Hacker News new | past | comments | ask | show | jobs | submit login
New in Chrome 73 (developers.google.com)
113 points by AliCollins 9 days ago | hide | past | web | favorite | 124 comments

The DevTools update is amazing: https://www.youtube.com/watch?v=uddZX9ZK6wY

Break point like console logging is a neat idea.

Xcode has similar functionality. It is the actual breakpoint, but you can just log the message (or make the computer to say it) and continue execution.

> Xcode has similar functionality.

And safari's own devtools inherited it years ago.

edit: looked it up, was introduced in Safari 8 (2014): https://news.ycombinator.com/item?id=8585122

This was already technically possible, you could add a conditional breakpoint with a console log in the input and it wouldn't "break" because console.log returns undefined (false).

Thanks for this! Extracting only used CSS was really helpful. I really needed this a few weeks ago and didn't even know it.

SXG[0] looks promising - allows signing (and subsequently caching externally) whole HTTP exchanges.

This may be useful for improving security, especially of CDNs. Binary Transparency seems to be one of the use cases mentioned in the spec[1] - perhaps someday this would be used for an unified scheme for signing application packages/updates, without reinventing the wheel every time.

[0]: https://developers.google.com/web/updates/2019/03/nic73#sgx

[1]: https://wicg.github.io/webpackage/draft-yasskin-http-origin-...

SXG is primarily designed for AMP so that the browser can display the origin in the address bar while the content is being served from Google.

Currently only one CA provides paid certificates with a special extension so that the cert can be used to sign SXG files [0].

[0]: https://www.digicert.com/account/ietf/http-signed-exchange.p...

As for binary transparency it's not enough to stamp the certificate (that's what CT logs do). The artifact would have to be stamped and published in a widely accessible source. Actually Binary Transparency doc published by Mozilla [1] creates a new regular certificate for new published binary thus utilizing CT infrastructure as it is today.

[1]: https://wiki.mozilla.org/Security/Binary_Transparency

If we're at Mozilla, it's also interesting to see what's their position on SXG [2]. There is only one spec there with that status there.

[2]: https://mozilla.github.io/standards-positions/

FWIW, Mozilla's current objections to the standard don't really make sense. See: https://github.com/mozilla/standards-positions/issues/29#iss...

It seems the real issue at the moment is that it just isn't a high priority for them.

Thanks for the link! Subscribed.

> SXG is primarily designed for AMP so that the browser can display the origin in the address bar while the content is being served from Google.

Displays the origin of the content?

Instead of ugly "https://www.google.com/amp/www.example.com/amp.doc.html" links as it displays them now when clicking AMP result on Google.com it would display "https://example.com/amp.doc.html" even though example.com was not contacted at all.

> unified scheme for signing application packages/updates

This is one of our longer term visions for the API, not there just yet.

This is a huge release. Supporting PWA installation cross-platform provides a viable alternative to Electron-style bloat for many apps, and shipping SGX is useful for anyone who dislikes the AMP UX. Kudos to the Chrome team.

Only for the 'apps' which are just websites but use electron because they want an installer. If you need native access, pwa is not it and won't be.

PWAs are getting more and more native access every day. File System access is a planned feature.

Oh $deity. If this is for real, we're all doomed.

For the category of apps I work on, this will finally make the browser a real alternative. (I need access to fast local IO).

sandboxed file system access. People use electron because they need things like registry keys, reading/writing files from anywhere on the system, interaction with specific kernal functions, and tons of things browsers cannot give anywhere the same flexibility to without becoming a major vector for malware. Not to mention a lot of electron apps, like Spotify, use their own c++ libs, like Spotify json, because browser perf sucks.

You cannot sandbox and have native access, full stop.

It's sad that Chrome gestures and shortcuts don't work in those.

Somehow, the update to Chrome 73 managed to ruin Chrome Sync on my work Mac. I have to "unpause" sync every time I run Chrome and I'm logged out of every account after closing the app, although I do allow cookies and don't clear them on close. My passwords don't autofill anymore, either.

What's the point of buying into the whole Chrome Sync thing if it randomly dies for no apparent reason?

Happened to me too.

It seems to be a wider problem by looking at the messages from https://support.google.com/chrome/threads?hl=en&thread_filte...

So, are these "Constructable Stylesheets" a non-standard Chrome-only web feature that Google designed and implemented, or what's going on here?

