Category Archive
The following is a list of all entries from the ajax category.
Web as a Platform: a step forward
Ok, no sweat, the shift from desktop-centric offline apps to web-centric online apps started long ago, and should go a step further with Google Gears, Firefox 3 [2] and other efforts which promise offline access to 2.0 webapps.
The Mozilla Foundation is working towards a webapp browser, so Webrunner [3] (Mozilla’s XUL-based lightweight app-wise browser) is now a Mozilla Labs project, code is in the trunk and named Prism, which ships a bunch of new features [4] (extension support, amongst others) and promise integration with Firefox, to make it easy for the user to “install” a webapp on its machine and access it without the need of a full featured web browser. Some of the benefits of “desktopized” apps in [1].
The border between desktop and browser seems to become blurry, and the day we’ll see the web as a platform to run applications upon seamlessly might be quickly getting closer. The project looks quite ambitious and sparkles a lots of interesting development scenarios in my head: they declare to be “working to increase the capabilities of those apps by adding functionality to the Web itself, such as providing support for offline data storage and access to 3D graphics hardware”.
To me, this looks much closer to the actual direction web/desktop development is going to undertake than other akin technologies’, and will endow developers with the best of both worlds, bringing the user experience right on the edge of the transition.
However, even if smart and far-seeing, this is just a glimpse: in the next years, no surprise, noticeable implications will likely come from the grid/p2p/multiagent systems, and distributed filesystems/OS research areas, whereas joint efforts and results will probably bring a real revolution in our perception – and use – of the web.
“Just” another step forward.
Links:
[1] Zoli’s blog
[2] Read/Write blog: Firefox 3 and offline apps
[3] Webrunner
[4] Mark Finkle
[5] ZDNet Blogs
Eclipse and iPhone development
Eclipse now supports iPhone development by an Aptana plugin, LeNettoyeur reports. I’ve been dealing with aptana months ago, in this post.
You might want to have a look, it’s got some quite interesting features IMHO.
Looking forward for the iPhone to ship in Italy as well 
.
Actually integrating Firefox + del.icio.us (at last)
I just installed this Firefox add-on (by Yahoo! Inc.), which actually allows to integrate your del.icio.us account into your Firefox profile.
Well, it’s quite straightforward so I won’t bother getting into the details, but it’s absolutely worth a try.
So far I used to use another extension which just allowed the user to tag a page and add it to his del.icio.us bookmarks: the feeling I had before was that my del.icio.us bookmarks were just separated from my Firefox bookmarks (and they actually were). Now I really feel they are a single, integrated set, since I can browse, add and search links in Firefox.
Some months ago I actually thought having such an extension would be nice, glad to see guys @ Yahoo! worked on it. 
Aptana Web IDE
It’s been out there for some months already, but I just discovered it through Yahoo! User Interface Blog: Aptana is a robust, JavaScript-focused IDE for building dynamic – AJAX - web applications.
It is available as a standalone app for Windows, Linux and Mac; and of course as an Eclipse plug-in too (which is in my opinion the solution most of people will choose).
A screencast – embedded in the Yahoo! blog post linked above – is showing Aptana’s most significant features, and its YUI support. Very nice.
Gmail: fetching perfection?
Gmail just launched an absolutely interesting feature that allows users to fetch mail from other accounts. I spotted the news via Techcrunch.
Yes folks, the mail fetcher allows users to access non-Gmail email accounts from within the Gmail interface, using POP parameters.
This actually turns Gmail into a web-based email client, which is a very smart move from a strategy-biased perspective too.
A big step towards the day the Web will be the platform apps will run onto.
[UPDATE] Privacy implications – on the long run – are not to underrate here, standing to Donna Bogatin @ ZDNet Blogs.
Revisiting the good old 80/20 rule: websites performances
The first of a series of posts @ Yahoo! User Interface blog about some experiments to spot where most of the time a user waits for a page to show up is wasted.
“Most performance optimization today are made on the parts that generate the HTML document (apache, C++, databases, etc.), but those parts only contribute to about 20% of the user’s response time. It’s better to focus on optimizing the parts that contribute to the other 80%.”
