Longhorn
Longhorn
An interesting item came up at the Recent WinHEC show. and something I mentioned what I thought of WinHEC and the Longhorn build. in a previous entry.
But listening to the Chris Pirillo show, great show. For the 2 shows with Robert McClaws brings up a very interesting point. http://www.thechrispirilloshow.com/help/20050502_robert_mclaws_on_x64_and_longhorn_part_1.phtml
http://www.thechrispirilloshow.com/help/20050503_robert_mclaws_on_x64_and_longhorn_part_2.phtml
What is a Power User???
A Knowledge worker is fairly easy to determine but you will have Power Users in that category.
Business user ditto.
Hard Core Gamers ditto and then some.
A Power user is hard to define because it covers way to much. but in essence as a Power User myself or how I look at a power user. Once I learn the Tool. take the training wheels away and give me full access. Turn off the wizards or allow them to be configurable at startup. Generally if it is an area I have no Idea on I will leave the wizards on the first few times to get the idea of what is needed. then turn them off.
I would like to go the step further and I am hoping that Moand brings that to fruition as the system moves forward and a safe scripting language becomes available system wide that. I can script some of my own needs.
But that is my view of the Power User. One that Uses the Computer to its full potential no matter what the underlying skill set is.
Home power users become the biggest issue as they have the widest varyance from the platform.
I would like to hear what Chris and Robert think what a Power user is. as during the show it wasn't clearly stated.
I do Agree with Robert that Of all the teams at MS that Windows (and perhaps Office) are the most secreative groups at MS.
There are two things I wish to point out. Most of the money made in Windows is from Business customers so that will drive most of the feature set. And businesses what stability so the 4 year turn around on OSes is a good fit for Business users.
Consumers tend to have need for access to the latest and greatest before hand. And that support for it in a 4 year cycle feels like it is breaking down. (lets just say that Blue Tooth support has a ways to go yet for those that have felt that pain)
Comments??
Interesting interview with one of the PM's from the MBF team. The much delayed Business foundation, which was linked heavily to WinFS.
http://www.eweek.com/article2/0,1759,1789299,00.asp
Interesting observations. MBF is part of the Orcas VS delivery. and it is expected in late 2k7.
Now where does this put Orcas at. I for one don't want to Shell out 300+ for standard 1200+ for pro (retail prices) and turn around a Year later and look at upgrading for LH.
Personally, Whidbey has been delayed Long Enough that if LH is released in 2k6 the designers for the platform should be built to work within whidbey. It would be interesting to see the Tools support evolve for both Avalon and ASP.NET on the “designer role“ but keep the cost in the realm of the Single Consultants realm.
So the whys of Orcas in 2k7.
It will give time for the Product to mature. Also the ability to implement version 2 of most of the DSL additions. Hopefully bring in a Small teams approach to some of the LIFELINE tools.
Delivery of a solution for the biggest issue we face in more software design. The mismatch between Objects, relational data and xml data. And the move to the universal data store.
I am sure these questions will be addressed at PDC 2k5. Although it would be nice to know before I drop the money on the Whidbey platform. or make a decision on which poison to pick on the MSDN subscriptions.
I am a small vendor. so in my role I am architech, test and the like. I do believe that team systems does have a Place. but give us a path to upgrade from.
An example would be that we already have source safe for small teams that can move up to Team Systems source control as the team expands.
it would be nice to have that for bug and workflow items.
the team already knows I would like to push source control all the way down to express editions. I personally believe that the Testing platform should start at express and propegate to a Test Team enviroment. IOW if we are going to give the tools for Hobbiests and dabblers at least give the tools to start down the road to proper develoment.
Oh well, got off on a tangent there. So back to the topic at hand. When do you want orcas and what feature set do you want. I personally am hoping that Orcas is now focused around the LH server delivery with particuliar view of delivering the WinFS platfrom for the Client and what ever they call the server vesion.
comments??
While those of us that couldn't attend wait for the presentations to show up on the web and scour through weblogs to catch a glimpse of the presentations.
I find it disheartening that there Seems to be a SERIOUS lack of understanding of the computer industry in the blogosphere.
First, maybe it is because I worked for 11 years in the semiconductor industry. have built a working 4 bit Computer from discrete compo nets. wrote interfaces between robots and the main system over rs232. I don't know if it is my hardware centric back ground that makes that difference. but when I look at the Name of a Conference that Says Windows Hardware and Engineering conference. I am expecting to get information on the boring stuff of computing. Boring but the Most important as this is the foundation that makes the exiting stuff possible.
While at PDC 2k3 and WinHec 2004 we were presented with a first showing of what MS wanted LH to be. (most of us then said it was to big of a chunk) a few things have changed in that time period.
We know that we were told that. so expecting that LH was further along (which it is in several areas btw) would be a fools goal. Most of the LH developers were repositioned to get SP2 out the door. and drive the sp1 support in 2k3.
But back to the point. Most of the new exiting stuff coming in LH build on the new support in the kernel. Express and Glass for aero will not work without LDDM drivers for example. The new File Filter stack in LH will require new filters written for it. (think all your AntiVirus, and other file filters. but even lower than that is the optimized streaming optimization made in the kernel for the file system.
