Hacker News new | past | comments | ask | show | jobs | submit login
Retiring Chrome Frame (chromium.org)
146 points by goddabuzz on June 13, 2013 | hide | past | favorite | 95 comments

Google may be continuing with the 'Don't Be Evil' motto but they sure ain't standing by a 'Don't Be Dicks' motto. What major software company announces a deprecation with 6 months of support. I have something going out relying on chrome frame plugin going out next week, now I have to go with my tail between my legs. Well lesson learned, can't rely on Google in the enterprise.

What were you paying for Chrome Frame?

Sometimes, it makes sense to retire products. And if you're not making money on the product, then it can be expensive to keep supporting it. I think six months is still a reasonable amount of advanced notice. No matter what, a small minority will be unhappy.

I think the real lesson here is: don't build something serious on top of a product you don't pay for and you don't have any guarantees on support.

I understand this completely. But moves like this is why you see Internet Explorer 6 still in the wild. Because C-levels can be confident that Microsoft will continue to provide baseline support, security patches, and forever. My boss didn't even get annoyed, he just said 'too funny' and reminded me of when he was telling me to just build on IE7 baseline.

6 Months in the Enterprise world is a barely time enough to sneeze... Anyways I am fine, its a lesson learned.

Microsoft will continue to provide baseline support, security patches, and forever

Actually they're going to stop providing IE6/7 support and patches in a little under a year.

IE6 was released 11 years ago, which is beyond forever for a browser version to be supported. No other vendor that I'm aware of provides support for browsers from that era.

The sad problem is, if you want to build a product for large corporates, a lot of folks are still stuck on old XP workstations with the default software installed (IE7 if you're lucky), and likely will be well beyond the MS support window. This is just part of doing business in that sector, but I think we can all agree that having to build for IE6/7 compatibility stifles a lot of innovation.

IE on Windows XP actually. IE follows the support lifecycle of the Windows version it is installed on since IE6 SP1, and IE 5.01 on Win2000 follows the support lifecycle of Win2000 too.

Will Microsoft continue providing baseline support for IE6 indefinitely? I doubt it. The top Google hit for "ie6" is http://www.ie6countdown.com, which is Microsoft's own site for tracking and celebrating the death of IE6. I imagine that once those numbers get low enough around the world, they'll stop supporting it.

>Will Microsoft continue providing baseline support for IE6 indefinitely? I doubt it.

Doesn't matter, they already have supported it for centuries in tech years.

One wasn't "paying" with money. But Google wants to move all of IT in a certain direction (where cloud and web based technology is acceptable). Therefore Google provided the illusion that it's possible to mix ancient IE6-IT with modern Chrome web apps. Developers could feel save that it's okay to use modern technology and still be able to reach old shops. And old shops might have started to adopt modern web apps, also from google. Now they are forced to think again how to keep the old IE6-intranet running next to some Google Docs stuff they bought...

So Chrome Frame helped Google a bit to extend a certain market for them by a little bit. Google still doesn't get paid, but they certainly didn't do Google Frame out of altruism.

> Now they are forced to think again how to keep the old IE6-intranet running next to some Google Docs stuff they bought...

The Chrome Frame retirement announcement points them to a solution for that, Google Chrome for Business [1] with Legacy Browser Support [2].

[1] https://www.google.com/intl/en/chrome/business/browser/

[2] https://www.google.com/intl/en/chrome/business/browser/lbs.h...

I was going to point this out as well, and it actually seems like a better solution. Default everyone to a modern browser with updates, and then throw them out to whatever you need to use for compatibility if needed.

Chrome Frame always seemed like a nasty hack which was unlikely to get much traction in companies that haven't updated from IE6.

> Now they are forced to think again how to keep the old IE6-intranet running next to some Google Docs stuff they bought...

Since they already switched from supporting IE6 to using modern technology as if IE6 didn't exist, then maybe they just won't bother switching back. Maybe it will be more cost-effective, at least for some of them, to finally ditch IE6. And that would be a win. The cancellation of Google Frame may be what is needed to break the feedback loop that keeps IE6 afloat.

He is saying he just learned that lesson. Free Google software isn't a reliable foundation for a business decision. Compare to Microsoft, who have a track record keeping their technologies alive and supported for many years.

The other lesson is: Use open source solutions. If a company drops support for something that's open source, you have the option to pay someone else (or DIY) to continue that support going forward.

