This post was taken from my blog, so be sure to check it out for more up-to-date content.
October 14, 2025 - remember this date, as this will supposedly (still not official) be the last day of IE 11 support. Why I'm telling you this? Well, because, as you might know, IE is one of these web browsers that cause headaches for many web developers. But is this really that bad, and should you even care about IE support in 2020?
A bit of history
Internet Explorer (IE) is a web browser, originally released by Microsoft in 1995. These were times when web browsers were at their infancy, and standards weren't widely-respected. JavaScript wasn't even a thing (created December 1995) and browsers often contained custom unstandardized features, addons, and plugins.
So, when Microsoft entered the market with IE in 1995, it didn't have much competition (other than Netscape Navigator). Not long after the original release MS started including IE for free in every new version of its highly-popular OS - Windows. This resulted in a drastic increase in IE usage, topping off at more than 90% market share in the early 2000s. Of course, it wasn't all without anti-trust and power-abuse controversies, but that's not important right now.
Dialogs and pop-ups like "Sorry, but this website only works on IE" went pretty wild. But, it all soon ended as Microsoft failed to improve its web browser and started introducing quirks and obstacles instead. The Web also became more accessible, and other web browsers entered the market (like Google Chrome in 2008). All of this, together with the rise of mobile (IE Mobile - there was such a thing), resulted in IE usage coming crashing-down to what now is ~1.5%.
Apparently, even Microsoft came to a conclusion that its web browser was failing. So, in 2015, together with Windows 10, it released the all-new Edge browser, with rewritten internals. However, low market share and "just install Chrome" mindset, proved too much for the new browser to handle. Microsoft had to act yet again.
Just in a few days, on January 15, 2020, a new version of Edge, with fresh new icon and Chromium at its core (the same one that powers Chrome, Opera and many other browsers), will try to once again to take back its userbase. The browser is available to download in beta right now, and I must admit - it feels OK. Kind of like Edge and Chrome combined!
Missing features
But let's get back to the main topic. Recently, when working on a redesign of this very website, I was considering what it would take to support IE. As it turns out - quite a lot! So, I'm sorry to all the <0.4% of my readers who use it as their main browser, but I'll stop supporting it in the near future. To justify my choice, let's go over a few features which a potential website that wants to support IE 11 (not even mentioning its older versions) would have to give up.
JavaScript
JS really exploded in popularity with the introduction of ES6 in 2015. Given that IE 11 was originally introduced in 2013 and replaced by Edge shortly after, in 2015, it's expected that it won't provide modern ES6 features. You might not even consider it an problem, as tools like Babel can handle the compatibility issues pretty fine. However, certain features cannot be polyfilled (replaced with "older" code). Also, as a lot of browsers support ES6, polyfilling it can only result in unnecessarily bloated code or complicated production pipeline.
EcmaScript 6
Based on the data from Can I Use, IE 11 doesn't support most of the EcmaScript 6 (ES6) features. This means no syntactical sugar like arrow functions or classes, but also more implementation-bound features, like Promises or WeakSets. Others, like (Weak)Maps, Sets and let/const variable declarations are only partially-supported. Naturally, features newer than ES6 are even less common (if they even exist).
There are a lot more of such examples, but I don't want to be nitpicky. Old versions of other browsers also don't provide certain features, but they're either frequently (and seamlessly) updated or not too popular.
Web APIs
While not part of the JavaScript itself, Web APIs allow for some really unique functionalities to be used on the Web. However, unlike syntax-related features, these, in most cases, are impossible to polyfill.
From the more relevant Web APIs, IE lacks support for Fetch API, Web Notifications and WebRTC. And while the Fetch API can be polyfilled with the use of XMLHttpRequest, the other two cannot. Thankfully, both Notification API and WebRTC are meant for modern, feature-rich web apps, that aren't even targeting IE in the first place.
There are also examples of Web APIs having only partial support. The most notable one is probably WebGL. WebGL 2 support is obviously not present, but that's expected. What's more important is the fact that IE 11 still requires "experimental-webgl" identifier to access WebGL context, rather than the standard "webgl".
HTML/CSS
If you want it hard enough, you can make a website without JavaScript. When considering Server-Side Rendering (SSR) or JAMStack (static websites) - it's not actually that hard. But there's no way of avoiding CSS, let alone HTML! And sadly enough, features-related issues got even here! Let's name a few of them.
