Min Vid is a great refinement over my video watching workflow...for years I've been using the always-on-top feature of my window manager on Linux to keep a video window open for watching tutorials on YouTube or for binge watching mindless shows while I do mindless grunt work that doesn't need all of my attention but also isn't easy to offload to a script. I dunno how "always on top" is such a seemingly obscure feature that it's never been standard on Windows or (I think) macOS. How do folks live without it?
Anyway, Min Vid means I can use that workflow even on Windows. I never used PiP on televisions, and never understood why it was a thing...but, on my computer I totally want a little video window in the corner!
I switched to Linux about 2 months ago specifically to take advantage of powerful WMs. My WM of choice handles the "always on top, small video window in the corner" beautifully. It also works with any app, and any video service, where Firefox's has to be custom tailored to different scenarios.
It all makes me wonder why both Windows and OSX are so unwilling to either explore more powerful window management, or at least provide APIs to enable power users to build their own.
Windows does provide the API to keep windows always on top. You can do that with AutoHotKey [1], or numerous other apps out there that have been working for years. Had used DeskPins [2] for quite a few years before realizing my favorite application StrokeIt [3] already allows it. If you don't like third party apps, you can very easily write your own.
>why both Windows and OSX are so unwilling to either explore more powerful window management, or at least provide APIs to enable power users to build their own.
You will wonder how many users have single-threaded, single-windowed mindset (hint: vast majority). No offense here, having multiple windows on desktop really distracts and confuses them, harming their productivity.
APIs are provided though, at least in Windows. Aston Desktop[1] is still working on Windows 7.
> It all makes me wonder why both Windows and OSX are so unwilling to either explore more powerful window management, or at least provide APIs to enable power users to build their own.
Take a look around at the desktops of the majority of windows and osx users. Not the power users, who are a small minority of the user base of both OS'es, but the "I merely _use_ a computer to do task _X_" type users.
Once you start to look, you'll notice a trend. An overwhelming majority of them will be running every single application on their desktop in window maximized mode. And this will be true no matter the size of any monitors they have attached. The closest they get to "window management" is to poke the icon in their toolbar/dock for whatever it is next that they want to see on their screen.
Both OS'es likely do have API's to allow for an 'always on top' setting, but with a userbase that is 99% "run everything full screen" at the outset there is no room for those users to make use of an 'always on top' window (because it too would be maximized, defeating the usefulness of the 'on top' setting). Therefore neither maker likely sees much value in exposing those API's.
Not really sure what you mean here, I choose a particular WM setup which best serves my workflow, and that's pretty much it, it's not as if you have to continously tweak your setup.
>This is yet another reason why GNU/Linux lost the desktop
Can't really see how it 'lost' the desktop, it never had it, nor was it ever in the runnning to claim it, same goes for OSX despite having a ton of product placement, straight up advertising and riding the huge wave of pro-Apple consumerism in the mobile space.
In short, I don't believe for a second that standardising on a particular UI would have made any difference in 'drawing the masses', if super wealthy Apple at it's prime couldn't, then neither would Linux.
>I imagine you are not old enough to remember the yearly "The year of desktop Linux" on Slashdot.
Sure I am, but that meme was (still is?) repeated as a joke, did anyone actually believe it ?
>Also in the old days it was fashionable to try all WM and post screenshots of our setups, specially on weekends.
People still do that, there's a reddit sub dedicated to it: https://www.reddit.com/r/unixporn ,but then again there's a reddit sub dedicated to just about everything.
Just like there are stamp collectors, train spotters, rock climbers, ballroom dancers etc, there are people who find enjoyment in tweaking/expressing themselves through 'pimping' their computer environment, I see nothing wrong with that although it's not my cup of tea.
After having configured my environment of choice according to my specific needs I leave it alone and instead waste my spare time on places like Hacker News ;)
Could you list your window manager of choice and also which others you explored (along with any observations)? I'm sure at least a few people here would be interested.
Not parent, but one answer I can give you is IceWM. It's a classic WM with a complete feature set supporting all advanced features like these (everything that is not tiling).
Xfwm (part of XFCE) but that's kind of useless w/o the other components around it such as the panel, the config, so you probably want the full desktop environment (don't worry, it's rather lightweight).
I also used Fluxbox and Openbox and like both very much, but they are kind of basic.
Working with any of those is both faster and easier than any of Windows or MacOS, for me at least.
i3wm is a pretty popular tiling window manager which supports what parent described - you can set a window to "float" (so it doesn't tile, you can move around it like in a regular wm) and you can also set it to be "sticky" (so it stays visible even if you switch to a different workspace).