Other comments here say Chrome Frame is open source, minus the MSI itself (just a "very thin wrapper"). https://src.chromium.org/viewvc/chrome/trunk/src/chrome_fram...

>What were you paying for Chrome Frame?

With the virtual currency of trust towards Google as a competent solutions vendor.

Six months in enterprise is not reasonable.

> What were you paying for Chrome Frame?

They can't have it both ways: if everyone thought like that, nobody would have uses Chrome Frame.

> can't rely on Google in the enterprise

I think some people read that as "Hey, that's a nice software ya got there, buddy. It'd be a shame if your dependencies went away. Good thing I got this service contract right here..."

As much as I hate to admit it, sometimes the policy to never use 3rd party software without an explicit support agreement is quite sane. But this is also a good argument for why open source and minimal complexity are good choices.

Chrome Frame will still work after January, just it won't be updated anymore. I assume that means you can still rely on it for your product, no?

No IT manager would permit an unsupported browser plugin to remain deployed. Who's going to fix newly discovered exploits?

No IT manager worth its salt should be relying on a free, unsupported, 'as-is' version of a hack in the first place.

And even less something that was always intended to be a quick workaround for users under very strict (or lazy) IT organizations that didn't upgrade to a decent browser in years.

Kudos to Google for creating and supporting it, until now. Time to take off the training wheels.

That's the beauty of Chrome Frame. In most cases, it gets "deployed" without needing any approval from IT.

Chrome for Business looks like a solid and proper upgrade path, but it does require a semi-progressive IT team.

These are the IT managers who haven't updated from IE6 yet. Something tells me they're not really that concerned about security in the first place.

Hah, this shows what you know! If a browser plugin for IE6 causes problems, we just remove more sites from the whitelist on the proxy! You can Google, but you can't follow any of the results. :-)

the same IT managers that are "supporting" deprecated browsers and os that will never be updated anymore ? Who's going to fix newly discovered exploited ?

I learned this when they killed google gears.

Your lesson should have been made loud and clear by Google App Engine.

What do you mean? Google App Engine is business as usual.

App engine had a pricing restructure a few years back that destroyed it as a business model for a lot of users. The worst was on Python where the language was missing scheduling features needed to make its use affordable... particularly nasty when Guido was working at google at the time.

Currently all those XP users can easily run Firefox or Chrome and are therefore not stuck to IE7, but I really wonder how long that will be possible. Will it be 1 or 2 years until Chrome won't run anymore on Windows XP? (Firefox has problems with the linker of older visual studio compilers already and any software compiled with a newer microsoft compiler will need the msvcrt update, unless you use http://kobyk.wordpress.com/2007/07/20/dynamically-linking-wi... )

Those XP machines will never be upgraded and many won't be replaced until there are no spare parts anymore.

One really can't trust Google to keep anything around for more than 6 months

Those XP machines can download an older version of Chrome if it ever comes to that. And if Google Chrome no longer provides support I'm sure Chromium will. XP is 10 years old. It's remarkable that Google does still support it (unlike Microsoft supporting their own software) so let's praise them on that.

Where do you download older versions of Chrome, please? I uninstalled Chrome 27 completely because it was unusable (slow, pale blue screens, crashy etc) on XP but Chrome 25 worked fine....

PS XP is 12 years old.

Many thanks! A few too many old versions, really ;-)

That older version wouldn't get security updates. And if it were to get them, but no new features, future web devs would forever be stuck with that old legacy webgl ;-)

But you are right that they should be praised for the past work.

If you're relying on more than 6 months of support (or any support really) from a free product in an enterprise environment, you're doing it wrong.

Microsoft will stop supporting XP in roughly a year. Then you either upgrade or expose yourself.

This is a huge deal for enterprise app developers. We've been able to develop on a modern web platform, but have our customers use our app with IE 6, 7 and 8 (via ChromeFrame). Now we're going to have to add support for at the very least IE8, since XP (still over 50% market share in business) isn't going anywhere anytime soon, despite Microsoft EOLing it. Most IT admins I've dealt with refuse to install any other browsers on their systems, but don't have a problem with browser plugins for some reason.

It seems like the enterprise grade solution would be to checkout the svn or git repo some time in the next 6 months and then rebrand it to "your company name here" frame.