It is:


They added two new methods to solve the FOUC (flash of unstyled content) problem.

It can easily become part of the standard if the feedback is positive.

Regarding signed exchanges:

"When the browser loads this Signed Exchange, it can safely show your URL in the address bar because the signature in the exchange indicates the content originally came from your origin."

- the original website will not see these requests in its logs. Can visitors even tell that their browser is not accessing the server that shows up in the address bar? Their privacy is being violated.

- when the original website deletes or modifies a resource, the old outdated version may still be distributed by a 3rd party

Analytics could be done using JavaScript. (Or maybe some other way of triggering a request?)

Alternately, depending on your perspective, it could improve privacy, by making it easier to create web apps that don't phone home and disclose your IP address. Perhaps distributed on a DVD? (I don't know if it supports full offline access.)

Yes, reading stale content is possible. But this model of distribution is not new, it's how ordinary software distribution works. It's like downloading software from a mirror site, but the signature is automatically checked.

Or consider what Cloudflare does and that you need to give them use of your private key to make it work. It's a trade-off.

This is just a crappy version of IPFS. NIH syndrome at its worse.

Desktop PWA update looks promising and considering the last update, allowed many to upload PWAs on Google Play store. I feel the promises are finally coming true.

Have written down my experience with Desktop PWA update: https://news.ycombinator.com/item?id=19377170

You should probably just say its a tutorial on how to install a PWA (i.e. how to click on the install button).

This update includes everything in V8 v7.3 as well: https://v8.dev/blog/v8-release-73

Is there a list of popular apps that are available as PWAs yet? Or info on if/when major apps will be available as PWAs (slack, spotify, etc)?

Lyft and Uber are PWAs.

Instagram and Tinder both have excellent PWAs.

Twitter, Starbucks, Skype

Spotify is a PWA

Has anyone else noticed a reduction in New Tab icons? I used to have 10 defined, but now it will only show 8 of them. OSX 10.14.3, Chrome 73.0.3683.75.

It also seems to sometimes not want to display the text for the New Tab icons in white and picks a nearly-imperceptible slightly-darker grey.

I briefly looked around and it seems like desktop PWA are not yet supported by Firefox. At least I haven't found a way to add an app to the desktop like we can do in Chrome. A bit disappointing.

On macOS at least, opening a PWA .app also launches Chrome, and quitting Chrome also quits any PWAs you have open. That's not a stellar experience.

Exactly like it was with Chrome “applications“ four or five years ago.

So Chrome supports dark theme, and PWAs can now be installed, but the PWA window does not support dark theme. Ironic. Two features developed in isolation?

What are they supposed to do with them, repaint Twitter and the others? Especially since these apps basically have zero system elements other than 3 buttons (to clarify I'm talking about MacOS).

PWAs can be themed by the app.

Intel SGX[0] is a specific technology supported by the CPU for encrypting application memory. Is the acronym used differently in the context with Chrome?

[0] https://software.intel.com/en-us/sgx

no, but to be fair it's pretty confusing.

intel sgx: software guard extensions

chrome sgx: Signed HTTP(????) Exchanges


* Dark mode is now supported on Mac, and Windows support is on the way.

I find it a little bit amazing that we now live in time where features in mainstream applications ship for the Mac before they ship for Windows.

You mean like how IE5 for Mac in 2000 had a variety of features IE for Windows lacked, including much superior CSS support?

There was so much more than that, too... Mac IE5 had full support for transparent PNGs, a download manager, an auction manager (!), and a bunch of different color themes to match Apple's product line. It was easily better than even MSIE6, and it's a shame that Microsoft essentially abandoned the product after release, along with the Tasman rendering engine it used.

It's because of MacOS's dark mode is a little more established the Win10's. It makes sense to release the feature there first.

How is dark more enabled, do you know? Is it automatically enabled while under Mojave dark mode, or? Thanks!

Yes, it's been automatic in Chrome Canary, so it's probably gonna stay that way.

Yes - my Chrome starts in dark mode when Mojave is in dark mode.

what about linux?

We need to drop Chrome and start using Firefox. I just switched to FF Nightly and it is really fast with GPU acceleration. Please download today!

I think the best browser definitely fluctuates over time. I started out with Mozilla c1995. Then have shifted through most of them at various points in time when I've become unhappy with my current choice. I went back to Firefox when they released their Quantum version and have been really happy with it. Haven't used Chrome much since then. But I'm always open to other browsers should something change in the future.

