Blog Archives

The Struggle for the Sould of the Web

Very interesting article, “The Struggle for the Soul of the Web” !
The author enforces the concept of the importance of Ajax standards (and, thus, the Open Ajax Alliance) as a mean to avoid that the web becomes the territory where proprietary solutions (see Flex and SilverLight) will flourish.

In developping his argument, on which I agree, the author makes an interesting statement:
More importantly, Flash and Silverlight work by installing a proprietary plug-in to your browser, thus opting out of the entire browser infrastructure. If you are a plug-in vendor, your incentive is to keep the browser as dumb as possible.
The worse the underlying browser is at rendering rich widgets and media, the more developers and users will want your plug-in. If you are both the vendor of a browser (say IE) as well as the proponent of a plug-in (say Silverlight), then the incentives get truly twisted.

In some way, what he says is very similar to what I have said since a while: we need a new generation of Browsers which are not constraining people from developping applications delivered through the web (see here and here and here for a summary of my opinion on this topic). In that sense, Chrome may be the start of an answer (unfortunately, I say, as it comes from Google instead than from the Open Source community…).
If we want to avoid the risk that Flex and Silverlight will dominate the Web, we need to address this kind of question, which can be summarized by what I found in this other article

We’re in a transition point between the Age of Web Apps and the Age of RIAs (in the web space, that is). And if you doubt that we’re at this transition point, or if you think that RIAs include web apps, ask yourself, does AJAX really give you “all the rich you need”?

Can AJAX really, as Jef Raskin famously stated [60], treat all user input as sacred? Is AJAX really the end all and be all of a Compelling User Experience? Or do we remember that applications used to run outside of a browser?

Of course, it is provoking. But the risk is quite present.

Google strikes back

So, here it is, the long awaited “Google Browser” (called Google Chrome, but the site should go online only tomorrow) has been unveiled in an unconventional announcement in the guise of a comic book.

For the moment, I hold any new comment. I read my old post (from last August). Let’s see if this move will actually make the battleground more free ( by removing the artificial obstacles that an evolution of the Browser technology found because of the war between IE and Firefox) or it will simply be a vehicle by which Google will transform its “presents” (GMail, GCalendar, G<something else>…) into “de-facto” standards.

The initial announcements explicitly thanks what Firefox and Apple Safari did and, more important, commits Google to open-source the innovations that are certainly present in the new Browser.

I suggest people to start reading this post from John Paczkowski, especially what he says at the end:

with its view of the Web as a Web of applications and its multi-process/multi-application design, Chrome almost seems more an operating system than a browser, doesn’t it? Funny, isn’t it. Google’s long been rumored to have been developing a browser and an OS. Who would have known they’d be the same thing ?.

Without having seen and tried yet the Google Chrome browser, I tend to agree with John on the fact that Google is probably shooting towards something that is more an RIA platform than a simple browser.

I would only ask a question. Given the “open source” nature of Firefox, why Google deployed another open-source initiative instead of joining the forces around Firefox ?

Let see when we will better understand how Google Browser is done.

User as center of the Universe

I am slowly catching up with some articles I read and over which I wanted to comment. I am dealing with this one SOA needs RIA – Burton Group, because there are few sentences I liked and because it lacks, in my opinion, a proper “end”.

The Value Hierarchy of Web 2.0So, here are the quotes I liked most:

  • “We firmly believe the user experience needs to be a first level priority at the same level as SDLC, platform languages, SOA and security.”
  • “If the business depends on people and people depend on information technology, then the interface between people and information technology — the user interface — naturally has to be very good. If you have an ineffective user interface, you’re going to have a less effective organization.”
  • “…people are the platform. IT is ephemeral. It continues to change over time, but what does not change in business is that the quality of any organization depends on the quality of its workers.”
  • If developers think the goal of SOA is to provide agility in assembling loosely coupled Web services into an application that provides real-time sales data to managers and marketers, they are missing a key component in the Burton view:  “The idea is to make user experience the end goal of any IT initiative and not an afterthought.”

http://hinchcliffe.org/img/useruniversecenter.jpg

I, personally, subscribe to all the above statements. They remember me a very nice article I read a couple of years ago, from Dion Hinchcliffe, titled The Web2.0 Trinity: People, Data and Great Software. The pictures in this post are both taken from Dion’s article, and I use them consistently in my talks around Web2.0 and the evolution of Desktop technologies.

