One of the exciting things about Windows Server 2008 (codename Longhorn), is the fact that you can deploy it in its most minimal form, known as Server Core. Of course, after reading a bit about this interesting new installation in a recent eWeek article, 'Longhorn' does more with less , all isn't as sweet as I initially thought. The first deal-breaker for me was the fact that .NET isn't included. Admittedly, Microsoft never said .NET would be the core of Windows before, but I still think the future of Windows is going to be .NET (or .NET's successor). It just makes sense. Well, unless you ask the Win32 zealots, of course. Apparently, .NET has so many dependancies, by the time you add them all, you're almost up to the full install Windows install. So, it sounds like .NET will only be in the full install. Ok, if that wasn't bad enough, it gets worse. One thing you should know is that, as I understand it, Server Core only provides a command-line interface. Some will like this, some won't. What does this tell you? PowerShell, right? Yeah, not quite. PowerShell depends on .NET. Granted, you may have seen this coming with the lack of .NET. The lack of these two makes Server Core useless to me. As much as I liked the idea of having a minimal install, I just don't see ever using that.
As Jason Brooks mentions, Server 2008 is a while away from producing the individual software packages you'll find in Linux installs, but this is definitely a step in the right direction. I can see this being further enhanced in 2010 or '11, with the next version of Windows. The future is definitely exciting; especially when you look at how far Windows has come in the past ten years. If you don't think it's come that far, take a look at other operating systems, like Linux distros as a whole, Sun Solaris, and/or Apple Macintosh. None have come even close. Of course, this is my opinion and I'm sure many would argue.
While I'm on the subject of the next version of Windows, I'd have to say that getting .NET into Server Core is a must. Well, maybe not necessarily in Server Core, but at least making it a small package to easily be added without a problem. Perhaps making .NET more modular might be an answer to that. Then again, I can only imagine the headaches that could bring. All I know is the heads better get together and figure out a way around this. I just can't imagine the "next generation" Windows without the "next generation" command-line. Seriously, who thought this one out?
Someone recently asked about Sidebar gadget development resources and I wanted to share the links that came out of it. Now, I haven't used all of these, so I can't speak to their value; but without further ado...
Ok, not really new, since I've been using Live Search for almost a year, now, but I recently came across this Live Search spin-off. Essentially, it's a Windows Live mash-up. Well, the beginnings of one. I have to admit I like it, but it needs some help. First and foremost, I wish it'd pull in the rest of the WL look and feel. They started to get the basics, but left out some of the dynamic features. This kind of makes me question who created it. The page says Microsoft, but anybody can write that. Either way, I like the intent. Hopefully, it'll be expanded over the coming months.
Update (11/23/2007): The aforementioned link no longer works. Going directly to the domain redirects to Live Search, now.
Eric Lundquist speaks to Ray Ozzie's description of Google as the "command line of the internet" in an old eWeek article. He suggests that this isn't quite true, arguing that the simplicity of search is nowhere near the same as the complexity of the command line. I have to say I agree with Ozzie. The truth of the matter is that search is the closest thing to the command line when it comes to the web. You type in what you want to do (or look for) and the search engine reacts. There's a deeper reasoning behind Ozzie's comment, but I won't get into that. I will say there's a lot to come and the next year will bring some of that to life.
Taking the general topic of the web a little closer to home, Lundquist poses 10 issues he believes must be addressed for Microsoft to truly compete in today's -- and tomorrow's -- web-focused world...
- Integration of online and offline offerings. Lundquist is definitely on-target when he mentions the gap between the two. Microsoft's missing the boat big time, right now. I'm very faithful that it will all come together, but it's taking too long, in my opinion. He also makes a direct reference to Google's productivity apps as "enterprise-level" here, but I think he's way off on that. These apps may be enterprise-scalable, but are far from enterprise-functional. There's a big difference. Google's a long way from getting there.
- Browsers are the new OS. Lundquist asks why Vista apps is being stressed more than browser apps. Let's face it, Microsoft is a commercial entity. Selling Vista makes money. Let's put two and two together, here. Some of this has changed with the new Silverlight announcements, but don't think that'll stop the push for applications to be updated for Vista and beyond.
