tag:blogger.com,1999:blog-2447174102813539049.post2375357816491925619..comments2024-03-27T03:52:14.195-07:00Comments on Room 101: SOBsGilad Brachahttp://www.blogger.com/profile/17934280339206214042noreply@blogger.comBlogger8125tag:blogger.com,1999:blog-2447174102813539049.post-11474229563164756762007-05-08T21:15:00.000-07:002007-05-08T21:15:00.000-07:00Yardena,The Rich Internet Application space is ind...Yardena,<BR/><BR/>The Rich Internet Application space is indeed heating up. I have been looking at this for the past 3 years, and had many conversations about it with Dan Ingalls. Dan is an amazing guy, and has some very talented people working with him.<BR/><BR/>Right now, Sun is pushing JavaFX (formerly F3), a project by Chris Oliver. It's based on Java and Swing, and so is politically attractive.<BR/><BR/>However, it's along way from these technologies to the SOB vision. Adobe's Apollo is a bit closer, but I think it's too early to say how things pan out.Gilad Brachahttps://www.blogger.com/profile/17934280339206214042noreply@blogger.comtag:blogger.com,1999:blog-2447174102813539049.post-83794045728670152512007-05-05T13:37:00.000-07:002007-05-05T13:37:00.000-07:00Oops, hit enter too soon... I meant to say, that g...Oops, hit enter too soon... I meant to say, that given JavaScript VM idea and Dan Ingall's SmallTalk background it may evolve in that direction. But I don't see any technical details about the project :-(Yardenahttps://www.blogger.com/profile/15649241856669571499noreply@blogger.comtag:blogger.com,1999:blog-2447174102813539049.post-73008151809441312802007-05-05T13:27:00.000-07:002007-05-05T13:27:00.000-07:00Maybe the Flair projectMaybe the <A HREF="http://www.infoworld.com/article/07/04/26/HNflair_1.html" REL="nofollow">Flair project</A>Yardenahttps://www.blogger.com/profile/15649241856669571499noreply@blogger.comtag:blogger.com,1999:blog-2447174102813539049.post-81035951161847739782007-04-09T16:39:00.000-07:002007-04-09T16:39:00.000-07:00Warren,If I were to build such a beast right now, ...Warren,<BR/><BR/>If I were to build such a beast right now, I would use Smalltalk. Lisp would work as well.<BR/>Lisp and Smalltalk runtimes remain the best suited for this sort of thing, because they are designed to allow live incremental updates to running programs, while being at least moderately performant.<BR/><BR/>The implementations of popular scripting languages remain immature in comparison, though this will change over time.<BR/><BR/>When (not if) Javascript implementations improve in terms of performance, they will be very well positioned as a vehicle on which to build on due to their ubiquity in web browsers. We see hints in this direction in, e.g,. Adobe's Apollo release, but it is still a long way from what I'm advocating.<BR/>One also needs some sort of standardized object store across browsers, and that will take time. <BR/><BR/>I don't expect any of this to happen tomorrow, but over time I see compiling something Smalltalk-like (conceptually; surface syntax would probably be more conventional) into a web browser/javascript combination as the most likely path toward realization.<BR/><BR/>Most importantly, The mindset of programmers has to change - and that is the hardest challenge of all. Today people are discussing the merits of things like OSGi, which are light years removed from where you want to be. So be patient.Gilad Brachahttps://www.blogger.com/profile/17934280339206214042noreply@blogger.comtag:blogger.com,1999:blog-2447174102813539049.post-48499295752824768642007-04-09T11:50:00.000-07:002007-04-09T11:50:00.000-07:00Hi GiladThe google video and your oopsla paper are...Hi Gilad<BR/><BR/>The google video and your oopsla paper are both fascinating. The end of versioning as we know it sounds like the only way out of this forest of complexity...<BR/><BR/>I guess I am curious as to how you see these ideas evolving into practical implementations, and how long do think that might take? <BR/><BR/>Is there a language or VM environment that is a good candidate to implement these concepts? <BR/><BR/><BR/>Thanks<BR/><BR/>WarrenWarren Strangehttps://www.blogger.com/profile/11832307765436664938noreply@blogger.comtag:blogger.com,1999:blog-2447174102813539049.post-21595325114154043482007-04-09T11:46:00.000-07:002007-04-09T11:46:00.000-07:00This comment has been removed by the author.Warren Strangehttps://www.blogger.com/profile/11832307765436664938noreply@blogger.comtag:blogger.com,1999:blog-2447174102813539049.post-70959579423123477372007-04-01T12:47:00.000-07:002007-04-01T12:47:00.000-07:00Hi Paul,Thanks for the comments. This post has not...Hi Paul,<BR/><BR/>Thanks for the comments. This post has not elicited much comment - perhaps because it is too 'blue' as you would put it.<BR/><BR/>On the other hand, when I present this talk in person to a hardcore programming language audience, it eleicits a great deal of comment.<BR/><BR/>I took a look at your blog posts. I agree with the general drift, though some of the technical and historic details are not quite accurate.<BR/><BR/>Overall, the trend is good. The 'pink' culture is in decline. Nevertheless, it has a lot of strengths, and my goal is to produce a good synthesis pink and blue.Gilad Brachahttps://www.blogger.com/profile/17934280339206214042noreply@blogger.comtag:blogger.com,1999:blog-2447174102813539049.post-59191163405869434162007-04-01T07:00:00.000-07:002007-04-01T07:00:00.000-07:00Hi,I watched the video, and it has had me thinking...Hi,<BR/><BR/>I watched the video, and it has had me thinking ever since. I am an agile coach, and a see a lot of parallels with the ideas in the video and Agile development. The common theme is change and how you deal with it.<BR/><BR/>The old idea of software as an artefact and of waterfall software development is that some how the development process is deterministic, fixed and pre-defined. This is OK for software that solves fixed problems like a dinks dispenser controller, but IMO when you move onto more people centric applications like the web, or business, the real challenge becomes creating software that deals with change.<BR/><BR/>So the idea of systems that evolve organically and where bits rot as you describe, is central if software systems are going to grow in complexity and deal with human scale problems like better communication and collaboration amongst individuals and communities.<BR/><BR/>I see late-binding and what I've labelled 'blue' OOP as central to this and have created a series of <A HREF="http://pab-data.blogspot.com" REL="nofollow">blogs posts</A> about Object Technology which were inspired by Alan Kay’s keynote OOPSLA speech in 1997 were he discusses the ‘pink’ and ‘blue’ idea planes. So far the responses to my posts haven’t been that positive.<BR/><BR/>I think generally, that programmers like the idea of a fixed, static, deterministic model, and aren’t that used to using the other side of their brains :^)<BR/><BR/>I look forward with interest to see what other tasty tit bits you have in store for us all.<BR/><BR/>Thanks for sharing your work. It is really appreciated.Paul Beckfordhttps://www.blogger.com/profile/16046651614960778254noreply@blogger.com