Going forward, there is another quote that my few readers may appreciate:

“We see the next step as RIAD, the rich Internet application desktop. Here you need to look at Adobe AIR, Google Gadgets, the Microsoft Widget Library, to see resident applications that provide you with a visual experience associated with RIA.”

This is even more close to what I have often written in my blog: moving beyond the browser (as we see it today) towards a mechanism where applications, delivered via the web, will be executed locally. GREAT !

What seems missing to me is the very last part of the article

In Burton’s view, the future of the UXP is in using Web widgets, portable chunks of code and gadgets, miniature objects that can be placed on a Web page to provide dynamic content.

With widgets and gadgets, real-time sales data is on the sales manager’s desktop without requiring him to do multiple click-throughs to find a table or chart, the Burton analyst said.

What I think is missing is the name to this approach, a name which already exists. It is called Mashups, isn’t it? What is needed is the possibility to define those widgets in a standard way and be able to mix and match them in different contexts: a Portal, a Mashup environment, a Rich Client, the desktop even….

How to be an instant Web me-2.0 developer

This article [1] [1] from Verity Stob [2] [2] at The Register [3] [3] made me laughing!  How true it is in many aspects.

I really liked it all, but I think I will use these two pictures in my next Web2.0 presentation to present the difference between Web1.0 and Web2.0:

Block diagram showing Web 1.0 program architecture Architecture diagram illustrating the confusion inherent in Web 2.0 applications

Of course, I was also laughing (and strongly agreeing) with those other comments:

  • Java Applets ?
    I bet Sun hopes that everybody had forgotten
  • Google Web Kit (GWT)
    Eughh! what were they thinking ?
  • Dojo
    Perhaps come back in a year, if they make a design environment to go with.

As to the last sentence on Dojo, I think that it is really something we may need to consider. We need to hide the complexity [4] [4] of Dojo behind some easy-to-use design environment which would make it possible for more people to enter the game.

Dreaming of Hiding the Complexity

Whilst the software products are geared towards making people executing things in a more effective way and allowing people to execute things that were not possible before (I agree, this is not always something good… we would live better without some of the software creatures…), I have always thought that the goal of the technology behind the software results (i.e. the technology that allows the production of software) would be to allow the artists (i.e. the developers) to do their job in the best possible conditions.

I remember how much I loved the VMS operating system (from Digital), the powerful CASE environment that was implemented on that operating systems (ah, Language Sensitive Editor…) and the Common Language Runtime.
I also remember how easy and natural it was, a life later, to develop distributed Service Oriented applications in the Forté environment (where Service Orientation and scalability was built inside the language framework itself). The motto from Forté was “Hiding the Complexity” and, indeed believe me, they couldn’t have been chosen a better motto!

Today I have read one of the “2008 predictions articles” and I was hit by the last item:

13. The next big thing. Software development will change to a wider use of code generators. Forget about heavy frameworks, regardless of what programming language you use.  In a simple case, use some XML style sheets combined with the metadata that describes your application objects to automatically generate the code for these objects. On a larger scale, the entire application may be described using metadata and XML, and an appropriate code generator will do the job. So programming will change from writing tedious code that requires lots of coders to describing the metadata and writing custom code generators.

I know, this will remain a dream: Rubik's Cube GameWhy steal the pleasure of fighting against the complexity of building a program that would let the author being proud of the many hours he spent in debugging it and in having a presentation that looks likee what he would have wanted ….?

 

Hiding the complexity and allowing the artist to express his creativity in addressing the solution to a problem (instead than in debugging, in challenging multithreading or fighting against the geometry manager) would be something nice to dream.

P.S.The Author has, also, some interesting observation on Java, AJAX and Flex/AIR.

Java on the desktop is already here!