- Google's offline weaknesses. Lundquist suggests that Microsoft should play on Google's weakness in offline data. I understand what Lundquist is saying, but I don't think this is a good strategy. Google's mantra was to "do one thing and do it good." Well, they did that with web search. With that popularity, they've tried to dig into other realms. The unfortunate problem there is that web search isn't the same as the other areas Google's tried to dig into. Sure, they've created some intriguing services and applications, but they aren't nearly as well thought out as their web search offering. Hell, they can't even get enterprise search right. Google hasn't been very good at taking in all the angles, so their weaknesses speak for themselves. If Microsoft steadily pointed fingers, that would simply back-fire.
- Online security. Lundquist wants Ozzie to explain the Windows Live security initiative. While I would never argue against it, I think this is more of something that needs to be proven, rather than described. Time will tell. All I really have to say about it is that the struggling OneCare service isn't doing a very good job of proving Microsoft's strong emphasis on security. As usual, I think this will improve significantly over time.
- Identity. Passport, and now Windows Live Id, is great an all, but I wish Microsoft would pick up and support something like Open Id. Lundquist just wants to know what Microsoft has to offer here and, well, I think that's a good question.
- Standards compliance. This is an area where Microsoft has been getting better and better at. I think we'll see this continue over the next few years and, eventually, I think we'll even surpass others in the same space, being more compliant than them. Of course, this all depends on consumers' desire for such compliance.
- Internal cross-pollinization. Knowing how online and offline teams are working together to build more integrated environments is something we're not seeing from the outside. Oh, wait... not from the inside, either. Actually, I know it's happening, but the vision of how that's happening should be shared a bit more. Microsoft's been good with transparency over the past few years, but they've also been careful about what they've shared due to the problems with Vista. The ideal balance has yet to be achieved. Who knows how long that'll take to perfect.
- Future of Windows Live. That's a damn good question. I can try to foresee some possibilities, but it's wide open. The first year or two has been all about defensive strategy as Microsoft plays catch-up. We all know Microsoft is best when in attack mode, tho. I truly see this as a turning point. I expect big things from the Windows Live platform. I just hope Microsoft will deliver the next generation experience it's capable of delivering.
- Timeliness. As I mentioned before, Lundquist is right on here. Microsoft has had a problem with over-envisioning the capabilities in its next gen products and services. I think the Windows Live platform has been better at that, with releases every 3-6 months -- I don't know what the actual release schedule is, but I think it's quarterly. The problem is that this isn't publicized. I'd like to see what is pushed so I can take advantage of the new features as they come online. Otherwise, the products and services just seem stagnant, like others have the reputation of being.
- Names and faces. Lundquist would apparently like to see more from the likes of those closer to the trenches than Ballmer and Ozzie. I can't blame him with this, but I don't know how much I care. Personally, give me the info. I could care less who it comes from. I may be off on what he's really getting at, but I'd rather hear the news than know who's behind it.
Ok, so this isn't anything huge, but I have to admit I like it. PageBull
is a search engine that simply returns web page thumbnails instead of links. While I feel like this is a completely novelty feature, I still like it. Give it a shot. I have a feeling you'll be intrigued by it, too.
Here's some pretty big news, if you ask me: security researcher Dino Dai Zovi has declared Windows Vista as being hands-down more secure than Mac OS 10.4 . I have to say I'm not surprised. I think the Apple community has been blindly professing Mac as being more secure simply because there aren't as many publicized exploits -- although, two big sets of security patches coming from Apple in the past month or so might shed some light on the vulnerabilities that are lying dormant. Let's face it, tho, a system with 100 users isn't going to be worth a hacker's time as much as a system with 100,000 users. Admittedly, I don't know the real numbers, but it is a huge difference. I want to say Windows has just under 90% of the desktop market while Mac has less than 5%. Who would you go after? Yeah, I thought so.
Despite the numbers, I'm glad to see a completely independent view. I'm sure there will be plenty more in the coming months and years. With the next release of Mac expected by the end of the year and Windows in 2008 (my personal guess), I see security coming up again. I think the real issue is that Microsoft learned a lot from past experiences and has changed for the better. People won't see that for a while, tho. Perhaps this is the first step in that direction.