HTML
From the HTML side, things look pretty good - if you consider partial HTML5 support "good". There aren't many features lacking - just those that were introduced a bit later than the standard itself, so it's fine.
CSS
CSS, however, is a whole another story. Mostly it's full of partially-supported, but really important features like Flexbox, Grid, CSS variables and viewport units (e.g. vmax). Some only require a prefix but others either lack certain functionalities or support older, incompatible versions of the specs. CSS can still be processed with tools like PostCSS, but it's still bad, considering that most web browsers fully support the listed features.
Case-study
For the purpose of this post, I left my Linux comfort zone and went on a journey to Windows 10's darkest corners - the IE 11. I must admit - the experience and performance of the browser itself leave a lot to be desired. I also couldn't help but reflect on all the memories and jokes that came to my mind. ;) Anyway, as we now have some knowledge about IE 11 limited feature set, let's browse a few websites and see how they work!
Areknawo
I'll start with my blog. In its current version, it works pretty well. Only the JavaScript part of it doesn't work. The ad at the top doesn't show up, the newsletter subscription box and the Disqus comments under each blog post disappeared and the AJAX page transitions don't work. All of this thanks to... unsupported ES6 template literals being used in my code!
Honestly, I don't think I'm going to fix this little issue - especially when working on IE-incompatible redesign. It just doesn't make much sense. This blog is directed towards web developers and "techy" people that usually use all the latest and greatest tools. Most of the targeted demographic simply doesn't use IE and even if... maybe only for testing. ;)
YouTube
Opening YouTube on IE feels like going back in time. Everything works, but the UI is simply outdated. Seems like Google left the browser with the last redesign that it's compatible with. Smart move. But for smaller organizations and companies, maintaining old versions of the same website might be somewhat inefficient.
GitHub
GitHub simply tells you that you're using an unsupported browser. Funny is the fact that it's now owned by Microsoft itself. But I don't blame them - they did the right thing. Anyway, you can dismiss the little dialog, but right off the bat, you'll be greeted with a broken toolbar. The login page also seems to be broken and shows a loader all the time. I didn't try to log in, as I'm a bit afraid about the security of that thing - there have been multiple security issues with IE in the past.
CodePen
CodePen also shows a dialog, but much bigger and "undismissable". It says that PRO members can use it with so-called Debug View, but as I haven't tested it, I've got nothing more to say.
CSS-Tricks
CSS-Tricks is an example of a simple website - it doesn't use much JS and features mostly text-heavy content. There's no dialog or info - just a somewhat broken page. Styles are off and things disappear, but at least the content and posts are readable.
Others
There's not enough space for this post to contain all the "anti-IE" websites. I still have some honorable mentions, though! Slack and Trello - very popular productivity tools simply don't allow you to log in. And even Apple's homepage layout is broken! Other pages look fine, although without fancy scrolling effects and you cannot buy anything unless you "upgrade your browser".
If you still want more examples than that, feel free to open IE 11 (if you're using Windows 10 - it's probably still there) and check out some of the websites you're usually visiting. Along the way, you'll feel the pain that I've gone through! ;)
Final words
The main goal of this post is to tell you that supporting IE is not that important anymore. You should feel a bit more freedom when it comes to using modern features. Especially when considering new projects.
I've heard of enterprises that rely upon IE-only code and cannot afford to upgrade. In my humble opinion, this is just bad design - no offense. The Web was and is a very dynamic place where you have to adapt and change in order to persist. If the design of your app doesn't let you do that, then it's not good. That's just my opinion. As a matter of fact, I even experienced such a website, which, funnily enough, didn't even work on IE 11! It required the previous version of the browser to properly work - even though none of them are supported anymore!
So, unless you're targeting a very broad or specific demographic, I'd say that you shouldn't care that much about IE. If supporting it doesn't cost you anything or doesn't limit your product's functionalities, then go for it! But, based on the features we've discussed, it seems the reality is often different...
Anyway, that's it! Let me know down in the comments what do you think about the whole IE support thing and if your website supports it. Also, if you like this post consider sharing it and following me on Twitter, Facebook or simply checking out my personal blog to keep up-to-date with all of the latest content. If you're interested, I also got a YouTube channel, which you can check out and drop a like or a sub there! As always, thank you very much for reading this piece, and have a nice day!
