The new HTML5 APIs may have triggered an arms race among desktop browser vendors to see who could deliver the best performance and standards compliance, but when it comes to anything but a traditional PC, developers should be prepared for serious challenges – or so says an Intel rep. "You're going to run into fragmented levels …
Something that will wipe out our market
not very good shock horror probe.
HTML5 running on low power ARM tablets/phones/smartscreens must look pretty awful to them.
Desktop range 350-450, evil phones mid-high 300's (350-390ish?) - doesn't sound particularly apocalptic, just clustered within the lower bounds of what to expect from desktop browsers, hardly a shocker and frankly better than the message of doom seems to spell out.
> the manufacturers of such devices seemed to have thrown a browser on there just to say they'd done it,
> without giving much thought to what content it should be capable of accessing.
Sounds fairly normal for a hardware manufacturer. Most of them seem to think that software is unimportant and put little or no real effort into it. It even extends to device drivers. I never use the disc that comes with hardware - I always give Windows first chance and if it already has drivers those are the ones I use. I'm certainly not going to install any other crap from a manufacturer.
> It even extends to device drivers. I never use the disc that comes with hardware - I always give Windows first chance and if it already has drivers those are the ones I use.
What an inverse piece of logic, which is the opposite of that usually recommended! Who is likely to know more about the operation of a device, the manufacturer or Microsoft? So who is the more likely to get a device driver correct?
If it's a generic product, like a USB Flash Drive or a hard disk, it isn't likely to matter one way or the other, but if there is any degree of unusuality about the device, then you want the manufacturer's device driver.
OTOH, who's more likely to know exactly how Windows is expecting a device driver to behave, a tiny usb device manufacturer or Microsoft? How many windows crashes are caused by "badly" written device drivers?
I find it's normally better to let windows have a go first, but that's just based on my past experience with hideous crapware installs that came bundled with various devices...
Re: @Anonymous IV
Gods do you remember the old Creative Labs installers? Not just the drivers but also a dozen more discrete utilities and widgets half of which put an icon in the notification area as well.
Where do you think Microsoft gets the drivers from? The hardware makers, you dolt!
The drivers on a Windows disc or from Windows Update have been through the WHQL process, which is costly but provides the level of testing enterprises prefer. This is why there can be such a version gap between the driver Windows offers and the latest for a device that is frequently updated, especially video cards. It would be horribly expensive and slow to have every Nvidia driver release go through WHQL but it isn't necessary since the customers who will only use WHQL certified drivers aren't concern with gaming performance.
I work in the web development industry and the irony of this is that a growing number of web developers (myself not included) are writing HTML5 applications, specifically targeted at mobile devices where, a the article points out, support is worse than on desktop browsers.
There is a lot of bandwagon jumping going on at the moment in this area (see also CSS3) with people deciding to use it even before the standards are truly finished, nevermind peoples devices and software being able to support what they're making.
Whilst I think it's a good thing that people are trying to move things forward and push boundaries, in a few years time we could end up with a load of systems/code which need updating because some developers couldn't wait.
Just an observation of mine. Any other devs that agree/disagree?
Re: HTML5 development
HTML 5 is the target to develop for because so many resources are being thrown at supporting it. At least we have better tools for coping with less "sophisticated" browsers than we used to have.
* we have to accept that it's a constantly moving target but also that this is not as bad as it sounds because degradability is built-in. The HTML 5 syntax alone is a huge leap forward.
* abuse such as using unprefixed css declarations for things which are only just in test have always happened and are an inevitable consequence not only of "lazy" or "stupid" developers but the pressure applied to them by customers with unrealistic expectations.
I would like to see a "development mode" switch in browsers which would default to off as a way to let people to try stuff out in the wild without force-feeding everyone with unfinished implementations
Re: HTML5 development
I don't work anywhere near web development but I think your observations are likely to be correct. Development decisions will usually be counter-productive if they are motivated by desires other than product focus. These dubious motives could be fear of being left behind, CV polishing, "because it's there", thrill of the new and so on. I would think that web development is particularly prone to this kind of thing because it is an exemplar of "hip and happening". The reality is that any app or site is going to be popular because of what it does and how well it does it rather than how bleeding edge it is. Some things are appealing because they are new and novel but I would have thought that quality of execution will win the day with users; better to concentrate on those aspects rather than fixate on the buzzwords.
Re: HTML5 development
we have to accept that it's a constantly moving target but also that this is not as bad as it sounds because degradability is built-in. The HTML 5 syntax alone is a huge leap forward.
If only this degrability theory was true. It's not. You cannot use the (useful) new tags because if you do, the presentation (styling) will be omitted by the vast number of useless browsers out there. As a result, you have to wrap the new tags in old tags (mostly div and span) and apple the style to these - at which point, why bother with the new tags?
The HTML 5 syntax isn't a huge leap forward - in some ways it's a headless rush backwards towards the "golden era" of mismatched tags. Take the example of browsers having to double guess what a web page markup really means... for example is "<p>foo<p>bar" two different paragraphs without closures or is it one paragraph embedded within another without closures? The gobsmacking idiocy of this backwards step must have the developers of the various web browsers clawing their eyes out.
I'd love HTML 5 to take off, but in reality it's not much more than an exercise in frustration - and that's before you start to get involved in the quasi-religious ideals of some of the actual writes of the standards and their own "lalalalalala (hands over ears) - not listening" approach to even exceedingly well reasoned and presented feedback.
IE benefits from the Windows update cycle
Where as all other browser benefit from a constant development cycle, so that most of HTML5 features are now being supported, regardless of operating system.
And with Microsoft changing their interpretation of HTML/CSS on each release, makes developing on IE a minefield.
// web dev, stuck on WinXP (at work) and IE8. :'-(
Re: IE benefits from the Windows update cycle
Indeed. Microsoft's brain-dead strategy of coupling browser versions to OS versions is a real roadblock:
Akamai has fairly representative (heavy US bias) figures of browser versions
IE 8 still at around 20 %. One of my customers where IE 8 is corporate standard is mulling moving straight from IE 8 to IE 10 but not before 2010!
er, what do I see there? 2010? I meant 2014!
PS3 browser ...
... is an absolutely shocking pile. Perhaps there should be some scoring that reflects on a company's commitment to keeping things like this up to date - it's too easy to sell something as having an internet browser when, within a few months or years at most "once contained it's own internet browser" is a more accurate claim.
Re: PS3 browser ...
Agreed, a truely HATEFUL browser - crashes more often than a drunk driver, taking the whole system with it - a few times it's even had to fsck the HDD of the PS3 after a crash!
Just as well I don't use it for much - but as it's been like this for HOW MANY YEARS NOW I can't see Sony lifting one little finger to remedy it.
Re: PS3 browser ...
I too have suffered from the browser from hell that is the PS3 browser. It is truly awful.
Intel needs HTML, gadget makes don't
The browser from hell forces developers onto apps.
Manufacturers of gadgets don't want you to use the browser.
And that is also why they don't even want you to have a browser choice otherwise their browser from hell would go to hell.
Re: The browser from hell
is that the one that sends all your personal data to Google?
IE 10 on Xbox 360
IE 10 is going to be added tot he Xbox 360 in the next big firmware update. I chose to not join the beta but I wonder how it scores compared to other non-PC platforms. Even IE 6 would be an improvement on the PS3 browser. Opera on the Wii is horribly constrained by the display and limited memory.
There is the ongoing conflict between selling native apps and selling the platform with more capable browsing. It'll be interesting to see how MS deals with the problem.
Chicken and egg or a turkey sandwich?
HTML is the open web so more likely it is that fat sow rolling on top of its piglets.