He seems to forget the fact that hardware sales doesn't translate into actual internet usage. The internet outside of iPhone, Android, and Palm is so terrible that barely anyone even bothers to use it. What do those three platforms have in common? Webkit!
The table does give that impression. Would it still give that impression if the width of each column were proportional to how many active, web-surfing users there are for each device/version pair listed? I'm sure the iPhone 2.x column, for example, would be a narrow sliver.
This statement may be true for your personal experience, but is in fact completely wrong.
Emerging economies such as India have huge mobile Internet populations. In fact in India mobile Internet outnumbers PC based internet 3 to 1 with a mobile Internet use north of 37million users, very few of which are using iPhones.
Actually, Fennec was the development name. The release is branded Mobile Firefox or Firefox for Mobile. Mozilla only puts the Firefox brand on finished releases.
Ok, so say you want to make a website that works reasonably well on all the major "smartphone" platforms that use WebKit. And furthermore, say you don't have the budget to get your hands on all of them, or even to go through a testing service.
Is there a WebKit-based browser that you can download where, if it looks good there, it should look ok on all these phones?
EDIT: Oops, I think edd's post below answers my question.
So the IE6 comparison is definitely the right way to sting developers into action. Nobody wants to end up being That Guy from 2001 again.
There's a few differences though. IE6 did a few things better (their box-sizing model is still more sane than the standard, box heights were nicer if unpredictable). But many things it just did differently, for no very sensible reason.
Mobile Safari isn't like that -- it's pretty damn standards compliant (tried a second ago, it passes most but not all of ACID2). That means anything you write that is completely standards-compliant will work, without modification, on any other similarly compliant browser, mobile or otherwise.
What is does do is a bunch of mobile-specific enhancement -- the viewport sizing, css gradients, etc. are all highly useful for the unusual constraints of mobile web development.
So, as PPK suggests, build the website in another browser first. Then enhance it for Safari using their extensions -- not the other way around.
First time I heard Quirksmode curse like a sailor. Regardless, the reason I can't develop for a Nokia, Blackberry, or HTC is because I don't have a Nokia, Blackberry, or HTC. I have an iPhone. It's pretty easy to develop webapps for it because if it looks good in desktop Safari, it will probably look good in iPhone Safari. I don't even know the browsers that other phones have. I know Opera Mini is on many phones and Firefox too. I try to make my sites work well with desktop versions of these browsers. But there's no way I can guarantee that my sites will work well on all the browsers everywhere.
Already I have to worry about:
* Firefox 3+ (all OSes)
* Safari (OSX, Windows)
* Opera 9+
* Chrome 3+
* IE7+
* iPhone
There are just so many browsers I have to worry about. And even though I code for standards, every browser still treats every little thing slightly differently. I realize making something work in Opera if it already works in FF/Chrome is probably not too difficult, but sometimes little things can take up hours. So sorry Quirksmode, despite your well-written article, I will not even bother with any other browsers just because you say so.
You know when I will bother? When my users tell me that my app doesn't work well in platform X, browser Y and they would really want it to. That's when I will go beyond my basic list of browsers. Otherwise I'm just wasting resources developing for yet another platform.
I think this article is aimed directly at you then. And clearly you disagree, sounds like because you don't feel like it.
A lot of us don't really want Apple (or Google, or x, or y, or z) to be a monopoly though. And perhaps the people who do follow the advice to consider the other browsers and will help keep things on an even keel.
I disagree not because I don't feel like it but rather because I'm not certain it is the best use of my resources to develop, test, and debug my app in every browser out there. I just don't have the time/money/resources to do it. And I highly doubt most startups do either. So you target the top 5-10 and stick to it till someone complains.
Top 5-10 what? Browsers? Because if Quirksmode is right, the fact that most mobile users of your site are iPhone users is a result of your decision only to do your mobile webdev for the iPhone, not a cause of it. As for sticking to it until someone complains - that's exactly what this article does.
But in the end, the loser is you, the web developer, because you're pushing potential visitors away from your site.
Users who use platform X and browser Y where your app doesn't work will simply switch to another app which does the same thing that was written by someone who _does_ care about all browsers.