Top comments (23)
Yes, yes and a thousand times yes, using an older browser is insecure for the user and to get a new one is free, you have a lot of choices. None would argue that not supporting Netscape is an accessibility problem, and as a community we should punish non standard implementations.
For years we suffered with IE only apps and we as nice guys and gals made an extra effort to support IE, we limit ourselves, we worked extra to appease IE users which only resulted in Microsoft keeping their practices. If no other free browsers existed I would agree, we should support IE, but it was never the case.
We as devs, kept IE alive, we kept it crappy. A bad product has to die and that process starts by stop being supported.
Better spend you time making your site readable by real accessibility tools, for people with visibility problems, make your sites lighter so you don't need a beast to run it, only add JS if necessary and as little as possible; that is well spent time, supporting old obsolete browsers when there are so many modern and free ones readily available, to me is not just a waste of our time is also hindering progress.
look at me, IE made me all riled up...
No one is willingly using IE. No one. They are either locked in frozen corporate / government devices or don't know any better. By dropping support you're not punishing a browser, you're punishing people that are not tech savvy enough to even know what a browser is, let alone install a new one.
If locked in corporate environments they should be pressured because old browsers are not only "ugly" they are unsafe and "corporate" can easily allow the install of other browser or even installed themselves.
If someone doesn't know better, they'll learn, people didn't know how to use an email, people didn't know how to use a mouse; if you suspect your target audience may not know that other browsers exists, add some links to other browser, but that is for a very specific niche of users, I wouldn't blame if dev.to doesn't support IE, if you don't know how to install a browser, you'll not understand a word on the site anyway.
I suspect nobody worries about Androids 1.5 browser, and that is a real lock in for people, because in an older phone you can't put a modern browser and many people only have access to older phones; unlike desktop browsers.
if you don't know how to install a browser, you'll not understand a word on the site anyway.is the most elitist thing I've read on dev.to so far.ehm, given that most of the content of this website is technical and 99% related to web development, I'm pretty confident that if you code for a browser you should know what a browser is. That's not elitist, is common sense.
A quick check on the definition of elitist
"but that is for a very specific niche of users, I wouldn't blame if dev.to doesn't support IE, if you don't know how to install a browser, you'll not understand a word on the site anyway."
Is fair that a woodworking site assume that their visitors know what wood is, getting offended by that obvious assumption is ridiculous.
I must say your comment is is the most selfrighteous thing I've read on dev.to so far.
I realize my mistake. I thought you were talking about ALL websites, not just Dev.to. my apologies.
no problem, IE is a controverted topic it makes me all riled up too :D specially because as a Linux user those IE only years where tuff.
the memories, the horror...
Haha indeed.
The new Edge has an IE mode for websites that require IE. Specifically in enterprise environments. You're punishing users and devs if you make concessions or avoid new features to cater to IE
While I agree that many popular brands and sites fail to support IE, I disagree that we should all follow suite.
I think you can only make that decision after having some data on your traffic. If your 10% or 20% of your visitors are coming from IE, is it really worth hampering their experience? I've worked on ecommerce sites with nearly 60% of their traffic coming from users on IE.
In my opinion, software exists to solve problems for humans and it should take into consideration the constraints of those humans; some people aren't going to be able to move from IE to a different browser. I believe our software shouldn't exclude those people when we have the means to support them.
Granted, I'm one opinion, I'm sure there are fine arguments against supporting IE.
I agree with you. I have not yet to meet the customer that is willing to miss any visitor.
I work mostly on enterprise customers and there on IE11 (not Edge) is still rolled out on Windows 10 when employee gets new PC. I don't expect this to change until Microsoft drops IE11 support entirely. Of course for hobby websites it is fine to drop IE support but for the rest of the internet supporting IE should be as important as supporting accessibility.
From our most recent stats we have 11% on IE11 so for us it's definitely an accessibility issue. We could of course lose those customers but that's anathema for our business colleagues, so we are attempting to measure the support cost against their value.. tricky!
