Web 2.0, RIAs, AJAX, and the semantic web (lowercase s)
Enough buzzwords for ya?
A long time ago, I remember having a conversation with Tantek Celic about how RSS/tags/etc have started to form a loose taxonomy of information. Tantek calls it the semantic web with a lowercase s. The reason this has so much momentum is because it’s a grassroots effort. Instead of creating a taxonomy the way a librarian would (top down), each person decides how to tag information, and tags tend to coalesce. He later joined Technorati, which just shows how strongly he holds these convictions.
Nowadays, there’s a lot of talk about Web 2.0, web mashups, AJAX, etc., which in my mind are all facets of the same phenomenon: that information and presentation are being separated in ways that allow for novel forms of reuse.
~
To some extent, this is closely related to service-oriented architectures, except that saying SOA makes it sound like you wear button down shirts, work at a large company with lots of consultants and so forth. Saying “web mashup” sounds like.. well.. the opposite.
Until the recent popularity of AJAX, the architecture of HTML-based Web apps and Flash/Flex based applications was quite different. In old school HTML-based apps, the controller logic and most of the presentation logic resides on the server, and feeds relatively dumb content down to the browser, which means that all the plumbing between the presentation and the back end services was all done behind a curtain.
Now, with AJAX-based apps, the client makes service requests and assembles presentation on the client machine, which is exactly what Flex/Flash-based apps do, and is one of the drivers behind Web remixing, as Jonathan Boutelle pointed out recently.
What would be nice to see at this point is a set of least common denominator interchange formats for common types of data and operations. Will this ever happen? Hard to say. Unlike with tagging, which is driven by the masses, APIs are driven by a few companies. The Flickr API is different from the smugmug API. What would cause that to change?
I think the only way this could change is through grassroots efforts, but I can’t quite picture how that would evolve. Perhaps some guerilla projects by people to write adapters between APIs? The FOAF project is one example of a group trying to standardize how information is represented, although I can’t tell how much progress they have made.
In any case, at the risk of stating the obvious, all this is great for RIAs. The trend of increasing separation between front end presentation behavior and back end services allows RIAs to get in the mix. Applications like Paul Neave’s excellent Flash Earth wouldn’t be possible without it.
>> Enough buzzwords for ya?
The only missing one is GUI, as in GUI applications. ;)
From reading of your posts, I get an idea that lot of this RIA stuff is still in its definitional stages.
Considering that GUI has been around for quite a while and the fact that it’s still looking for an eloquent deployment solution, why not start Zorn off with a GUI applications version ?
In the process, you would end up its mass introduction to the corporate world, where GUI applications are prevalent and the RIA ones are still around the corner.
Considering that Flex already has enough GUI controls, all you would really need to to is to repackage then within an Eclise IDE.
So, instead of waiting a year to year and a half for a product that still needs to be defined, within a quarter to a quarter and a half you could get the ball rolling by coming out with a slimmed down version that mostly the same masses want yesterday.
Thanks for the comment.
I think that Zorn in its first incarnation will allow people to put together relatively traditional (client server-like) application interfaces. However, the nature of Web delivery and the reality that RIAs need to live alongside traditional Web applications and content means that even if you deemphasize the “R” part of RIA, you will still need to think through some new issues. That is why, for example, the move toward service oriented architectures is so important for RIAs.
Hi Sho:
>> Thanks for the comment.
Don’t even mention it and especially when you consider that I’m just trying to accumulate enough participation credits for an an early alpha-tester invitation. :)
Kidding aside ;) I just happened to be on the Adobe site, where a whole series of brain flashes (pun inted) got activated.
I was watching a demo with a “Flash-Guy”(:TM:) walking around within PDF documents and that is when I realized that the biggest bang for the buck that MacrAdobe could deliver would be an “Integration Product”.
It would simply put together into a cohisive package, all of the existing wonderful products that already exist within your two fine organization.
Just imagine the power of being able to develop traditional GUI applications that are seamlessly integrated with documents management as well as cohesively integrated teaching.
Just imagine the power of such integrated solutions that would enable self-teaching for the end users and especially when you consider that we are in this electronic interactive age, where nobody reads documentation – not printed documentation, nor even online help.
Therefore, before going down some long winding path in search of a definition of some future RIA mouse-trap, why not first start out simply by re-inventing the old one that is sqeaking right in front of all of our noses ?
P.S. Does your marketing read any of this ? ;)
Tinic & Sho
Tinic & Sho: After pointing to Tinic Uro’s blog below, I realized some similarities/differences with Sho Kuwamoto’s writing — they’re in dramatically different places on their product cycles. Tinic is just wrapping up a round of development on the Mac…
I used to work at TIBCO Software, and an integration tier was their business actually. They built a HUGE collection of what they called adapters that normalized it to a common communication protocol/message framework. This was long before WebServices and XML became popular, so back then there really wasn’t much of a choice. But using their Adapters you could build tools that generically could connect to any backend without having to know the specifics (SAP, PeopleSoft, whatever..).
Even if I agree on most of the content of you article, I disagree, however, on the fact that FLEX is a client side solution. It involves translation form FLEX into flash at the server side.
Hi Dider,
What you say is absolutely true for Flex 1.5. With Flex 2.0, you will be able to compile Flex applications and run them without a server.
In addition, the Flex 2.0 compiler and SDK will be available for free, so you will not have to pay us any money whatsover to create and deploy a Flex application.
Does that answer your question? :-)
Data Mashups
Recently, I’ve been wrestling with the term data mashup. I think by now we all know what a Web mashup is, or at least have seen an example using Google Maps. But I’ve heard the term data mashup mentioned a