Was this motivated by something in the release notes, or are you just taking the opportunity to bang the Firefox drum? Should we expect this comment in every Chrome-related thread?

We don't "need" to do anything.

It's nice to have choiches.

Yeah, and one of these "choices" implies not having any more choice at all

You are free to make your own choiche, not to impose it on me.

Yes, you get the point, now you just need to enforce that because this is actually being forced by a ton of websites requiring chrome to work.

But the internet needs it. A one-browser monopoly will not be good for anyone.

Actually it is good for an average user who doesn't care about privacy.

This doesn't even make sense. Not caring about something doesn't mean it won't be an issue for you and the people around you.

We need to drop Chrome and start using Firefox exactly because of that.

You also have the choice to buy clothes stitched together by children in Southeast-Asia under horrible conditions, but we all know that a responsible consumer decision looks different.

Let's not pretend that using a web browser and supporting slave labour is even remotely comparable.

What I'm saying is, that "we have a choice" is a void argument if you have just been presented with a solid case that one choice is clearly better than the other.

no, we need it; or we will go back to a fancy version of Internet Explorer 6 (that is, chrome browser): proprietary technology implementation, by a top player, against any standard... and spying on users (but that's new, I don't think IE 6 was that vicious).

Web monoculture make the web worse for everybody.

It's funny, I see this comparison a lot and the people using it don't see how it misses the mark: the reason (the majority of tech) people hated SO MUCH on IE6 was not because it was proprietary, but because it was static and didn't bring get any advance, thus blocking the web's progress. Chrome doesn't have that issue, at worst you can claim it gives too much power to Google on decision about where the web goes, but not that it keeps it from progressing.

It was also not respecting the standards but only after a while (at release it was at the forefront of it), and that is also not a problem that Chrome can relate to.

I say that as someone who uses Chrome, enjoy it, but would also prefer if we didn't go back to a single engine being supported on the web (and I switched back to firefox on mobile, because of chrome lacks of support for ad blocking there).

By confusing what the fight is about and trying to make it about what it's not, I fear you make some people like me ignore it.

>It's funny, I see this comparison a lot and the people using it don't see how it misses the mark ...

You seem to be missing the link between monopoly and stagnation though. IE wasn't always stagnant. It brought us XMLHTTPRequest for instance.

I think the debate has to be about whether or not Chromium being open source changes the equation.

IE6 era's Microsoft had an interest in the web stagnating, while Google doesn't. If anything, they used their position to push forward new things that others wanted to slow down.

It appears to me that Google is pushing forward some things while slowing down others. Obviously, Google will never be at the forefront of limiting what advertisers can do.

My biggest worry with a Chrome quasi monopoly would be the erosion of browser making competence and financial resources outside of Google.

So whenever Google's vested interests clash with what users want, there would be no one left to step up. Chromium would be open source, but only Google employees would know the source well enough to do anything with it.

It's a worry. I'm not saying it's the same as with a closed source IE back then.

Progression != improvement.

When ad blockers won't work, when you get DRM everywhere and trackers will be embedded in the browser, the tech will have progressed. Yet it will not be the progress you want to live.

It's like if suddenly you had only one car maker and say "it will be easier to design infrastructure for it". Yes, it will be. But it also mean you won't get innovation for you, only for the car maker.

For google that would mean pushing techs serving the ad and data collection business of course. But also in general, that would mean building for big companies. Adopting standards that are mostly beneficial to people that have the money to buy patents, setup many CDN, have high end engineers, invest in specialized hardware, etc.

None of what you say corresponds to what I said.

"Chrome being a monopoly is bad because X" and I say there are lots of issue with only Chrome remaining but not X, it doesn't apply to Chrome. You answer by saying "but there is Y, and W, and Z, and ..." which is basically my point: no need to focus on what is not the issue.

IE6 wasn't at the forefront of standards compliance when it was released.

Mozilla (then at 0.9.x) and Internet Explorer 5 for Mac were both significantly better.

Chrome is just as bad but for different reasons. And perhaps even more dangerous (like slowly eroding privacy, or shaping the browser around Google’s revenue generating services). I think the analogy stands and the damage is similar in magnitude.


Was “for different reasons” too unclear? It’s right at the beginning of the comment. Different reasons but the impact is still dangerous.

