Why am I bitter?
1. It hands a lot of power over to chromium, regardless of how much it gets forked by Microsoft.
2. It would have put some very talented engineers onto a Chrome competitor, which I always like to see
3. It would have been a huge boost to Rust, which if it ever will unseat c++/c as a systems language target, will need all the help it can get as a de facto standard (you can argue if this is worthwhile or just pie in the sky dreaming, I'll concede on that)
4. I just plain think we need a V8 alternative for a nodejs/electron like ecosystem, in my opinion servo could be that. I know they are doing node-chakra-core, but this could have taken it to a whole new level.
Competition is good, in another words. Even if all my assumptions are wrong, I'd love to see an alternative history where they were laid to bare.
Now, I will admit, I am actually happy to see that they are on Chromimum instead of yet another engine, and I see all the upsides of this move too, and will definitely take advantage of them (with Microsoft Edge being my favorite version of Chrome right now)
Microsoft could have written any browser engine it wanted and then placed iOS/Safari like rules on other browsers, such as... "if you want us to allow your app to run in Battery Friendly mode, you must ONLY use our OS provided web engine."
https://msrc-blog.microsoft.com/2019/07/22/why-rust-for-safe...
I guess Microsoft was tired of losing the browser game, so it decided to back the winning worse this time. But I still believe it will be a mistake long-term, as Google will start controlling Chromium's direction with more of an iron fist.
I believe someone else mentioned in a previous HN comment about the ad-blocking API that it doesn't matter if Edge, Brave, or other Chromium-based browsers can "continue to support a Chromium API if Google wants to kill it", because in the end if their combined market share is tiny, most developers will stop supporting that API anyways.
I imagine the Chromium browsers will be going in the direction of Android, where "anyone can fork Android" but in practice you're dead in the market if you don't back Google's Android.
But the major improvement that made everyone switch was process separation for each tab. One tab could crash while the others stayed functional and a security vulnerability in one tab would have a more difficult time compromising the others.
It's also one of the reasons Chome uses so much memory, they failed to keep optimizing their copy-on-write paging behaviour so that tabs would share more memory among themselves.
Nowadays Firefox has process separation for tabs AND it's not written by an advertising company, so I like it better.
Which IE8 had around the same time (It may have launched after Chrome, but in the contemporaneous betas, they both had the feature)
What made people switch was a combination of everything google did having the sheen of "cool" around it, and the fact that I.E development was slow and shoddy.
I think that's when Google noticed that the IE monopoly was breaking, and they jumped onboard and took Konqueror's code and then launched their own browser. The same code that Apple used for Safari. So they basically came in on the hard work of Firefox and the KHTML team and took over the market by putting Chrome right on Google's homepage.
It's my opinion that Chrome didn't become popular because it was better, or because it was cool (no average user cares about that). It became popular by being right there on everybody's home page, right at the perfect time that people were catching on to the fact that IE had zero security.
Firefox broke the monopoly and freed us, and then Google and Apple came in and shat all over them.
The first time I saw break away tabs in Safari for Windows in 2007. Before Chrome was introduced.
Safari 3
https://lifehacker.com/safari-3-beta-now-available-for-windo...
Safari 5 http://cdn.osxdaily.com/wp-content/uploads/2017/08/safari-fo...
The safari 5 screenshot you have is EDGE with a safari installer in front of it.
In addition I also consider the marketing reach and the success of Android as important. Fwiw I never switched completely to chrome and quite regularly got a hint to try out a better/optimised browser.
Process sep per tab was a Chrome feature. Unless you count Lynx and separate instances per browser page ;-)
This would presumably be news to everybody involved.
Google's QUIC ("GQUIC") isn't very close to the IETF QUIC standard under development. The IETF went with TLS 1.3 as the cryptographic foundation (which didn't exist when Google built GQUIC) and most the details have changed considerably. You can definitely see a family resemblance, but for Google this was a proof of concept hack and for the QUIC WG they're building what is probably a successor to TCP. It's the difference between a five year old's doodle of a house and the plans you need to submit to get someone to approve it and then build the house.
The chairs are currently not Google employees (they work at Fastly and NetApp) and the editors on the main transport draft aren't Google either (Mozilla and Fastly). There are Google people in the room (such as Adam Langley of course) but there are people from the other browsers too.
