37signals to Drop IE6 Support | Hacker News Reader

37signals to Drop IE6 Support

37signals.blogs.com

77 pointsbdotdub74 comments

I think this says more about their customer base than it does about the state of IE6 in the market as a whole. There are lots of IE6 users still out there, many of whom are using corporate installations that they can't upgrade. If 37signals (did I spell it right?) doesn't need that market, more power to them. But the reality is that better browsers just aren't as simple as providing a list of "click here" installation links.

I agree. They customer base (I think) is mostly small businesses, which have the luxury of upgrading browsers.

Any effort to push the web forward (away from old browsers) gets a thumbs up in my book.

On the consumer Internet sites I have worked on in the past year, the stats have shown roughly 30% of users still have IE6. For the services that 37Signals provides, it's quite likely that this drops to just a few percent -- after all, their customers are tech-savvy enough to be using Firefox, Safari, and IE7. And when you've got just a few percent of IE6 users, it just doesn't make good financial sense to spend a disproportionate amount of time dealing with IE6's inane bugs.

Agreed on the customer-base point.

I'd guess their ie6 visitors are <3% - and I'd wager making stuff work for ie6 requires a disproportionate amount of their time, which could be spent serving the majority more effectively.

As a comparison, RescueTime's ie6 visits weighed in at about 3.7% over the last 30 days.

That's surprising (but great news!) that your IE6 visitors are that low.

I agree, I'd imagine they have comparable numbers to make a decision like this. It'd be interesting info to know (anyone from 37signals out there? ;)

One of the 280Slides guys said their IE6 # was 6% or so, as another data point.

Firefox 3 actually installs in some locked-down corporate networks. I was able to install FF3 (just not in /program files/), even though I can't install much else--including upgrading flash!

I can only assume that FF3 was cleverly designed to be installable more widely than most other programs, which refuse to install.

You know, I do not write software for Windows 2000. They are about 10% of my market, but I don't bother testing on 2000, because I choose to focus on the XP and up market. Nobody even questions this policy or finds it controversial. Why should a website not also decide to focus on 90% of its users so that it can make a better product?

Why should a website not also decide to focus on 90% of its users so that it can make a better product?

Because some people just don't want to leave a dime from every dollar on the table. That's all.

if only the math were that simple. Chances are that focusing on 90% of the market != 10% decline in revenue.

And that's exactly why some people want every eyeball they can get. Because they don't know.

37Signals isn't really in the "eyeballs" business.

If the time you save from ignoring an expensive 10% of your customers lets you grow your total market by 12%, you're into profit.

Because some people just don't want to leave a dime from every dollar on the table. That's all.

But at some point, if that dime has to come at an increase of 20% or 30% in effort, it stops being worth it.

Right. A good argument. Which you have to convert into a convincing argument with metrics. In the absence of other metrics, eyeballs wins.

Unless it costs $0.25 to grab that dime.

If you really think this, then you have no experience running commercial software on multiple platforms. The time spent making software compatible across platforms is way more expensive than the money you gain from selling on that platform.

For example, I have some shareware I sell. Brings me about $3000 a month. When I took Windows 2000 off the list, revenues dropped to about $2700 a month. However, about 50% of support requests could now be replied with "Platform not supported" because I could finally use new APIs that were only available on XP.

Your idealistic dreamy eyed attitude will be tempered by experience at some point.

Apple will also be dropping IE6 support in their upcoming MobileMe service. A very consumer oriented product.

http://www.tuaw.com/2008/06/23/mobileme-browser-requirements...

... yet another island of users who wouldn't have IE6 anyway. I'd be way more impressed with either Google/Yahoo/MSN dropping IE6 - THAT would be an awesome gift for all of us.

sorta. I mean the point of MobileMe is that you can access your email, calendar, etc... from any browser. But in this case, it means ie 7+. So it's not really meant for people who sit on their Mac at home perse.

Also, it is aimed at iPhone users, which is a fair percentage windows users.

