Saturday February 14, 2009
Saturday February 14, 2009
A time of transition and focus
It's been some time since I've blogged ... and past time for me to get back into the habit of putting thoughts to virtual paper. Part of the problem is a tendency to not blog when there isn't anything profound to convey. Personal foible, something to get over (consider it a New Years Resolution ... late, of course) ...
So what's change in the intervening half year ? Quite a lot, and I can't do justice for it all without writing many, many paragraphs ... coming right after this post (and in no particular order!)
Posted by brewin Feb 14 2009, 12:50:34 PM PST Permalink Comments [1]
Monday February 09, 2009 Last week, we (Sun and MySQL) said goodbye to one of the icons of open source, the main author and one of the founding members of MySQL, Michael (Monty) Widenius.
As he put forth in his blog Monty Says, Monty has moved on, and joined a company he founded (Monty Program AB). His stated goals ? To continue working on his Maria storage engine project, to actively work within the MySQL Community and interestingly enough open a restaurant (which even though I'm not a database developer, I still hope I get the discount he offers).
Working with Monty was an experience I will always remember. His passion about community, about open source and about MySQL was palpable. For anyone who knows him or had the chance to work with him, you'll also know that discussion about those topics frequently were lively and sometimes at odds with the established plans or directions. He loves to challenge assumptions and did so with vigor -- and whether his position was correct or not, whether he won the argument or not -- some of the value really wasn't in the end result anyway ... it was the journey, from which new insights and ideas would frequently spring to life.
I wish him well in his endeavors, and look forward to the day our paths cross again.
Posted by brewin Feb 09 2009, 12:32:16 PM PST Permalink
Friday August 01, 2008 One of the things which here in the U.S. we see every year is the arrival of the "Summer Blockbuster" ... usually the incredibly expensive to produce, star studded and special effects laden Hollywood movie. They usually start around Memorial Day and through Labor Day, a large number of these movies start arriving week after week (this year it was movies like "Hancock", "Batman", "The Mummy 3" and "Iron Man").
As of Thursday, July 31st, Sun reduced it's preview of it's own Summer Blockbuser, that is the preview release of JavaFX, now available with everything you need to start working with JavaFX (libraries, samples, documentation and an early developer tool).
Go forth an play ... soon to come will be a more fully rounded set of tools (including designer tools) as well as more components, samples and a runtime that allows you to deploy applications and content across the web, the PC as well as other devices like Java ME enabled phones.
Posted by brewin Aug 01 2008, 09:55:37 AM PDT Permalink Comments [1]
Wednesday October 31, 2007
Project Indiana Developer Preview -- Live!
This evening, it became official. The developer preview for Project Indiana went live. This is a landmark occasion and a tremendous accomplishment by a dedicated team of professionals ... both in Sun and others in the OpenSolaris community, once again demonstrating the commitment Sun has to Open Source and Solaris.
Download it, try it ... participate. Go get it today! Posted by brewin Oct 31 2007, 07:25:38 PM PDT Permalink Comments [1]
Friday July 20, 2007
JavaFX Compiler
An interesting thing happened this morning, and that was the launch of the JavaFX Compiler project. The JavaFX compiler will allow applications to be compiled to byte code, in the same manner as with javac, such that they will run on any compliant JVM but with significant performance improvements over what's possible using the interpreter. Chris Oliver posted some early benchmarks which demonstrate the potential of this new addition to the JavaFX universe.
The JavaFX compiler is built upon the same infrastructure which composes javac and what's perhaps even more significant is that this project is being launched into the open source community under the GPL license very early in it's development cycle. We at Sun truly want the participation of the community in the design, development and eventual deployment of this promising compiler and it is hoped that with the availability of the project and collaboration with a large development community, the entire JavaFX community will reap the benefits.
Posted by brewin Jul 20 2007, 08:43:50 AM PDT Permalink Comments [2]
Friday May 18, 2007
Leaner and Meaner
I'm in the middle of drafting a "problems with Java today" blog ... but Chet describes part of those problems (and the solutions!) coming in the "Consumer JRE" ... worth a look ...
- Bob Posted by brewin May 18 2007, 06:17:55 PM PDT Permalink
Tuesday May 15, 2007
Back to the Future
Last Tuesday at JavaOne, Sun launched what I think is a game-changing set of technologies, a new product family and as well a new initiative around improving client support in Java, one that I believe will fundamentally redefine how we express rich, compelling experiences to consumers leveraging the Java platform as the vehicle for the delivery of those applications and content. To understand the reasons why the announcements around JavaFX Script and Mobile are so important, one has to take a look at the evolution of modern web-based computing as it relates to client technologies, products and devices.
The question I will pose to you here, is whether a number of technologies introduced into the web are in fact, evolutionary or revolutionary ... and while at times it would seem that the entire ecosystem has been revolutionized periodically, I believe we can look back through history and determine that yes, while in the moment that may be so, over time we can see certain evolutionary trends.