The viability of this strategy depends on whether the Chrome/Chromium bits will continue to integrated with the "Frame" bits. I'm not getting a good feel for the architecture from just a quick glance at their repo, but I'd definitely like to find out how viable, long-term, a fork might be.

> Now we're going to have to add support for at the very least IE8 ....

> Most IT admins I've dealt with refuse to install any other

> browsers on their systems, but don't have a problem

> with browser plugins for some reason

Why don't you just deploy Chrome itself? Google has made a version of Chrome that can install without admin rights.

I have a feeling this was Google's plan all along: 1) get everybody dependent on Chrome with Chrome Frame, 2) make version of Chrome that installs without admin 3) remove support for Chrome Frame 4) $$$ profit from massive switchover to Chrome inside the enterprise as developers make an end run around IT by using the non-admin install.

Another great opportunity to pressure IT to switch to Firefox as the standard.

This is the opposite, this is just showing the c levels they were right all along, build on microsoft, 'no one every got fired for buying microsoft', because they will support their shit forever. I think i can get a supported visual fox pro environment if i want.

You can still buy a horse and buggy too. Does that make it the ideal option for transportation on the interstate?

I think you just come from a different perspective than me, i work in the enterprise for fortune 500 companies. The mindset and the process is just different. If the horse and buggy works, and is cheap, and doesn't have high maintenance costs, in the enterprise will gladly let the horse and buggy technology go until paste the day when we are flying spaceships to and from work.

I enjoy the witty snaps, but realistically, it just shows your out of touch with the enterprise space.

I am not talking client facing apps, all of those are cross browser up to date to ensure maximum visibility, i am talking intranet enterprise apps so hideous no one who isnt paid to use them ever would... these are the backroads that only the farmhands use. the overseers are like you want to throw an automatic carriage back here, but we pushed for it, and just as it rolls out i gotta tell em the automatic carriage company stopped selling gasoline, and they just look at me laugh, and say horses run on hay...

thats all it is aint the end of the world, aint even a big deal, i will write an active x control and embed it in the page to do what i need... horse and buggy style

Well... "enterprise" can sit still for as long as they want. I don't really care one way or another. Besides, eventually we'll be so far ahead that we won't even be able to hear them complain about the shrinking world they've built and refuse to leave. I've worked in government which traditionally is even worse than enterprise when it comes to things of this nature. But we had a pretty good IT department that didn't just dig in their heals and refuse to budge. Sure we weren't installing nightly builds or anything but we got fairly regular updates and stayed reasonably current. It is possible. Eventually it will catch up to you in terms of maintenance costs.

... and then these big enterprises are surprised when they get outmaneuvered by a startup using technology which less than 10 years old.

A cruise line is not going to get "outmaneuvered by a startup using technology which less than 10 year old.

Nor an insurance company.

Nor a bank.

Nor large agrobusiness.

The tech industry is not the only industry that exists in the world. There are many industries where software does not form the basis of competitive advantage. In those cases, IT is not the most efficient use of capital.

Who says the cruise line, insurance company, bank, agrobusiness and so on are going to get outmaneuvered by a startup? They're going to get outmaneuvered by another big company in their line of business with less-stupid IT policies.

Getting these sorts of issues (and bigger ones) right is something big companies call "IT strategy" and they'll often pay consultants real money for it. That's because, done well, IT can be the foundation of significant competitive advantages. Is using a more modern browser going to create an advantage all by itself? Almost certainly not. But the kind of IT organization that can't even handle something that simple is going to struggle mightily with the bigger fish.

Unlikely, as their competitive advantage does not come from IT.

For example, a bank competes on the basis of access to capital (i.e., deposits), which depends primarily on size, location, and interest rates. Given the choice between upgrading their computers and opening a new branch, it is much better to open a new branch.

There is still a ridiculous amount of enterprise web apps that are built for IE.

Just saying "switch to Firefox" or "change products" is simply not an option.

So? Old, poorly-maintained systems have all sorts of issues. Not working with newer browsers is just one of them. Consider these possibilities:

- Maintenance and support costs are notably higher than they would be with a newer system. It is worth noting that, technically speaking, Microsoft hasn't stopped supporting Windows XP. They're just raising the price and requiring a bespoke service agreement for XP support.

- Not working with newer browsers means not working with smartphones and tablets. A company could be passing up opportunities to improve productivity with mobile devices.

