Tech Off Thread

21 posts

Forum Read Only

This forum has been made read only by the site admins. No new threads or comments can be added.

Microsoft Object Spaces

Back to Forum: Tech Off
  • User profile image
    Cronan

    Does anyone know what has happened to Object Spaces? The MSN group seems to be dead, found something here.
    Man, I am sooooooo sick of writing my own object persistence to database ....

  • User profile image
    Cronan

    Oooer. Just found this.

  • User profile image
    SMac

    both J2EE and the Upcoming objectspaces rely primarily on Xml to table mapping.. why aren't Object/relational databases becoming more popular in the light of XML?

    would it be smart to implement a similar OO-like DB in a Relational DB using what would amount to a class table (class/schema), an object table (xml data/instance), a relation table (xml+instance) and indexes on xml elements which would be important to index?...

  • User profile image
    Natty Gur

    Java got implementation of Javaspaces such as gigaSpaces : http://www.gigaspaces.com/index.html.
    I feel sorry that .Net enterprises abilities are poor comparing to Java. Actually I post expertly on that issue today.

  • User profile image
    SMac

    Please explain more the shortcomings of .Net enterprise abilities versus J2EE?

  • User profile image
    Natty Gur

    Well, I think that I need to write long article to lay down all .net shortcomings in the enterprise field. To sum it up to few lines Java is in the enterprise business for long time. at least from marketing point of view java target was enterprises while Microsoft try to target more low level and large audience. The long Java experience in the enterprise market ends up with many Java specs and implementations that solve existing need in the enterprise field. I don't think at all that Java as a language and development tool is better then .Net on the other hand I do think that Java as a set of solutions got much more to offer to the enterprise then .Net. I don't ask from Microsoft to fill all of those gaps but I do except from MS to be part of open source projects that will create .net implementation for enterprise needs.

    I really concern because I don't see anyone in MS from the development teams (Not marketing stuff) that really see those shortcomings. For example I got a conversation about implementation of MVC in ASP.NET. MS guys try to argue that UIP (application block) is the right solution. They don't accept my arguments that controller should be separate from viewer. Now why on earth Java new studio creator should implement clear MVC implementation while I need to write my own?

  • User profile image
    Charles

    "I really concern because I don't see anyone in MS from the development teams (Not marketing stuff) that really see those shortcomings."

    Please elaborate on what these shortcomings are. What could we be doing better in the enterprise? Don't be shy. Long posts are fine Smiley

    Charles

  • User profile image
    Natty Gur

    mmm long post …

     

    I'll start with few words about myself to make the picture clearer. For the last 5 years I deeply engaged with competitive intelligence. Competitive intelligence is actually application of intelligence for the business community. Competitive intelligence aim is to forecast events, intensions, movements, etc' of people and organizations that might help your enterprise to make the right decisions. To reach competitive intelligence conclusion competitive intelligence software need to interact with all existing and upcoming enterprise software. The interacted software could be interior and exterior enterprise software's. Using the gathered data from enterprise systems and exterior systems CI application can make it own analyze and reach CI conclusion. As you probably guess implementation of such software is always force you to integrate with enterprise existing systems.

     

    Integration, from my point of view, is not only to consume data or logic from other systems. Consuming of visualization aspects, especially if those visualization aspects implements visualization algorithms, are also desire when integrating to other systems.

     

    So beside enterprises systems that I saw, I need to create my solution in such a way that I can use it in any organization regardless of the implementation tools and technologies that they choose. I decide to use HTML as the protocol that I will use to implement integration of my CI solution with existing and upcoming enterprise systems. Using HTML I can easily combine several data, logic and even visualization into one page and display it to the end user. I start to develop ASP based solution and migrate to ASP.NET as soon as it was available (I also use to help others and wrote couple of articles thus get ASP.NET MVP.). CI solutions aren’t simple, they usually collection of components develop as n – tired application. As complex web application I want to use application server and other enterprises services that I saw my clients implementing in their systems.

