[DanBC]

I can understand why you make the requirements for your app very clear.

I cannot understand why you'd chose to bounce potential customers from your website based on their browser. Isn't this just you failing to provide information to people who have made an effort to contact you and ask for it?

I ask this next question gently: Have you checked you're not bouncing people using assistive technology?

[kentbrew]

Speaking only on my own behalf and not as part of Lexity, for whom I no longer work, I'd have to say that I'm fine with the risk of losing business from people who are unwilling or unable to update from a seven-year-old browser, "assistive" or not.

Sorry to sound short about this, but it's the same sort of thing that happens when a road turns into a freeway. For everyone's safety, you don't get to ride your bicycle.

[aristus]

The OP means "assistive" as in "tools used by people who are disabled". It's not a choice to be blind.

[icambron]

What is the relationship between "assistive technology" and "old browser"? It would surprise me if that tech was only available for IE7.

[gte910h]

The apps they use probably have a web browser embedded in them. They may not have 5k for a brand new version

[jlawer]

No... generally its software like JAWS (http://www.freedomscientific.com/products/fs/jaws-product-pa...) that runs on stock windows.

No special hardware running antiquated hardware, just a stock windows system.

While I don't doubt there are solutions with a web browser embedded in them, the refresh cycle on such hardware is so old that people have moved off them on to more "generic" solutions a while ago.

[chris_wot]

I'd be very curious as to which apps these are. Eventually, they'll be left in the cold regardless. In at least two years time there will hopefully be nobody supporting IE6!

[DanBC]

Bob sells software. To run that software Ann needs X, Y, and Z.

When Bob tells Ann that she must have X, Y, and Z to run the software he is giving her important information. When Bob tells Ann (who is blind, or deaf, or has a motor-control disability (or just likes the software she uses), but who also buys the software her engineers use) that she cannot get information about the product that he sells because she uses the "wrong browser" Bob is being stupid.

You might not like the fact that people use different computers and different operating systems and different browsers and different settings on those browsers. You may not like the fact that WWW was an attempt to solve compatibility problems across those variety of systems.

But if you are trying to sell me something, and I visit your website to get information about what you are selling, and you tell me that I'm using the wrong browser to get that information - well, I'm probably not going to spend my money with you. This is especially true if the information you're giving is trivially easy to give as text with a few diagrams (and even diagrams can usually go with good alt and longdesc attributes.). That was available in HTML 1.x[1] I understand the desire to give everything sliding transitions and round corners and shiny overlays and etc. I understand that CSS and correct design help impart meaning that isn't available in plain text.[2] But it is incredibly frustrating to be told that you cannot get simple information because you're not using what someone else thinks you should be using.

Remember that it is trivially easy for me to use the browser you want me to use, on the the OS you want me to use, on the computer you want me to use[3] and yet still totally destroy the design by increasing the font size. People used to specify font sizes. Most people know realise that's a stupid idea. (Imagine a TV station that broadcast a programme at a fixed volume, overriding the viewer's tv volume control. You'd call that station stupid.)

[1] (http://1997.webhistory.org/www.lists/www-talk.1993q1/0182.ht...)

[1] (http://www.alanflavell.org.uk/alt/alt-text.html#howlers)

[1] (http://www.alanflavell.org.uk/alt/)

[2] A List Apart used to have a nice page about this, but I can't find it at the moment.

[3] Baffling that people say that a 1 GHz machine with 256 MB ram is insufficient power to browse a website, but people do say that. And not forgetting that IE on Mac used to be created by a completely different team to IE on Windows. God knows who the poor souls who did IE on Unix were. See also the unreasonable bandwidth requirements for many websites. Sure, do what you like for fun, but again if I'm buying something from you I want a quick responsive website even if I'm on a stupid slow Australian link, or on some dial-up in the US (which is still surprisingly common) or if I'm using my mobile broadband dongle on a train, with a poor signal.

[esbwhat]

This all sounds nice and reasonable, until you realize that these companies have to pay people to actually make it work in a browser that's 10 years old. Their website is a product, and they can choose who to offer it to. This is like complaining that this android app won't work on your 10 year old dumbphone.

Yes, it sucks that some screen reading software depends on outdated browsers, but the fault lies at the manufacturers of that software, not every single website designer in the world. IE7 doesn't work, even for simple layouts, it doesn't work. Furthermore, it's insecure. Maybe they didn't want their customers accounts being compromised because they use an insecure browser, and then have to hire more people to take care of that person who lost their account becuase they're using software a decade old. I don't go into a store expecting to be able to still buy heroin as my cough medicine of choice, either.

Why should everyone have to pay money to make stuff work in outdated software because the screen reader company refuses to pay money to make their software work with anything but outdated software? Why make compromises and waste time when implementing a new feature because it might screw up the viewing experience in your decade old browser? Why support that browser at all, giving people the impression the site is broken, and not their browser?

[DanBC]

> until you realize that these companies have to pay people to actually make it work in a browser that's 10 years old

Text and images don't work in ten year old browsers?

> Their website is a product, and they can choose who to offer it to.

And, as I've said more than once, it's fine if their product only works in some browsers and can only be used by some people. (Except, you know, there's some laws you need to comply with.)

But the product part of the website is different from the sales part of the website. Why not make a sales website that degrades gracefully to text and images?

> This is like complaining that this android app won't work on your 10 year old dumbphone.

No. It's like complaining that a listing of software for smartphones is unreadable on anything but the very latest generation of smartphones. (Except it isn't because all analogies are awful.)