Of those three, who do you think is most likely to jump first? Yahoo is very corporate-friendly, so I'd expect them to be the last to give up IE6 support. MSN might want to push everyone towards the latest IE version so Microsoft can finally stop supporting IE6 altogether; Google might push for Firefox, but then they might draw some antitrust criticism for that.

Since 37signals makes an API available for most of its services, it would be theoretically possible for someone to come along and build an alternative IE6 compatible front-end to 37signals' products. I expect, however, no-one will, demonstrating that there is no economic reason to support it.

I can't imagine such a service having good growth...

Many sources [1] show that IE6 still has more global usage than IE7. Though I'm sure 37signals' userbase is not represented well by global data, turning away users of perhaps the most popular version of the most popular browser is an interesting move.

[1] http://en.wikipedia.org/wiki/Usage_share_of_web_browsers

Where did you get that from?

I clicked through the links on that page and the most recent month had this:

http://www.thecounter.com/stats/2008/April/browser.php

What is really amusing about those stats is the 77 visitors still using IE 3. I don't even have a 12 year old computer that still functions, let alone a 12 year old browser.

No, 77 is the age of the IE3 users.

FWIW, the first one I clicked on was this:

http://www.w3counter.com/globalstats.php?date=2008-05-31

Not that it's the best source of data, but.

God, I didn't realize IE6 hit the streets in 2001. What is it that makes corporate cultures so slow in upgrading to IE7, especially if they are locked into the Windows platform?

I get the impression that many companies have internal applications that were written to be compatible with IE6... software written by employees or consultants who are no longer available.

The mere act of enumerating IE7 incompatibilities in these applications (let alone fixing them) represents a development cost that some companies will put off for as long as possible.

why can't they just let people install firefox, and use ie6 for internal apps, and firefox for external apps?

In some cases, because the company in question feels you shouldn't need external access at work.

If the company doesn't allow external internet access, it isn't relevant to anyone deciding what browsers their website should support.

Just because an employee shouldn't access the internet at such companies, doesn't mean he/she doesn't

At my Fortune 500 company, permission to use Firefox is only granted to programmers who can convincingly argue that they need it for compatibility testing on Internet-facing applications. It's IE6 for everything else. The reason is that Firefox is a security risk(!).

Because the people that do desktop support don't want to support more browsers than the minimum possible. I'd bet a decent proportion of desktop support calls are about 'problems with the internet' (read: browser) and starting every one of those call with "are you using IE or Firefox?" would most likely get answered with: "I dunno! I want to use Google!"

heh

"we got a problem here, uh yeah, the website is down"

http://www.thewebsiteisdown.com/

Would it be possible to write a proxy that converts IE6-HTML/CSS into good-HTML/CSS?

Then you could upgrade $(Initech) to IE7/Firefox, and slap the proxy between the users and the legacy web apps, and everybody's happy. When the internal developers upgrade the legacy web app, they can either keep it on the un-IE6-proxy, or upgrade its output and lose the proxy. IE6 can go away, we can stop wasting time with IE6 compatibility, everybody's happy, world peace, the end.

Disclaimer: I don't understand the nature of IE6 bugs very well at all, and I don't know if this is turing-complete or AI-complete or what.

Tricky, since most of these apps which can run in IE6 but not IE7 do bad things in JavaScript too. That means you can't just fix CSS issues in a proxy, because even if you could document every IE6 bug and how to fix it for IE7, the page could still change after it is loaded in the browser.

When IE6 was released isn't the issue, it's when IE7, Microsoft's replacement for IE6, was released. October 2006. We are quickly approaching IE7's two year birthday. Two years is still a long time.

Those guys are lucky to be able to do that. Our user base is still 35% IE6.

I just think that it's funny that this announcement comes on the heels of the "features are a one way street" posting...not that IE6 is a feature of any kind of course.

Its well worded. I imagine the original drafts would have been pretty harsh on IE6 (although 2001 - that was a LOOONG time ago, to be fair).

say that to mainstream users!