     

    The first service that I looked for is application server. You can't create complicate software solution based on hundreds and above classes, without any application server that implement inversion of control and aspect oriented programming. It doesn't matter if you create web or client server application you still need application server to deal with your classes. You need your application server to implement any level of container services. Container actually activates objects for you (by request) thus can implement interceptions that can activate predefined services (such as object pooling, transaction, clustering, security, auditing, logging, message driven components, persistence and others) or custom services for your classes. COM+ is actually build with inversion of control and aspect oriented programming and .Net introduce attributes that even support better aspect oriented programming. The problems are:

    1. COM+ is COM implementation that hurt .Net performance and there isn’t any .net container implementation yet.
    2. There are set of pre-defined services that COM+ container introduce and they are poor (for example there isn’t any clustering services) and you can't extend them with your own services.
    3. Java implementation introduces managed extension that control objects and contains data about object and object procedure monitoring. COM+ monitoring and management is poor comparing to java implementations.
    4. Java application servers also come with naming services and proxies. Naming services are like UDDI they let the client get instant of class by the class name. In contrast with .Net that let us create instance from class just by referencing class assembly, naming services let us get object from remote storage just by its name. This is very important service for enterprise systems where one component that one creates is used by all other systems. Using name services when I change my components all other systems don’t need to do anything in order to get the new instance (no recompiling or mapping reference in config files).
    5. Application server don’t just use containers for server side classes the containers also used for object proxies returned by naming servers, thus enable container services to client side proxies as well as server side objects.
    6. Application servers usually perform deployment task for hosting classes. For example application server will create proxy for hosted class and register it in the naming services.

     

    Application servers are important but not the only enterprise needs that java community creates proper solution whether as commercial tool or open source project. Most of those solutions got specification developed by sun and can be found at http://www.jcp.org. I'll mention just some of them that I see the need.

    1. ORM tools. Yes I know that Microsoft is going to ship object spaces with Whidbey but why not to ship such a tool with the first version or at least publish specification of ORM tool. If you are moving your development tool to OO why don’t you deal with mapping of relational database to objects?
    2. In memory database. Once there was IMDB but it gone …. On the other hand Java got JavaSpaces spec. with implementations such as GigaSpaces. IMDB disappear but the need for in memory database still exists.
    3. Rule base. I hope that I don’t have any mistake here. While Java community develops spec for rule base engine and Open source projects that implement rule base engine start to emerge. All I can find in the Microsoft community is that article on MSDN (http://msdn.microsoft.com/architecture/journal/default.aspx?pull=/library/en-us/dnmaj/html/aj1rules.asp).

     

    As I post earlier even MVC implementation of ASP.NET is an example for the difference attitude to enterprise development. While Microsoft published application block that implement MVC where the controller and the viewer are the same class Sun implements MVC with separate viewer and controller in their new development tool (that support now post back).

     

    As I already said I don’t think Microsoft should develop all of those solutions but I really think that they have to do something in order to push more enterprise oriented solutions to their development products.

     

     

  • User profile image
    Natty Gur

    I'll be happy to hear any MS comments ...

    I think it will be great if MS and we, as .Net community, cooperate to supply more enterprise solutions to the .Net developers.

    The first step might be a standard body that will set standard that solution developers should follow (by their own implementation). I'm ready to lead or join such an effort but I think MS should have strong involvement in such effort.

     

  • User profile image
    LazyCoder

    One of the first things I started to look for when I took on my latest task was a .NET version of JBoss, Gemstone or WebSphere. I couldn't find any. It certainly looks like the pieces are there too build one using the .NET Framework, but none exist right now. I think we could build an application server that would outperform any Java app server.

    Java application servers also come with naming services and proxies. Naming services are like UDDI they let the client get instant of class by the class name. in contrast with .Net that let us create instance from class just by referencing class assembly naming services let us get object from remote storage just by its name. This is very important service for enterprise systems where one component that one creates is used by all other systems.

    You could hack something together using WS/.NET Remoting and the Activator but it still doesn't handle the caching of the object, locks and so forth.

    As I post earlier even MVC implementation of ASP.NET is an example for the difference attitude to enterprise development. While Microsoft published application block that implement MVC where the controller and the viewer are the same class Sun implements MVC with separate viewer and controller in their new development tool (that support now post back).

    I've argued that the current ASP.NET code-behind model IS an MVC pattern and the View and the Controller do end up in different classes, but there are a lot of issues that go along with the implementation in ASP.NET. I've never figured out why they didn't pattern it more after the JSP model. Afraid of being a copycat? More likely because they started spec-ing out ASP.NET before JSP was released. Plus, that MVC model is only for ASP.NET.

  • User profile image
    Natty Gur

    The question is if we can join forces together with MS to build Application Server design for .Net and using .Net capabilities ?