Performance-aware web pages design is getting pivotal nowadays and, from a global perspective, the general load the network infrastructures have to handle would decrease significantly if designers kept in mind such issues.
Yahoo! performance researching indeed provided the following – quite predictable though – result: “…90% of the time is spent fetching other components in the page including images, scripts and stylesheets“.
This is, by the way, one of the preminent aspects the omnipresent AJAX technology has impact against, and in my opinion – almost – the only one to justify such enthusiasm: it is well known that asynchronous page refreshing saves a lot of time, and reduces the http server-to-client traffic.
Unfortunately the linked post does not deal with the impact of browser caching, which is very important as well; anyway, the big picture is always the same: websites performances increase as the number of HTTP requests get lower.
Get mixd!
Yahoo! launches a group texting service, techcrunch reports. The site name is mixd.
You can group-text your friends while you are out, share media, and see it on the web. This is not the first service of this kind, obviously (here it is a comparison among eigth available group texting services). But it looks nice.
The thing I keep wondering about though is just how would it be possible to use such a service here in Italy without getting broke in a few days, standing to our insane mobile phone tolls. 
Eleventh: don’t just AJAX everything
In a moment where it seems impossibile to get online without actually reading tons of blog posts or articles about Web 2.0, it’s nice to read some original point of view about the actual AJAX impact over industry and innovation (if any), considering the architectural implications and debating about which road should we go.
“Web 2.0 marks the dictatorship of the presentation layer, a triumph of appearance over architecture that any good computer scientist should immediately dismiss as unsustainable.”
In this post @ deal architect Vinnie Mirchandani deals with Bill Thompson’s article @ regdeveloper.co.uk, where the Web 2.0 “madness” is harshly criticized (with a bunch of – IMHO – nice politically-motivated metaphores ).
To be honest, I subscribe to this point of view: first of all I can’t quite understand where this enthusiasm about a late-1990 web development technique comes from; XmlHTTPRequest has been there since much earlier, it’s not rocket science, nor cuttin’ edge technology.
Furthermore, I definitely agree with Thompson’s invitation not to focus ourselves that much about the presentation layer: the way we present data is absolutely important, but the actual innovation is elsewhere to be chased. Engineering should face (and its actually facing) much more complicated challenges than asynchronous <div> refreshing.
Whatever, I believe there’s no chance to get a better conclusion for this post than Thompson’s one, so I’ll steal his one:
“The time has come to stand up and be counted, and we need people who can count in hex and see beyond the Web 2.0 hype. “
About digg actual traffic volumes: in medio stat virtus?
Digg CEO Jay Adelson has been interviewed by Michele Steele @ Forbes Video Network, Techcrunch reports. You can find the video embedded here in this post @ Techcrunch. Discussion topics were, among other things, eventual Digg acquisitions (rumors about that in the past days) and it’s actual market value, standing to web traffic assessment.
Adelson claims a huge number of unique accesses per month (circa 20 million accesses), combining RSS readers data with web traffic ones. This seems a bit unfair: RSS reader-generated accesses cannot always be considered as “unique”, and feed monitoring tools (at least those I know) are quite immature, in my humblest opinion.
Anyway, third party estimates look quite poor too: who’s gonna believe digg is only 1.3 million accesses per month worth?
Google: AJAX Search API and Web Toolkit
I just discovered through Google Code an interesting AJAX API set to add a dynamic searchbox to your pages. Take a look at it here, it’s very sleek, responsive and (most of all) won’t ever take visitors away from your pages .
The nice thing is you can even use the APIs to build – powerful – Google Search-based AJAX applications, maybe using Google Web Toolkit (GWT) if it fits your needs .
GWT, for those of you who weren’t on earth last weeks , is a compiler producing browser-compliant javascript code from java classes.
In my opinion it really relieves developers’ burden of writing and debugging thousands lines of javascript working on Internet Explorer XOR Firefox.
I believe after all there has to be some kind of formal proof that the intersection between lines of script code correctly working on IE and those working on Firefox is the empty set, and that the exact location of a bug in a javascript page is subject to Heisenberg’s indeterminacy principle, so GWT is definitely high in my “things to study” stack. 