- It may be difficult or (practically) impossible to scale an old, brittle system with the needs of the business, especially if the old system has issues with newer hardware.

- It may not interoperate well with newer systems in the organization, foreclosing opportunities to offer valuable products and/or services.

And so on. Letting parts of your IT infrastructure bitrot has long-term consequences and companies should realize that when they make IT decisions.

One of the biggest surprises has been the grudging increase in "open" support in web apps with execs and field support staff/salespeople adopting iPads.

It's easy enough to tell someone "you can't use Mac/Linux/IE9/FFox/etc. unless you can make a case for it," when they're not contributing to the top line. Kill arguments claiming that it will save money by asking if it will increase general support costs.

When the CFO wants to be able to access the intranet with her iPad, it's a different story.

This is quite unfortunate. Chrome Frame is one of the least intrusive ways to get WebGL support for visitors using IE8 and IE9.

For those who might be looking for an alternative, you can use IEWebGL (http://iewebgl.com/) for that. The FAQ says it should run with IE6+.

I'm using it for running games with IE10 at http://store.craftstud.io/games/ along with Three.js and it seems to work quite well.

...and 10 = (

Luckily, those IE 9 users are less of an issue if IE 11 includes WebGL as they tend to be auto-upgraded. The IE 8 users are a harder case.

If Chrome Frame is necessary for you, why not continue it as a separate open source project? Google has already helped you massively, it's not their job to manage your backwards compatibility issues for you...

Because they made it nearly impossible by only releasing half the code, not releasing any installer build, and only releasing parts of the build system.

All of the code necessary to build a working Chrome Frame and an installer binary is present in the open source repository. The code for the MSI isn't there, but that's just a very thin wrapper around the Chrome Frame and Google Update exe installers.

If you've got a copy of the MSI already, run it against Dark (the WiX decompiler) and you'll be pretty much handled a copy of the MSI source. Obviously you'll have some cleanup work, but it's a hell of a start. http://wix.sourceforge.net/manual-wix2/dark.htm

Yup it certainly is... too bad that the build is not usable without extensive knowledge of how to integrate it into IE. Something which a very limited subset of people know.

Why not set the date to April 2014, the end of support of XP?

That would be the perfect date to sunset chrome frame.

I'm surprised by the negativity - I wasn't aware that Chrome Frame existed, but after reading about it, I would classify it as a neat hack. It's basically a way of shimmying in an entirely different browser. That sort of thing is, to me, clearly a stop-gap solution. I wouldn't want to support such a thing in perpetuity either.

I don't understand the problem; Chrome can be installed on XP without administrator privileges, isn't that enough for Entreprise users?

99% of monolithic web intranet application only works in IE 7 (8&9 in compatibility mode), you cant expect your user to have a separate browser for one subsection of a web app, and IE for the rest of it.

Chrome in managed environments supports this via Legacy Browser Support, which allows administrators to select (specifically down to port and URL, if necessary, or just by hostname) which sites to open with IE with the rest defaulting to Chrome (or vice versa, as is most convenient for the particular environment.)

Some organizations don't allow it by company policy, regardless of whether it can be installed.

I've still yet to wrap my mind around the Chrome frame from an IT perspective. Even with thin clients, you could install FireFox or Chrome on your XP VM, and it'd result in less CPU overhead. I'll admit, IT is not my specialty nor is mega corporate politics, but I'd would venture to guess the cost of IE6-8 in IT overhead would negate any cost of deployment of Chrome or FireFox ESR.

Anyone care to enlighten me as to why?

Chrome Frame was brilliant: users still click on the blue e, all of the horrible enterprise loathware apps continue to work but pages which opt-in are suddenly much better looking and faster.

Large organizations, particularly those which train non-IT workers to follow consistent workflows like to avoid any changes. Chrome Frame was great because it was a minimally disruptive opt-in for a better web; Chrome for Business requires you to switch to Chrome for everything and whitelist legacy apps/sites – that might happen if someone at the C level is really scared about security but it's way more risk and the shops in question are among the most risk averse.

Sadly, it's not about logic in a lot of these cases.

Some I.T. policies also disallow chrome frame. Google had provided a policy loop hole which helped ie6 look like a legitimate option. Its not anymore...

I think its time enterprise I.T managgers update their thinking, and I think the less comfortable backwards compatible tools that exist, the better.