How I wish I could do that as well.

Why are we making this change? IE 6 is a last-generation browser. This means that IE 6 can't provide the same web experience that modern browsers can.

For example?

If I didn't know any better, I'd think that this was the same old "IT B.S." that we've been listening to for 30 years. You have to upgrade for our convenience but we tell you that it's for yours.

I know of many companies that don't want to pay maintenance. They say, "I don't want any upgrades and I don't want to pay for them. It works perfectly now. If I want anything to change, I'll ask for it and I'll pay for it. Why should this be any different that anything else that I own?"

Is this move for the benefit of the supplier or the customer?

Honestly, I've constantly see remarks that say you're lame or old fashioned if you still use IE6, but I sure would like to hear one of these posts provide some concrete data why. (And please don't say tabs or extensions, those are features many do not see a need for.)

Good points. However, the benefit of the supplier and customer can coincide. If it's easier for 37signals to make their products, they can lower prices or add extra features or add polish or something.

Maybe they consider ie6 support a feature, and they think their customers will prefer some other features instead.

benefit of the supplier and customer can coincide

Most customers understand this and are willing to go along for the sake of a valued supplier. If they said, "We need you to upgrade to make life easier for us to provide for you in the future," most would say OK.

But when they say, "IE 6 can't provide the same web experience that modern browsers can," inquiring customers want to know what that means before they commit. My comment wasn't because I disagreed with 37signal's move, it was because they never provided any supporting data.

One shouldn't need to include common knowledge as supporting data.

Clarify this?

It is common knowledge that IE 6 has serious bugs and problems. I'm not going to rehash them, because they were discussed in other responses to edw519's initial comment.

It's exactly this attitude that has given IT the black eye that we still haven't recovered from.

What's "common" to you may not be "common" to others. Especially your customers. Stop educating them and die.

It's exactly this attitude that has given internet forums the black eye that they still haven't recovered from.

Stop being harsh with people, and don't tell them to die.

-----

Sorry for the mirroring; I know it's often done in spite; it's just irresistibly amusing; and seriously I don't think it's good to tell people to die.

I think you misread his post. Pretty sure he meant:

"If you stop educating them, you will die."

At least that is how I read it.

Oops, you're right. I usually blame misinterpretations on the writer, not the reader. Sorry. I can't imagine saying what he thought I said.

Just an expression to emphasize how important it is to keep educating your customers, especially if you're in tech.

Sorry.

Why is IT, alone out of every other business division, expected to provide common knowledge reasoning behind their decisions?

Does it matter which one it is in the end?

If they used the first line, I feel like the customers would be less inclined to upgrade and be averse to the idea of upgrading. "We're the customers, you should make life easier for us."

The new browser technology will never take off if people don't start switching over, even if the use some finesse in their wording.

I feel like the customers would be less inclined to upgrade and be averse to the idea of upgrading

Not my experience. Customers will generally do whatever it takes to insure a good supply from a good vendor.

new browser technology will never take off if people don't start switching over

True. So what's so hard about telling them why you want them to do something?

It's not hard, but how many people will really spend the time reading about it?

In my experience, customers with a non tech background don't care about the details, but rather if it will be better or not. If it will, great.

But I agree, it's being slightly disingenuous to the people who do want to know why, or are curious what 'web experience' they are missing out on.

They are likely banking on IE6 users also often have FireFox installed too. Chances are they won't loose any usage over this and I believe that 37Signals has an amazing ability to generate hype they will see more customers not less.

Myself, I have a lot of nearly computer illiterate people using my system with over 30% IE 6. If I can't get something to work in every browser that people use I just don't use it.

Semi-transparent *.png and ":hover" support and lots of Javascript and CSS bugs

Thank you. Now we're getting somewhere.

Try submitting an AJAX get request with a variable than ends with UTF-8 data (chinese glyph) and has been fetched from the DOM.