Those users won't complain to you, they'll just switch.
If the effort to provide compatibility and an additional set of browser work-arounds exceeds the benefit given the insignificant share of traffic these other browsers provide then it isn't really a loss, is it?
> Those users won't complain to you, they'll just switch.
In my particular case, I only make things that don't exist already i.e. I'm not making yet another twitter-posting app or social bookmarking app. So I highly doubt my target users will switch simply because my app doesn't work nicely on a browser I don't even know about.
Also the solution to this would be to show a small notification if my app is being run in an untested environment and let the user complain with a 1-click form if they wish. I think most people who have a bad experience will let me know that is the case.
That bit of snark ignores the reality that a developer has resource constraints, and they constantly have to decide how best to deploy their time, money and energy.
Failing to address some platforms might well be a big win. Claiming it's a definite loss is absurd.
The original article's point was about progressive enhancement. If you're starting a project that requires, say canvas, it is your loss. If you start with tables and images, the site will work on a lot of browsers.
"First, so what? No, let me rephrase that: So fucking what? Since when does web development mean leaving 50% of your mobile users out in the cold? Since when is “I only support browsers with a large market share” a valid argument? (Answer: since we have an iPhone up our ass.)"
A device has more than 3x its market-share in traffic-share and we are supposed to ignore that? That doesn't seem realistic. This gap between market-share and traffic-share speak volumes about how far behind the other platforms are in UX.
Maybe developers aren't just obsessed with the iPhone: maybe they just don't want to develop for something that makes thier work look like shit.
* Those usage numbers (50% of mobile browser traffic)
could be a US-only statistic
* People on other browsers might not browse the web as much because
not enough effort is made to making web content usable on those
browsers.
Viewing your argument alongside the second point poses a possible chicken-and-the-egg problem. No one is using other mobile browsers because the experience sucks, but the experience sucks because not enough development is done to make it bearable (or pleasant). Not enough development time is being spent to make the experience bearable or pleasant because no one is using those other mobile browsers.
I guess I can bolster this with the usage scenarios that are probably happening:
People get their phone and try to use the mobile browser, but the experience sucks so they give up (or only use it whne they really need to get some info from the web).
Developers look at the usage statistics, see that no one is using the other browsers and proudly declare that the people whole own phones with those other browsers don't want to access the information that they are presenting to the world, so therefore there is no point in targeting them as an audience.
The problem is that this is the wrong conclusion to make. Just because people aren't accessing your content in a certain way does not mean that they don't want to.
No, you're not supposed to ignore that the iPhone has 3x its market share in mobile traffic. But you're also not supposed to ignore the other 50% of mobile traffic.
Lots of mobile browsers have iPhone in their UA strings to work around browser detects that obsessed web developers have set up. Do all traffic market share reporters work around that problem? Most probably do, but we can’t be sure.
This quote does not make much sense at all. Looks like ppk is so blinded by iPhone hate that he did not even bother to find out how these thing work.
iPhone was the first mobile phone which made ordinary web usable on mobile device for me—and that's for sites without no optimization whatsoever.
On the other hand most of optimization which makes web experience even more enjoyable on iPhone can be very simple (as simple as META element without any browser sniffing involved) and unobtrusive in the sense, that it does not affect other browsers.
If your mobile browser cannot render simple website no UA faking will help it. And I lay the blame solely on those browsers, not the webdevs who gave up trying to make sense of them.
What about calling mobile browser vendors to step up their game, instead of calling developers morons and wanting them to design to the lowest common denominator?
> iPhone was the first mobile phone which made ordinary web usable on mobile device for me ...
... while the rest of us did fine with Opera Mini on whatever "dumb phone" we owned. Try out Opera on one of Nokia's high-end Symbian phones one day (the N95 comes to mind). You'll be pleasantly surprised :)
- proportion of devices currently subscribed to a data plan;
- time spent online / number of pageviews per device;
- for businesses who don't plan to be profitable in the next couple of months, what OSes will be actually used for mobile web surfing in the future;
- which users have been trained by an AppStore-like system to easily spend money online.