I have been surprised when I read this article: James Gosling (Sun) : « Java sur le poste client n’est pas à la hauteur aujourd’hui ». It is in French, so I translate the title here:

James Gosling (Sun) : « Java is not ready today for the desktop »

Strange, isn’t it ? The “father of Java” who, 15 years after, makes such a big statement!Well, the reality is different, as we all know.
Eclipse is there and it is there since sometime now. Eclipse is no more only an “open development platform”, but has become ‘a platform for building and deploying rich client applications”: it is called Eclipse RCP. Many people are developing rich Java applications for the desktop (and for the mobile market also) based on Eclipse RCP:

And, not least, IBM is building the new generation of its products based on Eclipse RCP!

The Universal Managed Client for SOA, called Lotus Expeditor. A platform for building enterprise applications and enterprise mashups that bring the power of SOA towards the desktop and devices

The new Lotus Notes 8 client, which brings the possibility of building Composite Applications centered around the collaboration tools

Lotus Sametime, which provides a new frontier for Unified Collaboration and Communication

Sun may not be ready. But the world is not waiting in order to make Java evolving! And Java is bigger than a trade symbol.

Mashups, web2.0 and the SOA cake

I read a commentary around the recent Gartner 10 Strategic technologies to watch in 2008.
In this commentary, Evan Data Corp. Joe McKendrick and Software AG Miko Matsumura say, very high, that even in SOA is not explicitely spelled in the recent Gartner’s report, SOA itself is the basis for what we are building today and in the future. There are some interesting quotes from the commentary that I wanted to highlight here, as they have really a lot to do with what we do everyday.

  • The consumption patterns of Web 2.0 and Enterprise 2.0 are made possible by SOA in this view.
    “The architecture has no value until it’s expressed in consumptions patterns. …The underlying service is just a generic-kind of service, but it comes to life when you put an Ajax interface in front or some kind of cool mashup in front of it. Once you’ve got a platform of business services, you can make mashups or Web 2.0 or a ton of really cool things.
  • “Turning to another goodies metaphor, …SOA is invisible in the same way the recipe for a cake is invisible. Even the most proud baker wouldn’t stop people from eating his cake while he read them the recipe. The consumers of cake or Web 2.0 applications want to enjoy them not hear a dissertation on how they were made, he said.
  • The status of SOA today is similar to where e-commerce was in the late 1990s. At that time everybody was building e-commerce applications using e-commerce tools. “Now, we’re doing the same thing with SOA. We’re saying this is an SOA project or this is an SOA tool. Today, you still use content management and application servers and Java as a language and Web interfaces, but you no longer call it e-commerce because now it’s just apps. It’s just how we do it. We don’t really think of it as e-commerce any more, it’s just the typical pattern for applications these days. I think exactly the same thing will happen with SOA.”
  • “When you say SOA no longer matters, it’s everything that SOA enables that matters, I totally think that’s right because SOA is a way to achieve certain things from an architecture and an alignment and agility point of view,”

I like all these quotes, because they really make the point!

Going back to what Gartner asserts, I obviously like the presence of the following 3 items in the top-ten list:

  1. Business Process Modelling
  2. Mashups and Composite Applications
  3. Web Platform and WOA

My readers know how much I consider “Business Process Modelling”, at the point that I did not hesitate to say that it is the glorification of any SOA, the way in which Services could become useful from a Business Point of view. I am not sure, though, that BPM will emerge (finally!!!). Not because it should not deserver a shining place, but because of the power implications it brings into a company’s organization (who owns the process owns the power….).

In this context, though, the emergence of the Mashups and Composite Applications, may slightly change the picture. “They allow you to rapidly tailor the functionality you want in one place, without having to re-create the original”  is the quote from Gartner. I still think what I wrote last year in “Composite Applications, Mashups and Portals: relay race or team spirit?” . Through Mashups and Composite Applications, the user will become an actor in the SOA. SOA will not stop anymore at the beginning of the HTTP pipe on the server…. it will continue, it will encompass the desktop.

The user will be allowed to integrate what the “portal” gives him with tools and content coming from elsewhere. The “portal” will provide the official company process and the mashup will provide the creativity, the differentiator by which a user would tailor the standard process and add his own touch !

  • RSSRSS
  • Social Slider
  • RSS
show
 
close
rss Follow on Twitter facebook linkedin