Getting ready for the MIX 10 keynote to kick off and I have to admit, I'm getting pretty excited about it. While some Microsoft events can be a bit perfunctory, this MIX is setting up as an important event. In addition to the long-awaited unveiling of technical details around Windows Phone 7 development, the conference will offer new details on Silverlight 4 and Internet Explorer 9.
Make no mistake, the Windows Phone content alone elevates MIX 10 to the standing of a strategic conference for Microsoft. Which is kind of a amazing when you consider that the Microsoft Professional Developers Conference (PDC) was held only four months ago
Are you seeing what you hoped at MIX? What would you like to learn more about? Let me know and we'll try to get after it
Posted by Michael Desmond on 03/15/2010 at 1:15 PM0 comments
Application performance monitoring (APM) vendor dynaTrace the other day released the results of a survey it sponsored looking at the impact of heterogeneous environments on application performance and stability.
Now it's hardly news that a dynaTrace-inspired survey might find that mixed environments can impair application robustness. After all, such situations would merit the use of (surprise!) sophisticated APM tools to detect, diagnose and address these problems. What is interesting is that .NET/Java interoperability was singled out by the most respondents (53.8 percent, to be exact) as an obstacle to effective performance management. Virtualization, service-oriented architectures (SOA), and to a lesser extent cloud computing and software-as-a-service (SaaS) were also cited.
What does it all mean? Well, dynaTrace thinks that the ongoing success of.NET development is in part to blame, as once-exclusive Java shops find themselves slinging both .NET and Java code. As the survey notes: "While exchanging data between Java and .NET platforms can be relatively simple, it is more difficult to understand the behavior of heterogeneous systems from an end user’s perspective."
These kinds of challenges aren't likely to go away, the survey release notes, as Microsoft continues to ply an ever more open course. Witness yesterday's release at the RSA Conference of Microsoft's U-Prove cryptographic technology CTP under the Microsoft Open Specification Promise (OSP).
Does the dynaTrace survey jibe with your experience, or have you been able to manage the challenge of maintaining scalability and performance in mixed Java/.NET application environments? Let me know!
Posted by Michael Desmond on 03/03/2010 at 1:15 PM0 comments
The release candidate (RC) of Visual Studio 2010 goes to general availability today, probably around 3pm Eastern Standard Time, according to a blog post by Jason Zander
. The RC code gives developers a chance to kick the tires on the next version of Microsoft's flagship integrated development environment (IDE). The RTM of Visual Studio 2010 is due to roll out on April 12.
Once the bits go live, you can download the RC here.
Key Microsoft developer execs like Zander and Soma Somasegar have blogged about the VS2010 RC. Based on what they're saying, the ongoing complaints about performance in the beta versions of Visual Studio have gotten plenty of attention. The very early returns are encouraging.
Keith Patrick, responding on Jason Zander's blog, wrote: "Wow! I was disappointed (very) in the Beta 2 performance (5 second lag between me typing and the letters appearing on screen), but the RC is blazing. Ridiculously so."
Visual Studio Magazine contributor Dan Wahlin had this to say in a tweet: "The performance improvements with VS 2010 RC compared to previous builds are huge. Really happy with what I'm seeing so far."
Some developers are disappointed, however, with the lack of Silverlight 4 support in the Visual Studio 2010 RC. VS 2010 Beta 2 currently supports Silverlight 4 development, but Scott Guthrie, blogging on the RC drop, wrote that the RC only supports Silverlight 3 projects. He urged developers working with Silverlight 4 to stick with VS 2010 Beta 2.
"Silverlight 3 projects are supported with today's VS 2010 RC build -- however Silverlight 4 projects are not yet supported," Guthrie wrote. "We will be adding VS 2010 RC support for SL4 with the next public Silverlight 4 drop. If you are doing active Silverlight 4 development today we recommend staying with the VS10 Beta 2 build for now."
Posted by Michael Desmond on 02/10/2010 at 1:15 PM0 comments
Microsoft isn't offering any details, but Rob Caron, in a very brief blog entry
, says that Visual Studio 2010 will now launch on Monday, April 12.
