot of time working on Chromium compatibility. And Google was inside Microsoft's decision loop. Microsoft was in the position of having to port every arbitrary evolution Google made to Microsoft products...browsers and tooling.


They're implementing some features in Chromium, so they still have that division.


Well, I know one-person teams who implement individual features in a browser. Maintaining a fork, on the other hand, takes a much larger team.


Would have been nice if MS went for Gecko, would solve a lot of these concerns.

I was just starting to get that butterfly/stomach feeling for Edge, I don't think I'd ever leave FF for it but Edge was pretty attractive and I started to developed some warm fuzzy feelings for it. My initial revulsion has long since been replaced by a sort of curiosity which had the potential to grow into even more... That chromium news killed it though.


I don't think I ever tried Edge - what advantages did it have?


Starts fast, minimal UI, low battery usage, good built in pdf reader, fancy visual tabs (pull tabbar down, see miniature pages), set tabs aside feature is nice. But mainly it just feels very fast. This probably won't change with when they start to base on blink but that leaves the issues discussed here.


The sandbox in Edge also seemed well thought out: https://blogs.windows.com/msedgedev/2017/03/23/strengthening...


I used it few months and it mostly worked just fine. I had few crashes but those were rare. The main drawback was that their developer console was somewhat buggy, so eventually I switched back to Chrome because I needed it. Also 1Password extension for Edge was buggy as well.

And, oh my god, how many times Google asked me to install Chrome. They are really annoying with those ads which never stop to appear. I suppose that for many users that might be the reason to switch to Chrome, to stop that insanity.


> And, oh my god, how many times Google asked me to install Chrome.

Interesting - I've never seen that in FireFox


Microsoft should have put an anti-trust lawsuit on google for that. Google has enough of a monopoly that asking users to install chrome qualifies as abuse.


I opened microsoft site and it only showed me microsoft affiliated products. I think an anti trust lawsuit should be filed for that.


That is sort of equating a business with the road that leads to it.


Yeah, it's a shame they didn't open source MSHTML instead...


Technically it was the "Trident" engine


MS waited far too long to add things liken Ublock and Ghostery to Edge. I see they are available now, but for at least a year after its release, there were no such extensions in the app store. In this day and age, using a browser without an adblocker makes it hard to consider switching away from your current setup.


I'm not sure everyone sees how lousy Chrome adblockers are going to be after this set of Google proposed changes.

The API doesn't let you add to the list of blocked uris outside of posting a new release of an extension. Ouch.


They created a discussion thread, but the Chromium developers have been absent for a while: https://groups.google.com/a/chromium.org/forum/#!topic/chrom...


I'm not sure there's much to discuss. The API didn't provide any way to add an item to the blocked list. That's clearly deliberate, and has nothing to do with the stated goal of improving performance.

I can only gather that the actual goal is very different from the stated goal.


Firefox it is then. I have defaulted to Chrome because of DevTools and I'm just used to the keyboard shortcuts.

I have Firefox on OSX already, and it runs smoothly. Unfortunately mobile Firefox (while supporting all those extensions), has a jittery scroll issue on Android, so I use the Chromium-based Kiwi Browser (for night mode) or Brave (for built in adblocking).


Iirc Firefox Mobile's jittery scroll was fixed in the last couple months (along with fling velocity) so it might be worth checking again.


If we only count people who care about their browser, then the IE user share will be much lower. In other words Microsoft can do anything with IE and it won't reduce the usage much. And Microsoft knows this, even though it tried to gain technical edge for some time.


I recently replaced chrome on every company computer for Firefox. We shared a google account with passwords that everyone needed. But some fools that don’t understand their actions properly start deleting password by accident as the cleared the browsing history. Yes I know you have options when you clear history.

In Firefox passwords and browsing history / data is separated. So they can delete browsing history and they don’t accidentally delete the passwords anymore. And with a Firefox account all our passwords that employees need are synced safely.

In one day 150 employees changed to Firefox :D


>We shared a google account with passwords that everyone needed.

I am speechless.


I'm intrigued by the `needed` in there. Why would that ever be a necessity?


There's mundane stuff, like letting a small group of people log into Gmail for support@mysmallbiz.com

There's ways around that, but for some orgs, it would be unneeded overkill, and not protecting anything notable.


> but for some orgs, it would be unneeded overkill, and not protecting anything notable.

I can honenstly not think of a single venture where "support@" is not one of the most critical resources wrt privacy and security. On top of that, "support@" is typically the account that has a high churn rate. Where people move on and new people are hired. Of all the cases, I'd say that "support@" ranks amongst the top for need of proper account management.

That said, it's dead simple to grant jane@ and john@ access to an inbox in Google. Researching how to do this may take 30+ minutes. But getting it configured afterwards is really a two minute job.

