Hacker News Reader

Comment by EnderMB | Hacker News Reader

When I see heavy web pages like this it makes me worry about the state of front-end web design.

According to YSlow, the site came to 3.5MB, which is quite large when compared to the average home page. The site took at least five seconds to load on my fast dev machine, which is far too long.

The nail in the coffin for me, however, is that this site does not work in older browser, namely IE7. I've worked with large clients and we've made sure that everything works in IE6 and above because they make the cost back from sales within hours. A large company like airbnb simply cannot afford to not support older browsers and I firmly believe that they'll lose a ton on money on their lack of support.

All in all, I'm disappointed.

View on HN

I agree with everything you've written, but I couldn't disagree more.

Once the page loads it draws you in to wanting to visit nearly any of those places they've listed. Every shot looks like it was staged by an expert.

Or to put it another way, I would rather have someone who's trying to sell me steaks show me pictures of mother-watering, expertly cooked beef, rather than show me a drop down menu of the types of steaks they have. No matter how fast the page loads.

As for the Internet Exploder issue. In all honestly, anyone who is still using that browser probably wouldn't be very comfortable using AirBNB on their next vacation anyway.

Couldn't have said it better myself. That is some of the best landing page photography I've ever seen.

At the same time, that's what makes me wonder if it is legitimate or just stock photography. I know that it isn't, but from a first-timer's perspective, it almost seems "too good to be true."

Don't get me wrong. I love photography on web pages, but the page itself is similar to so many modern web pages. It's full of JS, it crawls on modern machines and just doesn't look nice enough to justify the size of the page. The page barely works with JS off, so I'd hate to imagine how a user with a screen reader would use it.

As for IE7 users, you may be right, but it is my experience that sites that are the size of airbnb can happily spend £10,000 to make your site work in older browsers, because they'll get that amount back from new customers very quickly.

Whether you like to develop for legacy browsers or not is irrelevant. It's poor practice not to.

"Whether you like to develop for legacy browsers or not is irrelevant. It's poor practice not to."

I would argue that it is poor practice to support them.

First is cost. Even if it only takes 10 grand to make a site "functional" in x browser, rebuilding an application to fully work with ie7 does not. If it takes 15-20% (or greater) more development time to add in support for IE7, but IE7 does not provide 15-20% increased revenues (it won't, unless you're making business software specifically for tech-agnostic corporations). You might say oh, it doesn't take 15-20% more time... just use jQuery and magic will happen. No man. It takes much more time.

Second is more philosophical. Assuming you are a consumer focused company like AirBnB, you customers are fully capable of upgrading to a new browser. There are no IT limitations on a personal computer. It takes all of 5 minutes and is completely painless. Stop coddling IE7 users; it is obsolete.

As far as pages working without JS: this is 2012, not 2002. If you still treat today like you treated 10 years ago, you really have no business complaining about anything on the internet. The internet is not for you, and neither is a site like AirBnB. If you think I'm being overly harsh, see my last paragraph. Installing FF, Chrome, or IE8/9 takes all of 4 clicks.

Those sound like nothing but excuses for being a terrible web developer.

"First is cost. Even if it only takes 10 grand to make a site "functional" in x browser, rebuilding an application to fully work with ie7 does not. If it takes 15-20% (or greater) more development time to add in support for IE7, but IE7 does not provide 15-20% increased revenues (it won't, unless you're making business software specifically for tech-agnostic corporations). You might say oh, it doesn't take 15-20% more time... just use jQuery and magic will happen. No man. It takes much more time."

Oh man, there is so much that is wrong with this comment.

First of all, users are more willing to buy from a service that works compared to one that, well, doesn't. That's just common sense. Secondly, if you really think that sprinkling some jQuery over a site will just make it work then I really hope that you aren't a web developer, because that mentality is exactly what is wrong with many websites out there today.

"Second is more philosophical. Assuming you are a consumer focused company like AirBnB, you customers are fully capable of upgrading to a new browser. There are no IT limitations on a personal computer. It takes all of 5 minutes and is completely painless. Stop coddling IE7 users; it is obsolete."

Again, if you are incapable of supporting older browsers it's probably because you're a terrible developer. Regardless, there is no philosophical debate here. Users can use whatever the hell they want and it is not your business to tell that what to use. It's entirely up to the client, and if there are users that are using IE6-7 it's probably because they are on old hardware or a machine that isn't theirs (i.e. a work or library machine). The limitation on a personal computer is just that. If your site takes ages to load are you going to suggest to the user to get a better internet connection or to buy a new computer?

"As far as pages working without JS: this is 2012, not 2002. If you still treat today like you treated 10 years ago, you really have no business complaining about anything on the internet. The internet is not for you, and neither is a site like AirBnB. If you think I'm being overly harsh, see my last paragraph. Installing FF, Chrome, or IE8/9 takes all of 4 clicks."

So, where do users limited to screen readers fall into your "the Internet is not for you" speech?

The reason many of us develop solutions that work across legacy browsers and screen readers is because we have the experience to be conscious of those who CAN'T view the web as you'd like to.