The release of Visual Studio 2010 had been pushed back from the original March 22 launch target date, as Microsoft worked to hash out concerns related to performance and virtual memory usage. The three week delay certainly falls in line with the initial description of the schedule change, which S. Somasegar, senior vice president of the Microsoft Developer Division, at the time said would amount to "a few weeks." You can read more details about the initial delay announcement here.
More details as we hear them...
Posted by Michael Desmond on 01/14/2010 at 1:15 PM2 comments
I've been talking to a lot of Silverlight developers and experts over the past couple weeks, as part of a cover feature we are doing on Microsoft's next version of its rich Internet application (RIA) platform, which was released in beta form
at PDC in November. Given how quickly Redmond turned the crank between Silverlight 3 and 4, I figured the beta might be a bit rough around the edges or missing a few critical pieces. So I was pleasantly surprised to hear that the current beta of Silverlight 4 is, in fact, surprisingly robust.
Ben Dewey, a senior software developer at consultancy twentysix New York, has been working with the Silverlight 4 beta since it came out. He said he hasn't encountered any problems and said there is no indication that features are missing. "I haven't run into any non-implemented exceptions or anything along those lines," he said.
Evan Hutnick, a developer evangelist at component maker Telerik, told me he's spoken with a number of customers currently using the beta release to develop Silverlight apps, and they "seem really confident in this beta right now."
I'm hearing these kinds of comments from just about everyone I talk to.
The current beta of Silverlight 4 doesn't carry a go-live license, so obviously it's way too early to start releasing software based on the current code. At PDC, Microsoft's Scott Guthrie offered a release time frame of the first half of 2010, but it certainly seems like the final version of Silverlight 4 could emerge sooner rather than later.
Have you been working with the Silverlight 4 beta? What are your impressions of what Microsoft has delivered? Comment below, or email me at [email protected]
Posted by Michael Desmond on 01/12/2010 at 1:15 PM6 comments
According to a blog post by Kathleen Richards, editor of Redmond Developer News
and executive editor at Visual Studio Magazine, a recent Forrester Research report projects that application development shops will face five key changes in 2010. The dominant theme? Dev shops will need to get more done with less.
The Forrester prognostication offers a five-point call to action for dev managers:
Embrace cloud computing
- Run your dev shop like a cash-strapped startup
- Be flexible in platform decisions
- Get after the user experience
- Seek and develop talent, rather than headcount
Read Kathleen's entire blog post
here. And check out the Forester report summary
Are you planning changes this year to get the most out of your application development and processes? Let us know below in the comments section, or email Kathleen Richards at [email protected]
Posted by Michael Desmond on 01/08/2010 at 1:15 PM0 comments
Since the dawn of the .NET era, Microsoft has too often been victim of its own orchestration. Intriguing features and capabilities would be cooked into platforms and frameworks, with few tools to take full advantage of them. From the release of WPF in .NET 3.0 to the frustrating bottlenecks facing Visual Studio developers working with SharePoint, the mismatch between tools and platforms has been an ongoing issue.
Microsoft, to its credit, has worked hard to address the situation. Once-interminable release cycles--which produced epic waits for SQL Server 2005 and Windows Vista among others--have given way to shorter, iterative releases that allow products to catch up to each other. Microsoft has also worked to provide toolkits to bridge the gaps, whether it was the ASP.NET AJAX Toolkit (code named Atlas) for Visual Studio 2005 a few years back, the SharePoint extensions to Visual Studio 2008, or the ASP.NET MVC tooling that Microsoft has been advancing.
With Visual Studio 2010 and .NET 4 on the horizon (albeit, slightly delayed from the expected March 22 launch date), Microsoft is prepared to put its tooling and platforms in something closer to lock step. Matured XAML tooling promises to ease Silverlight and WPF development in Visual Studio, even as Expression Blend offers compelling capabilities for designers. SharePoint finally gets first-class treatment in the VS IDE. And the recently updated Entity Framework 4 should bring stability to unsettled seas in the data access layer space.