Yeah, the enterprise problem. I don't have experience with such things, so I won't comment. But, IE support being not important "for hobby websites" is, in my opinion, not true. Not only because websites such as GitHub also drop it, but also because some of these websites might target IE-focused demographic.
can you remind me who owns Github nowdays?, ha, not even MS supports IE, even them recognize the annoyance. Windows should offer other browsers at install time and MS drop their browser efforts, why do they want a browser after all, they are collecting everything at OS level anyway. It even must be annoying for them to deal with all duplicated collected data. :D
for enterprise customers is just lazy IT, if you have an old SAP (or something alike) that needs IE, just get two browsers, you shouldn't use IE in the wild anyway, is too insecure; you can keep your IE just for your enterprise "IE only" website. The rest are just excuses from the IT department not wanting to do their job.
I just came to comment for nostalgic reasons as I have written web code for around 20 years & have watched its evolution & I'm finally happy to see IE almost gone. I remember the time when learning HTML/CSS/JS where the main advice was to code for IE only. It's great to see the exact opposite. I was also around for the transition from coding strictly for IE to creating a secondary rule set for IE (<!--[if IE]>). Edge is a damn good improvement, but I still stick to Brave or FF for my main. I am a dual OS user so I prefer cross-platforms as my mains. Great article. Many younger devs might not ever know the struggles of the fiercest browser wars lol.
In line with what some of the other people are saying, I have 10%...I say again, 10% of my user base for an app I manage that still uses IE. To say those people aren't worthy of a good experience is incredibly elitist. I chose my stack knowing full well I'd have to support older browsers.
As mentioned by Facundo,
you're punishing people that are not tech savvy enough to even know what a browser isThere are no carte blanche statements in web development, and saying stuff like is in this article is a detriment to what we are doing as developers.
For sure. I think Areknawo is working off the assumption of a 1.5% global usage stat though, I'd include IE support if 10% of my user base was on it too.
Supporting IE is not about IE support. Is about making webs and apps with progressive enhancement so any device with any browser can use it.
It's easy to focus on the IE market share, but what about the sum of all other minor browsers, including old versions of the evergreen ones? There are quite some people running frozen devices (government and corporate) that simple cannot upgrade, and when we check the stats, we can see that they normally have much greater share than IE.
Consider for instance the support for Grid. We can somewhat support the old spec that IE 10 / 11 uses, even automate that with PostCSS and Autoprefixer in the build. But for the sake of the argument, let's just consider their 1.52% as "unsupported"... well, if we check the stats for grid support at caniuse, we'll see that minor browsers + old versions of current browsers add up to a 6.47%.
Focusing in IE greatly underestimates the issue, and can be causing bad decisions.
This doesn't mean we should be coding like in the dark ages, we can and should be using those awesome new features of CSS, but always considering them as progressive enhancement layers. It doesn't even take that much effort, as mobile layouts are almost always done in a blocky way that could be served as the fallback too. So from the CSS point of view, I believe we should always be serving resilient, progressively enhanced layouts.
When it comes to JS, I totally agree that some of the features simply cannot be transpiled and this can leave us with no other choice but to drop support of older / lacking browsers. Same thing for the many, many web APIs. But we should be really careful when making the decision. Tools like Codepen can get away with no supporting IE at all, as they need the modern features for their system, and their audience is 100% web devs / designers. But for the average website, we can transpile most of the stuff, put some thought on whether we really need that non-transpilable feature, and consider which APIs are safe to use as progressive enhancement and which ones are simply scratching the itch of using the most recent stuff just for the sake of using it.
All in all, we should be really conscious and deliberate on the decision. It's easy to mask it as "we're dropping support for that archaic browser that's keeping the web in the dark ages", but the truth is that we are dropping support for the old, the poor, and those locked in frozen devices. It's people that we're leaving behind, not a browser.
Very insightful comment! I agree. I think it all depends on the demographic you're targetting. If you're willing to narrow it a bit - it's your choice. But you'll certainly never make everyone in the world happy.
It's all also region-depended. Here's a nice website that somewhat shows the usage of IE around the globe. If your website is targetting people leaving in areas where IE is still very much in use, you have no choice but to support it.
Wow, well done!
I was skeptical when I saw the title, expecting to read a list of complaints like I've heard from so many other devs, but look at all those facts & statistics. What a great resource, thanks for sharing.
The reason we have to support old browsers is because we keep supporting old browsers..