I think you should appreciate more how much the tens of billions of dollars Google has invested in Chrome has benefited the web and open source in general. Some examples:
Webrtc. Google’s implementation is super widely used in all sorts of communications software.
V8. Lots of innovation on the interpreter and JIT has made JS pretty fast and is reused in lots of other software like nodejs, electron etc.
Sandboxing. Chrome did a lot of new things here like site isolation and Firefox took a while to catch up.
Codecs. VP8/9 and AV1 broke the mpeg alliance monopoly and made non patented state of the art video compression possible.
SPDY/QUIC. Thanks to Google we have zero RTT TLS handshakes and no head of line blocking HTTP with header compression, etc now and H3 has mandatory encryption.
> Codecs. VP8/9 and AV1 broke the mpeg alliance monopoly and made non patented state of the art video compression possible.
Not really. That was done more by the greed of the MPEG alliance.
Back in the days when <video> was first proposed, VP8 was required to be supported as a codec by all browsers. This was removed as a requirement after Apple stated they were never going to support it, but the other browsers still implemented VP8 because it was codec free. Then Google implemented H.264 in Chrome. Mozilla only implemented H.264 in Firefox after it became clear that Google's announcement that they were going to rip H.264 out of Chrome was a bald-faced lie, making H.264 a de facto codec requirement for web browsers.
Having won, then the MPEG Alliance got greedy with their next version. H.265 upped the prices on its license agreement, and additionally demanded a cut of all streaming revenue. It got worse--the alliance fragment, and so you had to pay multiple consortia the royalties for the codec (although only one of them had the per-video demand).
It was in response to this greed that the Alliance for Open Media was created, which brought us AV1. I don't know how important Google is to the AOM, but I will note that, at launch, it did contain everybody important to the web video space except for Apple (which, as noted earlier, is the entity that previously torpedoed the attempt to mandate royalty-free codecs for web video).
The finer point is where these tens of billions came from.
All of it was ad money, and a lot of these innovations were also targeted at better dealing with ads (Flash died because of how taxing it was, mobile browsers just couldn't do it. JavaScript perf allowed these ads to come back full force)
The net balance of how much web technology advanced vs how much ad ecosystems developed is pretty near 0 to me, if not slightly negative.
Isn't webrtc broken in Chrome? Or did they finally fix that? It used to be that everyone supported Chrome's broken implementation, leaving Firefox users with the correct implementation out in the cold.
If you are referring to the standards-based "Unified Plan" vs. the Google proprietary "Plan B" for handling multiple media tracks in SDP, I believe that "Plan B" was finally phased out in 2022.
No, there isn't a need for appreciation. We all cheered at that time where Google was building a great JavaScript engine and a browser around that. But in hindsight it is clear, that Google was just running the old embrace, expand, extinguish playbook on a scale that we where unable to comprehend. We would've be just fine with Firefox, webkit and maybe Microsoft would have made Internet explorer somehow not total shit. Google captured the whole web as a market and we used the opportunity to build endless JS frameworks in top and went wild with all the VC and advertising money.
> Webrtc. Google’s implementation is super widely used in all sorts of communications software.
Webrtc uses the user's bandwidth without permission or notification and it used to prevent system sleep on macs without any user visible indication.
> V8. Lots of innovation on the interpreter and JIT has made JS pretty fast and is reused in lots of other software like nodejs, electron etc.
No matter how efficient they made it, javascript "applications" are still bloatware that needlessly waste the user's resources compared to native code.
> Sandboxing. Chrome did a lot of new things here like site isolation and Firefox took a while to catch up.
That's useful but only because the bloatware above. If you didn't give code running in the browser that much power you wouldn't need sandboxing.
> Codecs. VP8/9 and AV1 broke the mpeg alliance monopoly and made non patented state of the art video compression possible.
Could agree. Not sure of Google's real contribution to those.
> SPDY/QUIC. Thanks to Google we have zero RTT TLS handshakes and no head of line blocking HTTP with header compression, etc now and H3 has mandatory encryption.