Yup, i3 is what I'm using. I admit my wm investigation wasn't that thorough, I mostly chose i3 because coworkers use it and I really liked it right off the bat. I also tried xmonad and awesome.
And not that anyone will see this now, but most people in this thread were responding to my "OSX and Windows don't offer powerful window management solutions" specifically in reference to my comment about floating windows. That's just one example of many things OSX/Windows doesn't allow the user to do.
Not to be too dramatic, but i3 has made me more productive, more focused and happier while working. I now fear ending up in an environment where using Linux isn't possible. It's really that good. But really, tiling window managers are that good.
The frustrating thing is that Always On Top is implemented in Windows, just not exposed to the user in any useful way.
For years I used a simple ~30 LOC program that mapped Alt+F5 to maximize/restore and Alt+F6 to toggle Always On Top. Sadly, I seem to have lost it.
It really did wonders for my workflow, especially in the days before Dash or Aero Snap. Being able to have a little documentation window up was fantastic!
You can still do it even if the original application doesn't implement it. Find the main application window with WinSpy and just mark it as top most. I know its not convenient but it can be done :)
The worst part is that OSX does have native support for always on top windows, but just does not expose an API to set them so. Before I was using Afloat which injected to applications and gave you the control. Nowadays there is Helium browser and picture in picture. Still not ideal though.
I use PiP all the time when watching televisions, I switch between channels during ads.
About always-on-top in Windows I'm pretty sure nVidia drivers had this options years ago? I discovered TurboTop recently, it does this one job and it's great: https://www.savardsoftware.com/turbotop/
These new experimental features have quickly become my favorite part of Firefox. I've been enjoying Tab Center, Page Shot, Min Vid, and Tracking Protection immensely and have them installed on all of my Firefox instances.
I'd really love to see Test Pilot begin to sync these extensions (and preferably also their settings) across different Firefox instances through Firefox Sync though. Right now I have to enable them separately for each instance, and while it's not exactly the end of the world, it definitely is a huge pain point for me (especially the lack of settings synchronization) and a glaring blemish on the high standard of UX most of these experiments are obviously striving to provide.
I thought HN was supposed to have this "don't hate on people's projects" thing, but when it comes to Mozilla, anytime one of their projects is posted 50% of the comments are low-effort trolling.
https://news.ycombinator.com/item?id=12718876 is an example that involved not reading the article very carefully, then complaining about the fact that some things that are already extensions should be extensions.
An experimental, no, production feature I would love to have in every browser is a way to disable these excruciatingly annoying full-screen popups which some sites (including the one linked in the topic) show you on your first visit.
No, I don't want to subscribe to your newsletter, I don't want email notifications, I don't want to become a member (at least not until I've read a few posts first).
Make it possible for me to make those choices my global defaults and I'll be a very happy user.
Wow, a built-in solution for vertical tabs. I'm surprised it took this long. It seems to use a bit much screen estate per tab, so users with many tabs will probably still resort to plugins.
It works quite well. If you have many tabs open they do get smaller. I never run into space problems with it, but okay, I'm not a tab hoarder.
People might still use an addon if they like the tree style tabs, to have them in a threaded view. I would find it useful, but I also understand why it is not there yet, it complicates tab management and not every user will need it. But it would be a nice advanced option.
I did not see that option, no. The only option it currently has is to disable the wrong default setting to place new tabs at the top. And in the UI itself one can toggle autohide, which is very nice
What does "Tracking Protection" technically do? Unless it completely prevents browser fingerprinting, I consider it completely useless (browser fingerprinting being standard procedure nowadays, at least at the giant internet/tech corporations).
So the other thing with the Tracking Protection experiment is that we're soliciting feedback about how this feature breaks the web. We use a block list from Disconnect for Tracking Protection in Private Browsing and for Focus on iOS, but until now had very little understanding of how enabling this feature might affect browsing.
For the Test Pilot experiment we're asking users to tell us where and how the feature messes up websites so we can refine our Block list and improve the experience for everyone. We're already getting a lot of significant data and intend to share it soon.
How about removing referral headers, report back very common user-agents, accept headers, language headers, screen resolution, etc. Take a look at https://panopticlick.eff.org/
I think some confusion is that the article makes it seem like pageshot.net is separate from the experiment itself. Both the add-on and service are the same experiment, and even hosted out of the same repository: https://github.com/mozilla-services/pageshot/ – you could host your own, but I would actively discourage it now (unless you want to contribute) as we are making a lot of changes and it would be annoying to track.