Notably, Microsoft has come out of the gates preaching a synchronized vision for its Windows Azure cloud platform. At the Professional Developers Conference in November, Microsoft announced versions of the AppFabric middleware (comprising the Velocity distributed cache engine and Dublin app server extensions) as a common resource for both Windows Server and Windows Azure applications.
Microsoft has for a long time preached the benefits of leveraging .NET, from familiar tooling and skill sets to the ability to share code across projects. With the release of Visual Studio 2010, .NET 4, and updated apps and platforms like Silverlight 4 and SharePoint 2010, we're seeing Microsoft actually pay off on all the preaching.
Do you think Microsoft is doing a better job of synching its tools and platforms? What could Redmond do better? Email me at [email protected]
Posted by Michael Desmond on 01/04/2010 at 1:15 PM1 comments
If you've worked with or around Microsoft technology for more than a few years, you probably aren't shocked to hear that the upcoming launch of the company's Visual Studio 2010 integrated development environment (IDE) and .NET Framework 4 platform have been delayed
. When you are in the business of building incredibly complex tools and platforms, delays are almost inevitable.
But I have to admit that I was surprised to hear that the expected March 22 release of VS 2010 and .NET 4 would be pushed back "a few weeks," according to a blog post yesterday by S. "Soma" Somasegar, senior vice president of the Microsoft Developer Division. While VS2010 has had its share of performance and stability wrinkles, Microsoft has done an outstanding job of involving the developer community in the product's development cycle. The VS team delivered a very strong beta 2 in October, resolving a host of concerns produced by the first beta, back in May.
Microsoft isn't offering any details, but Somasegar in his blog post singled out issues related to virtual memory usage. He also cited the depth of performance-related improvements in the last beta, though it's not clear if general IDE performance is at issue here.
So what can developers expect to see next? Microsoft will add an interim checkpoint release to the VS/.NET review cycle, which Somasegar described as a "release candidate." That release should be made available under a broad Go Live license in February, Somasegar wrote.
Still, the question begs: How long will we have to wait?
Posted by Michael Desmond on 12/18/2009 at 1:15 PM3 comments
Stephen Chapman is long-time Microsoft watcher and author of the Microsoft Kitchen blog
(formerly UX Evangelist). He offered his thoughts on what he saw at the Microsoft Professional Developer Conference last month.
1. Was there much at all new in Sinofsky's Windows 7 presentation? Seemed like a long rehash to me.
Sinofsky's keynote speech was a snoozefest... unless you happen to have just started paying attention to Windows 7 between last year's PDC and this year's, in which case, it may have proved enlightening. The most exciting/revealing thing of his keynote was a small demonstration of a stripped-down Internet Explorer 9. Those of us who have paid attention to Microsoft since the days of Netscape know how important Internet market share has been to Microsoft, so in my opinion, if they get it right with Internet Explorer 9, I think they're finally going to have a solid foundation to start building up their Internet/browser empire.
Bing is great and anyone who says differently simply hasn't given it the fair shot it deserves. I'm not sure any search engine will ever replace Google as my home page, but there's a great chance that IE9 may very well find a place in my taskbar. What I'm getting at is a potential end to the Google + Firefox combination just about all of us use. Bing + IE9. Can you imagine?
Oh, and Sinofsky's keynote was also extremely depressing for all government and media attendees. See if this rings a bell: "Here's an amazing laptop! It's yours!" ...and then we all know where that ended up. (Ed note: Members of the media did not receive free hardware.)
2. Sinofsky was getting after developers to support Windows 7-specific capabilities like Trigger Start Services and the like. How good a job has Microsoft been doing of evangelizing Win7 development? What could they do better?
I think they've been doing well with it on a large scale, but I think they should be more proactive with their road shows. Windows 7 has a ridiculous amount of potential and, in a twist, it's actually showcasing much of what was capable in Windows Vista. Unfortunately, the negative stigma that goes along with the word "Vista" has kept developers a bit gun-shy.