Nobody in his right mind tries to surf with Windows Mobile if they can avoid it; the vast majority of Nokia or Sony-Ericcson phones, those which constitute the bulk of their sales, are barely usable surfing platforms, and are generally not sold with unlimited data plan; Spending money on Palm support supposes that you bet on their middle-term survival, a bet not everyone is willing to make; RIM is totally specialized into the corporate market, and is relevant if and only if you target this audience.
I agree right up to the comment about RIM: if you look at their recent efforts they are clearly holding their own in the teenage / budget smartphone market for families that want the email/txt capabilities of iPhones at a lower price point (or don't want to be on ATT.) I don't think RIM does a great job of serving this market but their price point is still well below Apple's and they're on more networks, and hopefully they will eventually get their act together in the browser area (although that will not help all the old units they have out there.)
Here’s the thing: developing for the iPhone means that you are developing for any browser that supports web standards. Why bend over backwards for browsers on phones with archaic web browsing capabilities. If anything, it will force vendors to get with the HTML5 and CSS3 programs.
And of course, once the vendors get with the HTML5 and CSS3 programs they'll release new firmware packages including the new version of the browsers for all the old devices that have archaic browsers, and then your users will all go and download the new firmware and install it on their phone, and, bingo, you don't need to worry about supporting the archaic browsers.
You're obviously new in town, so I'll point out the two flaws in your reasoning:
1) The vendors won't ship updated firmware for all the old devices they've already shipped to market. There are just too many different combinations of devices and firmware versions (each network typically gets a custom firmware build to include their links and logos), so the cost would be huge.
2) Even if they did ship updated firmware, most users won't bother installing it.
The mobile market just isn't like the desktop market - what you're effectively saying is that people can't use your web application until they upgrade to a new phone.
SMS:ing is bigger than (mobile) phone calls in Sweden, especially among young people — all they do is text each other, back and forth, even when calling might be cheaper. It's been this way for years.
Key reasons: receiving an SMS is free (I believe this to be true of most countries except the US).
Sending SMS as well as making calls is much much cheaper.
SMS are usually delivered the moment they are sent.
I can only tell you that SMS is big in India. And when I say "big", I mean so mind-bogglingly, utterly huge that you're going to have to grow an extra brain just to imagine the sheer volume of SMSs Indians send each day.
Besides, what will happen when the operators abandon the economically untenable flat rate for iPhone data traffic?
Er, what? Operators are expanding those flat rate plans to other phones (I have a unlimited everything month-to-month plan for $80 right now, on a non-AT&T operator).
While this article curses a lot to get attention, I think it will fall on deaf ears because I do not have other phones to test on. Unless a simulator or emulator is readily available for download on my platform of choice, I will not test against it.
iPhone and Android are pretty much covered with good WebKit implementation. Check in one of those and most likely your page will work in the other.
PPK claims that S60 WebKit is much inferior, but there's Opera Mobile 10 for S60 and for WinMo.
Blackberry doesn't have a good browser at the moment (although they are likely to get one). Opera is making an optimized version of Mini for it, though.
Practically all other phones run Opera Mini too. Version 5 uses the same engine as Mobile but it has a much lesser JavaScript capabilities. Personally I prefer it to the default S60 Webkit on Nokia 5800 ExpressMusic.
At the moment it's all seem not that complicated to me: check in Apple WebKit (Android/iPhone), Opera Mobile and Mini, and hope that other browsers will catch up eventually.
Mini can be run in Java emulator, and desktop Opera browser has a "small-screen" rendering mode (should work like Opera Mobile). Same goes with desktop Safari and iPhone.
> iPhone and Android are pretty much covered with good WebKit implementation. Check in one of those and most likely your page will work in the other.
This is tangential. I would have thought you were right, but in my anecdotal experience, it's very much not the case. I have an Eris; wife has an iTouch. Our browsing experiences are night and day surprisingly often.
I'm picking this nit mostly because it surprised the hell out of me.
