The first alphas of Longhorn looked AWESOME. An advanced 3D-driven windowing environment (with some very cool themes), a kick-ass sidebar with some very cool features, significant improvements to and total integration of .NET into the OS, and a relational file system that could store both files and folders (legacy hierarchical structure) as well as arbitrary data sets and a multitide of metadata and relations between files and data sets.
Vista today lacks most of the win-win that Longhorn had. The only remaining nicities are the sidebar (even the sidebar is only 'ok' now, IMO) and the Aero Glass look. Even WinFX, the underlying core development API of the OS, seems to have mostly slipped off the Vista radar (or rather, its no longer a Vista-only thing, which reduces Vistas value for me). And if I had to apples-to-apples compare the Windows Vista look with the MacOS X look, I sadly have to say that Apple still creams Microsoft in terms of look and feel. The Aero Glass look is ok, but it lacks the spit and polish of the OS X ui. (Maybe by D-Day Aero will look better, but between WindowBlinds 5.0, ObjectDock, and numerous other skinning and eye-candy utilities I can run on Windows XP, I think I'll still with XP.)
Security...great, I like it. Its better than before. People are still going to bitch and moan about it, and people are still going to try their hardest to find and exploit security flaws. I'd take Microsofts Security for their user volume over Apples any day. (Just imagine how secure Apples OS X would be if suddenly tomorrow it was the dominant operating system used by 95% of the market...the volume of known security flaws is directly proportional to the volume of users...within a month OS X would be so overrun by malicious code Jobs would have a stroke.)
Comparing Longhorn with the Vista of today, and the only thing I can say is "MEH!!!, I want my damn relational file system!". Nothing else matters. :P
I saw a lot of comments sying "vista is heavy" and "needs a great hardware".
I think most developer's computers are almost good. Something better than a usual computer. If we encounter requirement problem , then how many users this OS will attract? Most of the offices or companies I went had a much cheaper and slower machines(That's why C++ rules there yet )
How many jobs will use it or how long does it take to use? I mean if it's much expensive for an office to change hardware for say 20 machines. Does it worth to learn new programming features?
I think it will only move forward for most companies when the hardware is cheaper... which was the case for a lot of companies moving to Windows XP, because Windows 98 could easily run on a 333MHz CPU w/ 32MB-64MB RAM.
On the other end of the spectrum there are companies that are constantly staying on top of technology and will start moving to vista as soon as it is released.
...depending on what I'm looking at, but all-in-all, I would have to say 5 or 6.
WPF has some really cool new features, but they stuffed up its implementation in the way of perf, memory usage, complexity, meeting of practical needs, and access to lower-level features - very sad to see. WWF and WCF seem quite cool from what I've seen, but maybe a bit over-hyped. I've been saddened to see so many long-touted and cool features (like WinFS) being dropped, to the point where what remains (even though it's cool) seems very little in comparison.
But my biggest gripe is that it takes a minimum of 512 MB - or more likely 1 GB based on several reviews I've read - to even run it minimally! Considering that VS2005 chews up so much memory, and today's browsers are not too slim either, it seems pretty likely that I'm going to have to have 4GB of RAM to do my normal dev work. Given that I will probably be working with a Tablet PC in the near future, that is at least $800 if not $1000 for new laptop RAM if I use Vista as my dev machine's OS. Can I really afford it? Is it worth it? Should I really have to do it?
Considering that the next version of VS is probably going to be more bloated than the current one if trends continue, that WPF apps are likely to chew up a lot of extra memory, and that Office will undoubtedly also use more RAM, the future looks bleak indeed...
Or good old C# and Fluid when it comes out (I've implemented a quite stripped down version for a client already but still working on some of the killer features that will be in the real thing). My point is about Windows Vista, which many apps will be requiring in the near future, which will be the only truly supported version of Windows (besides the next one) in maybe 5 yrs, and which I may have to do dev work for clients with in the near future.
If they had kept the "database-like" file system (WinFS), THAT would have been something to get excited about. Just about everything else is "fluff" (mostly UI changes AGAIN, among many other changes to keep people migrating for another 2-3 years). I am sure the “nit-pickers” will drone on and on and on about all “this and that” innovation but really, when you boil it down… It’s mostly UI again.
The security won't be there. The UI, while quite beautiful, is just another UI, big deal. It’s like most things this day and age, too much surface and not enough substance.
The WinFS would have been something to get excited about. I was really looking forward to never having to worry about transferring to a new hard drive when the current one filled up. Think of how easy it would be for backups and restores… The whole OS/FS (possibly) could have been backed up in an easy to store format. That, my friends, would have been an innovation.
This idea (WinFS) is not new; Oracle came up with a concept like this back in 1995-1996 I believe. The idea of using a database as the OS's file system is great. Especially for OS's that can handle adding drives to a database as a raw device, this would have been a great step forward. No more C:\, D:\, blah blah blah. Nor would we have / mount point, /boot mount point or /whatever mount point. We would have had a single database that you just keep adding drive partitions too. Think of all the really excellent database tools that could have been used on this new file system. Archiving files that are older then… Just about anything you can do now with a database being done with the file-system. A giant step forward would have been achieved. All other OS’s would have had to fight to stay up with that one.
Instead we have status quo upgrades. I’d have been happier with them simply taking Windows XP PRO or even Windows 2000 the way they are currently and changing the file-system to WinFS. That would have been a huge step.
That was only for one beta, a number of months ago. User Account Control has been significantly improved since then, and while its still a little annoying, its certainly more secure than any previous version of windows. The deal with security is that its ALWAYS a tradeoff...you can't have your shiny secure cake and eat it to...either its behind the glass and vaccume sealed or not. The problem with Windows security isn't really the security, its everyone who bitches about it regardless of how good or bad it is.
I have to whole-heartedly agree on this point. I was extremely excited when I got my hands on Longhorn Alpha back near the beginning of last year. Despite the MONSTROUS memory consumption of WinFS (my test system booted using 800 megs of ram, out of 1gig total), it was REALLY REALLY COOL. WinFS was the only thing that made me want Vista in the first place, followed closely by the original sidebar. When they dropped both features, I pretty much made the decision to wait on an upgrade until whatever version of windows follows Vista.
I have a beta version of WinFS running on the XP box I'm typing this message on, and its been much improved since the original Longhorn Alpha. Its still got a LONG way to go, however. Something to be said for Microsoft over Oracle is that at least Microsoft is actively trying to implement a relational file system. Even if it doesn't get released with Vista, they have a base of code to work with, and even if its only released in pieces here and there, eventually we will have a fully relational file system in Windows. My only hope is that "eventually" is sooner rather than much much later. :\