  • User profile image
    Shaji

    I was at the Object Spaces Presentation at PDC and would like to comment on a few things. Traditionally the MS programming model focused more on getting the developer to build solutions using existing MS products to marry them into a cohesive whole to solve an enterprise problem (this may or may not work all the time). Which is one of the reason why Enterprise Architects looked down on us poor MS-Oriented programmer (or MSOP’s), as our solutions were invariably constrained by the products that MS offered. Java with J2EE suffered no such constraints and offered what ever the J2EE APP Server offered. Having said that code portability among J2EE APP server went down the proverbial gurgler as functionality increased. The documentation on IBM's site to port BEA Web Logic to Web Sphere is over 300 pages long.

    When .NET first come out it was a truly amazing (at least for me) platform and offered the enterprise developers a mechanism to break the product glue mentality. Then when I started using it, I found that it had all the features to do the job at least 90% of it. And the last 10% of the features it did not have took the 90% effort on my part. Also within the Microsoft developer community there is still a lot of focus on stitching products together and now instead of VB or C++ it is using VB.NET or C#.

     

    At the PDC I did ask a few people about implementing the MVC controller pattern in ASP.NET and got varying answers but none solved the issue. The classic Model 2 pattern remains an enigma (No I don’t want to implement the UI application block), mind you it can be implemented if you change the rules a bit, but then some one said "if one makes enough assumptions one can hang an elephant by it's tail off a cliff". Next came the question of Container Managed Persistence (CMP) or Bean Managed Persistence (BMP) Object spaces would have solved CMP in a kind of way but not really so the performance hit of 50% was not something I was prepared to live with. BMP is there is a kind of way as long as we rolled our own.

    Next came app servers and scalability using clustering. The one way I got around it was to use Lhotka excellent book on Business Objects. Which creates a kind of Object factory and a class in charge model using remoting. It takes sometime to do all the basic framework implementation as there is none provided by the framework. There is a possibility to implement clustering by having the remoting server on ISS with Windows Load Balancer. Again as in the previous post if only we had access to something like JNS (Java naming Service) all this would have been easier (Some may argue to use UDDI for the same purpose that comes with Windows 2003, it is not really the same).

    The other challenge in Enterprise Systems is that the Business Rules change constantly and any implementation using if then structure are asking for trouble. I would love to have something like JESS (Java Expert System Shell) to write my rules in and have it compiled to .NET IL which inturn will work on the objects dynamically. Then when the rules change just recompile one component and hey presto..I had heard that guys at Sandia were thinking of porting JESS to C#. In my last email exchange with the JESS’s author he said J2EE is superior to .NET and he is sticking to it J http://herzberg.ca.sandia.gov/jess/.

    This is my last peeve. In Visual Studio why is there no support for developing web services with contract first implementation. These days you have to first write a WSDL using XML Spy or something then use WSDL.EXE to generate the interface with /server switch and finally implement the abstract class. What a long winded way. Why not give developers the ability to develop the interface first and then the implementation. In a typical enterprise scenario you usually get the Interface contract before you can start implementing it. Also there is not support to write your own WSDL in VS.NET 2003 either (I mean a GUI front end).

    There are a lot more peeves but I will stop now. But having said all this and having used JBuilder + WebLogic and tried hopelessly to do a good UI in java I would not trade .NET for anything. I would rather build what is not there in .NET than abandon the platform. I firmly believe within MS the change or awareness for Enterprise Systems has started and it is only a question of time....

    Till then I will soldier on and hope some one is listening.

  • User profile image
    RichTurner

    Thanks for posting a detailed critique Natty.

    There are a bunch of subjects included in this mail and I would like to comment on one in particular: Application Servers.

    The short answer to your question is that Microsoft does have an App Server - it's called Windows! It offers a component hosting model in .NET ES/COM+, Queueing support via MSMQ, WebServices support via IIS, Object remoting via .NET Remoting.

    In response to the issues you posed above:

    A) COM+ Perf: If used properly, COM+ can host .NET Enterprise Services (ES) components such that they perform just as well as native COM+ objects (see my paper http://msdn.microsoft.com/library/en-us/dncomser/html/entsvcperf.asp for more details).