Another consideration is that we aren't only storing images, we're also storing information about the web pages. For instance, here's the JSON data for a clip of your comment: https://pageshot.net/data/aQOjNCw5wrjvveMC/news.ycombinator.... – of course you can still handle an image plus metadata on the local hard drive but it takes a lot of work if we wanted to support local and remote hosting.
Which leads to the question: why remote hosting at all? A lot of the use cases we see are for sharing. In Mozilla's research we've found that even people's archival habits tend to be built on sharing – people email themselves, send messages to other people for the primary purpose of reminding themselves, etc. My own theory is that to make something useful for your future self you have to go through all the same steps as if you were making it useful for someone else. So you have to pick out something interesting, make sure it is described well, etc. Anyway, we want to support sharing and cloud hosting is the best way to do that.
That said, this is an experiment, and what we're shipping with isn't necessarily what we're going to find, and we're open to suggestions. I'd encourage you to share any thoughts you have here: https://discourse.mozilla-community.org/c/test-pilot/page-sh...
One local hosting idea I've considered is an "all of the above" approach. In that case if you configured a directory to store your shots, then every time you take a shot we'd still upload it, but also same the image (by itself?) in the directory. That would be easy, and maybe it would tell us if this is a valuable direction (and if it was valuable then we might make it fancier than just dumping images). I'd be curious if that feature would make you more open to the concept?
What if I'm screenshotting, say, banking information or account recovery codes that I don't want uploaded to the public internet? As it stands now I feel the UI is misleading: I wouldn't necessarily expect clicking "Save" to send my screenshot to a remote server. "Upload", "Share", or "Save & Upload" would all be clearer. An additional "Save to Disk" button next to it, which popped open a file save dialog, would make Page Shot perfect for me.
The Page Shot service is run by Mozilla. It provides a download button so you can get a local copy of the image, but you might as well use the OS screenshot if you just want a local copy. Shots are designed to be easily shareable which is why they go on a hosted service. It is easy to run the open source backend yourself if that is what you would prefer.
Modern copies of Windows and OS X have native croppable screenshot apps that save to your hard drive, so I imagine there's a Linux variant on this idea as well. The convenience here appears to be saving the additional step that I always end up taking, of opening up imgur.com and uploading the image there.
They could have put an actual value-add over screenshot programs in there by offering to make a screenshot of the entire page but right now this is a terrible image hoster hooked right into my browser. So tired of this nonsense.
It is not an integration, it is a service hosted by Mozilla designed to make it easier to share things. You can run your own copy of the open source backend if you want.
The full page screenshot feature is coming in the next version in a week or two. We pared the product down to the absolute minimum for launch.
Yes, thank you, I don't want the MSN webhosting service in my browser. It's the equivalent of a Windows XP "publish to web" button only even Microsoft had the clue not to restrict it to their own service.
So Firefox is now picking winners? YouTube, Vimeo, and Pageshot get picked.
The features seem nice. The integrated to specific sites not so much. I know they say they'll add more video sites. That still means if youre not an established popular site you're basically 2nd class and your users may switch away because they can not use the feature.
That doesn't seem proper for a brower and certainly not one with Mozilla's M.O.
The Page Shot service is run by Mozilla. The back end is open source and it is easy to change the back end url that the add-on uses. (I work on Page Shot)
It would be nice if Min Vid would work on every <video> element. That would seem like the right way to implement the functionality.
The article describes some new additions to Test Pilot, a program to quickly get feedback on experimental features that might some day end up in Firefox proper. As such they start just fleshed out enough to prove the concept--you could say they're MVP versions of browser features. If Min Vid were to make its way into the default install I'm sure its compatibility would be expanded, for example.
A built in screenshot tool? It's like Mozilla saw chrome's built in screen sharing and said "we're Mozilla, anything you can do, we can do better, including browser bloat".
And somehow people still wonder why Safari is so popular for Mac users.
It's easy to dismiss Page Shot as just another screenshot tool. In fact, that's exactly what I thought of it at first.
After actually trying it out, however, it quickly became my favorite screenshot tool for web content due to its awesome UX made possible by having full access to the actual DOM of the page it takes screenshots of, namely things like being able to index all text in screenshots to allow you to search through them, and being able to intelligently limit the screenshot area to specific elements when you click on them. OS level screenshot tools simply can't offer an experience like this.
Hello had browser sharing built in for awhile, but nobody used it, so we took out the feature in 49.
Test Pilot is, at least in part, a program intended to let us develop ideas in the open without the cost and complexity of shipping to millions of Firefox users. Whether or not you choose to install these experiments is entirely discretionary.