The only reason I've came across why people shared Google accounts was "we have a business domain and we need to pay for every extra seat". Which is a valid excuse. I'd argue that its not a good enough excuse to lower your security for, but valid nontheless. For one, 2fa is almost impossible when sharing accounts.

Which is why having a "pay per seat" model for any SAAS is perpendicular to having proper security practices. You are not rewarding good security, but rather punishing it by letting organisations with proper separation of accounts pay more then the ones that choose to have as few as possible.


I've seen functionality where one gmail account can send an email so that it appears to be from another email address.

I haven't seen anything that allows a seamless view of the inbox/outbox, and a way of sending that doesn't accidentally use their normal email address if they forget to click a drop down.


Support emails fall into the "interesting" GDPR bucket often overlooked. I see only few companies actually share/keep track of that data bucket when doing a data request, or deletion request.


There isn't, the only excuse is lazyness and a complete disregard for any kind of security.


Lets say your company has a profile on a platform that uses a Google account login. How many Google accounts do you want to wager can be permitted to administer that profile?


On a properly written platform? As many as needed, RBAC exists for a reason. On a shitty platform? Zero, that shit isn't production ready.


And I'd bet you never worked in any OPS task.

Not every password identifies a user.


I've worked in ops for a decade. There is no legitimate excuse for sharing accounts/credentials.

Ignoring the myriad security issues with shared credentials auditing alone is completely ruined with shared creds.


What's the reasoning behind everyone needing that level of access?

We use 1password and have multiple vaults depending on what sort of clearance is needed. Need access to support emails? We'll add you to the appropriate vault. This way dev teams can be separate from marketing etc and no one can unexpectedly gain access to what they shouldn't be able to or wipe passwords they shouldn't have access to.


Small online tools that everyone needs to use. Passwords that needs to change often because of security settings of those tools.

Yes I know other solutions are there. Like 1password and step by step I will get there. At the moment they are unwilling to move to a password managers.

A Firefox account with synced passwords is a huge step forward. Before that there was a excel file with all passwords. Or worse post-it’s.


How about downloading and installing Firefox and seeing if it meets your needs? I totally understand if it doesn't, and then you should keep using chrome or safari or whatever, but I am a back end web dev and Firefox (+ edge) meet all my desktop needs.

Give it a try, you might like it.

And that'll have a lot more impact than testing cross browser regularly.


At my work I only work in Firefox, another Dev only works in Chrome, etc. By each developer only working in one browser all the time we get pretty good cross browser testing. This has led me to only using Firefox at home too since it works really really well.


Same here. When I joined my current team, I was told I should use Chrome, because nobody really tested the site on Firefox, so there may be some breakage. That made me double-down on Firefox use, just so that I could report if any problem actually occurs.


> And that'll have a lot more impact than testing cross browser regularly.

That depends entire on what you mean with "impact" here.

One site that does not work on Firefox has a great impact too. When that "one site" becomes every fourth site, then "we", the webdevelopers have handed Chrome their monopoly.


> That depends entire on what you mean with "impact" here.

Fair enough. I meant that using Firefox regularly as a web developer would have a larger impact on keeping the web free than testing periodically on Firefox. Why? Because testing periodically won't exercise the browser quirks as thoroughly as every day use (and sometimes time intensive cross browser testing gets sacrificed to schedule concerns). There's also an additional bonus of keeping browser share up, but that's not the primary benefit in my mind.


> Chromium will be the web’s de facto rendering engine

I think Blink is the rendering engine. Chromium is Chrome without Google branding

I think that the problem with Firefox right now is that it does not have something unique to offer. Before moving to the new plugin model, there were plugins that were not available on Chrome which was available for Firefox : DownThemAll, Tab wheel scroll etc. But in their desire to chase Chrome they threw it away.

I also seem to have an annoying bug in mobile where after a while long tap on link no longer opens the contextual menu. I wonder if I should report it since last time I reported a defect it turned out to be a problem with uBlock and this one takes a while to reproduce.

I think it might be a good idea for Mozilla to invest in good extensibility. It helped them a lot last time and is therefore a battle hardened tactic


> I also seem to have an annoying bug in mobile where after a while long tap on link no longer opens the contextual menu.

I encountered a bug matching this description on my S7. So far I discovered that it's correlated to the number of tabs open; around 70 tabs, the context menu appears after perceptible delay, which increases with the number of open tabs. I think I reached something around 90 tabs before I couldn't long-tap on links any more. Closing tabs makes the problem go away.

I'd say "go ahead and report it", but I too have uBlock installed, so I can't say if it's unrelated. uBlock was the very reason I switched to Firefox on Android anyway.