Again, it is not your place to dictate what users should be using. Regardless of their lack of browser support it is a wasteful site that is entirely dependent on JS and as such it is completely inaccessible.

It's comments like yours that give me "the fear" when talking about web development.

> Whether you like to develop for legacy browsers or not is irrelevant. It's poor practice not to.

Following that logic, it's poor practice to develop anything in HTML5 since IE8 and below won't support it without a JS shiv that won't work with JS turned off.

Half way through 2012 now, I've come to accept that anyone who is using a service like airbnb with IE6, or with JS turned off can just miss out. I'm not prepared to deny the 90%+ users a superior user experience in order to support buggy, insecure ancient browsers that should have died years ago.

It's time to look to the future. The IE6 crowd can be happily left behind.

According to statcounter ie7 was at ~1.5% last month. It dropped a full percentage point from the month before that. Even Firefox 3.6 usage was higher.

I think there is value in moving to faster deprecation. Plus even if it does make finicial sense for airbnb it is still a risky investment. Should a facebook, google, or wordpress drop ie7 support hard then all the sudden you've wasted money, developer time, and developer sanity.

Statcounter isn't your site. Everyone should look at their own site stats and make the case as to whether legacy browsers should be supported.

Regardless, for new projects we still follow the Yahoo Graded Browser Support baseline, like many other agencies do. http://yuilibrary.com/yui/docs/tutorials/gbs/

"WebAIM survey results [0] reveal 98.6% of screenreader users have javascript enabled," via Paul Irish [1].

[0] http://webaim.org/projects/screenreadersurvey4/#javascript [1] http://paulirish.com/2012/accessibility-and-developers/

It loaded for me in ~2 seconds, on a 2010 Macbook Air. I think we're at the stage where 'heavy' pages are permissible, and even advantageous at times.

It loaded for me in ~30 seconds on a 2011 Macbook Pro. I don't see what your loading experience has to do with anything. Different people have different connection speeds in different parts of the world (my connection speed varies within the same building depending on congestion, etc).

That said, all the fancy pictures do look very nice, and certainly draw you in like no other travel site that I've come across (when it eventually loads).

I don't see what your loading experience has to do with anything.

I was responding to anecdotal evidence with contrary anecdotal evidence. In the big scheme of things neither really means anything, but it would be pretty tedious if I had to test a statistically significant number of machines every time I altered a page layout.

It's still pretty heavily reliant on internet speeds. If you're using wifi at a coffee shop, or have a relatively congested work connection, the site does crawl -- though I can only attest to the latter case.

Thankfully, I use noscript to cut down the loading a good amount, but the parent's issue is a very significant one.

Your computer is barely two years old, and doesn't reflect your internet connection. Most people are on far older, and probably far slower connections.

"Most people" on the internet, or "most people" in AirBnb's user base? Because they could be very different. For once it would be nice to see HN give a site the benefit of the doubt in that they might have actually looked at their analytics and planned accordingly.

But aren't people like you the reason why we still have to support IE6 and 7? I think depending on your customer base, you have to support it, but I'd bet that a lot of Airbnb's users aren't using an IE6 or 7.

I currently work with a number of large clients in the travel sector, and you'd be very surprised.

Yes, it's restricting at times, but any developer worth their salt can write code that works in all browsers fairly easily. I've spent enough time on Reddit to know that a lot of developers refuse to support legacy browsers simply because it's too hard.

It is my belief that we'll have to support IE6-8 until XP dies, and I think that will happen when Windows 8 becomes the default install for most PC's. The real problem isn't older browsers, but older machines that need to upgrade their operating systems, and once that war is won the old browsers will drop like flies.

They can use FF/Chrome if they want to see the site correctly on XP. The faster people stop supporting broken browsers the faster they'll go away! If they can't/don't then they can suffer the trade-off knowingly.

For a successful startup that needs to grow at a reasonable pace (not too quickly) select browser support is one way to apply some light breaking. In five years if/when the business has scaled on its merits, those browsers will be history anyway. That very early (==expensive) developer time would have been wasted when it could have been focused on building a product people want to use.

Think of how Apple slowly removed restrictions to each version of iPhone so that they could handle the interest it generated. They simply couldn't have handled shipping 100 million the first year. So their focus worked out perfectly (despite all the bitching about cut/paste and At&t we had to listen to.) If they had wasted time on "perfection" it would have shipped for years.

Considering that a (considerable/significant) section of their target audience could be "on the move" and not have a great internet connection, not sure if the 3.5MB gorgeous looking site is a good idea.

It works for me in IE7 on IETester (mostly e.g. there's a correct mixed content error and the "popular cities" tab doesn't work). I assume they have looked at their stats and concluded that IE6/7 doesn't represent a large enough proportion of their market to support it. I also wouldn't regard airbnb as a large company.

More importantly than just being slow, the site crashes my iPad. After about 60-80% of the page loads, Safari closes to the home screen.

It crashed my browser (Opera) right after it loaded.

Reply on news.ycombinator.com