Simply, our user testing (with Windows users in particular) suggests it's a feature people find useful. We're using Test Pilot to learn a bit more, that's all.
I only mentioned screen sharing because of the comment about Chrome's sharing feature (of which I was unaware, but will check out).
You don't see a correlation between browser bloat and energy efficiency?
At the very least you have to agree that any amount of time spent on "bloat" features take away time that could be spent on improving energy efficiency.
I don't quite understand this correlation of user-centric functionality with bloat on HN. News flash: users want features!
Most users didn't care that Google Chrome had become a CPU and RAM dog the last few years. They cared that their favorite websites worked. Period.
Please recognize that while you want things like energy and RAM efficiencies to the hilt, this isn't what most users care about or even know how to detect/measure. Realistically speaking, how often are you in a spot where a power outlet isn't available? Most trains and planes these days have power outlets, and any of the popular browsers give you about 4-hours on a full charge. What's the beef?
Could it be that you want Mozilla to focus on your agenda as opposed to listening to their core user group? If yes, then by all means complain away. Otherwise, laud them for trying something new to entice users to use their browser vs. the competition.
> I don't quite understand this correlation of user-centric functionality with bloat
A web browser is inherently designed to browse the web, and display pages. Practically every OS in recent history has the screenshot capability, and almost all of them can run client-server type applications that allow functionality like screen-sharing, SSH, etc.
But somehow Chrome and then Mozilla have decided that their browsers should become these giant monoliths of largely existing user functionality. Remember when Microsoft tied Windows Update to MSIE. Nobody was claiming that was a good fucking idea, were they?
> News flash: users want features
News flash: users want things to be fast, secure and not kill their battery.
> Most users didn't care that Google Chrome had become a CPU and RAM dog the last few years.
[Citation needed].
> Please recognise that while you want things like energy and RAM efficiencies to the hilt, this isn't what most users care about or even know how to detect/measure.
"My computer is slow/hot/loud/lasts just 2 hours instead of 7 on battery when using Chrome" is very much something a non-technical user can determine quite easily.
> Realistically speaking, how often are you in a spot where a power outlet isn't available?
Battery life is just one advantage of energy efficiency. Have you felt how hot a laptop can get when it's chewing a lot of CPU cycles for an extended amount of time? More energy in = more heat out.
> Otherwise, laud them for trying something new to entice users to use their browser vs. the competition.
Um, no. Firefox started out as lean, high performance competition to MSIE, just after Safari was launched by Apple. The original trigger was SOFTWARE BLOAT in the Mozilla suite, the features of which were still more related than what Google and Mozilla have added lately. At some point that goal of being a high performance standards-compliant web browser has apparently fallen away in favour of a keeping-up-with-the-joneses approach against Chrome.
Chrome's reason for existence is to allow Google more control over their revenue source. If they don't have a browser they control, they can't push whatever new technology they decide will make their business do better.
Part of Google's business is continually pushing the concept that user's don't need an OS that runs apps locally - everything can be in a browser (i.e. ChromeOS/Chrome browser with its kitchen sink approach to functionality)
So while Google's reasoning is not commendable (fuck the user experience, we need people to buy into our business model) their logic in pushing every possible feature into Chrome at least is understandable, if not agreeable.
Mozilla has no business case to artificially "encourage" reliance on a web browser for all user functionality, so why give up the hedgehog approach (do one thing, and do it fucking excellently) for whatever the fuck they're doing now?
In an earlier version of Page Shot we took a capture of the DOM at the same time as the screenshot, and so the screenshot meant "look at this" while we had an archival version of the page as the full context. Unfortunately the DOM capture didn't make it into the first release, and so we no longer have that context; we'll probably bring it back in the future, but maybe not as a combined product.
I did expect the Activity Stream to show the history in some sort of tree view, where if you go back while browsing, and then hit another link on a previous page, this will be shown as a branch.
The three new features are really useful and I look forward to using them. I don't care about the others but I like seeing Mozilla testing new things. This is the right way to keep a browser alive. And keep paying attention to the basics, of course.
Update: About testing those features... I'm testing MinVid because they're logging basically only the name of the service I'm using, not the actual video. I'm not testing PageShot because they keep a copy of the screenshots: they're safe for office but they're not Mozilla's business. Actually, I could violate some NDAs by sending some of those screenshots to Mozilla. Worse, I'm not testing Tracking Protection because "When you engage with a prompt, we also collect information about the page you are on and the tracking domain that is being blocked". Mozilla won't know what I'm browsing.
Page Shot, like all features in Test Pilot is opt in. We have no plans to force anything on anyone. A big part of why we built test pilot in the first place was to give us a platform to test UX without going all 'Leeroy Jenkins' on Firefox users every time we come up with a new idea.
That's great, thanks for elaborating - I had missed checking out what "Test Pilot" actually was.
That said, what will happen to Page Shot if it's deemed a good/decent UX and it's graduation time? Will it be an installable addon? Pre-shipped? Baked in?
I guess it's something to be decided in the future, yet it makes me curious today.
why would you want to remove that feature? i mean, you can probably remove the button to be out of sight easily. if it does nothing when you're not using it, what's the deal?
The browser already renders the page, it's a pretty small step to saving it as an image. After all, there's already a screenshot tool in the developer tools.
Does your screenshot tool let you capture whole webpages, or just what's in the viewport at the current moment?
All these tools are lazy loaded and don't affect browser speed when disabled. The most you're complaining about is the existance of a small bit of extra code in the Firefox download.
well, but people usually don't ask for making e.g. the print feature optional because they don't use it and want a lean browser... i mean, you need to draw a line somewhere, sure, but different people will draw it differently.
The 'print' feature is a platform-provided thing. Firefox isn't shipping printer drivers or a printer UI. It simply gives the OS printing subsystem content the user wants to print.
Funnily enough the platform also provides a screenshot system. From another comment it seems like this was driven mostly by user feedback from Windows users, because the ~3 different ways of doing a print screen on Windows are not sufficient and they demand more.
And giving the OS printing subsystem the content the user wants to print requires thousands lines of code in Firefox. Thousands of lines of code many people don't need, while many others do. Just like with any other feature. My parents don't need bookmarks or even tabs, so should we get rid of them in Firefox?
Btw. the screenshot tool of Windows can't capture a webpage properly, only the visible area.
When is Mozilla going to channel its resources into building a JavaScript engine that can keep up with Chrome's so that the web is actually usable in FF again?
actually, on 64-bit, IonMonkey is within 100 milliseconds of V8 either direction. With backtracking, it's within quite a lot less than that most of the time. On some benchmarks, it's even faster than V8. Source: https://arewefastyet.com/overview/#/machine/29
I must say i was a Chrome user for the past years and changed back to Firefox for approx. 6 months now. Aurora and Stable.
But after thinking that i can cope with a little bit of sluggishness and slowness it's just too much of a difference in responsiveness overall. Chrome starts much faster, tabs open much faster, tab switching is faster.
In Chrome it is _instant_. In Firefox it is fast but it's noticeable to the point where it distracts from fast keypresses. Sometimes i am pressing keys faster then Firefox is ready to accept and Firefox misses keys.
Yes, firefox is fast but it's not yet right where Chrome is.
Today, i tested the experimental features but in the end, i switched back to Chrome.
I just hope Servo will be great and wait. Until then, i am running Chrome again.
P.S.: I've used FF and Chrome on Windows and Linux, and on both OS, Chrome beats Firefox in UI responsiveness.
Have you filed bugs on Mozilla for those? There's more than a JS engine to a web app; you could be running into layout or style system issues just as easily... But the important part is that we can't fix performance issues we don't know about, so please report performance issues if you encounter them!
...Some webapps just flat-out won't consistantly run on FF. Given that FF is fully standards compliant, some of both of these problems could be attributed to shoddy web developers.
Then again, FF could be at fault. I don't really know in this case.
"fully standards compliant" is definitely overselling it. Firefox aims for that, but doesn't always succeed. (Disclaimer: I work on standards support in Firefox, among other things.)
That video popup thing reminds me that browsers really, really need mpris2 support. It is super annoying not being able to use hardware / OS level multimedia keys on the focused / current webpage because there is no browser support for pause / play like that when every other media player in the universe makes sure to support them out of the box.
Is there a way to enable casting to Chromecast with Firefox? I haven't found any viable solutions as of yet - but if there were I would jump to Firefox instantly.
It's a WIP. Still, you don't need to wait for the integrated sandbox, you can sandbox Firefox as a whole using tools available for your OS (eg. firejail for Linux).
IMO Mozilla would be wise to spend time on improving Firefox’s memory footprint and responsiveness. As it is it’s slower and more cumbersome than Chrome (at least on Linux). Some of these features are nice, sure, but perhaps they should be provided as extensions? The core user experience is browsing, and right now that’s a huge pain.
Anyway, Min Vid means I can use that workflow even on Windows. I never used PiP on televisions, and never understood why it was a thing...but, on my computer I totally want a little video window in the corner!