It's also a binary protocol that cannot be debugged/tested via plain telnet, which places a barrier to entry for development. Perhaps enhances Google's market domination by requiring their libraries and via their control of the standard.
> > Codecs. VP8/9 and AV1 broke the mpeg alliance monopoly and made non patented state of the art video compression possible.
> Could agree. Not sure of Google's real contribution to those.
They were not the only contributor (I was the technical lead for Mozilla's efforts in this space), but they were by far the largest contributor, in both dollars and engineering hours.
> No matter how efficient they made it, javascript "applications" are still bloatware that needlessly waste the user's resources compared to native code.
Well that's just biased. Saying application is bloated (which is often not true) is the result of an entire ecosystem, has something to do with an interpreter, is ridiculous. Any qualified software engineer can see the fault in such a comment. You probably know that as well.
Is have to agree to be honest. Whoever decided to run JavaScript in the backend should be committed to a mental institution. JavaScript is a nightmare. But you can't tell a man something his paycheck depends on him not knowing.
>Webrtc uses the user's bandwidth without permission or notification and it used to prevent system sleep on macs without any user visible indication.
>No matter how efficient they made it, javascript "applications" are still bloatware that needlessly waste the user's resources compared to native code.
>No matter how efficient they made it, javascript "applications" are still bloatware that needlessly waste the user's resources compared to native code.
So should we not deliver advanced sandboxed cross platform applications for any platform, and instead deliver unsandboxed native code for all possible platforms? ActiveX called, it wants to say thanks for the endorsement and that it told you so.
And no more zoom meetings because somebody's Mac might not go to sleep? I'm with you on that one, brother!
>There actions back then fitted the Don‘t be evil motto.
Disagree with that. All the privacy issues people have problem with now were already a problem in 2007. But being the media darling along with Submarine PR Google didn't get much bad press.
There were lots of other things too, including their site breaking Firefox as well as Chrome, their promise not to make another browser.
Of all the things you've mentioned, the only one that genuinely stands out to me as a positive contribution from Google—something that wouldn’t have happened had Chrome never existed—is the codec situation.
They leveraged their scale and influence for good in that instance.
That said, it’s not as if other browsers weren’t already making independent strides in optimisation and innovation.
In fact, I sometimes wonder whether Chrome has actually steered the browser ecosystem in the wrong direction, while simultaneously eroding a lot of the diversity that once existed.
> That said, it’s not as if other browsers weren’t already making independent strides in optimisation and innovation
Honestly I can't believe that anyone who was around when Chrome came out would say this. IE7 was around, and terrible. Firefox was trying hard, as was Opera, but web tech has become infinitely better with Chrome around, and Google funding it. Without Google funding Firefox as well, Firefox would be nowhere near what it is today.
Aside: it’s impressive how the whole blog post does not mention a single detail of what they actually did to achieve these performance improvements. Code changes, really?
The point of deterrence is that you spend the money so the chance of Russia doing something like a land invasion of Europe is decreased.
The much more likely scenario that Europeans are wanting to prevent is a limited invasion in the Baltics or Balkans in order to politically divide and damage western democracies.
You're being overly dramatic. Germany has been spending low amounts of money on its military for two and a half decades. Circumstances changed and spending needs to be increased to adjust to the new situation. This is not militarization, it's the state fulfilling fundamental purpose of providing external security for its citizens. Spending is going to be less (relative to GDP) than in the 70s and 80s.
The agenda after WW2 was to not allow germany to militarize itself again and to keep the soviets out. Now it seems that germany is allowed to take it's safety in it's own hands again. Maybe there is more trust (western values?), maybe americans want to pull out? I still find it an astounding development. Maybe because it is an european project. Becoming adult. And the US has a bigger fish to fry.
There is definitely more trust toward Germany then toward USA or Russia now. Wither way, even if there was less trust, it does nit matter. Russia is very active threat and America is hostile toward Europe, threatens annexation of part of it ... while being friendly to Russia and other authoritarian regimes.