    B) COM+ Services are poor. Not sure why you state this. COM+ offers a broad range of services that control object hosting (inproc/out-of-proc), lifetime (JITA), pooling, security, restartability, etc. When you say clustering, do you mean clustering or load-balancing? If the latter, Component Load Balancing is available through ApplicationCenter ... but more on this shortly ... Smiley When it comes to extensibility, what kinds of extensibility would you like to add to the COM+ model?

    C) We know that customers want more monitoring control over their running services ... more on this shortly ... Smiley

    D) One of the problems with distributed component registries is that they rarely offer the level of versioning support that one may require. Also, there are perf' issues that need to be carefully considered - you don't necessarily want to go across the network and lookup the physical location of every component in your system when you want to create a new instance. Again ... more on this in a moment.

    E) COM+ provides a powerful proxy story. While is may not match Java in implementation, it is a powerful mechanism. Is there something in particular that it does that you don't like, or that it doesn't do?

    F) Accepting that COM+ does provide an abstracted naming service, it's also clear that COM+ provides proxy-generation infrastructure.

    So I've mentioned "more on this shortly" a few times above. Why? Well, we are listening to our customers and all the points you raise above (along with many others) we're taking very seriously when it comes to designing new Products.

    So what's coming next? In short, "Indigo". This is the codename for our next-generation .NET application platform that makes developing Service Oriented systems MUCH easier. "Indigo" is a very powerful, flexible app platform that provides an entirely managed environment for hosting services implemented in .NET, abstracting the physical hosting model and networking/protocol/capability support from the service's code. "Indigo" also supports the most important and stable elements of WS-* including WS-Security, WS-Addressing, WS-AtomicTransactions, WS-ReliableMessaging, along with WS-Policy, etc.

    So, essentially, "Indigo" will meet all your requirements:

    A) Indigo offers a variety of hosting models - .EXE, NT Service, IIS, DLLHOST, etc. "Indigo" is brand new, entirely managed code which means that there's little interop to get in the way ... unless you explicitly interop with COMponents.

    B) Indigo provides a huge variety of services and its architecture is entirely pluggable so you can add capabilities of your own if you wish.

    C) "Indigo" will support a vastly superior management & monitoring story over our existing app infrastructure.

    D) "Indigo" services are essentially WebServices on Speed! Because we'll be supporting WS-Discovery/WS-Addressing, services will be location independent (you specify the URI of the target service and "Indigo" will look it up for you at runtime". The runtime will also deal with caching pre-referenced service addresses etc., to minimize nw hops.

    E) Proxies are managed by the "Indigo" runtime infrastructure.

    F) "Indigo" will aid deployment by making service hosting transparent and hook into any WS-Discovery/WS-Addressing infrastructure.

    "Indigo" was first announced at PDC in LA last fall and will be nearing its Beta1 later this year. Keep your eyes peeled on MSDN for more details.

  • User profile image
    Natty Gur

    RichTurner wrote:

    Thanks for posting a detailed critique Natty.


    I just want make it clear that my intensions are to help me and others to build systems using MS technology. I don't know when you need to stand against Java guy and convince customers that MS got better solution for their enterprise. Believe me, it's not easy task! I'll comment your replay but generally I'm still thinking that we are missing something in the enterprise solutions that we (MS community) have to offer. It's more wide then Application server it's about Enterprise development attitude. You can see it starting from specification through application server ending with MVC 2 Implantation in windows/web forms.

    RichTurner wrote:

    The short answer to your question is that Microsoft does have an App Server - it's called Windows! It offers a component hosting model in .NET ES/COM+, Queueing support via MSMQ, WebServices support via IIS, Object remoting via .NET Remoting.


    well I don't think that operating system is application server. Indeed most of the enterprise need exist in windows (some are still missing) but I still need to write code to tie services together. Application servers on the other hand host classes for me and supply those services by using aspect oriented programming. Application server should help me to write just classes to apply my application business logic and DAL. Deployment, pooling, clustering, transactions, logging, audit and any other services for classes that my organization might want should be provide by application server for me. Furthermore application server implemented as one software can be port from windows XX to longhorn without any change in the applications. You just need to adjust your application server to Longhorn. So I think the idea of coupling application server and OS is not so good.

    RichTurner wrote:

    A) COM+ Perf: If used properly, COM+ can host .NET Enterprise Services (ES) components such that they perform just as well as native COM+ objects (see my paper http://msdn.microsoft.com/library/en-us/dncomser/html/entsvcperf.asp for more details).