Well, now that Windows 7 is out and it has received some fantastic fanfare, I think developers are going to see value added in the fact that what they write for Windows 7 will more than likely work for Vista. I think that would be a great point to emphasize, but I think Microsoft is done with Vista. One -- maybe two -- more Vista service pack(s) and they will completely wash their hands of it. But I digress.
Lastly, I think one of the coolest features for devs to play with is multi-touch. Unfortunately, I think it's also probably the most useless right now. I think Microsoft realizes that, so what did they do? They put a multi-touch platform smack-dab in their hands. Not only did they give them a device to develop on, but they gave them a device that will probably be quite affordable. Give the developers a real device that can realistically be adopted by many consumers. I could be wrong about that whole thing, but for as awesome and groundbreaking as multi-touch has the potential to be, I've never really seen it as something that would take off beyond the enterprise/professional realm.
3. Have you had a chance to look into some of the IE9 optimizations? Seems like taking advantage of the underlying hardware and accelerated DX stack is a no brainer. Is this something devs should be excited about? Any reason why other browsers can't do similar things (via published APIs, etc)?
So, the IE9 thing... incredibly exciting! Developers should be unbelievably excited about it. This is the type of thing Microsoft needs to get back in the game with IE. Granted, I think people forget how much market share Microsoft still has with IE, but IE9 has the potential to bring it back home. If they can succeed with meeting all the standards and passing the ACID test, etc., I think they're going to be well on their way to getting on top of their game... and maybe more so than they've ever been! Microsoft has been doomed in Internet market share for such a long time now, and I'm really happy for them that Bing is such a solid product.
It seems like Microsoft is finally onto something.
In regards to others pulling off hardware accelerated browsing, the issue lies not with how to get it to work, but how to get it to interact with all the browser technologies that exist out there. For instance, how do you render something in a frame that already renders itself? Flash is a good example, or perhaps even Silverlight. I think developers should get excited as hell about this, but I also think they should wait until Microsoft irons out the issues that might make devs think, "wait a minute... how the heck am I supposed to code for this?" For what it's worth, I think Microsoft has something unique going on here and for them to demo IE9 so early in the game, they're obviously confident in what they're doing.
4. The Silverlight 4 stuff was huge. Local file system access, out of browser execution, printer and clipboard support, drag and drop support and mouse wheel inputs. That doesn't even touch on performance optimizations and better integration with the .NET stack (via ADO.NET Data Services and the JIT CLR). Is all this as big as I think it is?
The Silverlight 4 stuff was HUGE!!! I think Silverlight 4 will *FINALLY* take Silverlight out of its "Flash wanna-be" status and catapult it into its own respectable technology. I mean, not that it hasn't already commanded respect, but Web developers need to just go ahead and jump on the Silverlight bandwagon.
What started as a boring PDC for me, ended with me questioning if we didn't somehow just attend MIX 09 Part Deux! It has me wanting to attend MIX 2010 REALLY badly. This feels like 2010 is going to be the year of Microsoft where the Web is concerned. Bing is doing much better than any previous search engine Microsoft has cooked up, IE9 sounds like it's going to be the solution to every problem we all have with any and every browser of our choice (no matter how powerful the computer), and Silverlight 4. I don't know about you, but this makes me want to jump on learning to be a Web dev and immerse myself completely in Microsoft's technologies!
Oh, and Silverlight for mobile. That is going to be massive. With everything Microsoft mentioned in relation to the Web here, as well as noting they would discuss Windows Mobile 7, MIX is going to be the event to attend. If Silverlight 4 is as great as it seems to be for Windows, I can't help but wonder what they have in store for Windows Mobile 7 where Silverlight Mobile is concerned.
5. What are your thoughts on the ability to compile an assembly once and have it run on both Silverlight 4 and .NET 4?
Genius on Microsoft's behalf to make this possible. When this came up, I had two good dev friends of mine comment on how incredibly excited they were about this feature. Personally, I've never coded anything that made me even put these two together as being something I wanted, but like I said, knowing two devs directly who were ultra-excited about this, I would love to hear what your reader base thinks.