Wouldn't it be hard to target Nokia/RIM/etc? They have dozens of handsets. Different screen resolutions and even different rendering engines. Apple just made it easier by keeping the render/resolution consistent for the last 3 years. The other factor is Nokia/RIM/Microsoft really lagged behind Apple, Palm and Android in offering a modern standards compliant browser on their handsets. I think the assumption is people buying a Windows Mobile phone for example don't really care much about the web. (or they would have bought a phone with a modern browser) Why go out of your way to please them? We know iPhone/Touch uses really like the mobile web.
I have Opera on my Nokia mobile. Just as I did on my last Nokia mobile.
And yes, mobile phones come in different shapes and sizes, just like monitors come in different shapes and sizes. Design flexibly, so that your app is usable by people accessing it from anywhere.
Any phone's browser(s) should be capable of displaying the regular non-mobile site. I assume the author is referring to the iPhone specific/enhanced sites out there.
All stats I see (and I looked at quite a bit of them) have iPhone traffic at 50% or more of mobile trafic. See e.g. Morgan Stanly: http://techcrunch.com/2009/10/21/how-the-iphone-is-blowing-e...
While I agree you shouldn't _exclude_ the smaller browsers, it's not strange to optimize for your largest market share.
PPK is right Nokia sells a lot of phones. However, people who buy nokia's aren't _using_ them to browse the web: They use them for phonecalls. Iphone users do.
I'm not sure what he's talking about. I don't mean for that to sound snarky. I just feel like I'm being yelled at when I honestly don't know what I did wrong. When I start building a web app I start with standards, assuming that my early adopters will likely be using Safari, Firefox or Chrome. Mobile Safari usually Just Works(tm). Is he talking about custom iPhone stylesheets or what?
When you spend money developing a mobile app, you have to future proof it so that keeps up with the apps/sites of similar functionality, otherwise people will end going to your competitors (obviously that statement depends on your service, but its certainly the case in my sector). People with smartphones wont want to use a site that is so stripped down it would work well on a crappy handset.
We are sitting on the edge of a large explosion of smartphone devices and (potentially) users (with lots of Android handsets coming out this year). In the UK you can now get a Android powered device for £20 / month with a free handset. If you want a smartphone (and therefore probably interested in using the web), you will be able to afford one this year (at least in Europe!). This is going to be huge.
The iPhone is the standard they all will judge themselves against, and therefore the iPhone is the benchmark device.
I would suspect that the 50% smartphone / 50% other phone traffic statistic will be way higher towards smartphones in 12 months time.
I had a Nokia phone for years and then a Motorala phone and then a couple of windows mobile phones from HTC and others. But, in none of them I have ever downloaded an app, leave alone having the interest to code up an app. With my iPod touch however (don't want to deal with ATT) I have downloaded dozens of apps and feel good to code on it (though I had to buy my first Mac for this and had a hard time with Objective C's syntax).
I used to work in the windows mobile development team at Microsoft and at that time used to have a dismissive tone about iPhone and said it was just an expensive version of HTC Touch. But, once I started using it and downloading apps, I found the real difference. It is in a whole different league and engineers in other phone companies know that. (I can speak for atleast one of them)
Are people actually designing for Webkit mobile? I thought the big breakthrough with Mobile Safari was that basically every website on the web just works.
Okay I have to turn the phone sideways to read Hacker News comfortably, but apart from that it generally just works. Mobile enhanced sites are nice but if we had to rely on them then we wouldn't be getting any mobile web at all.
If the other phones can't render that non-iPhone optimised sites then they've got bigger problems than not being able to see the iPhone version, because only some vanishingly tiny percentage of the web is available in that format.
(I believe Opera's command of the market on lower end phones was partly due to very smart reformatting of standard webpages (done both server- and client-side.)
He is missing/ignoring the obvious. Although the iPhone may only be 15% of the smartphone market in terms of units shipped, it represents something on the order of 98% of the market in terms of user dollars spent on applications. This is the only number that matters if you are selling an app. The number of installed platforms is meaningless.
At least in the US, a lot of designers have an iPhone (and so do their friends). I think it's also likely that the iPhone was the first mobile device they have ever used the internet on. It's easy to fall into the trap of thinking that since me and all my friends only use the iPhone everyone else does to.