Winhec is all about getting hardware vendors on board to write drivers to support these changes. They don't need to see the final outcome of what is planned they just need to see enough to know that it will be worth while to support it.
My main disappointment with WinHec 2k5 is that WDF wasn't at a final format so that Hardware vendors are not writing old style drivers for xp x64 and the x64 bit servers. then turn around later to write wdf based drivers. especially those that could be User Mode Drivers now to get them out of the kernel and increase stability.
One of the folks asked why to upgrade to LH a while back. with so many features going down level now (speaking of winfx, avalon and Indigo)
One of the Main reasons to upgrade to LH will be the significant upgrades coming the kernel. Running a winfx app on LH as compared to xp and 2k3 will be no comparison. especially if you need to use any streaming technology.
Beta 1 will be a step in the right direction although it will generally still be in the lower levels of the OS. PDC build will be closer to what is expected for general consumption and of course Beta 2 will be feature complete(feature complete doesn't not mean spit and polished).
I remember reading a blog entry a while back that stated that Scoble stated that the sidebar was being dropped from LH, at the time I didn't have time to fully investigate.
I guess tomorrow we will find out if the next winhec 2k5 build as it will probably become to be known has it or not.
I have 2 concerns about it. As I sit and look for Wide Screen LCD Monitors, I find it interesting that in the HDTV market they are easy to find in the higher sizes the tend to drop to edtv modules at the “15”in range. The reason I bring this up was that the article seemed to hint that the reason that Side bar was being dropped because of screen real estate issues.
A HD Monitor, (or a better description) a Wide screen LCD monitor that allows for 16:9 aspect ratio with the capability of displaying ATSC Signals at 720p 1080i (and hopefully 1080p (hoping I am remember all of the lines correctly) would account for most of the Lost screen real estate given the wider screen as compared to the standard 4:3 platforms in use today.
In particular. HD MCE's will be pushing for Wide screen formats, but I believe there will be a “need” for it as xbox 360 issues in the era of HD gaming. Actually I believe that MS would be dropping the Ball in the PC arena if they were not pushing for HD gaming in the PC area also with the Release of LH.
what are your thoughts
I have been thinkng more on WinFS since I posted the last post.
Several sites still list WinFS as going in beta 2nd half 2k6. Which was based on statements that MS made in regards to the platform.
I was thinking more along the Lines of a r2 release cycle which would be 2k8 for client 2k9 for servers if history maintains itself.
I am now thinking along the lines of the sql server.next. I don't think ms will wait for 5 years again to release the next sql server.
The data aspects that would be built into a C Omega based CLR would fit in quite well in such an enviroment, not to mention (unless w3 has become such a quagmire that it can't release standards any longer) xquery should be finished.
So it makes sense that a statement that WinFS would not be part of server by MS (makes more sense if WinFS is a client side api and they will have a different name for the server side api, but lets assume that WinFS is a Unified api)
So with that said. WinFX version 1 will be based on .net 2.0 for the most part. and be in the first release in client and server release of LH. Closely followed by sql server.net which will have .net 3.0 in it with the C Omega technologies. at that time the unified store will roll out. this gives times for an upgrade path for exchange 12, etc etc to be imlemented and tested and have a very clear platform
I believe this will be released a year after server and before the R2 server platforms. but probably with the R2 of the client.
Douglas
First this story is pure speculation built on information presented on what is publically available on WinFS.
Since Longhorn originally was only going to be a client release. WinFS was focused on the client picture. This was pretty much confirmed by Bill Gates in the Q&A at the Office developer conference. Watch the Keynote from that conference.
Given that focus, WinFS is a client side workgroup enabler, IOW it is has designs around a domainless workgroup and excells in that enviroment. It breaks down in a domain Server Client enviroment where a Central IT department will want total control.
After PDC 2k3 there was an outcry of putting WinFS on the servers. which brought us to the situation that WinFS was going to be in Beta, when LH client was released. and released when the LH servers were released a year later.
Here is some interesting issues. it was during that time and then the announcement that WinFS wouldn't be ready that certain items were released from the Language development teams and the Data teams. That item was a Compiler from Microsoft Research called C Omega.
There is some pain with OO languages interacting with xml platforms and relational platforms. (Hmm sounds like WinFS doesn't it) If that is resolved at the language level it would make for a more performant data access layer.
I personally believe from information presented and some observations. (the Office developers conference points to it, listen to the interview). We know that the Universal File store will be based on a modified form of Sql Server. Share Point Services appear to be the first item that MS claims to be built on a universal File store. It will modify to the real one as it is released.
On the Server side. we know that in 2k3 at least AD is built on the Jet engine. and it was just announced that exchange 12 will have a modified jet engine for it. for a Universal file store to work both of those will have to change sql server based store. We know that isn't going ot happen by LH server in 2k7.
So given the state. WinFX 2.0. will have a clr that will be based on a data implementation that C Omega has shown. what it looks like in the end will be different that C Omega. as it will be cross language solution. we know that it will be cross language based from the interview with anders on Channel 9.
I believe that by the R2 dates of LH client and server that we will see the implementation and conversion to the Universal file store. The client side will have athe WinFS API. server side will have a different api name but will be compatable some what. the new Data tier and persistance layer (known as Object spaces) will also have a different look and feel.
Comments.??