    Well I already read your article and the really (from my experience) is a little bit different. In fact all the systems that I wrote used COM+ and other services. I found out that COM+ especially as server application has bad effect on .Net objects. I also published article on that subject (http://www.codeproject.com/dotnet/complusnetpracticalapp01.asp).

    RichTurner wrote:

    B) COM+ Services are poor. Not sure why you state this. COM+ offers a broad range of services that control object hosting (inproc/out-of-proc), lifetime (JITA), pooling, security, restartability, etc. When you say clustering, do you mean clustering or load-balancing? If the latter, Component Load Balancing is available through ApplicationCenter ... but more on this shortly ... Smiley When it comes to extensibility, what kinds of extensibility would you like to add to the COM+ model?


    well, I think that clustering, fail over, resource pooling, logging, audit and mostly the option to add my own services are missing (naturally I can list more). I want the ability to set in config file that certain attribute that I create (based on given interface Smiley ) should called given service that I wrote. My user (programmer) just need to decorate his class with attribute, all other work should be done by Application server.

    RichTurner wrote:

    D) One of the problems with distributed component registries is that they rarely offer the level of versioning support that one may require. Also, there are perf' issues that need to be carefully considered - you don't necessarily want to go across the network and lookup the physical location of every component in your system when you want to create a new instance. Again ... more on this in a moment.


    Yes, you right. But don't decide for me what I need and what good for me. Give me the opportunity to choose whether I want to use naming services or version attitude approach to get class instance (remember extensibility Smiley, let me configure it). Any way naming services return type to the caller and not physical locations. It really is nice if we can deploy just type without assembly.

    RichTurner wrote:

    E) COM+ provides a powerful proxy story. While is may not match Java in implementation, it is a powerful mechanism. Is there something in particular that it does that you don't like, or that it doesn't do?


    DCOM ? why DCOM for .Net applications? I'll also would like to see a way to intercept and add custom interceptors to proxy activity (extensibility Wink ).

    RichTurner wrote:

    Indigo


    It is hard for me to say something about product I cant play with yet. I've been in PDC and MVP summit and all I heard about Indigo is on it Connectivity nature. I understand that Indigo is combination of Remoting, Web services and COM+. but I never saw or heard something about COM+ aspect of Indigo in details. I'll be happy to hear.

    Again I don't want to offence or attack any one. I just want to help as much as I can to see more robust MS solutions. I think that most of the work should be do by MS community (as open source or commercial solutions) together with MS. I'm just trying to do as much as I can (Write open source solutions, help others, write about it ..).  Who knows maybe sometimes I can do it within Microsoft Smiley .

  • User profile image
    dsuspense

    Natty, I think you should take a look at Spring.NET http://www.springframework.net/. It is going to materialize to be a very good enterprise level framework for .NET, as Spring http://www.springframework.org/ has become for J2EE. Spring's aim to simplify and reduce common J2EE enterprise app problems is by creating a framework that provides IOC, AOP, a Bean Factory, MVC and Plugable Persistence. It will provide even more advantage than a .NET "app server", since you can program to the framework (higher abstraction level) and use whatever underlying app server you like to provide your basic enterprise services like transactions, logging, persistence etc. The fact that .NET does not really have one cohesive Enterprise API such as J2EE is another reason I see Spring.NET fullfilling a void that is definitely needed in the Enterprise .NET space.

    Also, EJB has fallen out of favor in the J2EE world, and now JDO and Hibernate based persistence mechanisms are becoming more widely adopted. EJB 3.0 will be approaching more of a JDO-like persistence model to achieve CMP.

    For good reading material check out DotNetGuru, and especially this article on lightweight IOC containers from a .NET perspective: http://www.dotnetguru.org/us/articles/ioc/ioc.html?POSTNUKESID=11265df57d0ba

    -D

  • User profile image
    Natty Gur

    dsuspense, hi

    thanks for your comment. I already saw Nspring couple of weeks ago on theserverside.det and read Sami article (which is one of the nspring developers). I'm waiting to see the final product .....

    Anyway tier target is light container. I still thinking that is a strong need to container. Maybe the best way to achieve it is to take the challenge and Immigrate JBOSS to be 100% .Net kosher Wink .

    should we publish it on theserverside ? Wink *2

  • User profile image
    jwanagel

    Hey Natty, I'm reading your posts in this forum and I completely resonate with a lot of your thinking and comments in this area.