6. Are you following the AppFabric stuff at all? What I'm hearing sounds like a fairly compelling pitch. Write your .NET apps for higher-level platforms like WCF and WF, and AppFabric will provide the common abstracted app interface to make your stuff run seamlessly on both Windows Server and Windows Azure. How convinced are you about this? (Obviously it's early, since the Windows Azure version of an AppFabric CTP won't see light until 2010).
Honestly, I've been so into Microsoft's web announcements that I haven't even looked into AppFabric yet. All I've gathered from it thus far is exactly what you noted. I'm going to dig much deeper into it this week and gain a further understanding of exactly what it is and how plausible it really seems (or, as you said, if it's just a compelling pitch).
7. Can you provide any general thoughts on this PDC? Can you compare to last year's session in terms of the messaging Microsoft was bringing as well as the energy of the dev audience? Is there a lot to be excited about?
Silverlight and IE9. When first announced, I was quite underwhelmed... but then, I really started thinking about it and how far-reaching the implications are of hardware-rendered browsing and Silverlight 4 (as well as Silverlight for Mobile when they discuss it on the plane of Windows Mobile 7).
As for everything else, it was nice to see how far they've come with Azure and how much they've been participating with the community and their partners to fine-tune it, but Microsoft really just drilled home cloud computing even more and dug deeper into Windows 7 features like the Ribbon UI. Like you said before, this PDC was mostly just a recap, recap, recap.
Posted by Michael Desmond on 12/01/2009 at 1:15 PM0 comments
I'm crawling toward Burbank Airport after a three-day stay at the Microsoft Professional Developer Conference (PDC) in LA this week, and I have to say this was certainly an interesting PDC. Unlike last year's confab, which played host to a diverse array of strategic launches that ranged from the high-stakes debut of Windows Azure to the beta launch of Windows 7 to some heavy visioneering around Oslo, PDC09 had a more tactical feel.
As Andrew Brust noted in his Redmond Review blog report after Day 1 of the show, PDC09 felt more like PDC08 R2, a careful scoping and refinement of the far-reaching and far-flung messaging from last year's show. Windows Azure has matured remarkably, with its fleshed out AppFabric middle-tier built around the Velocity distributing caching engine and Dublin app server extensions. AppFabric for Windows Server debuted in beta form at the show, and will arrive as a CTP for Windows Azure sometime next year. The common infrastructure, Microsoft promises, will give Visual Studio developers one target to work toward, whether they reside on premise or in the cloud. But there's a lot of work yet to be done.
Most devs I talked to seemed to appreciate Microsoft's year-long effort to refine the Azure roadmap, and felt this second PDC went a long way toward helping the .NET community grapple with the impending cloud platform. Microsoft announced Windows Azure would go to production on January 1st, with official customer billing picking up on February 1 after a month-long trial period.
Still, Microsoft managed to make a show of it on the second day, giving away free, touch-enabled Acer laptops to all show attendees. The surprise giveaway may have looked like something out of an Oprah broadcast, but I have to say, it worked. A lot of folks were talking about the Windows Touch UI the last couple days of the conference.
More important, Scott Guthrie surprised attendees by announcing the launch of the Silverlight 4 beta.
As Silverlight versions go (and there's been a few of them in the product's short life span), version 4 is huge. The new features in this version read like they came straight off the list of "what in Silverlight angers devs most," with support for out-of-browser execution, local file system access, printer and system clipboard support and lots more. Microsoft partners I spoke to were enthused, in no small part because the devs they were talking to raved about Silverlight 4. But the inevitable question kept coming up: Why would mainstream developers choose WPF for desktop development when Silverlight 4 is around? It's a very good question.
This year's show wasn't as well attended as PDC08, but vendors I spoke with were genuinely excited about the quality of the audience. PDC tends to draw the deep thinkers and strategic decision makers out of the woodwork, and PDC09 -- despite coming on the heels of PDC08 -- was no exception.