I don't agree. The iPhone was at least my 4th or 5th device that had some kind of "mobile web" function. Most folks had smartphones, blackberries, etc before they got an iphone if they had any interest in the net. The problem was, the reality of tying to actually _use_ these devices to access web sites was such a horrible UX that you didn't do it unless there was absolutely no other option. In comparison, lots of us will idly web browse on an iPhone while waiting on line at the store, etc -- something that probably would never have happened on a Blackberry.
Even in 2000, developers had Netscape Navigator, so developing for IE because they had to and NN because they wanted to was the best option (unless distorted by being paid to meet other requirements).
Now, developers have iPhones and like iphones, developing for other devices isn't so personally pressing.
Perhaps this will force other manufacturers to improve in the way we always wanted to do to IE6 but couldn't?
Here in Norway, Nokia has the largest share of the mobile phone market overall, and their smartphones are quite nice (and popular). At the moment, I'm trying to decide between buying an iPhone and a Nokia N97.
Get your hands on an N97 before buying one. The one I worked with last week seemed to crash just by looking at it. I literally had to pull the battery 3 times in the first hour, and it crashed on things like opening the SMS inbox. Don't know if I had a bad model, but it was one of the worst experiences I have ever had (and since I do quite a lot of mobile dev, I have seen quite a bit).
Maybe. But I'm not looking to match the iPhone. I'm just looking for a good phone. I currently have an HTC S710, and I like the full QWERTY keyboard, so that's a factor in favor of the N97. But, on the other hand, there are a lot of iPhone apps out there. So, I'm still on the fence.
Not really. N900 is tech oriented, not consumer oriented. N97 is squarely aimed at the consumer market as per the iPhone, and has the Ovi store for applications.
Nokia, as far as I'm aware, is the biggest cell phone company in the world. Here in China, when I see someone pull out a cell phone, it's a decent bet that it will be a Nokia.
And those phones have a web experience worth targeting? I understand that Nokia has a lot of phones out there, but the vast majority seem only tangentially related to the discussion of smartphone apps.
It's fairly easy to kick together a "mobile" version of a site. Just use these simple CSS rules:
img { display: none; }
body { hideous-eyebleach: true; }
For many phones, just having 0 style, 0 images and a few text fields is enough. Their browsers are barely capable of mediating an experience beyond that (let alone displaying it). These don't really represent a valid target for this discussion, I think.
Well, given that Nokia S60v3 browser _is_ webkit, I don't think that the results with your suggested stylesheet would be satisfactory. Would you accept it on your iphone?
If the phone is worth targetting, then target it. I'm just saying that arguing Nokia's total phone market share seems like a sort of sketchy way to extrapolate priorities for smartphone-enabled websites.
Does the Nokia S60v3 have large market share worth targeting? The reason Android and iPhone markets have such gravity is that they're large and relatively homogenous consumers of rich web content from the mobile space; maximum bang for minimum buck.
I really think that this article is just wrong, and its willingness to confuse rich-web-enabled smartphones with phones that can barely display unstyled webpages is what drives me to that conclusion.
Sorry, you're the one who's confused -- the article is quite explicit that the marketshare numbers given are for smartphones, not the entire phone market.
Except that we're talking about smartphones capable of delivering a web experience that is competent enough to be concerned about experience.
There are smartphones and then there are post-2006-smartphones, which actually do real web browsing. I strongly suspect that Nokias numbers here include many phones less capable than the one that was raised in objection previously. Is there independent evidence to the contrary?
Everyone and their mom and pop in India, those who have nicer phones, have a Nokia. But, yes, you're right, here in US I don't see them very often. I'm only taking about the smartphones.
The cheaper phone market in India is pretty fragmented.
This guy is sounding more and more foolish, and pretty wacky. he used to be cool. What the heck happened to him I wonder? Maybe he can't get an iPhone where he lives.
Screw this article, inaccurate and not true for other mobile platforms. They are over saturated with developers just as iTunes is. We do not need any more dev's.
He seems to forget the fact that hardware sales doesn't translate into actual internet usage. The internet outside of iPhone, Android, and Palm is so terrible that barely anyone even bothers to use it. What do those three platforms have in common? Webkit!