    There is a group at Microsoft (called PAG) which is focused on providing guidance on building enterprise business systems using the Microsoft platform that you should check out if you haven't before (http://www.microsoft.com/practices).

    I work in the PAG group and someone at MS forwarded me this thread because they read your message and thought PAG would be the right place to direct you to.  The group actually has some projects in-progress that relate very much to some of the things you touch on, so if you're interested in learning more about them I can elaborate on that.

    I find the topic of Application Servers and the Microsoft platform very interesting because it's one of my personal hot buttons and often a huge debate.  People's reaction or response to the topic tend to fall under one of three categories either: "Microsoft doesn't have an app server", or "Windows is the app server", or "Microsoft has 5 to 10 different app servers depending on how you count them".

    Unfortunately each answer is in ways both correct and flawed depending on where you stand or are looking at it.  So I'll just expand on some of my thoughts around the topic without specifically trying to answer it. Smiley  It would be interesting to hear if the following comments make logical sense to you or not.

    A basic primitive definition of an Application Server is something that can host and run server side application/business logic.  You touch on a lot of the additional features or value adds of an Application Server, but hosting logic is the most fundamental one.

    Hosting server side logic means running in some process.  Microsoft has several possible processes you can use for hosting server side application/business logic.  IIS, COM+, a straight custom Windows Service, Exchange, BizTalk, SQL Server, etc.

    IIS and COM+ are normally what come closest to qualifying as what people typically think of as an "Application Server", with probably a custom Windows service being third, and BizTalk, Exchange, and SQL server following.  I've certainly seen people use all of them in various ways as an Application Server (i.e. some process to host their server logic).

    One of the most frustrating things right now for people creating enterprise business applications on the Microsoft platform is choosing their host process aka Application Server.  Do they use IIS?  Or COM+?  Or BizTalk?  Or a custom Windows Service?  There are feature differences and tradeoffs with each.  IIS is really very good, but if you're not building a web application, then it's not so good.  However a lot of its features would be great even for non-web applications (monitoring, scalability, and management tools).  COM+ has some good management and scalability features as well, as well as some monitoring capabilities, but the problem of having to use DCOM as the only way of communicating with your COM+ application server is problematic.  A Windows Service has no monitoring or scalability features, and minimal management (a start, pause, stop interface).

    The other problem is that each is tightly coupled to certain things that force you to be making the decision based on criteria you don't want to be basing your decision on.  If you like using IIS as your App Server for management and scalability reasons, but need to be processing messages from something like MQSeries, you are out of luck because IIS can only be used for processing HTTP or WebService requests.  You would actually have to use a custom Windows Service in that scenario, since COM+ only takes requests (outside machine) via DCOM.

    You also mention some features that none of our "potential" app servers support.  So this is where the "Microsoft has no app server" camp comes from.  Since they don't feel IIS, or COM+, or anything else meets their minimum feature bar to qualify since it's lacking X feature.

    Jonathan

  • User profile image
    Natty Gur

    Jonathan Hi.

     

    I'm happy to see concrete answer!

     

    I'm aware to PAG activity and visit PAG site regularly. I still think, and probably you, that there is still lot left to do. I'll be happy if you can contact me by mail so we can discuss ways of cooperation. I really want to feel that I'm doing something and not just yelling J.

     

    Anyway I'll be happy to hear anything that you can elaborate on publicly or share with me (I'm under NDA (ASP.NET MVP)).

     

    I agree that ASP.NET, COM+, BizTalk, etc' host server logic but I think that there sould be one mechanism that host my classes in every one of those product. There should be one container that holds my written application logic class and provide them the same services. I can configure where that Container would run. Working this way, my code is just aware to Container services (it indirect awareness anyway, since my code use Aspect oriented programming – AOP to request container service). It doesn't meter if I run my code eventually in one program process or another, it transparent to me (even if I move to different operating system). I can accept that there are several server side applications that might run my code but I want them all to be able to host container or run my class from out of process.

     

    Your comment about the frustration of choosing the right application even supports my argument. If ASP.NET could serve the same services that I can GET from COM+, I could use ASP.NET or COM+ and the COM issue wasn’t rise at all. If I wasn’t clear my Application server or any application could host container the same way that I can run COM+ application in or out of process.

     

Conversation locked

This conversation has been locked by the site admins. No new comments can be made.