For example IE6 was not reporting my location to MS despite me turning the option off. Just because some people are ok with that doesn’t make it less of a problem. It just makes it a different (bad) one.

And in the passive aggressiveness [0] of your comment you didn’t even think that there are far better reasons why IE6 sucked. If feels random and based less on long term use experience.

You write as if you’d be ok with Chrome stabbing you in the eye as long as it’s not a CSS hack, the single reference issue all browsers are benchmarked against.

[0] https://news.ycombinator.com/newsguidelines.html

I've read it and I think it's way overblown. All I care about is solid support for standards and Chrome shines it that.

I picked them because I could test if you actually experienced it. Chrome at the moment is the best we have but it does come with strings attached. It has different problems indeed, surprise surprise, however I don't think that's that a big issue at all.

> All I care about [...]

You should have started with that instead of the jab. Not only would it have been clear that it’s a personal opinion not a matter of fact (Chrome is better for me) but it would have also saved you the time of coming back later with the real single biggest issue of IE: standards.

The issues Chrome pose for most people right now are real. You don’t care about them for your personal reasons but disqualifying them because they’re not yours is very dismissive. You never know when your interests change.

> All I care about is solid support for standards and Chrome shines it that.

Except for cases when it doesn't feel like supporting the standard in question, or it claims their support is sufficient. A recent example could be support for <input autocomplete="..."> [0].

[0]: https://www.reddit.com/r/programming/comments/ar1qj1//egl52w...

I think you are wrong, or at least, I don't share your impression at all...

I was there when IE 6 was a thing, and no, nobody was complaining that IE was too static. That wasn't the main reason at all. One of the main problems is that it was too dynamic, but in a proprietary and unsafe way (ActiveX).

Read, for example, this article from 2002, when "Phoenix" (aka Firefox) was launched: https://www.geek.com/news/mozilla-launches-new-faster-browse...

IE was pushed down the throat of everybody, and the main consumer complaints was "it's bloated".

For developers instead, the craze of the day was to pass the Acid test: we are talking about CSS implementation. We are talking about open standards.

Open standards where the main concerns for developers: http://davidnaylor.org/blog/2005/04/web-standards-acid2-test...

Acid 2 and Acid 3 where a the main thing during the browsers wars (https://en.wikipedia.org/wiki/Browser_wars). And they were ALL about standards.

It is fair to point out that the negative effect of a google monopoly would be quite different from what happened with IE6

The problems IE6 had are vastly different from the problems Chrome presents. But I would argue the Chrome's are potentially worse. IE6 was a big problem mostly for developers who were perhaps better equipped to deal with them. Chrome not only has the chance to become the "singl;e browser" (if it's not already there) and direct the web wherever but it also poses problems directly to the users who are most definitely not equipped to handle them.

I see this as a much bigger issue than IE ever had. It's not technical so a simple patch in your web app won't fix it.

I would love to do that, I'm testing new releases, but every time I try, my machine wants to explode https://www.reddit.com/r/firefox/comments/7g6k9n/firefox_qua... , https://www.reddit.com/r/firefox/comments/asdkxf/firefox_on_... , etc...

My main problem with Firefox is that it eats my batteries, otherwise I would have used it as my main browser.

Same here. Cannt use it on my mba 2012. It just to heavy on resources. Chrome handles it without more loosely.

If you want to get away from Chrome my tip is to use Opera. Same engine underneath & can handle same extensions (with a proper extension installed) but without Google.

Opera also has a battery saving feature.

Opera is also closed source and has been bought by some Chinese consortium, so that may not be the best option if you want to get away from Google for privacy reasons.

Correct. I just think that Google is a bigger problem than this Chinese consortium. If I practically could I would avoid both.

Tried running Edge for a while, which is gentle on battery, however Edge is just lacking.

Using Opera only superficially sidesteps the big Google problem, as you still contribute to the Chromium monoculture that reinforces Chrome's dominance and benefits Google.

Yep, sad state of affairs. Sad day when Microsoft succumbed to Chromium monoculture.

However when I develop I always develop for Firefox first, but the Firefox developer tools are starting to age as well.

At least the performance issue of Firefox on Mac is being worked on and should hopefully be resolved soon: https://bugzilla.mozilla.org/show_bug.cgi?id=1429522

Even though PWA:s are good in theory, will it only increase Google's grip on the web?

Has the web standards become so complex that it is inevitable that we will have a Chromium monoculture?