I hate IE6.

Aren't there libraries that handle that kind of stuff for you?

Lots and lots of CSS bugs. So many developer hours have been wasted debugging IE6 issues it's obscene.

I got fed up and decided not to make things work anymore with IE, when I realised I had a folder of little tweaks that I was including in everything called 'IE Fixes and Workarounds'

While, yes, it probably is more convenient for 37signals than it is for the customers, it would be beneficial for the customer going forward.

There are a lot of CSS hacks that need to be done for IE6, which makes developing annoying and tedious, and takes away time from developing potentially better features.

Also, they can start writing code that capitalizes on better standards support and newer technology that IE6 does not have, giving the users a better app to use.

"I don't want any upgrades and I don't want to pay for them. It works perfectly now. If I want anything to change, I'll ask for it and I'll pay for it. Why should this be any different that anything else that I own?"

But with hosted apps, you don't own (or license) anything. You're paying for access to something that will change over time, and may go away.

The ability to snapshot your environment and stand still while the world moves forward is dying. Even if you stick with desktop software (where you can install version 4.5.2s1 and stick with it), if it has any connectivity features, a vulnerability will mean you have to upgrade sooner or later.

Not sure how this applies to "30 years" of "IT BS". Previously, with shrinkwrapped, if you bought something, and it worked fine, you were in a position not to upgrade, and not have to pay for it. And if it works fine, you don't need support for it either. (this, of course, ignores the network effects of everyone else upgrading other than you, which forces you to, but this is not the provider's fault, other than creating built-in obsolescence).

Now, you maintain a relationship with the provider of your web-based software (one could argue that Internet Explorer does not provide a good "web" experience because it is so broken), it doesn't come in a shrinkwrapped package that you install and use. How this is different from anything else the customer owns is that the customer doesn't own anything in this case, they are paying for a service (technically, because of the law and license agreements, they don't "own" shrinkwrapped software either, but that's a side point). The service they were paying for ("ability to use the service on IE6") is disappearing. It's not economically viable for IE6 to be supported anymore (especially for user who are not interested in paying for maintenance) -- are IE6 users part of the long tail?

It's not 37signals that is forcing you to upgrade to use their service. It's Microsoft, who wrote the busted browser to begin with. It's also Microsoft for not easily allowing multiple versions Internet Explorer to co-exist on the same computer, thereby making upgrades more difficult to people to deal with.

As for if it is a benefit for the supplier for the customer, it's a benefit for both. Customers get a better, more consistent experience (including features deployed faster, quicker load times, both because there is less code to maintain), and suppliers do less work for providing the same kind of product.

(obligatory car analogy) Does anyone complain the same way about parking garages that have height limits, and about how they are losing customers because they don't cater to customers who drive oversized vehicles? They just choose not to support the incompatible customer.

This is just laziness. They should support IE6 until it is the minority of installs on all machines.

37signals has always extolled the mantra of usability. But, this move puts them entirely against that. You need to work within the constraints of the user, and one of them, unfortunately, is IE6.

There tends to be a very fine line between "laziness" and "increased productivity". Having to spend umpteen hours supporting legacy browsers can be seen on both sides of that fence.

From an operational perspective if they feel that this won't have too extreme an impact on their customers I think it makes a ton of sense from a productivity standpoint.

Huh.

So do you think it's better for most users if they can fix bugs faster, respond to support requests faster, build new products, add new features (<gasp>) etc? I think it is. And I think blowing off IE6 will allow them to server other folks better.

Time is always allocated somewhere-- the time they save by giving up IE6 support can go to more constructive/forward thinking pursuits.

Only 37s has access to their web stats-- if ie6 is vanishingly small within their paying userbase, it's a smart move.

At some point forcing users to upgrade is to their benefit. Giving them support makes them think there is nothing wrong with using IE6 when there are better alternatives available.

No one complains about IE5.5 not being supported by most sites anymore, but people still use it.

Reply on news.ycombinator.com