So, it is nor like there would be another choice.
German is at least teaching about own atrocities as about atrocities. Other countries have massively bigger problem to admit their past might have been ugly.
The historical german inability to correctly read geopolitical currents, anticipate their domestic dynamics, or understand international perceptions of Germany seems to be intact. It's true that The Atlantic Council and other US cultural-state entities aren't making it easier, but this is egregious.
No one looks to Germany for leadership on rule of law, or human rights, least of all Germans.
I did not said Germans are perfect. They are massively better then USA, they are overall trustworthy enough to trust them.
Your last link is about AfD being supported by Elon Musk. And second about other parties not wanting to cooperate with AfD - and while fascism is up everywhere, in America there is basically no functional opposition.
Sad to meet someone who doesn't care that they're going down the same river and over the same waterfall so long as someone else is going a little faster.
I know exactly what's in the links, what I don't know is why you believe it's logical to assume that every other human being must be a cheerleader for one group of corrupt authoritarians or another. Supporting either government as such is shameful and your descendants will be as ashamed of you as you are of yours.
I believe it's fine now because there's been another, better, solution that doesn't rely on spending so many $$$ on it; the EU. USA is probably not at all as worried about Germany as a country going crazy again.
Ironically, Hitler wanted to establish a third Reich - basically a new Roman empire that spanned all of... Europe. Funny how these things goes. Old wine in new bottles.
The API looks nice on the surface, but this will be expensive to operate due to Ruby and Rails’ lack of support for concurrency. Everything is blocking, which is not a great match for the async nature of interacting with these models.
Well instead of looking at the two first hits of Google I spent several years on a platform team of a multi billion dollar company using mostly Rails and worked on solving real world problems caused by Ruby/Rails’ design choices which lead me to believe that Ruby concurrency as of today is hot garbage.
They need fundamental breaking changes to the language to fix this, which means people won’t be able to use their beloved pile of 438 gems that haven’t seen a commit in 7 years. If I had to bet, I’d say the language is dead. It might still be a nice niche language for easy prototyping, but the world has moved on to async/await (js/python/Rust/C++) or stackful coroutines (Go).
Unfortunately, 5 years after the release you linked, almost none of this has made it to Rails or even to relatively new libraries this post is about. The reason (imho) are unfortunate design choices in how the language incorporates concurrency - it’s just not well done.
This will synchronously block until ‘chat.ask’ returns though. Be prepared to be paying for the memory of your whole app tens/low hundreds of MB of memory being held alive doing nothing (other than handling new chunks) until whatever streaming API this is using under the hood is finished streaming.
Yes, it does. Ruby has a global interpreter lock (GIL) that prevents multiple threads to be executed by the interpreter at the same time, so Puma does have threads, they just can’t run Ruby code at the same time. They can hide IO though.
Concurrency support is missing from the language syntax and this particular library as a concept. This is by design, to not distract from beautiful code. Your request will make zero progress and take up memory while waiting for the LLM answer. Other threads might make progress on other requests, but in real world deployments this will be a handful (<10). This server will get 10s of requests per second when something written in JS or Go will get many 1000s.
It’s amazing how the Ruby community argues against their own docs and doesn’t acknowledge the design choices their language creators have made.
I don’t know why “number of companies using language X” is a metric that is used here. Wordpress is serving 43% of websites on the internet as of 2025, so we should all be learning PHP!
Webrtc. Google’s implementation is super widely used in all sorts of communications software.
V8. Lots of innovation on the interpreter and JIT has made JS pretty fast and is reused in lots of other software like nodejs, electron etc.
Sandboxing. Chrome did a lot of new things here like site isolation and Firefox took a while to catch up.
Codecs. VP8/9 and AV1 broke the mpeg alliance monopoly and made non patented state of the art video compression possible.
SPDY/QUIC. Thanks to Google we have zero RTT TLS handshakes and no head of line blocking HTTP with header compression, etc now and H3 has mandatory encryption.
reply