> should hopefully be resolved soon

I want to believe, but I heard that multiple times and still can't even open simplest html websites without hearing the pain of my machine.

It had been blocked by webrender if I remember correctly, now the work on fixing this issue has actually started. Markus Stange is a capable engineer as well so there's hope.

Firefox team needs to wake up and add support for PWA.

I thought PWA was just service workers and implementing routing on the web app level.

What's missing in Firefox?

"Integration" into the underlying OS (showing up as a desktop icon / homescreen shortcut)

Drag and drop url to desktop. It does the same thing. Firefox should get a head start and create a pwa store.

Please, respect other people's choices.

Is it still a choice if the user didn't think about choosing?

Not possible for me to start using Firefox until it starts to support WebSocket frame inspection..

You know, it's not like you couldn't use one browser for work and the other one for everything else. I certainly don't need to inspect WebSocket frames while booking a flight, reading the news, chatting with family on facebook, or shitposting on HN.

Having Firefox and Chrome installed is still less than the bunch of Electron based apps that are so beloved on HN.


And one more thing - 2, 3 and most exciting 4 works on Android too.

But not 5)? I think if you watch youtube in FF Nightly with uBlock installed you should not see any ads?

uBlock Origin is not ideal. But there's NewPipe at F-Droid for YouTube, it's also without the proprietary YouTube-API or any of Google's (proprietary) play-services. https://f-droid.org/en/packages/org.schabi.newpipe/

Did you forget

1.5) Start IE

Have they reverted the horrible UI?

Could you at least go into a little detail about what you think is horrible about the ui? That way you're contributing to the discussion.

The new design is space wasting. Check the URL bar for example, you type something and you get suggestions. The distance between each entry is huge. I don't have a touch interface on my desktop. It should be configurable. It was, until they removed the flags.

Tabs: with the old edgy style you saw "more" and was easier to click. This might be a personal choice though.

Account: Previously (on Windows) your Google account was displayed next to the minimize/maximize/close button. Now it's next to the icons for addons. Wasting space.

Difference between Linux and Windows: On Linux (Fedora, Ubuntu) you can use the scrollwheel to cycle through tabs when hovering over them. Doesn't work in Windows.

We need less PWAs, not more. Many people are already confused enough when it comes to grasping the difference between websites and native applications. Another "standard" created to consolidate Google's position in the market.

I'm not familiar with the details of implementing a PWA and how much it's tied to Chrome specifics (that would certainly be bad), but I generally disagree with you.

I feel like way too many companies are making native apps just for the sake of users being able to install them and have that icon on the home screen. Most of the time the app doesn't require any native-only functionalities, it could function as web app just fine, so it's a bunch of unnecessary overhead. And mobile platform owners love this as the growing ecosystem means more lock-in.

PWAs seem to be a nice middle ground solution to this problem IMHO.

I get your viewpoint. I definitely don't want to be harassed by a website to "install" their PWA on the desktop, like how websites already harass me to "install" their app on my phone because it allows them to do more tracking.

However, if PWAs allow us to ditch electron for good, I'm all for them. Each electron app shipping its own browser copy and having full access to everything on my computer is a security nightmare.

I wish Servo was mature enough to serve as Electron replacement.

PWAs aren't specific to Chrome, they are also being supported by other browser vendors.

In fact, PWAs are UWP apps on Windows, having access to all native UWP apis without additional FFI, or having to deal with the likes of Ionic, React Native and friends.

I'm also confused about how privacy enhancing browser extensions work within a PWA. I've yet to get a full answer on this (it seems as if no one really knows) but it seems to me as if a PWA is more friendly to trackers and less friendly to users of the browser.

I work on the project - PWA's on desktop are essentially just tabs without the Chrome at the moment, so it all goes through the same plumbing right now. It's not more or less friendly to trackers, if you have an extension the extension should still run in the same context.

Is that an implementation detail or in the design spec?

I appreciate the explanation, thanks.

If you want to keep trackers out, the only real option is not to use any Web based app.

There are plenty of tricks to work around ad blockers.

I remember an early version of this from the iPhone2 days. The standard allows running the app offline, not supporting the desktop when mobile had this feature feels backwards. I don't see these types of comments about Microsoft locking people into windows, chrome is managing to be both an application, and an OS.

PWAs are allowing better access to the device, things like offline access and File System access soon. Its more than just an icon for a website on your home screen.

Applications are open for YC Summer 2019

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