Live data from Hacker News

Tab Unloading in Firefox 93

hacks.mozilla.org

61–70 of 283 posts

Re: Tab Unloading in Firefox 93

#61
post #36
post #22

What I want implemented.. Chrome/Firefox whatever... is that instead of unloading and reloading when I go back to the tab I want it to export basically an HTML file and when I click back to that tab, I want it to reload the HTML file instead with some notification/marker that it is a "cached" version. There are times where I'll leave a tab in the background that I don't want refreshed but when it gets unloaded, poof,…

This! I feel like Back used to get instant results all the time, and now it's almost always a big delay, which is a huge pain -- plus, it loses your position on infiniscroll and other dark patterns. Part of it is caused by site coding, but part of it is browsers treating "back" as some species of new page load/arrival/selection. And I know it's possible to get us back to the old days! The functionality I want is almo…

Back basically works like it always did on static sites. Blame the fashion for single-page apps and endless scrolling for it being broken.

Re: Tab Unloading in Firefox 93

#62
post #13

I've never had Firefox crash. I guess they are reaching for those edge cases on power users who keep a billion tabs open. Maybe I'm missing some sort of productivity hack that involves leaving 100+ tabs open.

I'm going to reply with a comment stolen from a different HN discussion: The fact that people (myself included) have 100s of tabs open is a symptom of just how broken bookmarking is on the modern web. Bookmarks are a miserable way to create a placeholder for your thoughts. It's too static, must be maintained, and generally prone to link rot. I'm a scientist. I open and read 10-20-100 links for a given topic depending…

Not going to expand on your point too much, but I've commented similarly in the past, and I heavily agree with you. There's so much room for innovation in browser UX, Bookmarks and tab archival never really evolved and it feels like we're just stuck in the past.

Part of the problem is that it's tricky for people to innovate on bookmarks in an accessible way, because I'm not going to do heavy browsing in a 3rd-party Chromium fork that doesn't have Firefox's privacy tools or support for Ublock Origin.

But more experimentation with bookmarking systems would be very welcome. In the meantime, here I sit with my 977 open tabs.

Re: Tab Unloading in Firefox 93

#63
post #18
post #13

I've never had Firefox crash. I guess they are reaching for those edge cases on power users who keep a billion tabs open. Maybe I'm missing some sort of productivity hack that involves leaving 100+ tabs open.

Firefox runs on systems with 512MB RAM. [1] Can you imagine keeping a Slack tab open on a system like that? Never mind 100 tabs. [1] https://www.mozilla.org/en-US/firefox/93.0/system-requiremen...

To be fair, I tried running it recently on a machine that I had artificially limited to one core and 512MB RAM; it was not actually usable, even for static pages. WebKit browsers didn't really do any better, though, which surprised me.

Re: Tab Unloading in Firefox 93

#65

Earlier quoted context omitted.

I'm going to reply with a comment stolen from a different HN discussion: The fact that people (myself included) have 100s of tabs open is a symptom of just how broken bookmarking is on the modern web. Bookmarks are a miserable way to create a placeholder for your thoughts. It's too static, must be maintained, and generally prone to link rot. I'm a scientist. I open and read 10-20-100 links for a given topic depending…

Bookmarking systems are completely outdated, but I'm not sure what they should be replaced with. What would help you in this case so you didn't have to worry about losing the tabs if you closed them?

Maybe a sufficiently good implementation of history search could replace the need for bookmarks entirely. Instead of just matching on page titles, match on content as well. Rank results according to relevance and time spent on the page.

Or alternately, maybe bookmarks could become an extension of existing tab management features. Something like Chrome's tab groups or the Tree Style Tabs[1] extension for Firefox, but with the ability to "archive" tabs or groups of tabs into a searchable repository for later reference.

Just some ideas...

[1]: https://addons.mozilla.org/en-US/firefox/addon/tree-style-ta...

Re: Tab Unloading in Firefox 93

#66
post #50
post #22

What I want implemented.. Chrome/Firefox whatever... is that instead of unloading and reloading when I go back to the tab I want it to export basically an HTML file and when I click back to that tab, I want it to reload the HTML file instead with some notification/marker that it is a "cached" version. There are times where I'll leave a tab in the background that I don't want refreshed but when it gets unloaded, poof,…

It'd have to be more than just the HTML, it'd need to be everything the page used. Basically, all the data you'd get by clicking the "download HAR" button in the DevTools Network panel. Including some special sauce on top to replay any XHRs (assuming you want to get to the last state by replaying... otherwise the JS page state would need to serialized some how)

Couldn't the browser just dump the current DOM and JS state onto a disk? Treat it like a computer hibernating.

Re: Tab Unloading in Firefox 93

#67
post #36
post #22

What I want implemented.. Chrome/Firefox whatever... is that instead of unloading and reloading when I go back to the tab I want it to export basically an HTML file and when I click back to that tab, I want it to reload the HTML file instead with some notification/marker that it is a "cached" version. There are times where I'll leave a tab in the background that I don't want refreshed but when it gets unloaded, poof,…

This! I feel like Back used to get instant results all the time, and now it's almost always a big delay, which is a huge pain -- plus, it loses your position on infiniscroll and other dark patterns. Part of it is caused by site coding, but part of it is browsers treating "back" as some species of new page load/arrival/selection. And I know it's possible to get us back to the old days! The functionality I want is almo…

AFAIK whether page will be restored or reloaded depends on its cache headers. I remember breaking back button in one of my projects, because users asked for it, as they expected fresh data for some reason. It was not SPA, just old server rendered website, so I disabled caching via response headers and it worked.

Re: Tab Unloading in Firefox 93

#68
> Firefox will respond by unloading memory-heavy but not actively used tabs.

A dropped tab cannot always be reloaded to the same content. No thanks, I'd rather get more RAM or swap space for paging.

The tabbing feature of "close all pages to the right of this one" is almost good enough for me.

What I would like is some UI where all the tabs are compactly shown as rows in a table. Here, you could select tabs and move them up or down, delete them, or farm them off to their own window.

Re: Tab Unloading in Firefox 93

#69
post #13

I've never had Firefox crash. I guess they are reaching for those edge cases on power users who keep a billion tabs open. Maybe I'm missing some sort of productivity hack that involves leaving 100+ tabs open.

How is that possible, I get FF crashes literally every time I load a moderately intensive website (eg. Google Keep). I still use FF because I can't abide Chrome. My PC is admittedly a bit older (only 6 gigs RAM) but still.

6GB is ample memory if you're not loading tons of tabs. I run Firefox on Raspberry Pi model 1s and while you have to limit yourself to a tab or two and have plenty of patience it does still work.

What doesn't work is trying to compile the browser on the Pi. The build process has some really gigantic source files that cause the compiler to choke.

Re: Tab Unloading in Firefox 93

#70
post #22

What I want implemented.. Chrome/Firefox whatever... is that instead of unloading and reloading when I go back to the tab I want it to export basically an HTML file and when I click back to that tab, I want it to reload the HTML file instead with some notification/marker that it is a "cached" version. There are times where I'll leave a tab in the background that I don't want refreshed but when it gets unloaded, poof,…

I can’t speak for Gecko or Blink/Chromium, but I don’t think this would be difficult to implement in a WebKit browser as long as some limits are involved.

Just save the tab’s contents in a webarchive file, then when restoring load that webarchive and inject some JavaScript that makes any kind of interaction aside from scrolling load the page’s live counterpart (with a confirmation prompt, perhaps).

This would also come with the advantage of the cached version showing up in browser history for “free”, so if you inadvertently go to the live page undoing it is as easy as hitting the back button, and if you decide you want to preserve the cached state indefinitely all the browser has to do is copy the webarchive file out of temporary storage to wherever the user indicates.

Post reply on HN