Is the browser the best platform for applications?
A Curious Conversation
While at SXSW, Garrett, Matt and I (mostly Garrett and Matt) got into an interesting conversation about the future of web-services and service based apps. The discussion floated around thin, smart, and thick clients, and whether the browser (which Matt commented was originally built as a document reader) is the right platform— moving forward —for these rich applications. Garrett argued that the browser has been adopted, and is constantly evolving. Matt wondered if we wouldn’t move forward more quickly with a new platform for web apps, which of course opens up discussions around (1) who should build this platform, and (2) who has the power (M$, cough) to get a new platform adopted.
The discussion went something like this: Desktop apps can offer richer UI experiences. Web-based apps offer social sharing and collaboration features, mashup potential, yada, yada anywhere and on any platform or device without installation or update issues.
A Deeper Conversation
All this brought to mind two separate presentations I’ve listened to lately (both freely available from IT Conversations), which give some depth to this discussion:
First, The Platform Revolution where “leading developer and business evangelists discuss and debate the very definition of the words ‘platform’ and ‘developer’ in a Web 2.0 world.”
→ Check out The Platform Revolution
While there’s no clear point to take away from this conversation, there are some nice little tidbits. One that struck me was Halsey Minor’s (CEO, Chairman, & Founder, Grand Central Communications) comment about Enterprise’s reluctance to embrace web services or web2.0 paradigms. He says:
“There are those people who are practical on one extreme and those people who are purist on the other extreme. The practical people don’t do anything because they see all the problems with any sort of new technology. The purist don’t ever do anything that works because they don’t realize how many practical people there are in the world that they have to deal with… We try to be practical purists. We know that this new architecture is coming but we also recognize that there are a lot of ideas from traditional, say, enterprise software that aren’t going to go away, like people want reliability, they want service level agreements, they want guarantees…”
This quote struck a cord. Since joining with Geniant, I’ve been exposed to the Enterprise side of things—and it’s appalling. The Enterprise level Software Apps I’ve encountered are in the dark ages—pre usability days! This is something you’ll probably see quite a few posts about in the upcoming months, starting with this one.
Second, Orchestrating the Stack is a more heavyweight presentation from the often dead-on technology strategist/analyst Geoffrey Moore.
→ Check out Orchestrating the Stack
If you don’t recognize the name, you’ve definitely used or encountered his ideas—most notably his ‘Crossing the Chasm’ model illustrating the technology adoption curve and the gap between early adopters and mainstream acceptance
In this presentation, Moore describes where the tech industry might be heading in this decade—not this year or next—but farther out in what he callas ‘horizon three’. He describes a technology stack that includes everyone from Microsoft and IBM to Intel and Cisco.
If you listen to this presentation, you’ll definitely need the slides, as he constantly refers to them. This is one presentation to listen to if you want a view of where things might be headed—a view that offers much more than the buzz and hype around Web2.0 apps. He makes some pretty interesting prognostications.
Enjoy!
Comments closed for this post.
On Mar 17, 01:38 PM Matt said
I’m not sure I quite agree with your summary of my conversation with Garrett.
We both agree that web apps are great, and have a thousand advantages over traditional desktop apps. One of their drawbacks, though, is that web browsers aren’t very good at providing rich, responsive user experiences. They just aren’t capable.
We both agree that in a few years, web apps will be better than they are now at providing that rich user experience (at least when viewed on a Desktop PC—you’ll obviously never have get drag-and-drop on your cell phone).
Our only point of contention was this: He thinks we should keep shoehorning our increasingly-complicated applications into the web browser, and that over time, the web browser will evolve into something that can give the rich user experience that’s so familiar to users of desktop apps. I think it’s kind of dumb to spend the next ten years forcing the web browser to do something it wasn’t designed for. I think we should let the web browser continue to be just a document reader (and let people use it to submit the odd form), and create a different kind of client that’s like a web browser, but designed from the ground up to be an interface-presenter rather than a document-reader. I think my approach is faster, smarter, more direct, offers better backwards-compatibility, and produces a better end result.
The only real hurdle for my approach is that you’d need the interface-presenter installed on a certain percentage of computers to make it even worth developing for. Fortunately, there are 2 or 3 companies (the best candidate is M$ lol, but I think Google could do it, too) that can build an interface-presenter and get it installed on enough machines to give it the critical mass it’d need to take off.