Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Lightspeed – A Browser Experiment

blog.mozilla.org

Post reply on HN
Print view View original

121–130 of 184 posts

Re: Lightspeed – A Browser Experiment

#121

Post by lemieux »

ChikkaChiChi wrote: post #116

Not a bad start. This can be pared down even further: * Adjust search results based on your own clickthrough rates. If you bounce out of a site a few times (hello, quora) then it grays out or falls off. Sites you go to and stay on become more prominent. * Kill downloads. I know this won't be popular, but the ability to download a thing isn't necessary for everyday use. There would be a heightened awareness for the us…

> No cross-site loading of any kind; all source material MUST come from the domain you are on. This would seriously break some sites but it would close large gaps in security and tracking.

This would probably break almost everything. Who doesn't use some sort of a CDN nowadays?

Re: Lightspeed – A Browser Experiment

#122

Post by asadotzler »

gsnedders wrote: post #102
bane wrote: post #9

At first I thought "oh no, another distraction for Mozilla", but then I sat through the presentation and thought there was some really cool ideas (amazebar!). In fact, I think it can be simplified and paired down even more. Even better, some of the assumptions here even make the software portable (as in doesn't really need to be installed and could just run off a usb drive/out of my dropbox). Automatically being in a…

> - get rid of loved sites/bookmarks altogether. Frequent usage should just percolate autocomplete suggestions to the top. This design doesn't really need them, and tbh I've barely used bookmarks in any browser in a long time. Opera tried getting rid of bookmarks with Opium, and very much was forced to go back after the initial release.

Why did they fail? What forced them to go back? What could have been done differently in the implementation to have resulted in a different outcome.

Re: Lightspeed – A Browser Experiment

#123

Post by asadotzler »

Pacabel wrote: post #86
Flimm wrote: post #62

Earlier quoted context omitted.

If you're connected to the Internet you have to update your browser all the time just to stay minimally secure.

Are you suggesting that the updates themselves never introduce new security flaws?

I don't see any such suggestion. The fact is that updates fix many known flaws -- flaws that are or will be exploited after an update is released and the details of the flaws are fully disclosed (to the world, including the bad guys) so not taking an update is the same as being zero day'd by the software providing the update. It's not a good idea to avoid taking fixes for known and already or soon to be exploited flaws because the update may or may not have introduced new as yet unknown flaws.

Re: Lightspeed – A Browser Experiment

#124

Post by Zikes »

lemieux wrote: post #121
ChikkaChiChi wrote: post #116

Not a bad start. This can be pared down even further: * Adjust search results based on your own clickthrough rates. If you bounce out of a site a few times (hello, quora) then it grays out or falls off. Sites you go to and stay on become more prominent. * Kill downloads. I know this won't be popular, but the ability to download a thing isn't necessary for everyday use. There would be a heightened awareness for the us…

> No cross-site loading of any kind; all source material MUST come from the domain you are on. This would seriously break some sites but it would close large gaps in security and tracking. This would probably break almost everything. Who doesn't use some sort of a CDN nowadays?

A better alternative would be to promote the use of Content-Security-Policy, perhaps by requiring it be used in order to load source material from an alternate domain.

Re: Lightspeed – A Browser Experiment

#125

Post by asadotzler »

yarou wrote: post #33

The design choices really look interesting. But shouldn't Mozilla focus more on eliminating the ever persistent memory leak bug that has plagued it since forever? I can't even use Firefox on a netbook. I end up installing Midori.

Bug number? Firefox memory usage beats all the other browsers by considerable margins in my usage and has for at least a couple of years now.

Re: Lightspeed – A Browser Experiment

#126

Post by hackuser »

nemothekid wrote: post #24
cypher543 wrote: post #16

> Auto-updates are always on I really hope this isn't going to become a standard thing. There are still plenty of people like me who are stuck with something like satellite internet, which forces us to micromanage our data usage. I can't afford to just have things auto-updating whenever they want.

It was the default in Chrome, and I believe its the default in IE as well. It has its upsides as well, you generally only need to target the latest stable version of Chrome.

> It has its upsides as well, you generally only need to target the latest stable version of Chrome.

That is an upside for developers; for users, there is an additional cost to using your website (i.e., updating Chrome, which they do seem to make as easy as possible).

Re: Lightspeed – A Browser Experiment

#127

Post by Zikes »

xj9 wrote: post #120

I think Mozilla should focus on making Firefox fast and easy on memory usage before they start trying to "innovate" in the UI space, even if it's just conceptual.

This is assuming that everyone at Mozilla is capable of focusing on speed/memory. I'm betting there are several dedicated UI developers and designers who would not be able to meaningfully contribute to the core browser, and conceptualizing things like Lightspeed is their primary job role.

Re: Lightspeed – A Browser Experiment

#128

Post by hackuser »

gioele wrote: post #28
cypher543 wrote: post #16

> Auto-updates are always on I really hope this isn't going to become a standard thing. There are still plenty of people like me who are stuck with something like satellite internet, which forces us to micromanage our data usage. I can't afford to just have things auto-updating whenever they want.

> I really hope this isn't going to become a standard thing. I think people are starting to realize that you cannot always live on the bleeding edge, even then the edge is not that sharp. IIRC Firefox ESRs (extended support releases) do not update automatically. Or they update automatically but only to other infrequent ESRs.

> IIRC Firefox ESRs (extended support releases) do not update automatically. Or they update automatically but only to other infrequent ESRs.

I'm pretty sure you can disable automatic updates in any version of Firefox.

In the ESR editions, you certainly can disable them. It's an essential feature -- imagine being the manager of 20,000 desktops when they all auto-update at once, consuming all the bandwidth, simultaneously stopping the work of every employee in the enterprise, creating hundreds or more support calls, and breaking the corporate intranet.

And yes, ESR editions update only to other ESR editions.

Re: Lightspeed – A Browser Experiment

#129

Post by hackuser »

izzydata wrote: post #108

The one thing I find very weird is how this guy designing this and other people are saying "Well, I can't use this for work of course because I have hundreds of tabs open." Wouldn't make so much sense to try and solve that problem in particular? It seems like everyone does this, but it seems worthwhile to ask why everyone does this. Do you have a tab open because you don't want to lose unsaved data? Do you want a tab…

> I believe it can be integrated into a much simpler navigation than looking at hundreds 16 pixel icons squeezed side by side horizontally.

Try one of the vertical tab bar extensions (i.e., the tabs are stacked on top of each other in a sidebar) such as Tree Style Tabs. It displays many more tabs (mine displays ~40 before it overflows) and you always can see their titles.

Re: Lightspeed – A Browser Experiment

#130

Post by shurcooL »

A very specific thought on this:

I love the fact that there are no settings because of one advantage it offers (it has disadvantages too, like any trade-off).

Namely, that you can always be sure of what settings you've changed. The answer will always be none, since there are no settings.

Removing settings altogether is not the only way to achieve that; for example see how Sublime Text does its settings, you have a very easy to read explicit list of all settings you've changed. So resetting to default or undoing a bad chance is as easy as removing unwanted entries from that Preferences.sublime-settings file.

Having less choices is nice for when you don't want to change anything, since you know your settings are optimal defaults and you don't need to spend time on making sure that they are. If you use 100 apps and each offers 100 choices, that's 10000 settings to ensure are optimal. If you don't, they likely won't be.

99% of software pushes towards more choices, more settings, more customizability, which adds to cognitive overhead of using said software if you care about having optimal settings, so I'm glad to see an experiment that boldly pushes in the opposite direction.

Post reply on HN