> I also seem to have an annoying bug in mobile where after a while long tap on link no longer opens the contextual menu

I thought I was the only one. I don't have uBlock installed, but also don't really bother reporting bugs because mobile Firefox doesn't appear like a supported project to me at all (where can I find ANY documentation?).

It happens to me occasionally, but I can't say anything about my number of open tabs because it just displays "∞" (wtf man?).


Maybe temporarily disabled Ublock next time you notice the behavior and test? My experience has been positive when submitting bug reports for Firefox.


> I think that the problem with Firefox right now is that it does not have something unique to offer.

Sounds about right. Because they are not tied down by allegiance to a platform, they're going for an ideal. This might pass the smell test if Chrome was anywhere near IE6 share levels or was generally a bad web citizen, but they're not really - save for decisions that come from Google thinking about its ad businesses and how to enable more of that through Chrome.


Stop using Google Spyware Browser. Google has abused its market position for a long time now, and we are seeing the impact. Forced browser logins, and now proposed deprecation of Ad Blocking API's are the two that bother me the most. The Fox is cool again, Mozilla has your back.


I don't think Chrome is a spyware for me, but I am still trying to get to Firefox, if only because I don't want chrome to be the only browser remaining in next 10 year and then turn to spyware. Firefox is much more open on the moral issues and on most instance has better track record than Google.


>Mozilla has your back

That's kinda funny when you consider where Mozilla came from... the ashes of a thing called Netscape... who tried to rule the world while a couple grad students were busy plugging away at Stanford, doing it the right way.


Okay, please state the supposed crimes of the ashes of Netscape because I think a lot of people here aren't following your line of reasoning.


You guys can do whatever you want to "diversify" internet browsers, but the writings on the wall. Not only is the majority using Chrome, but all minorities other than Firefox are now derivatives of Chromium. The average tech illiterate consumer isn't going to say "You know what, this isn't right! We can't let Google take control as the dominant layout engine!". They want something relevant that performs well, and Firefox hasn't been relevant in years. Stop wasting your time and drop it. You're ancient browser will crumble just like Netscape, IE, and Mosaic.


> You're ancient browser will crumble just like Netscape, IE, and Mosaic.

The exact same argument could be made about IE around 2000, yet now you list IE as ancient and crumbled, which completely undermines your argument.


This critique amounts to "it's not working so give up". Thankfully minorities in the past haven't always been so hopeless.

Every vote counts.


Yes but this isn't an election, it's a choice of software. Regular people do not care enough about internet browser dominance. Software projects like these are largely supported by the size of their user space and corporate backing. Your voice will never be heard.


Much like voting in politics our usage does add up. If a significant portion of users abandon a browser then the producers will notice.

It's only hopeless if we all agree to do nothing.


I am unsure how a .exe file "crumbles".

Most websites are already broken, due to anti-adblocking, newsletter signups, GDPR consent, etc. I don't really care if most websites requires Chrome, most websites are broken anyway.

Plus this is Firefox. It has been down this patch before, heck it was born in it.


I use Edge and ( seriously ) all you guys did was mock me for it.


What's the best way to port a puppeteer-based suite to one that uses Firefox?

Is there an npm package that will include up-to-date versions of all components without further environment setup or use of external services?

Ideally, with great support for taking snapshots and running headless or headed in simple fashion?



I use this now, and love the ease of installation and api. If it's up to developers to use Firefox it'd be great to port our test suites.


I've been a happy Conkeror user for years. Recent changes to Firefox seem to have broken it, so I'm sticking with the older version for now. Chrome/IE/Edge/Safari/etc. aren't FOSS so I don't touch them.

Firefox and Chromium are getting increasingly slow and bloated, plus their chrome takes up a bunch of screen space, doesn't follow the system theme particularly well, and they rely far too much on the mouse (there are extensions which help, but they're still not particularly good).

Out of the two I would still recommend Firefox over Chromium, since Mozilla seem more ethical than Google, contribute less bloat and churn to Web APIs, etc.


If you insist on using Chrome but back this cause, set your user agent to Firefox.


You gotta be the change you want to see. Stop using Chrome/Chromium/Opera/Brave/etc/etc. Stop using Firefox. Start using forks.


This rings true with me. I want to use Firefox and ditch chrome, but until it gets less laggy on macOS, I don’t think I have a choice. I do use safari for day to day browsing and move to chrome only for development.

Multiple successful browsers is a good thing for the web. I hope that doesn’t just die away.


I agree with the post, but it should be noted that there's a big difference between Chromium -- an open source project -- and IE 6 being the de-facto browser engine on the web.


I think the difference is mostly theoretical. Consider the proposed extension changes: no one seems to like them, but Google will be able to push them through anyways. Maintaining a Chromium fork is a huge undertaking that only a few companies would be willing to invest in.


