Monday, July 5, 2010
Discovery and AMP
One of my favorite new features in Eclipse Helios is not really about the Eclipse tools at all but how users get those tools.
The discovery mechanism was originally built by the Mylyn project and the folks at Tasktop. But it was quickly and enthusiastically taken up by individual Eclipse projects and by the foundation marketing team. If anyone needs an example of how Open Source works when it really works well, just take a look at this bug. It's a beautiful thing.
Last year had already brought enormous improvements to the Eclipse application delivery (P2) process, so there was solid technology behind all of this. What all this means is that now in the space of a little over a year we have gone from a situation where it was -- quite honestly -- really frustrating for even very experienced users to find and install the various bits of the platform they needed to one where brand new Eclipse users can grab a single distribution, launch it, get the additional functionality they need and deploy it. And I mean, like, for real.
I wrote a post back in February about the idea for an Eclipse Application Store -- and shouldn't have been at all surprised to see that that was hardly a unique vision. Eclipse can, should and I think will be seen for what it is -- a truly cross-platform application environment in which small and large application vendors can innovative on a level playing field. Think the best aspects of Flash and the Apple Store combined but for real applications and with a secure, robust platform and independent, respected community. (And hey, with E4 maybe it will even run on the iPad.) In fact, if it is possible for an open source "collective" to have a secret strategy for world domination, then this could it. The Eclipse Marketplace is the key wedge for that -- and as Ian Skerrett pointed out in his Helios presentation, the only real missing piece is the store mechanism itself.
The really interesting challenge for everyone will be how to position the Eclipse platform to application consumers. Can we create an end-to-end, seamless, low-effort path for even the most casual users to obtain the Eclipse platform and install our applications? That kind of effort might even involve rebranding and redirecting some of our efforts to focus on this more consumer oriented space but on the other hand the potential for eco-system growth is not incremental, but orders of magnitude.
Back a little bit closer to earth my real personal favorite adoption of the Discovery mechanism has been the effort that Cédric Brun and others have put into the Eclipse Modeling Project discovery mechanism. For those who aren't familiar with it, the Modeling Project isn't really one big homogenous project, except in the sense that it all begins and ends with EMF. Instead it is a whole ecosystem in itself with set of tools that are diverse targeted and targeted at many different users and use cases. In some cases there are even tools that meet the needs of the same set of users in different ways, or put in a more narrow way "compete". The Eclipse Modeling distribution was getting bigger and bigger and most users simply don't need all of the different tools that it offers. So rather than have an ever-growing distribution -- the "bigger is better" approach -- the Modeling project elected to actually reduce the a core set of technologies and instead provide a simple, easy to use and -- I think this is the most important point -- engaging mechanism for installing other components as needed.
Which brings me to the original point of this post which is that the Agent Modeling Platform is now available in the modeling discovery mechanism. (One of the sort of obvious but overlooked aspects of a discovery mechanism is that because it allows offerings to be defined externally, it balances the predictable and robust release process with a much more dynamic capability delivery process.) This has made the AMP installation process easy enough for my mother to use. (OK, we'll see about that, but she's managed to figure out how to email photos from Windows XP...)
I've even made a movie about it:
The discovery mechanism was originally built by the Mylyn project and the folks at Tasktop. But it was quickly and enthusiastically taken up by individual Eclipse projects and by the foundation marketing team. If anyone needs an example of how Open Source works when it really works well, just take a look at this bug. It's a beautiful thing.
Last year had already brought enormous improvements to the Eclipse application delivery (P2) process, so there was solid technology behind all of this. What all this means is that now in the space of a little over a year we have gone from a situation where it was -- quite honestly -- really frustrating for even very experienced users to find and install the various bits of the platform they needed to one where brand new Eclipse users can grab a single distribution, launch it, get the additional functionality they need and deploy it. And I mean, like, for real.
I wrote a post back in February about the idea for an Eclipse Application Store -- and shouldn't have been at all surprised to see that that was hardly a unique vision. Eclipse can, should and I think will be seen for what it is -- a truly cross-platform application environment in which small and large application vendors can innovative on a level playing field. Think the best aspects of Flash and the Apple Store combined but for real applications and with a secure, robust platform and independent, respected community. (And hey, with E4 maybe it will even run on the iPad.) In fact, if it is possible for an open source "collective" to have a secret strategy for world domination, then this could it. The Eclipse Marketplace is the key wedge for that -- and as Ian Skerrett pointed out in his Helios presentation, the only real missing piece is the store mechanism itself.
The really interesting challenge for everyone will be how to position the Eclipse platform to application consumers. Can we create an end-to-end, seamless, low-effort path for even the most casual users to obtain the Eclipse platform and install our applications? That kind of effort might even involve rebranding and redirecting some of our efforts to focus on this more consumer oriented space but on the other hand the potential for eco-system growth is not incremental, but orders of magnitude.
Back a little bit closer to earth my real personal favorite adoption of the Discovery mechanism has been the effort that Cédric Brun and others have put into the Eclipse Modeling Project discovery mechanism. For those who aren't familiar with it, the Modeling Project isn't really one big homogenous project, except in the sense that it all begins and ends with EMF. Instead it is a whole ecosystem in itself with set of tools that are diverse targeted and targeted at many different users and use cases. In some cases there are even tools that meet the needs of the same set of users in different ways, or put in a more narrow way "compete". The Eclipse Modeling distribution was getting bigger and bigger and most users simply don't need all of the different tools that it offers. So rather than have an ever-growing distribution -- the "bigger is better" approach -- the Modeling project elected to actually reduce the a core set of technologies and instead provide a simple, easy to use and -- I think this is the most important point -- engaging mechanism for installing other components as needed.
Which brings me to the original point of this post which is that the Agent Modeling Platform is now available in the modeling discovery mechanism. (One of the sort of obvious but overlooked aspects of a discovery mechanism is that because it allows offerings to be defined externally, it balances the predictable and robust release process with a much more dynamic capability delivery process.) This has made the AMP installation process easy enough for my mother to use. (OK, we'll see about that, but she's managed to figure out how to email photos from Windows XP...)
I've even made a movie about it:
Subscribe to:
Posts (Atom)
Popular Posts
-
Or something like that.. In this article, I'm going to describe an easy way to create wizards for end-users to create their own plugin p...
-
Hi all. I wanted to let folks know about a new project proposal I'm considering, and ask for your feedback and support. I'm really h...
-
The resource model is one of those aspects of the Eclipse Modeling Framework that is extremely powerful yet can seem way too complicated for...
-
Many of us who've spent time traipsing through the Eclipse wilderness are probably familiar with this story... Tis a beautiful, sunlit...
-
How much has Gerrit changed software development? Well, consider that a couple of years ago most of us had never heard of Gerrit, and now it...
-
I've been spending quite a bit of time lately playing with the EEF framework. C'est magnifique! (No, I don't actually know any f...
-
In my post earlier this week I sighed about the incompatibilities between GPL and EPL, took the FSF to task for their (in my eyes anyway) d...
-
I was excited to see Apple's announcement about the OS App Store. It even got me thinking..what if I wrap up that nice little piece of s...
-
In my last post , I described the current state of Eclipse Gerrit adoption. If you haven't read that post, you might be as surprised as...
-
“So the final lesson of 1918, a simple one yet the one most difficult to execute, is that those who occupy positions of authority must lesse...


