In a sudden about turn, analysts at Forrester Research have decided AJAX technology is not the best solution to rich Internet applications after all. In a report, Forrester has recommended businesses should resort to vendor-specific platforms such as Adobe Systems' AIR and Microsoft's Silverlight because AJAX can't deliver the …
What they do with
their credibility is their business .I don't know of anybody using anything but AJAX so it's hard to see it not working.
Persona usage intensity
Gotta love those analysts. The free extract of the report says:
"Strategy professionals at application vendors should become familiar with the concept of persona usage intensity"
Never heard of that one. Panic. Must find out. But wait, "persona usage intensity" returns only a single Google hit - to the Forrester report extract.
So I can only decypher this intriguing concept by spending the best part of €800 on the full report!
Now THAT'S a professional strategy.
Depends whos paying!
I am surprised anyone takes these "independent" analysts seriously anymore. When they are not writing thinly disguised marketing brochures for whoever pays the most they get things so wrong it is often worth buying the research to determine what NOT to do.
Take a look at any Gartner magic quadrant from five/ten years ago and count how many "Leader" and "Visionaries" are still in business.
And boy do these people pick losers over the years they championed OSI, DCE, CICS on UNIX, CORBA, Itainium etc. etc.
Who gives a F**k what these muppets pump out as advice. AJAX? Isnt that like a few years old now, it works ideal to change a combo box without a postback!
right, for the wrong reasons
It's not power users who have the problems with AJAX websites, it's those with a disability. This is the bigest factor preventing it's commercial acceptance in most western countries.
Most AJAX (and other web 2.0 technologies) fail to provide the level of accessibility mandated by the Disability Discrimination Act and it's equivalents in other countries. The unenlightened may not care about this, for themselves and for their personal websites - so long as it looks Coooool, dude. Organisations are more responsible and take other needs into account.
A glance at the real world page on the Prototype website suggests that major commercial organisations are comfortable using an open framework:
Meanwhile, Dojo is sponsored by Sun and IBM and even a less well known framework like Mootools is used by Chrysler, CNET and the W3C. I'll start paying attention when organisations likes these drop open AJAX libraries in favour of an interface built in Silverlight.
Developers tend to be power users
So most developers are in a good position to have a valid opinion on what works and what doesn't, they have the advantage of seeing it from both sides of the coin.
Analysts are just regurgitating whatever they heard from the last marketing droid to nobble them.
Get paid to state the bleeding obvious
AJAX is just a stop gap. It is unavoidably slower than anything running a persistent connection and spitting byte code and objects to a runtime. No browser compatibility worries either.
That'll be € 1000. And don't forget to say thank you!
Notice me! Notice me!
@Pete. Not so. With an Ajax site you can force it to change the colour scheme (e.g. high contrast) or increase the text size even if the designer didn't build that in to their site. With Flash and Silverlight you have no control over those elements that could help people with visual impairments.
Ajax sites should provide greater compatibility across different platforms including modern mobile devices *without* downloading extra software. Many large corporates lock their desktops so you can't install or upgrade 3rd party software. Ajax sites load quickly and are quicker to develop for thanks to the ample of quality frameworks and libraries.
I think the report was done just to get the author noticed. These days you've got to say something alarming or highly questionable to stick out from the crowd. A trojan horse in most web 2.0 apps will create a global tsunami before 2012. Wibble wibble!
I keep hearing about this, but i have yet to see any real details.
On the other hand in a work environment i am asked by current, old and new clients about Ajax up to 10 to 15 times a day.
I have also been doing lots with Flash and Flash Remoting (which is pretty much ajax in another coat).
When a proprietry system is both available and well used / called for, then i will think about listening to one of these reports, but until then i may as well seek commercial IT advice from the milkman, these "analysts" have no integrity or perceivable experience in the field.
>>> With an Ajax site you can force it to change the colour scheme (e.g. high contrast) or increase the text size
So how does that help with a screen reader, or where a user needs to use TAB to move through the sections on a page as their vision is too poor to use a mouse.
As long as some developers think big bright text is all that is needed, we'll get inaccessible sites.
There are additional recommendations that need to be engineered into a site which is 508 compliant that makes provision for other special needs browsers. For example, it is recommended that a hidden anchor be posted at the top of the page to permit unsighted users to skip any headers and navigation and take the focus to the main content of the page.
So it is possible to have your cake and eat it too. The state of the art is not yet to the point where this type of engineering is common practice.
Why do you even report this stuff?
"In a sudden about turn, analysts at Forrester Research have decided [ ... ]"
... that whatever it is that whoever's paying them is selling this week is the one and only true way, and far better than whatever it is that whoever was paying them last week was selling.
I can't even begin to imagine why ElReg (or any other media outlet, for that matter) would print items about reports from analysts unless you were /beyond/ desparate to fill column inches. Frankly, these firms should be prosecuted under the trades descriptions act, because starting from a pre-ordained (since paid-for) conclusion and then attempting to reason backward to try and fabricate some kind of justification for it is just not what the word "analysis" means.
All these reports are shallow garbage, full of empty buzzwords and PHB-speak, and the only actual factual piece of information you can possibly derive from any of them is "The person selling (whatever this report recommends) has paid (this analyst) to write a report recommending (whatever this person is selling)".
Printing these stories undermines your editorial integrity. I look forward longingly to the day when you announce you will no longer be giving free advertising-disguised-as-editorial to anyone who can be bothered to send you a press release.
My bets are on SL2
I've been using Silverlight strongly for the past 6 months, after JS/AJAX/Ruby/Perl/PHP for the past 5 years.
There is no comparison as far as client side speed, using the Silverlight 2.0 beta vs. JS, from my own personal tests.
The immense amount of manpower we put in at my day job to get YUI/YUI-ext with AJAX, to perform poorly, sucks.
Yes, there are a lot of things that still lacking. There's no decent Monolight for linux yet. There's no developer tools for the Mac, and the Windows ones are behind, compared to Adobe's suite.
But these things are changing. Just last week Silverlight didn't work on Opera. Now it does. Setting up to host on Apache? Took me a minute of google searching and 30 seconds editing httpd.conf to add xaml and xap mime types.
But these things don't matter to 99% of those who will be visiting your website.
A good rundown on Flash vs. SL2 is here:
My usual pro-MS bias notwithstanding, Silverlight is very cool, and only going to get cooler. They've got a startling amount of .NET 3.5 into the ickle runtime already, including LINQ and WCF. I'd be amazed if they didn't get the DLR into a future version, so people will be able to write apps in various Iron implementations (Python, Ruby, Scheme(!)) as well as C# and VB.NET. Just imagine: you could be writing the client-side and server-side code in the same language!
There are some nice AJAX frameworks out there, granted, but I've yet to see one that comes close to what Silverlight can do. Couldn't comment on AIR or Flex as I've not played with them.
I agree with all the comments about analysts though.
- Fee fie Firefox: Mozilla's lawyers probe Dell over browser install charge
- 20 Freescale staff on vanished Malaysia Airlines flight MH370
- Did Apple's iOS literally make you SICK? Try swallowing version 7.1
- Neil Young touts MP3 player that's no Piece of Crap
- Review Distro diaspora: Four flavours of Ubuntu unpacked