Maintaining a fork is much less of an undertaking, than reimplementing all Chromium quirks. Hence plenty of Chromium forks exist, but only one last barely competing browser.


No nontrivial Chromium forks exist, to my knowledge. Do you know of any?


I don't think Opera, Vivaldi, Brave, Yandex browser are trivial by any measure. Heck, even something small and independent, like ungoogled-chromium, is in no way trivial.


I haven't checked, but it was my understanding that – outside of the UX – these are all very shallow forks.

Does anyone know more about this?


Yes, these are all very shallow forks that mostly “just” change UI. That guy doesn’t understand what nontrivial means.


I never ever used chrome: I used Opera, and after Opera 12 I am using Firefox. What am I missing?


This could be a huge opportunity. We have a web monoculture now. It's only going to get increasingly so over the next few years. Google will let the platform stagnate, as it has little to no reason to compete. It's time to come up with something better. Something designed for both documents and applications.


> Google will let the platform stagnate, as it has little to no reason to compete.

If they do, they will be replaced, just like iexplore was. So they won't stagnate unless they are stupid and want Firefox to take over.


> If they do, they will be replaced, just like iexplore was. So they won't stagnate unless they are stupid and want Firefox to take over.

From a practical perspective, ten years of stagnation is not very different from indefinite stagnation. Even if there's a light at the end of the tunnel, we're going to have to spend a huge chunk of our lives suffering through it.

And with Google, stagnation is probably the best case. If they continue to "move things forward" as a browser monopoly, they'll probably force many user-hostile things (like their recent proposal to cripple ad blockers https://bugs.chromium.org/p/chromium/issues/detail?id=896897...).


Do you have any starting point?


Maybe something similar to what we have now, but exposes much lower level aspects of the browser and gives you more fine grained control over it. WASM is a good starting point in that it lets you use whatever language you want. Now, we need a browser API that will let us draw directly to the window if we want to. I think a lot could be accomplished by creating the UI portions of a site programatically then dumping the content into a "renderMarkdown()" function.

That's just my idea, though. Hopefully, the current sad state of the web and the coming browser monoculture will spur others to experiment too. I know some gaming companies have experimented with the distributing and executing of x64 binary executables using the browser. I know others are looking into ways to chop up and do progressive loading on binary executables.


The appification of the Internet qualifies here, with Google Fuchsia running on Chromebooks and ARM-based desktops being a good opportunity for them to move most Internet based activities in apps on a platform they control.


lol, no one complains when apple shoves webkit down your throat or google does the same with mobile chrome. the battle has already been lost mostly-who uses mobile firefox really?

on play store, by installs firefox - 100million+,chrome - 1billion+. firefox is great but it's awkward focus on freedom over usability has led to it's downfall imo


It's obviously a matter of money, politics, and things beyond my imagination but I can't understand why Firefox didn't double down on Servo (and friends) for mobiles.

At one point there was talk about Samsung being a partner and Firefox/Servo becoming the default browser on the Samsung phones. Which would make a perfect sense, since every non-Apple phone gets its performance from multi-core SoC, and Firefox would be the only browser taking advantage of that. Now, after Google getting their hands slapped by the UE regulators, this Samsung/Firefox marriage could be even more fruitful. And yet, the Servo team got assigned to some pointless VR experiments, and it turned out Android was never a priority. Again: I can't understand why, oh, why.


Servo was a test bed, a place to experiment with new technologies on a minimalist browser. The successful underlying technology was/is being moved into Firefox (desktop and mobile).

There's also a new (I think?) mobile browser in progress. I'm not sure how it fits into the larger picture and information is a bit scarce, they just started releasing beta builds.


Yes, Apple gives its 1.3 billion iOS device users zero choice in browser engine and gets a pass.

Google lets any browser engine on Android but they're evil because their browser, Chrome, is #1

If you want a choice in browser the first place you should be pushing is to allow them on iOS. Every other platform, Android, Windows, MacOS, Linux, you get to choose. I for one would actually like to be able to run real Chromium on iOS to get access to the 100s of standards that Apple has yet to implement. I'm sure lots of Firefox lovers would also like to run real Firefox on iOS as well.


You are missing the part where nobody forces users to choose iOS.


No one is forcing users to choose chrome either. Except all the advocates who suddenly realized that firefox is fighting a losing game


Lots of people complain about your two examples. Google is even changing their Android licensing, making it easier to not ship Chrome by default, because the EU told them off for it.


100 million is actually pretty good. Isn't Chrome the default for many vendors, not to mention what people are most familiar with in the first place?

(I use mobile Firefox myself, it's fine.)


> firefox - 100million

Actually, that's more than I thought. I've always thought that Firefox on mobile is virtually nonexistent but it seems to have more users than any european country (except russia). In any other instance an install base of 100 mil would be wonderful.


installs != active users




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: