Slashdot Mirror

Can't read the article?

There are few reasons because of which third party browsers are not so comfortable with the iOS environment. First and foremost may be the hurdle of not able to carry their rendering techniques and javascript engines over to iOS

Can't use Google?

http://news.cnet.com/8301-1023_3-57439936-93/browser-choice-a-thing-of-the-past/

http://www.buzzfeed.com/jwherrman/the-problem-with-chrome-for-ios

http://support.mozilla.org/en-US/questions/945460

It's like you're not even trying. I don't think a "rogue" mod is responsible for your karma.

I can feed trolls with the best of 'em. Burn, karma, burn!

The fact that the rendering engine would be Gecko on their PC and WebKit on their iPhone just doesn't fucking matter.

Apple limits third party IOS developers to UIWebview, while Safari gets to use the Nitro JIT javascript engine. It's an automatic performance disadvantage for any aftermarket browser. That fucking matters.

It really shows that Mozilla's focus is on themselves and software developers, not on the consumer end user, who has been running Firefox on their PC for years now and Safari on their iPhone for years now and just wants a Firefox interface and bookmark syncing on their iPhone.

No, it shows that Mozilla is smart enough to recognize and avoid pitched battles with Apple. Why fight to have a weird mutant version of their flagship project on a closed device, damaging their brand with artificially limited performance and a rendering engine that doesn't act like Firefox?

If that is Mozilla's focus, then they don't belong on iOS and good riddance.

Mozilla's focus is on opening up the web. You're right - they don't "belong" on closed, controlled iOS. They will, however, try to encourage Apple to let them in.

On iOS, the end user is at the top of the hierarchy, and software developers and content producers all work for the user. The user already has an HTML5 renderer in their iPhone, they already have a TCP/IP stack. You do not need to replace them to build a browser, and in fact, it is much better security that you can't replace them. That is what is best for the consumer: a secure renderer that is highly-optimized specifically for their device.

Who decides what's in your interest? If it's Apple, then Apple is at the top of the hierarchy, not users as you say.

As a user myself, I value the ability to use Firefox over Chrome on my Android device. With Android, I can decide what's in my interests. The defaults work for "most consumers", and for everyone else there is a measure of freedom.

There are plenty of reasons that software monocultures are bad, and Google is your friend there.

There are hundreds of 3rd party browsers on iOS, many with very innovative features. Like Skyfire, which converts Flash Video to ISO standard video on a server and essentially enables you to run Flash on iPhone or iPad. There are browsers that are exploring lots of gestures, or deep social integration.

Cute little user-interface experiments are one thing, but that's all niche-market small time stuff. Deep social integration and gestures? Tee hee. Calling a UIWebview wrapper a browser is kind of endearing.

Mozilla is missing out on all of that because they are pouty, entitled developers who want their feet rubbed and cheeks kissed before they deign to bless us with their bloated, mangled code.

You realize that Firefox is the best browser on the memory usage front, and near tops in performance right? If your gut feeling about Mozilla is based on a 2006-era opinion, you might want to look at what they've done lately.

And of course, Mozilla knows better than Apple what Apple users want. As if.

Most users want options and the ability to use their devices as they see fit. Mozilla has only ever supported users' rights. Apple can't say that.

And finally, Mozilla's hypocrisy: note that the one and only HTML renderer on Firefox OS is Gecko. And Firefox OS has zero 3rd party browsers as of right now.

Hey now, third party browsers can just wrap Gecko (actually, it's more like just opening an IFRAME, since the UI is all HTML.) In your world, using the system renderer is a good thing, right? What are you complaining about? /s

In all seriousness though, it could be done with some work. I

You can get 16GB of RAM these days for less than $100.

But can I put 16 GB of ram into my mother board?

The limit is not how much memory I can afford.
The limit is the physical hardware. Or rather, The FIRST part of the limit is my physical hardware

The second part of the limit is even simpler: I don't run a single app as my whole machine.

I don't run a single app as my whole machine. My machine is multitasking.

Right now, what do I have running?

Gui:
Finder, Process monitor, Temperature monitor, Terminal (With 10 different windows on 7 different desktops), TextEdit (with something like 70 open files that I need to go through, review, and take action on), Mail, Quicktime (mostly just sitting there, with 4 parts of a video I'm making open, waiting for me to start recording the next segment), MultiMC (just the launcher for minecraft), Skype (so I can can get contacted by the people I play with), Midi Audio Setup (so I can monitor the sound, and switch between hardware speakers and soundflower, as well as adjust volume levels -- which I can't do if I use a multi-output device, and the volume keys adjust the wrong thing if I'm using soundflower), NBT Explorer (just sitting there; has something up that I'll be using in my next video segment), and a Flac to m4a audio converter (again, just sitting there.)

That's just the BACKGROUND gui stuff.

There's also a mumble server, two minecraft servers (both currently ctrl-Z'd; testing memory usage/reclaimation shows that Mac OS does a crappy job of paging out unused memory, reducing the effective memory available to me).

And ... FireFox. Firefox that wants to use ridiculously high amounts of memory. https://bugzilla.mozilla.org/show_bug.cgi?id=814173 -- and remember, Firefox is currently the only browser I know of that has something like NoScript.

What else am am I likely (but not at the moment) running? iMovie doing quicktime exports to compress video. Yea, it stinks that I have to load a giant GUI program to compress video, but attempting to compress it from Finder gives me a "one-size fits all, we won't even tell you what settings we're using, and we don't care if you don't like the looks, we know what works on YouTube no matter how old our settings are or how much YouTube has changed" output that is of no use to me. (Seriously, if I'm recording 480p screencasts, for a system that wants 480p uploads, that streams to users at 480p, why would I want a 540p encoding?) Or Mumble -- a great audio/voice processor and recorder, that sadly wants to consume very large amounts of CPU activity even when there's no voice activity -- so while I have murmurd (the mumble server) running all the time, I only use mumble during play (And leave skype up so my cohorts can reach me).

Do I need to shut down the browser when playing? Given that I may need to refer to documentation, or report issues to various forums/bug trackers? Should I shut down email for a trivial gain of space?

You can get 16GB of RAM these days for less than $100.

But due to stupidity from Apple, that 16 GB of ram requires 16 GB of hibernation space to be on my root partition. It has to be in /var/vm/swapfile, and /var/vm must be a directory -- neither /var, nor /var/vm can be symbolic links to a second partition on the primary drive. So I would have to be able to repartition my system drive -- which doesn't even work.

So ... I have to have temporary files (/var/tmp, /var/folders), swap files (/var/vm), hibernation, etc, all in a partition that was sized big enough for the operating system, and expected growth -- back in 10.6 when XCode was on a different partition, etc. -- and then Apple decided that developers must have the entire development environment on that partition a

A github employee presented on Air Mozilla recently: https://air.mozilla.org/scaling-happiness/

I use a Firefox extension called LeechBlock.

Basically, you give it a list of domains to block, and when to block them.

For instance, I usually get my best work done between 5:00 and 10:00 PM. So, I gave LeechBlock a list of websites on which I typically waste my time. Between 5:00 and 10:00 PM, it only lets me waste 10 minutes of my time every two hours.

When I go over my time limit on time-sucking websites, it shows me a page informing me of my condition. I configured LeechBlock such that I can click on a button, and see my time-sucking page anyway. LeechBlock serves as a gentle reminder to get back to work.

It is rather configurable. For instance, there is a setting that allows you to disable the add-on manager during a given time interval. The idea is to prevent yourself from turning the extension off.

The extension is incredibly customizable and surprisingly effective.

There is a similar extension for Chrome.

With luck there will eventually be a push for a standardized tablet platform that is open enough to permit users to select their own OS.

That standard platform is the Android kernel.

Meanwhile Plasma Active, Salifish, and Tizen are based on a traditional Linux platform, and the Mer project hopes to be the common core distribution for them.

For the tiny fraction of users who "select their own OS", device popularity and an unlocked bootloader matter far more than standardization. If you buy an unsuccessful phone, it won't have a community providing images for it and jailbreaking its bootloader if necessary.

The standardized platform is vital for all these also-ran OSes to get lots of apps. Aaron Seigo's post about standardizing the QML compontents across KDE Plasma, Jolla Sailfish, BlackBerry 10 and Ubuntu is a good sign, but they still suffer from inconsistent device APIs and different packaging requirements. That's where Firefox OS has a theoretical edge: apps for it are just web pages with a manifest. The number of web developers (incuding "app" developers who just put a wrapper around an HTML app) is orders of magnitude more than QML developers.

Most likely this will come from the second tier Chinese manufacturers who would benefit most from a common reference standard.

They don't push for anything. They ship Android.

With luck there will eventually be a push for a standardized tablet platform that is open enough to permit users to select their own OS.

That standard platform is the Android kernel.

Meanwhile Plasma Active, Salifish, and Tizen are based on a traditional Linux platform, and the Mer project hopes to be the common core distribution for them.

For the tiny fraction of users who "select their own OS", device popularity and an unlocked bootloader matter far more than standardization. If you buy an unsuccessful phone, it won't have a community providing images for it and jailbreaking its bootloader if necessary.

The standardized platform is vital for all these also-ran OSes to get lots of apps. Aaron Seigo's post about standardizing the QML compontents across KDE Plasma, Jolla Sailfish, BlackBerry 10 and Ubuntu is a good sign, but they still suffer from inconsistent device APIs and different packaging requirements. That's where Firefox OS has a theoretical edge: apps for it are just web pages with a manifest. The number of web developers (incuding "app" developers who just put a wrapper around an HTML app) is orders of magnitude more than QML developers.

Most likely this will come from the second tier Chinese manufacturers who would benefit most from a common reference standard.

They don't push for anything. They ship Android.

At least the people from Mozilla is trying to get their APIs approved by W3C and spread to other vendors. Check out https://wiki.mozilla.org/WebAPI

You don't even know what telemetry in Windows means.
You get one notification after installing asking you explicitly if you want to opt in to help improve Windows by sending telemetry information.
Even if you opt in, those calculations you see in the post are done locally and only the stats are sent to the server.
Those stats do not include people who haven't enabled telemetry.
It's similar to Firefox's dialog here:
http://support.mozilla.org/en-US/kb/send-performance-data-improve-firefox
OMG FIREFOX IS SNOOPING ON YOU.

You're welcome to prove me wrong.

This has nothing to do with sending packets when you search local files.

Not only that, but on his other point, the memshrink project took off, Firefox has been using significantly less memory than other browsers.
On my system, for 5-10 tabs, Firefox uses about half as much memory as Chrome. For a large number of tabs, Chrome explodes to gigabytes of memory while Firefox doesn't go up by much at all.
Not to mention tab groups make organising that large number of tabs a lot easier.

https://blog.mozilla.org/nnethercote/category/memshrink/

but no automatic identification of origin affiliation by subdomain can be done simply without understanding of per-domain policies from the TLD down to the first level at which all subdomains are affiliated

Yes, this is major breakage in browser specifications. It's also a solved problem.

It has been necessary for years because cookie security relies on it. The solution is a bloody great list of effective TLDs. It's not pretty but it works.

There's a big difference between going to a web site and being able to run it offline, vs. downloading then running a setup.exe (and re-installing the Java or .NET runtime you got rid of in 2011). HTML5 delivers a universal zero-install runtime that eliminates any "installation" step, and when the user is connected there is no "upgrade" step either. It ought to be the future. I may never get a Firefox OS phone, but I'm looking forward to its app stores and Mozilla's advocacy to make any web page an app.

Correct, here is the validation documentation: https://developer.mozilla.org/en-US/docs/Apps/Publishing/Validating_a_receipt

"A usual time to validate is when the user starts the app."

Where's the Linux version?

Oh wait! Found it!

https://download.mozilla.org/?product=firefox-19.0&os=linux

The same reason people asked Mozilla for a longest update cycle: ESR http://www.mozilla.org/en-US/firefox/organizations/ to have time to test the changes on your environment before you apply an update that will break your old applications

Sadly, doesn't look like it, yet. *Sigh*

I just *love* scrolling through all my bookmarks looking for one I know I have in there. Yes, I know, I can use the search bar to find it, but that doesn't really help when I just want to browse bookmarks of a particular type...for example, restaurants, or online stores.

In the past, I have stored these bookmark types in their own folders, then when I'm feeling peckish (for food or toys :), I'll just open the relevant bookmark folder and see what looks 'appetizing' today. With FF mobile, there is no way to even sort your bookmarks manually to group similar types together, much less have them tidily tucked away in their own folder like you can on the desktop client. You are stuck with some weird arse ordering system invented by Mozilla (new bookmarks go to the end of your bookmark list...almost?)

This (apparently deliberate) design choice just baffles me...why, Mozilla? Why take away folder and bookmark management tools for mobile devices, where people are constantly looking for ways to optimize screen real estate? Why??

IE10 does exactly that. If you open YouTube videos to background tabs, they won't start playing until you activate the tab.

For Firefox, I recommend installing the Flashblock addon, which allows you to start Flash plugins manually by clicking them.

It does on Windows, unless you set various options that aren't really supported and are likely to break when the big theme refresh lands.

The new UI is happening, too. See, for instance, this blog post about the changes to customization that will restrict what you can do with the UI. Or the UX branch, which has curvy tabs already.

Extensions can probably address many of the problems they're introducing, but -- particularly with the theme changes -- it's really getting to the point where an actual fork would be easier.

Pretty sure they're running on Androids guts, so kinda?
(I think they took out the UI/Java/whatever layers and are using the Linux kernel that Android uses, plus their own UI layer. See info on Gonk.)

There's an add-on which does this:
https://addons.mozilla.org/en-us/firefox/addon/requestpolicy/

It's annoying at first, since every site you visit needs setting up. And I really wouldn't want to try it while I was buying anything. But it can be useful and usable.

Mozilla publish the release schedule...so 2013-06-25

https://wiki.mozilla.org/RapidRelease/Calendar

IMHO, the next step is to block referrer information to third party sites. E.g. if example.com loads a script from gstatic.com, then the HTTP_REFERER header is not sent to gstatic.com. There's almost zero collateral damage (one captcha service doesn't work), and companies like Facebook and Google no longer get to know every site that most internet users visit.

I agree whole-heartedly with this sentiment, but it might cause more grief that most would guess.

Over the last year or so I've played around with blocking the referer header from being sent at all, to any websites. 99% handle this just fine, but every now and then I'll come across sites that fail, and in various ways. Sometimes I get a useless error message from CloudFlare, and sometimes the page will simply render blank, like this one (in this case because TypeKit issues a 403 when requesting the CSS if the referer is missing).

I have no idea why some sites rely so heavily upon an HTTP header which is not required to be present at all. I'd love to see a browser start to do what you suggest and exclude the header in 3rd party requests because it would force sites to treat the header as it was intended (advisory only) and would also make it easier for those who want to block sending it entirely.

Extension

Because I also block cookies by default, for the cookie lifetime I have "ask me every time", which only prompts when I quickly toggle allowing cookies to add a site to the whitelist. So for a button to quickly toggle them on and back off I use Toggle Cookies which also keeps 3rd party cookies blocked by default even when allowing.

Ghostery manages to bloat my memory, so I use the lightweight about:trackers extension instead.
Also, instead of NoScript to avoid XSS I use UserCSP

Ghostery manages to bloat my memory, so I use the lightweight about:trackers extension instead.
Also, instead of NoScript to avoid XSS I use UserCSP

IMHO, the next step is to block referrer information to third party sites. E.g. if example.com loads a script from gstatic.com, then the HTTP_REFERER header is not sent to gstatic.com. There's almost zero collateral damage (one captcha service doesn't work), and companies like Facebook and Google no longer get to know every site that most internet users visit.

Well, have a better look before spreading lies. For kernel, web browser and libreoffice

Kernel:
Stable - 2.6
Testing - 3.2
Unstable - 3.2 (released July, 2012)
Current version - 3.8
Those 3.6/3.7 files seen in your link? Experimental. Yeah you could make it work, but then you aren't running Sid anymore. Not entirely. And if you run too much experimental for too long, something is going to end up horribly broken.

Firefox/Iceweasel:
Stable - 3.5
Testing - 10esr
Unstable - 10esr (released March 2012)
Current Version - 19.0

Libreoffice:
Stable - 3.5
Testing - 3.5
Unstable - 3.5 (released February 2012)
Current version - 4.0

YOU DO have very recent packages available, even right now, during the freeze of testing. I haven't checked DE and X, since I don't know what you run (eg: which graphic card, and which environment you like).

I run XFCE on testing, not that it matters.

XFCE:
Stable - 4.6
Testing - 4.8
Unstable - 4.8 (released Jan 2011)
Current Version - 4.10 (released April 2012)

Like I said, Sid isn't bleeding edge. Of the packages here, the newest in Sid is the kernel: 7 months old.

It's also worth noting that drivers receive unblock from the release team so that they can enter stable.

Well that's great (and I genuinely mean that), but a bleeding-edge enthusiast would only see that 3.2 != 3.8.

The patch is not exactly a one-liner, because the implemented behavior is not as straight-forward as just "block 3rd party cookies".

It's "block cross-site cookies from origins which I've not visited yet as a 1st party websites and have already 1st party cookies from".

This means, for instance, that Facebook, Google and Twitter gets likely a free-pass to track almost anybody.

And that once you (accidentally or not) click any ad box, you give a free-pass to its advertising agency too.

Firefox 17 not 18 correct? You should see it immediately. Here is a sample image with FF 16 vs. Safari.

https://bug674373.bugzilla.mozilla.org/attachment.cgi?id=631507

Don't pull things out of your ass, they fixed dozens of memory leaks: https://blog.mozilla.org/nnethercote/category/memshrink/

User-defined Content Security Policy

Tizen joins Blackberry 10, Firefox OS, webOS, and Windows 8 in saying "Write HTML5 apps for our platform". Unfortunately these are all also-ran platforms, but it does make it easier for PhoneGap to target them along with turning HTML5 into Android and iOS native apps.

So where are these HTML5 apps? I don't want to have to connect to a a web site and hand over my personal details to maintain a list or edit a photo in my browser. I should be able to try out any application in my browser, and if I like it "pin it" to run locally. I hoped FLOSS developers would step up and develop these, but they seem stuck in the 90s arguing irrelevancies like GTK vs. Qt and Python vs. C++.

Instead there are hundreds of thousands of "apps" that are nothing more than HTML5 packaged a certain way, all dumped into a few needlessly platform-specific App stores.. It's a travesty of the principles of the web, and for no good reason. At least Mozilla has the right vision:

The Mozilla Open Web Apps project proposes some small additions to existing sites to turn them into apps that run in a rich, fun, and powerful computing environment. These apps run on desktop browsers and mobile devices, and are easier for a user to discover and launch than Web sites. They have access to a growing set of novel features, such as synchronizing across all of a user's devices.
...The only thing you have to do to create a Web app from a Web site is to add an app manifest. This is a JSON file that describes your app, including its name, its icons, and a human-readable description.

They should apologize to you for what, pray tell? For updating their *free* web browser more often than you'd like?

Rather than ranting, why don't you go here if the update schedule is keeping you up at night:
http://www.mozilla.org/en-US/firefox/organizations/all.htm

It works well, but I've always been annoyed that Firefox doesn't just dish stuff off to the built-in Mac PDF renderer - which is resident all the time and is necessarily snappy.

There used to be a plugin to integrate Preview into FF & render PDFs in the browser.

Unfortunately FF4 in 64-bit mode broke it, and FF18 seems to have killed it completely. Pdf.js is clunky, slow, broken on many PDFs, and the cross-site restrictions of js mean it's useless on many academic journal sites. The other recommended alternative, Schubert|it PDF Browser Plugin, is a piece of shit.

It's the same thing PDF Viewer does... maybe the same project? I don't follow such things. It works well, but I've always been annoyed that Firefox doesn't just dish stuff off to the built-in Mac PDF renderer - which is resident all the time and is necessarily snappy.

Whenever someone mentions NoScript they should also mention RequestPolicy. Try RequestPolicy for yourself and see it block all the stuff that NoScript is letting through.

Regarding NoScript, I just wanted to add that RequestPolicy is to be used in addition to NoScript.

"And, finally and most importantly, Firefox has a zillion useful extensions. Like NoScript"

Whenever someone mentions NoScript they should also mention RequestPolicy. Try RequestPolicy for yourself and see it block all the stuff that NoScript is letting through.

The technical bits of 1.1 are availaible in the NSS-library (the library created by Netscape at the time I believe and now developed by the people who develop Firefox).

The technical bits for 1.2 exists too, but I don't know if they still need more core, I believe they are under review.

The Firefox parts are almost there:

https://bugzilla.mozilla.org/showdependencytree.cgi?id=733647&hide_resolved=1

The problem is really with all the webservers which still don't properly work with it.

Which forces a new TCP-connection with 1.0 (which means adding the 1.1 or 1.2 is probably useless if an active attacker can mess with your traffic and force fallback to an older protocol).

We banned it from our company after waiting years for various memory leaks to be fixed.

That was fixed. The Firefox memory heap is now divided into "compartments", and Firefox 15 changed memory management to be more aggressive at purging compartments associated with closed pages.

TFA links to blogspam, below is the actual release note list from Mozilla

http://www.mozilla.org/en-US/firefox/19.0/releasenotes/

Come on, guys.

Everyone locked into ActiveX did so willingly. MS never held a gun to anyone's head and the problems inherent with it were obvious from the day of release.

If it's backed by force of law, does that count as a gun? That's exactly what happened in South Korea.

Which cookies do you "have to accept"? And, which ones must you keep forever?

I accept few cookies, almost none of those infamous ever-cookies, and they are almost universally deleted when I close my browser.

https://addons.mozilla.org/en-US/firefox/addon/betterprivacy/ One among several tools that are useful when preventing trackers from tracking you.

I've always thought that having GUI files is the way to go instead of in code. I'm fine with XML (FXML in this case), but I'm sure some others have gripes and may prefer property files/etc. But how nice would it be to have an XML standard for all GUIs? Then all you have to do is load one XML file across GTK+, Qt, X11, Windows, Cocoa, and even OpenGL. Example:

<window width="300" height="300">

<edit width="100" height="20" value="Type name." />

<button width="50" height="50" value="Submit" />

</window>

Then do the logic in whatever language you want. I know it's a pipe dream with several problems, but damn it would be nice.

http://www.mozilla.org/keymaster/gatekeeper/there.is.only.xul

I have Flash disabled in Firefox, but still occasionally open Flash content in Chromium. 99.9% of that is for videos on YouTube and Vimeo. For YouTube it's roughly 75% flash-only videos and 25% H.264-only videos. More like 90%/10% on Vimeo.

They don't ship H.264 support on Desktop due to licensing costs. There is work and plans for Firefox to support GStreamer for codecs they don't ship, but it's not here yet. See bug 794282 and its dependencies for details.

The Mozilla plugin check tool can be used in any browser, and reports Flash on IE10 on Win8 is still "outdated": https://www.mozilla.org/en-US/plugincheck/
But the tool can be inaccurate for some browsers. At this time it does show Flash on Chrome as up-to-date. Chrome also bundles its own Flash. Firefox shows as OK too, after you update. If you try to update Flash in IE10 you get a notice that Flash is bundled, but it also says you can install it if you really want to.

My advice is: always internationalize your software — make it possible for people to adapt it to their languages and local cultures. You might think it's not strictly necessary as of now, but I believe software needs to be prepared to grow and be extensible in the future, cultural adoption included.

While English is a kind of de-facto language on the Internet and anything around its technologies, that doesn't necessarily hold true for programmers.
If you want to see a real life example, head over the Mozilla Developer Network, where you can find technical content available in several languages. Those pages are not translated for the sake of translating, but because they answer to an actual need from programmers. You can find another example in the Firefox Developer Tools — which, by the way, are not a requirement to translate in order to have an official localized build on the release channel, but are actually translated into lots of languages.

My advice is: always internationalize your software — make it possible for people to adapt it to their languages and local cultures. You might think it's not strictly necessary as of now, but I believe software needs to be prepared to grow and be extensible in the future, cultural adoption included.

While English is a kind of de-facto language on the Internet and anything around its technologies, that doesn't necessarily hold true for programmers.
If you want to see a real life example, head over the Mozilla Developer Network, where you can find technical content available in several languages. Those pages are not translated for the sake of translating, but because they answer to an actual need from programmers. You can find another example in the Firefox Developer Tools — which, by the way, are not a requirement to translate in order to have an official localized build on the release channel, but are actually translated into lots of languages.

What ads? Use GMail through a protocol actually designed for email with the MUA of your choice and there are no ads.

They can target all the ads they want. I'm not in the beaten zone of their futile barrage of spam.

Webmail users, and anyone else who signs into Google for web-based services, have volunteered for this abuse. I think of it as natural selection in action.

https://developer.mozilla.org/en-US/docs/DOM/XMLHttpRequest

XMLHttpRequest.mozSystem -- Read only -- boolean -- If true, the same origin policy will not be enforced on the request.

It is false by default in every release of the browser I could get my hands on.