> Why should everyone have to pay money to make stuff work in outdated software

This isn't about screen reader software. This is about compliance with web standards for information that is an excellent candidate for standards compliance. Remember that I'm not talking about web apps or software here - I'm only talking about websites describing software, websites that give information to potential customers.

> Why make compromises and waste time when implementing a new feature because it might screw up the viewing experience in your decade old browser?

Because these are not apps that I'm talking about. It's information. Why would you take something simple like text and images and a bit of CSS and make it hard for people to use it on the device of their choice? Why do people make websites that can't be viewed on a mobile phone? What features do they absolutely need that can't be provided on a mobile phone? Usually, it appears to be a mistake they made.

> the viewing experience

FUCK the viewing experience. Seriously. My life has been transformed reading pages of plain text. My life hasn't been transformed by a really nice CSS border. I want to read information. I don't want some idiot designer to mess it around. Let designers do the hard work - give it a nice font, give it a nice clean layout, make it useful. (And this is where good CSS is important - it's hard to make things clean and minimal.) See this example of an error, from the only other tab I had open, Chrome on OSX Snow Leopard. Not an obscure browser.

(http://imgur.com/9Ts2b)

The link to, I think, the links to view // communications are obscured by the grey background to the menu bar and search box. (Notice also that the top line is partly dark grey while the second line is light grey, leading to a confusing mix of colours for menu items.)

> I don't go into a store expecting to be able to still buy heroin as my cough medicine of choice, either.

A weird analogy. Say you went into a store to ask about information for blood glucose testing meters. You ask the clerk for information. She shows you 4 machines. She gives you manufacturer leaflets, and prices, for 4 machines. But you notice there are 5 machines on the shelf.

"What about that one?" you ask.

"Oh that," she says, "that's the new Blood-o-matic-9001. It's a great machine. Do you have a computer with Windows 8 on you?"

"Well, er, no, I don't." you reply, wondering why she'd ask such a thing. You only want information about it.

"That's a shame," she says "we'll only give information about that machine to people using Windows 8. Come back when you have it and we'll give you information."

Can you see that the app (the meter) can be flash-bang-whizzy and can set limits on the use because that's what apps do, but that it's stupid to set limits on what is essentially a bit of text and graphics just because you wish to force users to "enjoy" the viewing experience of transitions and curved borders.

I guess, because you've mentioned it, that I strongly agree that people should not support broken browsers. But that's very different from not allowing websites to gracefully degrade. I apologise for not being clear enough to convey my point. Communicaing my meaning is something I need to work on.

[kentbrew]

Key phrase, there: "it's insecure." Dreadfully so, and by encouraging its use, we keep the Internet unsafe for everyone.

Still waiting for someone to tell me exactly which brand of "assistive technology" absolutely requires IE<8, and why Chrome Frame can't fix it.

[kamjam]

Have you gone to the site in question? I agree with DanBC, however I don't expect you to fully support all users if you don't want to, that's your perogative, you know your customer base better that we do so it's for you to decide what browsers and technology set you want to support.

I think what DanBC is saying is if you are not going to support something, then tell the user why and they need. The user gets redirected to http://lexity.com/aiee.html which if you visit in IE6 or 7 puts you in an infinite loop... works fine in browsers that actually the site actually supports, kinda ironic.

[kentbrew]

Whoops, I need to let those guys know that aiee is broken. Thanks!

[talmand]

"IE7 doesn't work, even for simple layouts, it doesn't work."

What in the world are you talking about?

[kamjam]

From this I get 1. A server side developer (C#/Java/PHP/Whatever) trying to be a HTML designer 2. A n00b HTML designer that's not been around long enough to know the different IE box models and workarounds needed.

I'd guess 2. You can get most layouts to work in IE7, and with a lot of effort, IE6 even. Agree on the security aspect, but bear in mind that the majority of home users probably have Windows Auto Updates on so they should be running IE8 (I know IE7 was a forced update for XP users, not sure if IE8 is...) So that means either a user has turned off auto updates, OR mroe likely it is a business with policies in place.

There's nothing on that lexity site (at least from the demo) that couldn't be coded to work in IE7 with an conditional css include.

[esbwhat]

Okay, sorry. Yeah obviously layouts will work, if you make them work. But that's the point, you have to make them work. IE has a weird box model, haslayout, buggy float handling, etc.

Of course it's possible to make things work, in case of a simple layout, most of the time not even hard (note that this wasn't my point at all), but if you code in a standards-compliant, normal way, without doing anything to especially accomodate IE7, there are a number of situation where things will just not work, while working in every other browser.

Is this a problem? Yeah. I would say so. People shouldn't be supporting something and thereby supporting the continued use of a product that doesn't work unless you coddle it. What if someone is a linux user? Are they supposed to buy a windows license just to coddle to users who are hurting the web with their browsing choice?

I'm by no means an expert designer, but I've been doing HTML stuff for a decade or so, from simple to complicated. I can and do get my layouts running in IE7, but that's not the point. The point is that I shouldn't, and if someone doesn't want to, they shouldn't be criticized for that. I'm pretty tired of running a VM to test my layouts as well.

[talmand]

Although I understand what you're saying, it sounded to me as if the person has direct experience with IE7 not working with "simple" layouts. In fact, it reads as if the person is saying that IE7 just doesn't work at all despite the design and layout. I'm curious as to what that means because based on my experience simple layouts are not much of an issue for IE7 and complex ones are not that bad. The claim that it "just doesn't work" requires further explanation regardless of the person's experience.

[seanp2k2]

Solution: help create better free, open-source accessibility software. This is my goal.