Simon Willison’s Weblog

Subscribe

19 items tagged “chrome”

2023

Web LLM runs the vicuna-7b Large Language Model entirely in your browser, and it’s very impressive

A month ago I asked Could you train a ChatGPT-beating model for $85,000 and run it in a browser?. $85,000 was a hypothetical training cost for LLaMA 7B plus Stanford Alpaca. “Run it in a browser” was based on the fact that Web Stable Diffusion runs a 1.9GB Stable Diffusion model in a browser, so maybe it’s not such a big leap to run a small Large Language Model there as well.

[... 2276 words]

2022

Data-driven performance optimization with Rust and Miri (via) Useful guide to some Rust performance optimization tools. Miri can be used to dump out a detailed JSON profile of a program which can then be opened and explored using the Chrome browser’s performance tool. # 9th December 2022, 5:19 pm

Paint Holding—reducing the flash of white on same-origin navigations. I missed this when it happened back in 2019: Chrome (and apparently Safari too—not sure about Firefox) implemented a feature where rather than showing a blank screen in between page navigations Chrome “waits briefly before starting to paint, especially if the page is fast enough”. As a result, fast loading multi-page applications become almost indistinguishable from SPAs (single-page apps). It’s a really neat feature, and now that I know how it works I realize that it explains why page navigations have felt a lot snappier to me over the past few years. # 22nd May 2022, 2:50 am

2021

Beginning in M94, Chrome will offer HTTPS-First Mode, which will attempt to upgrade all page loads to HTTPS and display a full-page warning before loading sites that don’t support it. Users who enable this mode gain confidence that Chrome is connecting them to sites over HTTPS whenever possible, and that they will see a warning before connecting to sites over HTTP. Based on ecosystem feedback, we’ll explore making HTTPS-First mode the default for all users in the future.

Chromium Blog # 14th July 2021, 7:14 pm

2020

Intent to Remove: HTTP/2 and gQUIC server push (via) The Chrome / Blink team announce their intent to remove HTTP/2 server push support, where servers can start pushing an asset to a client before it has been requested. It’s been in browsers for over five years now and adoption is terrible. “Over the past 28 days [...] 99.97% of connections never received a pushed stream that got matched with a request [...] These numbers are exactly the same as in June 2019”. Datasette serves redirects with Link: preload headers that cause smart proxies (like Cloudflare) to push the redirected page to the client along with the redirect, but I don’t exepect to miss that optimization if it quietly stops working. # 12th November 2020, 1:44 am

html-to-svg (via) ‪This is absolutely ingenious: 50 lines of JavaScript which uses Puppeteer to get headless Chrome to grab a PDF screenshot of a page, then shells out to Inkscape to convert the PDF to SVG. Wraps the whole thing up in a Docker container and ships it to Cloud Run as a web service you can call by passing it a URL. # 7th May 2020, 6:01 am

2020 Web Milestones (via) A lot of stuff is happening in 2020! Mike Sherov rounds it up—highlights include the release of Chromium Edge (Microsoft’s Chrome-powered browser for Windows 7+), Web Components supported in every major browser, Deno 1.x, SameSite Cookies turned on by default (which should dramatically reduce CSRF exposure) and Python 2 and Flash EOLs. # 24th January 2020, 4:43 am

Come version 80, any cookie without a SameSite attribute will be treated as “Lax” by Chrome. This is really important to understand because put simply, it’ll very likely break a bunch of stuff. [...] The fix is easy, all it needs is for everyone responsible for maintaining any system that uses cookies that might be passed from an external origin to understand what’s going on. Can’t be that hard, right? Hello? Oh...

Troy Hunt # 3rd January 2020, 4:22 pm

2013

What data structures are used to implement the DOM tree?

You may enjoy this post from Hixie back in 2002 which illustrates how different browsers deal with incorrectly nested HTML. IE6 used to create a tree that wasn’t actually a tree! http://ln.hixie.ch/?start=103791...

[... 49 words]

2012

Why doesn’t Google Chrome use the Google Search Engine spellchecker?

Because the Google search spellchecker runs in their data centers, while the Chrome spellchecker runs on your local machine. Sending everything you type in to your browser to Google’s servers just to check spelling would be an unnecessary privacy violation, especially since local spell checking does a good-enough job.

[... 65 words]

2009

Web Sockets in Tornado. Bret Taylor has a simple class making it trivial to experiment with the Web Sockets protocol (now shipping in Chrome) using the scalable Tornado application server. He also raises the million dollar question: what will existing load balancers and proxies make of the new protocol? # 31st December 2009, 11:54 am

A piece with a lot of screenshots about the close tab behaviour in Google Chrome. If you click “close” with your mouse, Chrome doesn’t resize the remaining tabs until you mouse away from the area. This means you can click “close” multiple times without having to chase the close button. I hadn’t noticed this, partly because Chrome doesn’t do it if you hit Command-W. They even switch the position of the close button in RTL languages such as Arabic. # 11th December 2009, 9:19 am

Chromium OS User Experience. The 2 minute UI concept video is probably the best way to understand the ideas behind Google’s Chrome OS. # 19th November 2009, 10:12 pm

Given the security issues with plugins in general and Google Chrome in particular, Google Chrome Frame running as a plugin has doubled the attach area for malware and malicious scripts. This is not a risk we would recommend our friends and families take.

Microsoft spokesperson # 24th September 2009, 4:49 pm

More technical details about Google Chrome Frame. It’s implemented as a Browser Helper Object, uses IE’s cookies, history and password-remembering, includes the WebKit developer tools and appends “chromeframe” to the regular IE user agent string—though not apparently the Chrome Frame version itself. # 23rd September 2009, 10:20 pm

Ask browser users, and they’ll tell you the overwhelming reason why they can’t upgrade to a more modern, standards-compliant browser is because their work won’t let them. Ask IT departments why this is the case and they’ll point to the six- to seven-figure costs of upgrading turn-of-the-century Intranets written to work in, and only in, Internet Explorer 6. Google have provided a way for websites to opt out of IE6 (and even IE7) support without requiring enterprise-wide, Intranet-breaking browser upgrades.

Charles Miller # 23rd September 2009, 3:08 pm

In the past, the Google Wave team has spent countless hours solely on improving the experience of running Google Wave in Internet Explorer. We could continue in this fashion, but using Google Chrome Frame instead lets us invest all that engineering time in more features for all our users, without leaving Internet Explorer users behind.

Lars Rasmussen and Adam Schuck # 23rd September 2009, 9:59 am

Introducing Google Chrome Frame. Here’s what Alex Russell has been up to at Google: An IE plugin (for 6, 7 and 8 on all Windows versions) which embeds the Google Chrome rendering engine—sites can then opt-in to using it by including a X-UA-Compatible meta tag. Seems to be aimed at corporate networks which mandate IE for badly written intranet applications—they can roll this out without retraining users to use another browser or breaking their existing in house apps. # 23rd September 2009, 9:57 am

2008

The story behind Google Chrome. Superbly researched by Niall Kennedy—a detailed overview of the staff and acquisitions that went in to Google Chrome. # 4th September 2008, 1:50 am