But fair enough, to say they wrote it all is a mischaracterization of events.
Respectively. Meaning private=safari, tabs=opera.
https://www.quora.com/What-is-the-history-of-tabbed-browsing
Off-topic, but I don't think its about being cocky. For me, the most difficult thing to keep in mind about information technology is that it is entirely a construct of the human mind and we haven't yet devised methods to validate it against reality.
We see things as they are now and think "well, of course it has to be that way." That kind of view is true in a lot of fields--a rocket ship has a certain design because it must have a certain design. You can't put the engines on perpendicular to the ship or the fuel storage under the rocket exhaust.
But you can do that kind of thing with computers and might not even know it. I started to get a sense of this a couple years ago, when I first learned about the works of Ted Nelson and his contemporaries. Ted is critical of things like the modern hyperlink which, at first, seemed to me like criticizing the ocean. He even disapproves of the word "computer".
The idea that computing can and should be fundamentally reinvented has been the most exciting discovery of my life.
This was one of the more depressing discoveries of mine.
"Reinvention" works by the main stream of this industry having knowledge half-life on the order of decade or less. Instead of building on top of prior work, we keep reinventing the old stuff; sometimes better, but usually very poorly - I attribute this to bad incentives. In the 70s they had time to think about what they're actually doing. In 2010s we have constant race to the market, and "move fast and break things" culture, which leads to layers of crappy abstractions that keeps relative performance of our software on the level of 1980s while delivering little usability on top.
According to this research article [1], QUIC is unfair towards TCP and consumes more than twice its fair share of bottleneck bandwidth.
[1] [PDF] https://conferences.sigcomm.org/imc/2017/papers/imc17-final3...
Just for reference
https://en.wikipedia.org/wiki/Private_browsing
It was available for 3 years in Safari before it became available in Chrome and four years before Firefox.
i think the author is a bit confused about this:
1. every email provider in existence scans emails for spam purposes. if they don't, they lose market share since their product becomes unusable.
2. gmail stopped scanning emails for ad purposes a while ago.
Unlike how a normal service is normally supposed to store my password (hashed, but with no copy of the original password anywhere, EVENT THE LOGS) the password manager DOES retain a plain text copy of a password, not JUST a hashed derivative.
Maybe don't use 1FA for your passwords?
https://support.mozilla.org/en-US/kb/use-master-password-pro...
That said, can you still copy the profile to a new computer and have it autofill? Physical access to data on a hard drive is often end of game, unlike cloud only hosted stuff that streams. Im guessing you cant copy a firefox profile with lastpass or 1pass or dashlane and have the new computer start autofilling without reentering the master password?
Firefox also has Firefox Lockbox now for storing passwords in the cloud.
I've been using it without any issues for several years now.
Now chrome is getting credit for in private browsing and tabs? (Safari and Opera respectively. Phoenix/Firefox popularized them long before Chrome was a twinkle.)
>It is mind-boggling that basic tools like these are not yet standard features in browsers.
How long have browsers offered to save passwords? Seems like a lot of bad history in this article.
>The way existing password managers work, the vendor’s liability even if they are hacked is minimal. They don’t hold any passwords!
HUH? Yes they do.. They absolutely hold a copy of your password ... in the cloud.
>Sure, Chromium and the V8 engine are open-source, but they still belong to Google in a way that, say, TCP/IP or IMAP does not.
Im gonna feel really bad breaking the QUIC / HTTP/3 news. Cuz google totally wrote the new transport layer. They wrote the last one too (SPDY HTTP/2.) With this much market share, what google implements and forces the hands of the standards organizations.
>This is why Microsoft’s decision to kill Internet Explorer, once the leading portal to the web, and replace it with a Chromium-based successor says so much about its new direction as a company. I suspect that decision had more to do with clearly understanding Microsoft’s strengths and weaknesses than anything else.
Its primarily for two reasons. 1 is battery life, they want to embed blink into the os so programs dont need to bundle it (electron included.) 2 is because 99% of the web just targets blink now. I dont think you can credit "the move to ssas" alone with Microsoft finally abandoning IE derivatives.