For instance, going back a few years, we see the change from what were classic "Terminal Applications", that is extremely thin clients connected to (usually) large rooms of computers to that of Client/Server architectures ... typically represented as those emerging as part of the PC "revolution".
(there's that word again, "revolution")
This diagram represents that shift between Terminals and Clients, and for this I will use roughly the amount of code applied to each side of this binary equation ... code on the client, versus code on the server ...
Stepping forward a bit, we see the emergence of the Internet ... or the modern web. And while not nearly as severe a shift in cycle as that between Terminal Applications and those in the Client/Server space, we do see a move away from the Rich Clients which dominated the early PC life cycle with that of applications typically architected solely around the web browser ... which in the early days was static HTML delivered to the client by the web server.
Enter the modern "Web 2.0" era, and the rise of the Rich Internet Application ... as more was required of the modern web client, the injection of intelligence into the richer web client caused once again, a swing back towards more code executing on the client, such as JavaScript and other languages which get pushed to the client by server-based applications. So it's not a true re-biasing towards clients per se, although they are certainly richer for things like AJAX, etc., but more like server-controlled or server-managed clients ....
There is a well established engineering principle typically used in designing aircraft, a function of aerodynamics referred to as stability. Basically, the stability of an aircraft in flight to remain in straight, level and upright flight and to return to that attitude without corrective action by the pilot. The overall tendency of an airplane to do so is called dynamic stability, which is preceded by a series of damped out oscillations ...
Stability can be positive ... which means that as a result of those oscillations, the plane will return to stable flight. Stability can also be negative, which means that as a result of those oscillations, the plane will move it farther away from stable flight ... basically, unstable flight. Most modern military aircraft have that tendency ... without computers, they are in fact, a flying brick ...
So what forces operate against this cycle of the client and the server ? The current trend is one of increasing stability (less oscillation) ... so what are those possible forces which are in effect dampening the wild swings biasing towards client or server and serving the function of creating a more balanced result ?
For one, created code. If we look at this force over time, the drive towards the assembly of applications as opposed to the creation of new code has increased. The rapid rise of interest and prevalence of scripting being used to tie together existing components and functionality especially with modern RIAs is one example of this trend.
The other is the demand on rich interactivity, which includes the use of rich content and media. The need for compelling and rich UIs, coupled with the need to deliver and manage this content in real time as well as the intelligence required in the modern client application for interacting with the modern web is a force which places a high demand on richer clients.
A third force is one which is, in many ways, the "great equalizer" .. and directly effects and has driven the modern web, that of collaboration and the need to bridge the requirements and capabilities of the increasingly richer clients as well as the data and services present on the server. This force is manifest in this drive to continue to improve both the user experience as well as the performance, scalability and in general overall usefulness of the web ... and collaboration and integration is integral to that forward progress.
The result, of course, is this notion of the "web as a platform" ... a platform that provides the wealth of services needed by increasingly richer clients and an increasingly large set of services and data available underneath.
We are today living in the world of the rich internet application built on the web as a platform, yet because of the need for something more, struggling to somehow build a richer, yet more tightly integrated client into that model. A client which is increasingly integrated with the underlying web as the platform for collaboration, but also increasingly integrated with it's host environment, whether that's a desktop, a mobile device or other more consumer-oriented platforms.
So the question remains ... what next ?
What's the point of increasing balance in this cycle of web computing ?
That Integrated Rich Client is the next step in the evolution of the web computing client, currently dominated by the rich internet application, which today leverages things like AJAX to provide for that richer experience people are demanding. The Integrated Rich Client is about the following:
- A rich, compelling user experience that, available as a modern desktop application, content which should include not only static visual assets, but time-based media and dynamic UI effects
- That same experience on the desktop and as close as possible with the browser
- Full integration with the web platform as is available today with the modern RIA, which includes the ability to continue to leverage traditional "web" assets (even within the context of a rich application) ... so HTML, Script and CSS have to work
- Full integration with the host environment (example, the desktop and system on a desktop or functional elements such as a telephony stack on a mobile phone)
- The integrated rich client is not bound to a browser. It may be a stand-alone application which can run on any consumer client device ... for instance desktop/laptop, phone, kiosk, TV or other media receiver or player
- The integrated rich client, while a networked application / device, shouldn't require the network at all times to be functional ... the ability to continue to provide a functional and usable application and experience when disconnected (either because of intermittent, partial network connectivity loss or lack of network) is critical
- A similar (or even in some cases identical) experience for a given application should exist, regardless of client
We are now at the point where the confluence of available technology and consumer demand makes this all possible.
At JavaOne, we demonstrated elements of that solution, technologies such as GlassFish V3, Derby in the browser, Java SE futures and JavaFX Script. Each provide a component piece of where the future lies within the integrated rich client. Within products coming from Sun, such as JavaFX Mobile, you'll see the realization of those technologies in a fully integrated distribution, initially for mobile devices and eventually on a broad spectrum of devices ... from set top boxes to desktops. We'll go back to the future, but a future different than the "thick" of old, one that leverages the modern (yet evolving) web and the rich set of services and data underneath, utilizing the rich Java clients that will be built upon a leaner JRE and dynamic and compelling user experience.
Posted by brewin May 15 2007, 03:28:53 PM PDT Permalink
Monday May 14, 2007
FIC and OpenMoko - Thank You!
Something was inadvertently missed during the avalanche that was JavaOne, and I'd like to make sure that the world recognizes that one of the more visible players in last weeks JavaFX news was not mentioned (no excuse for that, my apologies to everyone involved).
While the photos used to represent JavaFX Mobile clearly displayed the label and manufacturer of the device that Sun used on stage (with Jonathan, Rich as well as myself), proper attribution wasn't made to the company which helped make that presentation possible and I want to make sure that credit is given where credit is due. You all may be wondering "where'd we get that phone?" and "who is FIC?".
FIC (First International Computer) is the company which created the hardware device that Sun used at JavaOne last week. It is a compelling and very powerful mobile handset that served as the host for the JavaFX Mobile runtime at the show. It should be noted that while we installed JavaFX on a number of handsets, in most cases the Neo 1973 phone performed better and had a brighter and more compelling result than some of the other phones we used. It bodes well for the future when that device becomes available to everyone.
What may also not be as well known is that FIC also founded and is the driver behind the OpenMoko project, which is delivering an open source, Linux-based mobile software system running initially on the Neo 1973 handset. As all of you know, we at Sun completely support all such initiatives within open source and for community benefit ... the folks at FIC and those who are part of the Open Moko Project are to be commended for their thought leadership here.
So to FIC and the OpenMoko project, Thank You ...
Posted by brewin May 14 2007, 02:39:00 PM PDT Permalink Comments [1]
Friday May 11, 2007
NASA World Wind is Live !
For everyone who has been waiting to get to the NASA Java World Wind application and SDK, the site is now live!
Go to the NASA World Wind site and start to play ( I just started myself )
Go out there and innovate and create something new with this revolutionary and embeddable Java technology Posted by brewin May 11 2007, 07:00:00 PM PDT Permalink Comments [1]
Yeah, yeah, yeah ...
So I've been getting blasted recently about updating my blog ... given the mad rush leading up to JavaOne and the event itself (and those who attend this week long event know that it's not a dawn-to-dusk thing, it's a dawn-to-dawn thing ... ), I haven't been able to finish a complete posting to save my life ... but now that the week is over, expect to see those completed soon ... of course I'll cover some of the messaging from my keynote, JavaFX and as well some commentary I'd like to offer on the Java platform itself. Stay tuned ...
Posted by brewin
May 11 2007, 12:50:21 PM PDT
Permalink
Comments [1]
Sunday March 18, 2007
What's "Good Enough" ?
I've been reading many blogs of late discussing the merits of the Web UI (usually in the context of AJAX) and whether they are suitable replacements for the classic Desktop app.
Joel Spolsky concludes we're doomed to a life where everything is web-centric and we're just going to have to get used to dealing with what is (IMHO) a sub-standard user experience.
I'm fairly sure most have read Bruce Eckel's Hybridizing Java which concludes that the future is Flash as opposed to HTML. The UI is certainly richer, but the cost of being locked into Adobe and the hybrid which is Flex doesn't make me warm and fuzzy either ...
Meanwhile, Jeff Atwood concludes that the path of least resistance is what will eventually win (and no, not happy about that prospect either), meaning the Web UI. Like Jeff, I wonder why we have to settle on second best.
Personally, I hope we're just in this weird state of discovering (or re-discovering) ourselves and will arrive at a state where the Rich Desktop re-materializes as the UI of choice. We've made a ton of progress over the years in rich graphics, 2D, 3D, hardware (and software) acceleration ... where we've failed (I believe) is in making all that technology accessible to "mere mortals" and as a result:
- We've made it way too hard for your average developer (or worse a designer) from being able to develop a reasonable UI, one with a reasonable user experience that performs (at all)
- Impossible to rapidly iterate on a design, meaning that in the end we either end up with a crappy UI which is largely programmer-built or a sub-standard app built by a designer which integrates poorly with the rest of the system (or is unmaintainable, usually because of automated tools)
- Arrived at systems which are largely hacks, for instance the plethora of XML-derived variants (where XML might be "OK" to some for layout, but lousy as a programming language)
I personally believe we're on (or will be on) the path that takes us to the RIA of tomorrow. Historically, users/consumers accept "Good Enough" as a new technology or product becomes available ... but only for a while. Witness the simple cases of latency, where we've gone through multiple cycles where high-latency was acceptable on early network-based systems but as the technology and volume requirements grew, users required increasingly lower-latency for their (then) desktop apps. With the web, the cycle continues and while for now users are willing to deal with "Good Enough", that trend will not continue. Users will want a richer experience, they will not want to be bound to the idiosyncrasies of the browser model, they will want more direct interaction with their desktop and hopefully they will in time reject being bound into proprietary models for achieving this.
For one such possible future, look at what Chris is doing ... (more later on that specific project ...)
Posted by brewin Mar 18 2007, 09:31:50 PM PDT Permalink Comments [1]
Five things
This one I've been procrastinating ... but I just got tagged for the second time, first by James, and now again by Steve (and no, I'm not going to do mine with video ... Steve obviously has waaay too much time on his hands ...). Time to get it over with I guess ...
1. My first computer job was working for a small retail shop when I was still in high school. My job was fairly simple: I would take an IBM Selectric typewriters and convert them into printers for computers (the process is somewhat described here, but as usual ... it was more complicated than that ...). In principal, it was a grand experiment (for this little company anyway) where people would bring in their Selectric and I would spend the entire day "grafting" solenoids into the machinery ...
2. When I was much, much younger I was in the U.S. Marine Corps (and I think I need to chat with the Commandant at some point ... simply because the web site uses Flash, not Java ...). For the majority of my time, I was in armor (meaning tanks). If you've ever met me, you have to wonder how I ever fit in one of these "wee" beasties ...
3. While in the Marines, I spent a year living in Okinawa, Japan. Has to be one of the best places I've ever lived in (I lived in the northern, most rural part of the island). It's essentially a Japanese version of Hawaii and I lived their with my wife and oldest son (who was born on the island) while we were there. While in Okinawa, I did a stint as (which is odd if you read Rich's blog) a lifeguard (technically Water Safety Instructor), teaching folks, including other lifeguards how to do everything from swim to survive for long periods in the water (which is a good thing for a surfer to know ...)
4. The most interesting place I've surfed so far is Costa Rica. Spent many an hour driving across miles of dusty roads looking for good surf (the term road is being generous). That being said, the most interesting and best parts of the trip was getting to surf by boat, places like Witch's Rock and Ollie's Point (the latter named by the locals for Oliver North .. evidently they knew he was using the location to ship weapons to the Contras before Congress did).
Hopefully, an equally memorable trip is forthcoming (thinking of a 2 week trip driving down the Baja peninsula to places like Scorpion Bay, Todos Santos and Cabo San Lucas either after JavaOne or in the early fall ... although Vinicius Senger of Globalcode in Brazil keeps threatening to take me surfing next time I show up there for a Sun Tech Day)
5. Lastly ... I hate this virus (because in many ways, that's what it is, although obviously it does have some interesting social consequences ... as with all viruses :-) ) called the "Five Things" ... so in my own way of trying to treat this malady, I'm not going to pick anyone else ... so in this thread, I hoping this will be the end of the line ...
Posted by brewin Mar 18 2007, 07:37:49 AM PDT Permalink
Monday February 12, 2007
Speaking of the evolving web ...
For those developing for "Web 2.0" or evolving your applications to take advantage of next generation technologies, I would strongly suggest a visit to the new Web Developer Resource Center on SDN. Here you can find the latest and greatest in Scripting, AJAX technologies, tools and the optimized AMP platform components for Solaris. A definite "must see" ...
Posted by brewin
Feb 12 2007, 09:33:36 PM PST
Permalink
Friday February 09, 2007
And no, I don't "hate" Web 2.0 ...
A recent article leads with the rather interesting title "Sun CTO Of Software Bob Brewin Hates 'Web 2.0'" ... reading on, it becomes apparent (I hope) that it's not Web 2.0 the concept I "hate", but the term itself. I've paraphrased Hal Stern before on this topic, which is that "any number after the term 'Web' is redundant"... and in discussion with a number of engineers, they also typically pooh pooh the term "Web 2.0" as more of a marketing concept (but the technologies and potential for what's evolving in the this evolving web is oh-so-cool and represents the current and future of the web).
The Web is an evolving thing. We've gone from largely static systems to dynamic ones which both allow for user generated content and input, but can change as a result. Is that the end state ? Absolutely not. Thus the term I use when describing the web is the "evolving web", which I think is more appropriate. Posted by brewin Feb 09 2007, 01:45:56 PM PST Permalink
Sunday February 04, 2007
What is Web 2.0 ?
So James recently tagged me (so the next time you read this, I'll have my five things for your enjoyment), but as I was putting some of those thoughts to virtual paper, the following was forwarded to me and I had to share.