What were your impressions of this year's PDC event? Are you planning any efforts based on what was revealed at the conference? Email me at [email protected]
Posted by Michael Desmond on 11/19/2009 at 1:15 PM2 comments
Wandering the Professional Developer Conference (PDC) show floor today, I ran across the Microsoft Research section, and spoke with Maden Musuvathi, a researcher in the Software Reliability Research group at Microsoft Research working on testing solutions for concurrent code. He was demoing a prototype product called Microsoft Cuzz -- shorthand for Concurrent Fuzz testing.
Of course, fuzz testing is used to smoke out flaws in code by blasting random or otherwise unexpected data input at application interfaces, with the goal of kicking off program failures that might otherwise go undetected. Cuzz does something similar, shaping thread interaction in concurrent C or C++ Win32 code to create conditions likely to produce hard to reproduce failures.
As Musuvathi explains. Cuzz determines the mathematical probability that a concurrent code failure will occur by observing the number of threads and the number of synchronizations. Other issues, such as locks, mutexes and complex synchronizations don't impact the calculated probability, Musuvathi said. Developers inject the Cuzz DLL into the binary executable and then run the test. Cuzz tracks each thread it activates, logging the thread activations that result in a crash.
Musuvathi said that developers at PDC have lamented the difficulty of finding and resolving concurrency-based flaws. It's a lament Microsoft's own teams have shared. Musuvathi said another concurrent code testing app -- called CHESS
-- was initially developed to help parallel code projects at Microsoft, including Parallel LINQ and Task Parallel Library (TPL). Musuvathi described CHESS as a unit testing tool, whereas Cuzz can be broadly applied to detect hard-to-find flaws.
Cuzz remains a project, but Musuvathi is hopeful that the tool can find its way into developers hands down the road. He also says the program will likely be advanced to address managed and 64-bit parallel code down the road.
Posted by Michael Desmond on 11/18/2009 at 1:15 PM1 comments
With ASP.NET 4 cooked and nearly ready to ship, the folks at Microsoft turned their attention to what will come next in the Web programming framework. During the one-hour ASP.NET Futures session held at the Microsoft Professional Developers Conference (PDC) in Los Angeles at Monday, Jonathan Carter and Scott Hunter offered details on the direction developers can expect ASP.NET to take.
The ASP.NET team has set three broad goals for the next version: Reduced complexity, support for the HTML 5 standard, and improved performance and management.
On the complexity front, the pair walked through several new features that promise to provide simpler ways of getting common tasks done. For instance, the new SmartyRoute feature promises to do away with avoidable Page Not Found errors, by adding intelligence to the site when receiving URL requests. SmartyRoutes allow developers to identify the page or resources users want to access, even if they enter the wrong extension. SmartyRoute will also walk up path segments in a URL to locate the first valid ASPX file. Developers will be able to write one line of code, one time, to establish intelligent routing and ensure that context is provided when an invalid URL is input.
Carter and Hunter also walked through a series of ASP.NET Helpers that will provide streamlined image manipulation, better email authentication, and efficient management of long-running background task based on minimal code. They also demoed an improved file upload progress dialog box, which provides accurate context by tapping new resources in the ASP.NET framework.
In the area of HTML 5 support, the two showed how ASP.NET developers can easily enable local storage access for Web apps, enabling offline productivity and allowing developers to save changes made to the server back to the local store or vice versa. ASP.NET will make these activities "extremely trivial" by automatically tracking user data to enable context.
Also discussed was support for new markup tags in HTML 5, including the ‹video›, ‹audio› and ‹menu› tags.
The session addressed the always-relevant issues of ASP.NET performance. One issue of concern is the processing of multiple HTTP transactions and file downloads, which can produce unwanted latency. The team hopes to introduce a mechanism, called Sprites, that automatically combine multiple images into a single image package that is downloaded in one step. The client then separates the images and renders them properly on the page.
Hunter and Carter also talked about the Velocity Distributed Cache platform and how this cluster caching resource might be used for general ASP.NET caching.
Posted by Michael Desmond on 11/18/2009 at 1:15 PM17 comments