Here's a brief quote on what Dino said regarding Windows vs. Mac security...
I have found the code quality, at least in terms of security, to be much better overall in Vista than Mac OS X 10.4. It is obvious from observing affected components in security patches that Microsoft’s Security Development Lifecycle (SDL) has resulted in fewer vulnerabilities in newly-written code. I hope that more software vendors follow their lead in developing proactive software security development methodologies.
You may be interested in some of the other questions/answers from the interview, but this is what I found the most noteworthy. Now, I'm looking forward to hearing reactions of some of the Apple community. I'm sure they'll dismiss it, but it should be entertaining to see how this plays out.
Well, I was off on my guesses (1 , 2 ) at what was going to be announced at Mix '07 . In the same breath, I'm kind of disappointed. Here's what I've seen...
Jasper is another new project , announced by the ADO.NET team , which aims to simplify application development by providing a built-in data layer. Granted, I haven't seen anything about this, but I severely question it. Sure, it'll be better than what most people are doing, but is it enough? I can't say for sure without digging into it; but I'd look at it pretty hard before assuming it is. I like to have a lot of control over my applications, so building a data layer is just worth it, for me. I'm fine with the additional time it takes, which isn't all that much in my opinion, because it allows me to customize it to meet project-specific needs easily. This is by far the most flexible way to go. Of course, the problem is the amount of time it takes. Like I said, we'll have to see. I'm interested, but skeptical.
Apparently there's an early edition of Blend 2 available which supports Silverlight. This was an obvious progression, but I didn't expect it to happen this quickly. Heck, I didn't even know Blend 1 was released...
oh, wait, it hasn't!? While completely understandable, it's kind of an odd move. Let's face it. Will you buy Blend knowing Blend 2 is on its heels? I'm sure there will be some sort of promotion for all Blend 1 owners; but that hasn't been announced, yet. I just hope it's better than the typical upgrade cost. In the end, I'm thinking this is probably going to be more of a 1.1 release than a 2.0 release, but a necessary one nonetheless. My hopes are that they'll go for a small upgrade fee of $23 or so, like they did for the VS 2002 to 2003 upgrade. Only time will tell.
Dynamic language support has been made official; but again, expected. I will say I didn't expect this to be a separate runtime environment, but I can imagine there might be some technical issues limiting the dynamic nature on top of the CLR. I'm interested to see how these two ciblings grow up together. In the same breath, I'm curious as to the number of different versions of .NET and the CLR. We're quickly approaching half a dozen, if not more.
While I was thinking Silverlight itself could be the "big announcement," it wasn't. I wasn't too far off tho. Three Silverlight announcements came out: (1) Silverlight will support .NET on all it's supported clients, most notably including non-IE browsers and Mac; (2) there's already an alpha release for this; and, (3) Windows Live has a new Silverlight Streaming service to give you 4GB of streaming content. Ok, this last one is definitely a marketing move, but one you can benefit from, if you have content to push. But I do have to touch on the fact that, before Silverlight 1.0 is out, we've got preview bits for a follow-on release. This is unheard of. I don't think I ever remember seeing this happen with any other companies. Not that I'm saying Microsoft is a great company simply because of this, but I do think it's a strong testament to the direction the company wants to move in. This will be very interesting in the coming year as Silverlight starts to push into the Flash space. Then again, Flash has only pierced a niche market. I'm very confident that Silverlight will surpass that small portion and make it into the true web application space. Heck, I'm already seeing viable options for Silverlight in a number of apps I'm privvy to.
So, the Silverlight on Mac announcement was probably one of the most surprising to people, as it seems. When a friend first said, "CLR will be supported on multiple OSes," I got very excited. Then, when he clarified that it was within Silverlight, I just responded, "Oh. I already knew that." I mean, seriously, wasn't that known? Maybe I just assumed it. I'm not trying to take away from it; I definitely think it's a fantastic progression. I just expected this. Nonetheless, there was some good stuff that came out of Mix. I ultimately decided against going this year, but I may have to rethink that for next year.