I'd never heard of Chrome Frame and had to look it up. Apparently it is (was) a way to run Google's web browser inside Internet Explorer:

> Google Chrome Frame is a plug-in designed for Internet Explorer based on the open-source Chromium project.


Anybody familiar with the architecture of Chrome Frame? In terms of how Chrome/Chromium is integrated, I'm wondering if it would be viable to maintain a fork of the Chrome Frame code or if there are "concessions" made for Chrome Frame in the Chrome/Chromium code that, if they go away, makes it impossible for Chrome Frame to function.

It would be a fair bit of work but it wouldn't be impossible. There are really only a couple of places in the main Chrome code base that are CF-specific - the automation interface and the external tab container.

Sounds like a business opportunity for someone. Even based on this thread there seem to be a few shops that would pay for support for it.

Does this mean just Chrome Frame is vanishing, or does it include the features that enable http://code.google.com/p/chromiumembedded/ to exist/work?

Chrome Frame won't be available for consumers to install any more. Chromium Embedded is a separate project and they can maintain whatever patches they need to make things fly, methinks.

This seems more like Google is retiring support for users of old Microsoft products... considering they're offering both a newer browser, and a newer option for running web sites that requires an old browser... you're only boxed out if you insist on continuing to run an old IE, rather than the newer options available from either company. How often does Microsoft create products to support other companies' legacy users?

hahahahahahahaaa you have no idea the vast amount of windows code that is emulation layers to ensure that other companies legacy users can continue to use programs designed for dos, windows 98 etc.

Google doesnt have to do it, but like i said thats why the c levels insist on microsoft even when their products arent anywhere near the most elegant, feature rich, or performant.

That's supporting their own legacy products, not other companies'.

Nah man, he's saying Microsoft goes to extreme lengths to keep old 3rd party software running on new versions of Windows. For a long time (maybe still), there was a principle at MS that upgrading to a new version of Windows should never break a 3rd party app.

I can't speak to the emulation layers, but there are tons of verified stories about MS writing special code that actually watched the running processes and adjusted the behavior of the OS. They'd look for popular apps that utilized bugs/quirks in older version (which were remedied in new updates) and essentially emulated that bug for that one process to keep it working in the new version of Windows. Crazy, crazy stuff. Sim City was a famous example.

From Joel Spolsky: "I first heard about this from one of the developers of the hit game SimCity, who told me that there was a critical bug in his application: it used memory right after freeing it, a major no-no that happened to work OK on DOS but would not work under Windows where memory that is freed is likely to be snatched up by another running application right away. The testers on the Windows team were going through various popular applications, testing them to make sure they worked OK, but SimCity kept crashing. They reported this to the Windows developers, who disassembled SimCity, stepped through it in a debugger, found the bug, and added special code that checked if SimCity was running, and if it did, ran the memory allocator in a special mode in which you could still use memory after freeing it."

I'm no fan of MS, but that example speaks to a respectable dedication to a principle – rightly or wrongly.

Yes, Microsoft maintains tons of legacy support for user's running old software on newer platforms - but that's essentially support for their own legacy platforms. They don't go out and provide tools for say, pre-OSX Mac users to get the latest Microsoft Office. Chrome Frame is essentially software that runs on top of Microsoft's legacy, deprecated browser and allows it to have modern functionality.

As the post points out, Google Chrome does have features similar to what you describe to allow backwards-compatibility with legacy sites that need an old browser. That's not this.

Sometimes HN discussions are really funny.

People are up in arms about a free product from a company with an established track record of killing products, having nearly zero customer service and not caring about financially hurting businesses who use their tools (no-recourse account cancellations, etc.).

That's just too funny. What, you didn't know this could happen?

I know my parents are going to die, but I'm still going to be sad when it happens.

They simply want people to use Chrome, makes business sense now that they have a stronger usage. They have the option to run old IE apps within Chrome to hopefully get more corporate feudal empires to switch. They are also giving some time to migrate off it rather than just shutting it off.

Something BIG is going on at Google!!

They close one product after another, to free all resources for something. But what?

Google is actually the NSA.

As a programmer this makes my decision easier if I should invest time in Google products like Go and Dart.

No, even if they are open source, it isn't the same if Google retreats from the project.

Just terrible - most of my Enterprise users can't upgrade.

Is there an open-source alternative to Chrome Frame?

Looks like it's time for a fork.

Chrome frame is the open source alternative to chrome frame.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact