Update on LINQ to SQL and LINQ to Entities Roadmap

Published 29 October 08 05:36 PM | dpblogs 

Since the release of LINQ to SQL and the Entity Framework, many questions have been raised about the future plans for the technologies and how they will relate to each other long term.
During this week of PDC we are now at a point, with the announcement of Visual Studio 10 and the .NET Framework 4.0, that we can provide more clarity on our direction.
 
We have seen great momentum with LINQ in the last year.  In .NET Framework 3.5 we released several LINQ providers, including LINQ to SQL which set the bar for a great programming model with LINQ over relational databases.  In .NET 3.5 SP1, we followed up that investment with the Entity Framework enabling developers to build more advanced scenarios and to use LINQ against any database including SQL Server, Oracle, DB2, MySQL, etc.
 
We’re making significant investments in the Entity Framework such that as of .NET 4.0 the Entity Framework will be our recommended data access solution for LINQ to relational scenarios.  We are listening to customers regarding LINQ to SQL and will continue to evolve the product based on feedback we receive from the community as well.

Tim Mallalieu
Program Manager, LINQ to SQL and Entity Framework

Comment Notification

If you would like to receive an email when updates are made to this post, please register here

Subscribe to this post's comments using RSS

Comments

# insurancesitesfind » Blog Archive » Update on LINQ to SQL and LINQ to Entities Roadmap said on October 29, 2008 9:19 PM:

PingBack from http://www.insurancesitesfind.com/?p=6939

# Jens said on October 30, 2008 3:37 AM:

So you guys are actually admittingthat Linq To SQL is a dead end ?  Thanks a lot.  Linq To SQL has taht 'it just works' attitude to it and is the underpinning of our new project.  I coud never ever persuade my boss to go to Entity Framework.

And if you're really listening to your customers regarding Linq To SQL, we'll ga and read up all posts on the internet.  I think there are plenty of things we'd like (I for one have a long wish list, with on number one a sync tool for the designer). There as a time that MS aimed at the small and medium sized businesses, and Linq To SQL is perfect for those, never forget the basis of your success ...

# int19h said on October 30, 2008 4:28 AM:

Can we have a SharePoint Entity Framework provider, please? please? ;)

# John said on October 30, 2008 4:51 AM:

A appreciate the desire to have a single 'Linq to DB' framework, but I hope the proposed Entity Framework will offer full Linq To Sql compatibility? Allowing a painless transition to people who don't need all the extra muscle of the framework. I'd rather do OR mapping myself, and use Linq To SQL as a simple way to grab the data only. EF is current a long way from what I'd need!

# Matt Brooks said on October 30, 2008 6:15 AM:

I agree with John. LINQ to SQL offers a great light weight data access option with a number of key extensibility points and 'get-out' options when it doesn't provide the required functionality.

It is fine if MS have bigger plans for the Entity Framework and I'm glad to hear LINQ to SQL will continue to be developed and supported. You say you will listen to community feedback and evolve the product accordingly but what we lack at the moment is an official list of potential features, like a vNext roadmap.

I think at least knowledge that something like this is in the pipeline will ease developers anxieties regarding their investment in LINQ to SQL.

# Kristofer said on October 30, 2008 1:20 PM:

@Jens,

The number one item on your wish-list is here: http://www.huagati.com/dbmltools/

# Stephen Redd said on October 30, 2008 2:10 PM:

What I want is a light version of EF that does what LINQ to SQL does. It should enable the same simple usage scenario I have with LINQ to SQL, but it should use the same logical model and API as EF.

Most importantly it needs to be fully compatible with newer MS technologies that are planning to build on top of EF down the road.

The simpler usage scenario is especially useful to web developers... Most web apps, even the large ones, really don't need THAT much abstraction between the relational schema and the logical one. We don't need all the fancy mappings, the inheritance, and other advanced stuff that EF provides.

Sure, we don't care if it CAN do all those abstract things... we just don't want those features getting in our way, complicating how we use the entities, or putting shackles on what we can do or how it performs because some enterprise edge case scenario has a conflicting requirement.

Most web programmers just want a simple way to model our data and automate the persistence stuff so we can unleash LINQ in our apps without having to spend too much time worried about how to push data around behind the scenes.  

But we don't want to be left out of the loop with future tech that builds on the EF and LINQ to Entities... and it looks like MS has a LOT of stuff building on EF in the works.

Oh!, Please give us a conversion tool to help move existing LINQ to SQL apps over to LINQ to  Entities and EF....

Right now EF just isn't ready so many of us are sticking with LINQ to SQL until we can see if the next version of EF and LINQ to Entities is better. It would be nice if we can be assured that we aren't going to be stuck facing a massive manual code migration down the road.

# David Hayden [MVP C#] said on October 30, 2008 2:59 PM:

Based on the recent announcement on the ADO.NET Team Blog : "We’re making significant investments

# Community Blogs said on October 30, 2008 3:23 PM:

Based on the recent announcement on the ADO.NET Team Blog : "We’re making significant investments

# Colin Jack said on October 30, 2008 3:23 PM:

"We are listening to customers regarding LINQ to SQL and will continue to evolve the product based on feedback we receive from the community as well."

I really hope this is the case, if it is then I'd recommend considering the following two links:

http://codebetter.com/blogs/david.hayden/archive/2008/10/30/linq-to-sql-gets-kicked-to-the-curb-needs-a-good-home.aspx

http://codebetter.com/blogs/ian_cooper/archive/2008/07/02/showing-some-support-for-linq-to-sql.aspx

# LUTI @ Microsoft said on October 30, 2008 7:59 PM:

Bom, já têm um tempinho que nós vemos um enfoque maior no Entity Framework que no Linq

# The Other Steve said on October 30, 2008 10:31 PM:

This makes no sense.  Are you actually listening to customers?

# Saud said on October 31, 2008 2:20 AM:

I guess, they spent all the money on silly Vista ads!

# Reva said on October 31, 2008 3:49 AM:

LINQ to SQL is too good to killed like this, At least put it out on codeplex

# Milan said on October 31, 2008 4:41 AM:

Hi,

"We are listening to customers regarding LINQ to SQL and will continue to evolve the product based on feedback we receive from the community as well."

Can you please show me anybody from the team bloging about LINQ to SQL? Or discussing anything with community? Giving some responses? Doing any development after 3.5?

DamienG entered LINQ to SQL team 4/1/2008. After that you will see only 3 notes about LINQ at his blog...

Why is not possible for you to do some support?

And the main: "why there is never any response"? You publish "Update on LINQ to SQL and LINQ to Entities Roadmap" 10/29 - every information about roadmap is important everytime. And look - two day later it is discussed here and on different blogs - but there is NO RESPONSE from you or anybody else from your team. I don't think that you are showing respect to your customers which invests into your technologies

# Muhammad Mosa said on October 31, 2008 5:15 AM:

Sounds like everybody is excited a little bit! I am not a Linq to SQL fan but I admit it is a cool and very light and easy to use. So why ADO.NET team simplify this us! Are you going to continue developing and extending LINQ to SQL or not? It is only one year old technology if it is killed now better than being killed after few years! I am not asking to abandon it of course I am just wondering what is going on? puting the project on CodePlex as open source would be great idea of course.

Entity Framewok is much more advanced, designed to support different DB providers. Why I would use LINQ to SQL while I have something similar (but LINQ to Entities is not as powerful as LINQ to SQL) that support different Databases?! The only reason I can see was at the time of LINQ to SQL there was not EF, it was Beta and that is why no body really paied attention to it. And in EF v1 still LINQ to SQL is much better than LINQ to Entities.

So the post need to be clarified LINQ to SQL will remain and last forever or will die?!

# Ian Cooper said on October 31, 2008 5:22 AM:

LINQ To SQL supports important domain first modelling ideas like persistence ignorance that were woefully lacking from EF. It is a shame that it is not raised as the solution for customers who want that option more.

It is tragic that the EF team did not pick up on the design understanding shown by the L2S team.

It is tragic that the Data Services team continue to mischaracterize L2S as a RAD tool.  

It is tragic that support for features like fine-grained object models and exposed provider model, trumped as key advantages of EF, will not be added to L2S.

The journey from L2S to be a ral competitor to Nhiberate is much shorter than the journey for EF.

It remains the best offering from MS in this space.

If you want to know what I believe you could do to support the product, please follow this link: http://codebetter.com/blogs/ian_cooper/archive/2008/07/01/architecting-linq-to-sql-part-10.aspx

# Sidar Ok said on October 31, 2008 8:20 AM:

Linq to SQL is indeed one of the most useful tools in my toolbelt, but I can't say the same with EF. EF is a great example that if you don't listen your community & customers, you are going to get no confidence. Microsoft should have taken some lessons out of this already.

What we need is, support for other models than TPH and enabled provider model, and that's it. Why doesn't anybody hear us again ?

# Kristofer said on October 31, 2008 8:51 AM:

What is missing from Tim's post above is something along the lines of: "We have learnt a lesson from the EFv1 fiasco and so we're actually in the process of integrating all the goodies from Linq-to-SQL (copying-and-pasting code) into EFv2.".

He didn't say that, but he demoed several features during his PDC "Entity Framework Futures" session that look like they have been copied-and-pasted straight from L2S into EF. Watch this if you can spare 90 minutes and you'll see what I mean:

http://channel9.msdn.com/pdc2008/TL20/

# VistaDB .net database blog said on October 31, 2008 9:23 AM:

Interesting blog post about it . And some related information on Stackoverflow posts . The basic gist appears to be comments made on the ado.net blog that state the Entity Framework is the only thing getting major developer time for Visual Studio 2010

# Colin Jack said on October 31, 2008 9:24 AM:

"We have learnt a lesson from the EFv1 fiasco and so we're actually in the process of integrating all the goodies from Linq-to-SQL (copying-and-pasting code) into EFv2.".

If MS needs to learn from its own failures rather than learning from others then thats OK. Much slower than if they'd asked people, looked at existing practices/patterns but probably a lot more fun for the people involved.

Even accepting that though putting those features into EF is, in my view, not necessarily as good as having L2S. Depends on whether you think the whole EF foundation is sound, and I've heard nothing so far to make me think it is.

What bothers me most is that rather than learning about things like DDD/SOA/REST/TDD and then seeing how they can apply and improve on them they instead constantly try to come up with the next huge paradigm shift. Most of the time though its flavoured with the same old data-drag-drop style thinking:

1) DDD - Drag tables onto designer

2) REST - Generate resources from DB

3) SOA - Entity services/SOAP/WSDL

4) Web Design - Bind tables/datasets to controls

5) EIP/SOA - Drag and drop integration with Biztalk

6) DSLs - Graphical DSLs (drag-drop, you get the idea)

Does seem like some teams are listening though, look at MVC/JQuery on the Web side but thats more about MS listening and following rather than trying to drive the entire industry down some path that Redmond have decided is appropriate.

# Jason Short said on October 31, 2008 9:34 AM:

Microsoft has always said tha the Entity Framework was the ONLY provider model for LINQ.  That means LINQ to SQL was doomed to be a proprietary implementation quickly eclipsed by EF.  

EFV1 does need work, but that IS the model of the future from Microsoft.  They just didn't ship in the box with VS 2008.

The late ship, and the confusion with people already starting LINQ to SQL over a year ago I think will lead to a lot of negative backlash towards EF, even though this has always been the plan.

Entity Framework third party providers are just starting to come out.  The model is going through a rev 2.  I think it is here to stay.

# OnTheBlog said on October 31, 2008 9:44 AM:

An interesting post has arrived on the Ado.net team blog with regard to the future of LINQ to SQL. Although not coming outright and saying it, it seems that LINQ to SQL is likely to be deprecated in t ...

# msp said on October 31, 2008 10:00 AM:

I think this is a smart move.  Having two things try to do essentially the same thing is just room for confusion.  Do we use A and all it's simplicity of client programming or do we use B because it actually works with our more complicated persistence layer needs?

People trying to make that decision are in a lose-lose.  

Merging the two products into one hopefully means we'll get the best of both worlds.

# anon said on October 31, 2008 10:25 AM:

why wait or depend on MS, diy maybe? idk

# Joshua Thomas said on October 31, 2008 10:41 AM:

If Microsoft is essentially depricating LINQ To SQL they need to make it open source, so those who use it are not left high and dry based on the new flavor of the day syndrom.

Whoever thought releasing LINQ To SQL was a good idea should be fired. Better to have one solution than competing solutions.

# Travis said on October 31, 2008 10:41 AM:

EF is a *far* superior product, but it does not come close to nHibernate.  So does Linq To Sql dying bother me?  No, I say good riddance to the POS.  

Why all these shops are scared to use anything that is not from MS is beyond me...

# Gil Fink on .Net said on October 31, 2008 11:20 AM:

The End of LINQ To SQL? Today I read a few blog posts that were published regarding  the announcement

# Colin Jack said on October 31, 2008 11:23 AM:

"EF is a *far* superior product"

WS* star is *far* superior to REST according to one school of thought too though, choices. L2S seemed to me like a good foundation, a good v1 product.

# WS said on October 31, 2008 11:39 AM:

Good thing we just invested a lot of time and money into dead-end tech!

There is no way we will touch EF after this...

# David Hayden [MVP C#] said on October 31, 2008 12:33 PM:

This is my last post on this subject, I promise, and unfortunately it has to be a little more blunt as

# Community Blogs said on October 31, 2008 12:41 PM:

This is my last post on this subject, I promise, and unfortunately it has to be a little more blunt as

# kevindan said on October 31, 2008 1:14 PM:

The team is simply out of the touch with Linq2SQL users. Hence there is the plea. Here is one feedback.

If it's a burden to have one MS team develop on two products, either have another team working on Linq2SQL, or release the technology out to the community. As a Linq2Sql user, I am very disappointed if MS keeps it closed without significant new developments.

The technology is too good to be killed, the investment from everyone should be preserved.

As one of the customers, I WANT LINQ2SQL to be continuously developed with significant new features, no matter who is going to do the work.

# Alex said on October 31, 2008 2:09 PM:

Disapointing... When you say to all your friends that "Microsoft is changing" they came with this.

Well, time to move back to Subsonic...

# dpblogs said on October 31, 2008 3:00 PM:

Thank you for your comments. We apologize for the delay in response as we were at PDC this week, but wanted to get the post out in line with our PDC sessions. We would like to reiterate that we are listening to customers regarding LINQ to SQL and will continue to evolve the product based on feedback we receive from the community, but that as many of you said, we are making a significant investment in the Entity Framework and many other MS technologies are planning to build on the Entity Framework going forward.

We would like to assure you that as we invest in the Entity Framework going forward, it is our priority to ensure it enables the same simple, lightweight usage scenario you currently have with LINQ to SQL and will continue to be fully compatible with newer MS technologies that are building on top of the Entity Framework down the road. For more information on some of the features we are working on to enable this scenario in the Entity Framework please check out the EFDesign Blog and PDC’s EF Futures from this past Tuesday.

Elisa Flasko

Program Manager, Data Programmability

# leo said on October 31, 2008 4:05 PM:

I've met a lot of smart people at MS but the ado.net team is definitely not among them. thanks for deceiving me.

# Brian Adams said on October 31, 2008 6:48 PM:

It has been obvious via the posts that there was an internal spat and the vision of entity framework was somehow threatened by  Linq to SQL.  It is a shame that this resulted in killing the "it just works" solution.  As a customer Linq to SQL is straight forward and gets the job done.  Entity has another learning curve and requires me to manage more things than I necessary want.  Unless you plan on making entity framework EASIER and QUICKER to impliment than Linq to SQL, you are making a mistake.

# HerbT said on October 31, 2008 7:38 PM:

The simple fact that you MS people keep stressing your 'significant investment' in EF whenever the subject of LINQ to SQL comes up undermines any credibility you might have to the subject. You sound like political advisors on the evening news trying to dodge questions that will make your candidate look bad. We are not fools you know. We realize that there must have been some kind of internal turf war going on and somehow you guys won and are now pushing your next ill conceived, over designed, behemoth down our throats.  So, great, some other yet-to-be-named MS monstrosity is going to be tied to this one, so that we can sink even faster with two boat anchors instead of one. LINQ to SQL is still the best choice. It is simple and elegant and nearly perfect except for a few minor features and other providers. Yet instead of investing where appropriate, you invest a whole lot more for a train wreck. Is there no accountability at Microsoft? You can't be drunk on the stock anymore, so it must be something else. All I can image is oversized egos need oversized projects. Get a clue and talk to REAL customers, not just the corporate muscle heads that approve purchase orders for SQL Server.

# Mark said on October 31, 2008 8:17 PM:

If you do dead end the project, you should MSPL it so the community can have it and do what they want with it.

# Dmitry said on October 31, 2008 10:29 PM:

I certainly agree with all the comments that LINQ to SQL is a simple, elegant and extensible solution that just works. It has a lot of potential. It definitely should be released to the community if Microsoft does not want to support it.

If LINQ to SQL were to support multiple database providers, ability to use multiple databases (at least on the same SQL server) and had value object support, it would have been an excellent ORM.

I'm not going to use Entity Framework; at least not until it gets a lot more lightweight and be persistent ignorant. I'd rather go with NHibernate 2.1 which is going to support LINQ.

# EntLib said on November 1, 2008 12:45 AM:

本文翻译自Tim Mallalieu, Program Manager, LINQ to SQL and Entity Framework 在ADO.NET Team Blog 上的文章。在.NET 4.0,我们将对Entity Framework 做出显著的投入,推荐Entity Framework 作为LINQ对关系数据库的数据访问解决方案。同时,我们正在倾听用户对LINQ to SQL的意见,同时基于来自社区的想法,继续发展LINQ to SQL。

# Jay said on November 1, 2008 3:42 AM:

You really are a bunch of idiots! We have just engineered our whole product using LINQ to SQL.  Can Microsoft make their damn minds up! Arrrggghhhhhhhhh

# Hadi Hariri's Blog said on November 1, 2008 3:53 AM:

Linq to Sql is dead. Did you abstract well?

# Ori said on November 1, 2008 7:02 AM:

Elisa,

Why should we trust that Entity Framework will be supported in the future? When the ADO.NET team decides on a new technology in a year or two, then Entity Framework will be tossed aside as well.

How can MS platform developers trust you after this?

Ori

# Whatever is my name said on November 1, 2008 7:56 AM:

The only reason to stay with MS technology at all has been LINQ to SQL. Killi it and I simply will revert back to Delphi.

But you decision do not surprise me at all - if something works let's drop it and go for something that has problems - that way we can suck more money for 'fixes', 'updates', 'new releases'. More that usual typical Microsoft. You turning this into massive disappointment :( After all seems MS doesn't care about people opinion. I have been witness of that attitude for last 25 years. I'm really bored of you guys. After all seems your 'technology' plus 'attitude' sucks! Might be it's time for Linux?!

# Neil said on November 1, 2008 9:31 AM:

This decision is plain crazy.

In EF you have a product that has received a negative response from the developer community (including a lot of MVCs) not seen since .NET launched.

In Linq-to-sql you had a product that was basically liked by the developer community and where a lot of people and shops had invested time in learning the skills to use it and incorporating it into their application design.

As others have said bringing Linq-to-sql up to parity with NHibernate would have been relatively trivial.

What you have done is significantly piss off a large part of the developer community that others in Microsoft try so hard to cultivate. Why will these people possibly trust new technologies in the future, particularly those such as the EF, which is already so incredibly disliked (for very valid reasons).

The only hope MS has to rectify this situation is to release Linq-to-sql out to code plex as other have suggested. I worry that MS will not do this though as there is a very significant chance that were they to the community would take it on and turn it into a real force in the ORM  space, one which would represent yet another nail in the EF coffin.

I feel very vindicated right now in persuading my company to adopt NHibernate, and battling against those who even yesterday were suggesting that the MS solution was the only one that could be trusted to be supported properly over the life of the application.

# Trent said on November 1, 2008 12:28 PM:

I just watched the video where they talk about EF. It does look like they adopted some stuff from linq to sql. They made it easier to get started without needing to edit the mapping. This is good really.

The problem I have is that they did this on top of entities that in my opinion is way behind linq to sql in use of linq, speed and quality of sql that is made at runtime. What is worse, they try to tell us that linq to sql is not dead, but they also told us before that the time to use it was when you had simple mapping and developed with rad style. Now they are trying to add all those same features to entities. I dont get it. They say they support linq to sql but are constantly upselling to linq to entities. How is this? both are free to me. why so much upsell to product that is made worse?

It seems to me that they make linq to sql like candy to lure me to MS and then try to switch me. bait and switch. why do they try to do this? it is all the same. i don't want to switch. why do they cause me so much trouble? it will now cost so much to switch my boss will blow his top.

# Damien Guard said on November 1, 2008 1:20 PM:

@Trent we are doing all this work so that if you start to exceed the LINQ to SQL feature set it won't cost so much to switch and your boss won't blow his top.

[)amien

# BlogServiceHost.Create() said on November 1, 2008 1:28 PM:

Negli ultimi giorni la notizia sono rimbalzate. Forse troppo. Ad una prima lettura non mi era sembrato

# jmorris said on November 1, 2008 7:41 PM:

Wow. After spending much frustration using the EF in a project, I finally realized that it was a shoddy, have baked product and converted over to L2SQL. L2SQL has been a pleasure to work with especially the tool support and extensibility. I guess its time to switch over to NHibernate...

# Caber said on November 1, 2008 11:29 PM:

For those LINQ to SQL fans, take a look at SqlTac(www.SqlTac.com). It extends LINQ to SQL with extension methods. Support for REST, WCF Window Messaging and Obfuscation attribute right out of the box.  Over 25 different options.

# TracyR said on November 2, 2008 7:36 AM:

It appears to me that there is room for both technologies at the current time and into the future. For the developers who develop using sql direct to the database and only for sql server then Linq-to-sql is the obvious choice.

For developers who develop in large IT shops that require stored procs for everything and a much larger base of connections then I believe that the Entity Framework will be best for them.

Fear not that if Microsoft stops development and support for Link-To-SQL in the future I believe there will still be a number of independent developers picking up the mantle and developing L2S into open source. I also have confidence that Microsoft will place this into Open Source if they are not continuing development.

However, as advice to Microsoft it appears that the LinQ-to-SQL technologies more closely utilizes the same predicates as LinQ-To-Entities . I for one will never go back to anything other than the LinQ strategies, predicates and projections I am currently utilizing. Let's face it the comparer class combined with Except, Intersect, Union and Distinct makes the use of a backend database provider agnostic. After all, it is cheaper to have several middleware servers than even one additional SQL Server.

Good luck on EF but I for one will continue using the Join concepts of LinQ in my middleware servers.

# Steve W said on November 2, 2008 7:52 AM:

Linq to Sql is an inefficient, bloated pile of poop. Great great for scipt kiddies and those who can't write their own queries, though. DBA's everywhere will party hardy if it gets dropped.

# Ross said on November 2, 2008 9:48 AM:

This is very disappointing for anyone who has made an investment in learning and utilising LINQ to SQL. LINK to SQL has a major advantage in that it is very lightweight and easily accessible for people who do not need the extra mapping functions of the EF or nHibernate.

# Bill said on November 2, 2008 12:51 PM:

We also have engineered LINQ based solutions for both small and large companies based on LINQ *assuming* MS finally had provided a stable ORM solution.  This is looking to be a wrong assumption!  How can MS pull the rug out from folks who have now make significant investments in this new technology!?  How can MS abandon something that works perfectly fine and after so many years of marketing hype?  Worse yet is to leave LINQ to SQL adopter's hanging without a clear concise indication of future direction.  We've been burned enough in the past with the flip-flops that most of us now anticipate the worst, which is that LINQ to SQL development will stop, and effectively phased out through lack of support in new products and platforms.  Behavior like this is what pushes people away from MS and towards the many alternatives...LAMP here I come.

# From the outside in said on November 2, 2008 2:54 PM:

What do you think the following post from ADO.NET team blogs says? http://blogs.msdn.com/adonet/archive

# Rohland said on November 2, 2008 4:30 PM:

We have been very happy with Linq to Sql and have deployed a few solutions already. It is really disappointing that we don't have a clear roadmap from Microsoft that outlines the future of Linq To SQL. For one, I would like to see providers for other popular database technologies such as mySQL.

If EF is the future then I can only request that Microsoft release a series of practical migration guides that outline best practices for migrating projects that use Linq to SQL.

# Integrated said on November 3, 2008 1:36 AM:

Uh, hello the great thing about Linq to SQL was the fact that it is integrated! Let's not criticize others for there skill level since you were there once, eh?

# NYK said on November 3, 2008 2:14 AM:

Now I have full confidence that my decision to adopt NHibernate for my whole team is a right one.  Other than CLR and ASP.NET, I won't trust any other team under the .Net Framework umbrella.

# Kole said on November 3, 2008 3:32 AM:

I.m dissapointed

If this blog means dead for Linq to Sql, I'm not going to use EF. In addition, in the future I'll not jump into any MS proposed technology, untill it's mature enough and widely accepted

# Sergey Chavlytko said on November 3, 2008 8:49 AM:

After Microsoft killed it’s the best product FoxPro - I was in shock. What can I use the next?

Fortunately, Microsoft introduced LINQ in .NET 3.5. It was a big relief for me and thousands other developers who need simple, fast and reliable solutions for our clients in RAD development...

Now you are killing LINQ... I believe decision already has been made. We can only beg ADO.NET team to do it more "human friendly" and transparent for working with data as much as possible.

Please, do it... I will pray for your success and wisdom...

# Jamie said on November 3, 2008 9:45 AM:

"listening to the community"... so you mean EF will be just like vista? wondeful.

i'm just getting into LINQ, and it blows NHibernate out of the water as far as learning curve is concerned. would hate to have to go back.

# AndrewSeven said on November 3, 2008 6:59 PM:

From the examples I've seen of EF, this lets us go back to writing sql-like statements inside strings instead of writing code.

# AlexLiu said on November 3, 2008 10:50 PM:

几个国外大牛对ADO.NET小组发布文章的评论。和大家分享。

# Dennis van der Stelt said on November 4, 2008 3:47 AM:

A bit late, but it was still in my drafts folder, I just had to finish it. Timothy Mallalieu presented

# Chris said on November 4, 2008 5:05 AM:

I hope the new version of EF supports detached(offline) object states

# naturalight67 said on November 5, 2008 1:18 PM:

Man, I'm still lovin' ado.net 2.0!   I have invested some time and money into Linq2Sql, and I would hate to see it go.  I think I can still take one of my books back to the store.  I feel like I may have burned some money, but I do not feel burned by the ado.net team.

You guys rock.

# MichaelC said on November 5, 2008 3:23 PM:

I am so ticked off by this.

I had to beg, plead, bribe, and grovel to get my organization to adopt LINQ-to-SQL and now Microsoft goes and pulls the rug out from under me.  Hell, I even got them to buy books on the subject for the development team.

What am I supposed to say to my project manager about this?  Hey boss, you know all those man-hours I convinced you to add to the project so the team could master Microsoft's newest, hottest technology?  Well...  turns out it was all just an April fools joke.  Surprise!

Seriously, Microsoft, WHAT... THE... F#@%?!

# Nannette Thacker ShiningStar.net said on November 5, 2008 5:22 PM:

When you opened up your Visual Studio 2008, you may have seen the ASP.net News on the Start Page promoting

# Ivara blogs said on November 6, 2008 1:44 AM:

Vakar ADO.NET komandas blogā atradu interesantu rakstu par LINQ to SQL un LINQ to Entities nākotnes plāniem

# Swiss MSDN Team Blog said on November 6, 2008 10:41 AM:

I always get the same question about the future of LINQ to SQL. Finally the ADO.NET Team, which is supporting

# Charles Miles said on November 6, 2008 12:34 PM:

I really hope you continue to devote resources to Linq to Sql - the posts and comments make it really unclear what 'evolve the product based on feedback' means and make me nervous about its future.

Maybe sometime in the future you can evolve the Entity Framework so that it is as light/simple as Linq to Sql (wow, that sounds like a serious challenge) - but the tough detail is that Linq to Sql is good for lightweight use NOW (generate with SqlMetal, extend with partial classes, query via Linq -> awesome) and is attractive and functional enough it is being used and deployed!

For me Linq to Sql gives a beautiful/light/quick way to get great/easy db access that I would normally hand write sql for and maybe write quick DAL classes for - it is NOT subbing for nHibernate or other complete ORM products. Hard to imageine how ties to the Entity Framework help me quickly make a lightweight DAL for a simple application.

I hope Linq to Sql is evolved, supported and a first class technology going forward.

CM

# David Jade said on November 6, 2008 5:27 PM:

I too am disappointed with this decision as we are heavily invested in LINQ to SQL in such a way that makes transitioning to LINQ to Entities impossible, due to L2E’s lack of flexibility and subpar support for advanced SQL generation. I really hope that you do listen to the community and realize that for many of us LINQ to SQL provides an alternative yet valuable and lightweight way to access SQL data when a full-blown ORM is not desired (or when someone else’s idea of what an ORM should be doesn’t match our needs). In this regard I don’t view LINQ to SQL as a competitor to LINQ to Entities but rather as an alternative for those who want a solution that is lightweight and flexible with superior close to the metal SQL query generation.

While it is clearly necessary to invest in LINQ to Entities in its current state, it seems to me that it is evolving in a different direction than what LINQ to SQL was designed to fulfill. Along the way many design tradeoffs in L2E are being made, some of which are making L2E less flexible and less capable than the current version of LINQ to SQL is.

For starters, I would encourage you to take a long and hard look at what is currently possible with LINQ to SQL that is not possible with LINQ to Entities. Some of this is outlined here: http://mosesofegypt.net/post/LINQ-to-Entities-what-is-not-supported.aspx (especially the current limitations on LINQ query expression evaluations and the boundaries between client-side method invocations and server-side query generation). While workarounds may be possible many of these require trade-offs that the current LINQ to SQL does not require (i.e. more client-side evaluation and less intelligent SQL generation).

I would also encourage you to look at Rob Conery’s work on the MVC Storefront (http://blog.wekeroad.com/mvc-storefront) with LINQ to SQL to see how he is using it to build a data access layer, particularly this post: http://blog.wekeroad.com/mvc-storefront/asp-net-mvc-mvc-storefront-part-2 . His series of posts in particular show just how flexible LINQ to SQL can be.

If L2E is to supplant L2S in the long run then it must step up to the plate and enable the same scenarios and flexibility that LINQ to SQL currently enables, otherwise we as the users of these technologies will be at a net loss in flexibility and design choice. Until L2E can take on these scenarios with the same flexibility I really don’t want to hear about LINQ to SQL being a dead-end technology. That to me sounds simply premature.

David Jade

# Ben Baird said on November 6, 2008 7:43 PM:

I left Microsoft development tools behind starting with the debut of .NET because of this attitude toward developers. Glad to see my decision being reinforced :)

# Luke Breuer said on November 7, 2008 2:08 AM:

I am disappointed with the decision to obsolete Linq to SQL.  I hope you follow up on your promise to "listen to customers" and reconsider the decision to drop this lightweight-but-useful technology based on the massive amount of feedback you are getting.

Microsoft has a propensity to ignore the need for lightweight tooling and technology.  Compare ASP.NET WebForms to ASP.NET MVC.  Microsoft actually recognized the desire for a lighter-weight framework that purposely does _less_ for the programmer.  Less framework tends to mean more flexibility -- this can be a good thing.  Please do not leave developers such as me and all the people above stuck in the mud of overbearing frameworks that enforce constraints that are detrimental to certain use cases.  

# sgvlfogca said on November 7, 2008 10:57 AM:

970UBo  <a href="http://ngbihyiuxzyu.com/">ngbihyiuxzyu</a>, [url=http://rwfdobsnipvf.com/]rwfdobsnipvf[/url], [link=http://wnjusujqtjtw.com/]wnjusujqtjtw[/link], http://ihsiybnbewwr.com/

# ScottK said on November 7, 2008 11:12 AM:

I have found LINQ to SQL to be an incredibly lightweight yet powerful tool.  It is horribly mischaracterized as a RAD tool (even by the ADO team) but used appropriately can be extremely flexible and conducive to developing domain centered applications.  There are people out there who understand the potential of this technology (stand up and take a bow, Ian) and MS should let people like these take the product forward, whether as an MS offering or not.

# Mark Kamoski said on November 7, 2008 3:36 PM:

Is the Entity Framework out of Beta? I just Googled for it and came up with Beta 2, circa 2007. Gah.

# LINQ in Action roller said on November 7, 2008 3:39 PM:

All -- Please help. I keep hearing about how Linq-To-Entities is now preferred to Linq-To-Sql, such as

# Chad# said on November 9, 2008 4:12 PM:

So with the recent news of LINQ to SQL going DOA a lot of those who have developed applications around

# #.think.in said on November 9, 2008 5:33 PM:

#.think.in infoDose #6 (3rd Nov - 8th Nov)

# Chad Moran said on November 9, 2008 11:06 PM:

I recently wrote a article regarding ADO.NET EF v LINQ to SQL which you can find here...

http://weblogs.asp.net/chadmoran/archive/2008/11/09/ado-net-v-linq-to-sql.aspx

Microsoft I think this is a terrible mistake.  I think what you need to do instead of having EF v L2S you should have EF -extend- L2S instead of having it replace.  You're wasting everyone's time, money and hard work with your inconsistencies and constant mind-changing.

Get your act together, and soon.  A lot of people are now paying attention to Microsoft as a platform for development with a lot of the good things you've done lately like having MVC, MEF and ASP.NET AJAX being developed on CodePlex.  Now is the last time you want to mess things up.

# BusinessRx Reading List said on November 10, 2008 9:13 PM:

&lt;p&gt;Tim Mallalieu, PM of LINQ to SQL and LINQ to Entities, recently &lt;a href="http://blogs.msdn.com/adonet/archive/2008/10/29/update-on-linq-to-sql-and-linq-to-entities-roadmap.aspx"

# K. Scott Allen said on November 10, 2008 10:12 PM:

Tim Mallalieu, PM of LINQ to SQL and LINQ to Entities, recently announced: “…as of .NET 4.0 the Entity...

# Peter said on November 11, 2008 3:17 AM:

I can't believe that you are going to drop L2S just like this after 1 year.

It's quick, intuitive, powerful and just working.

Developers invested a lot of time in learning this technology and implementing it in new projects.

I'm strongly convinced that you should invest in this technology, implementing new features in L2S that developers have been asking.

If you confirm that you are going to drop this technology, I suggest everyone just to move to NHibernate, that now supports even Linq To NHibernate, and at least it's open source and can be carried on by the community.

I hope something you might change (again) your mind about L2S.

# Kaveh Shahbazian said on November 12, 2008 2:06 AM:

This behavior of abandoning technologies so carelessly, bring high level of doubtfulness in using the new technologies. For example I have not started to use WF and WCF yet. What if they are disappear in next version? Microsoft always have technology collisions because being property is somehow in conflict with being productive. LINQ to SQL is a very well designed piece and I see why it must be abandoned: It is too high level to be totally dependent on SQL Server!

And is not WF a rival for BizTalk? And WCF a rival for Windows Azureus service oriented design?

This kind of decisions makes .NET development too expensive and  too risky to rely on in enterprise.

I really hope Microsoft "understand" this.

Best Regards

# John Walker said on November 12, 2008 3:47 AM:

Cruel joke, right. Seriously. Punch in the gut. Thanks.

# Ike said on November 12, 2008 10:42 PM:

As was stated above, I believe that EF is a superior product and I am using it successfully in production applications with one exception.  Linq to SQL gives me the ability (whereas Linq to EF does not) to add a table-valued function invocation to a Linq statement and join that table with other tables, etc. in my Linq statement.  This allows me to do something very powerful that I utilize extensively (and that isn't available natively).  I have a set of TVFs that simply convert an XML input into a table of the specified SQL type and, when joined with another data set, allows me to one-up the SQL "IN" functionality.

Long story short...ax Linq to SQL, but please, please, PLEASE don't throw me in that briar patch!  No seriously, please allow me to invoke table-valued functions using Linq to Entities (which I CAN'T DO TODAY!!! @$!#).

M$ rocks...

# ray linn said on November 18, 2008 8:27 PM:

I don't like the Entity framework at all. If I had to make a decision to choose one of the ORMs , why I do not choose Nhibernate over Entity Framework, It is stable, compatible with hibernate.

LINQ to SQL is a baby , i would like to see the improvement on it, instead of a totally new framework. I hope the feature if we could just use the general collection for one-many or many-many mapping...

# Dan Waldner said on November 18, 2008 8:59 PM:

Let me ask you a question - if you guys are not going to continue technologies past the first generation, how the hell are we supposed to figure out which technology to base our decisions on if you're just going to yank them out from under us.

We've developed our newest technologies based on the lightweight Linq to SQL framework and now you're telling us that it's not going to be supported in 4.0?

You have no idea how many people you've succeeded in pissing off.

# .net positiv said on November 24, 2008 10:28 AM:

Microsoft favorisiert Entity Framework vor Linq2Sql

# Zlatan's Blog [MVP SharePoint] said on December 1, 2008 3:03 PM:

Following bunch of announcements about .NET 4.0, ADO.NET team makes a not so surprising announcement

# Mike_BB said on December 3, 2008 5:00 AM:

1 problem - if i have table t1 (id int primary key) and generate it"s pk by sequence, not identity, that attribute use?

StoreGeneratedPattern="Computed" not use in case primary key!!!

2 Why absent  partial void Insert ...How customize insertd, update, delete?

# James said on December 3, 2008 5:48 PM:

I have nothing constructive to add to this other than the fact that you've totally screwed over lots of my projects.....  awesome... thanks Microsoft.

# Pedro Rainho said on December 4, 2008 9:50 AM:

WAR between Entity Framework Vs LINQ to SQL

# Jason Short said on December 4, 2008 10:34 PM:

I think the comments here will bear out that a lot of people never understood LINQ 2 SQL to be a temp solution or a partial implemenation.  

http://visualstudiomagazine.com/features/article.aspx?editorialsid=2583

From that post it sounds like everyone was supposed to know it was a "risky" technology and an early adopter only short term strategy.  

I knew that because I went to the PDC and TechED 2 years ago.  But no one writing those pretty cover articles for Visual Studio Mag or MSDN ever said "don't do this if your project is going to live more than 1 year".  I have been yelling for that entire two years that people needs to look at EF as their long term solution, but no one ever listened and generally gave me a ton of opinions why L2S was "superior"...  

I personally have no problems with EF.  It works about as well as L2S, but lacks the docs or community support.  Maybe that will change now that is a finally THE access method.

# Mike_BB said on December 5, 2008 4:55 AM:

Hi! How include master table in group expression? E.g. in AdventureWorks: vaster table Person.Contact, detail table Sales.SalesOrderHeader. Group

context = new EF.AdventureWorksEntities();

var query = (from p in context.SalesOrderHeaders group p by p.Contact.ContactID into g  select g)

and we write

var result =

               from grp in query

               select new

               {

                   id = grp.Key,

                   Number = grp.Count(),

                    FirstName= (from s in grp

                               where s.Contact.ContactID == grp.Key

                               select s.Contact.FirstName).First<string>()}

In  LINQ OK! But in EF s.Contact == NULL!!!

# Ian Blackburn said on December 7, 2008 1:48 PM:

Working with Data in Silverlight 2 (Entity Framework, Ado.Net Data Services, and DataGrid)

# Bayer White, Connected Systems MVP said on December 7, 2008 4:14 PM:

What Is The Future To "LINQ To SQL"

# Mike_BB said on December 8, 2008 7:33 AM:

How add in ConceptualModels business object (EntitySet) that no direct mapping in database? E.g. business object with grouping. Simple solution - create view in database not use!

I am sorry my bad english!

# Daniel2ThePoint said on December 13, 2008 12:20 PM:

I&#39;m not dead or gone, just had a lot of work to do. Have started to record a lot of screen casts

# henryjwr said on December 16, 2008 2:45 AM:

Busyettt...............bener nih LINQ 2 SQL nggak ada kelajutan Road Mapnya.... gimana nih om MIC &amp;

# Thomas DeMille Interactive said on December 17, 2008 12:11 PM:

considerable issues moving from Linq to SQL to Entity Framework

# VB for Dummies said on December 18, 2008 3:18 PM:

Just finished up a presentation on Visual Studio 2010 for ICC's MS Dev special interest group. A good time was had by all. I used the card deck that is in the Training Kit, and added a few salient slides. We had a lot of coversation about Linq, WPF and

# VB for Dummies said on December 18, 2008 3:32 PM:

Just finished up a presentation on Visual Studio 2010 for ICC's MS Dev special interest group. A good time was had by all. I used the card deck that is in the Training Kit, and added a few salient slides. We had a lot of coversation about Linq, WPF and

# Dave R. said on December 24, 2008 8:58 AM:

It's plain from all the feedback here and elsewhere that many people find LINQ to SQL powerful and simple, whereas the Entity Framework is currently not a plausible alternative.

With this in mind, would you please reconsider? If you obsolete such a promising technology just because it wasn't built by your team, you risk alienating many thousands of early adopters who find that it fits their needs and who were looking forward to future development.

# Moon said on January 1, 2009 2:03 PM:

Linq to Entities doesn't even support the Create Function Import of a stored procedure that returns nothing! If so, where is it?

There are tutorials and videos all over the web for linq to sql but next to nothing on linq to entities.

I am so frustrated with this technology. It has put me weeks behind in development. I rewrote all of my linq to sql over into linq to entities and getting past the stored procedures has been a nightmare. It is impossible!

And I can't get support on it. There's not even an asp.net forum dedicated to linq to entities. Not that anyone ever answers you over there anyway.

Also, try combining ado.net with linq to entities. The changes you make in running a stored procedure are not updated in the entitie model and now you're retrieving old data before the sproc was run.

Also, forget updating your entity model without entirely deleting it and starting over. It simply has so many limitations that it doesn't update the model entirely and only on a few things. If you change your database, just delete the model and start over. So forget any customizations.

Please put some effort into this technology and "fix" it NOW to run with stored procedures. Give us a patch. Were we told it we were just microsoft's free beta testers? Because this technology does not work and forget trying to use it in a real world application. Do I have to wait for 4.0 to actually get a technology that works?

By then, I'll have moved to a working ORM solution and when that happens you won't find me coming back to it.

How can you release a technology as a "working solution" when it doesn't even support something as simple as running ANY type of stored procedure?

# TerryLee said on January 4, 2009 11:47 AM:

原因近日不断看到有关“LINQ已死”,“LINQ玩完了吧”的言论,甚至于更有牛人说出“程序开发最终会回到本机代码上”,暂不说这些言论是否正确,且先来看看各位的惊人言论,下面仅摘录部分:言论1...

# TerryLee said on January 4, 2009 12:31 PM:

近日不断看到有关“LINQ已死”,“LINQ玩完了吧”的言论,甚至于更有牛人说出“程序开发最终会回到本机代码上”,于是便有了本文:

# Allen said on January 5, 2009 6:58 PM:

This is such a horrible idea and it smacks heavily of a new person coming into a position and wanting to create his own 'baby' as it were.  The linq to sql was a really good idea that deserves to be improved on while the entity framework is an overly cumbersome unit of garbage that is of questionable need.  Why not just improve the linq to sql and keep moving it forward instead of yet again creating something new and forcing us all to switch to it?

# mknopf said on January 7, 2009 11:01 AM:

I recently started on a Proof-of-Concept project using ADO.NET Entity Framework. The project had an existing simple data model which had tables, views, stored procedures, and user defined functions.

After many many hours of frustration with using Entity Framework I finally reached a breaking point, where the execution of a simple stored procedure required the installation of an "Extension Library" and a large block of code necessary to execute each Stored Proc. This was and is absurd, Linq to SQL makes this very easy (along with everything else).

I ripped out all the entity framework code and replaced it with Linq to SQL, in a fraction of the time I had accomplished everything i set out to do with the Proof-of-Concept. This proved to me that Entity Framework is not production ready and I will not use it in the future until I see sufficient proof that this is no longer the case.

PLEASE DO NOT KILL LINQ TO SQL!!!! You will be making a HUGE mistake.  

# Al said on January 7, 2009 4:23 PM:

Could somebody translate this blog post into English, please?

# SAL said on January 11, 2009 10:42 PM:

Linqtosql微软不主动更新了,最近比外国的技术博客上看到的。不知道这信息是否真实。但是本人刚刚有了解了一下linqtosql,他对存储过程的支持不太好,不能返回数据集,如:我的一个通用...

# Ben said on January 14, 2009 8:28 AM:

Hibernate and EF are only good for bloated development environments who have no competative advantage in the market. Personally i think productivity in the industry has dived in the last 5-6 years . LINQ2SQL was a nice solution that fitted 90% of the apps that we need to write not the huge monsters that need massive development and support teams for little business gain except to make developers/architects good that they have made the "perfect product".

# TT said on January 14, 2009 8:06 PM:

The lack of response from the ADO.NET team is curious at best... You guys come off as a bunch of aristocrats.

If some new MS products are built on top of entity framework, why aren't the dao layer pluggable so that you could use Linq2sql or NHibernate for that sake. Seems like bad design

# TT said on January 14, 2009 8:06 PM:

The lack of response from the ADO.NET team is curious at best... You guys come off as a bunch of aristocrats.

If some new MS products are built on top of entity framework, why aren't the dao layer pluggable so that you could use Linq2sql or NHibernate for that sake. Seems like bad design

# Bill Blogs in C# said on January 15, 2009 5:13 PM:

I’d hoped to completely avoid any discussion surrounding the future (or futures) of Entity Framework

# Chris Williams said on January 15, 2009 10:10 PM:

Amen to mknopf.

I can't believe that you're putting out any kind of "the future is Entiity Framework" message when EF in it's current incarnation is so brain dead.  Am I just not supposed to develop any sort of data application until you get your sh*t together?

Stored procedures for example!  I can't tell my dba not to use them.

# jay greasley said on January 16, 2009 2:42 AM:

Is this not just a nail in the coffin of closed source software?

# Olivier said on January 25, 2009 12:38 PM:

If LinqToSql was to not be supported any more, would-it be possible to have its source code on CodePlex?

Or at least, to open all this "internal classes" programming model so that third-parties could integrate other vendors databases.

When looking at the way LinqToSql is made, it is obvious that extensibility points are provided. Just make them public!

# vaccac said on January 26, 2009 4:13 PM:

This is going to make so many people switch to NHibernate

# emvik.net said on January 28, 2009 6:42 PM:

ADO.NET Entity Framework OR LINQ to SQL

# emvik.net said on January 28, 2009 6:50 PM:

ADO.NET Entity Framework OR LINQ to SQL

# Himadri Banerjee said on January 29, 2009 3:31 AM:

I think the Project lead of ADO.NET should be fired. Is this a freaking JOKE?? What in the hell are these people thinking? They should give back the L2S to the C# team ( Hats off to them for creating something so wonderful) or the ASP.NET Team. Don't you guys even learn from history? Do you all work for the same damn company? Why kill the most beautiful thing that was created. L2S was simple easy to learn and more people switched to MS platform because of that. This is a crime to tell us now to  switch to EF model...Why would you force us like this? Just because you did not create it? Man you guys should not be in the IT business and go flip burgers and I personally will tell all our channel partners and everyone to switch to Non MS platform if you drop L2S and make us use EF model.

# Russian SQL Server Club said on January 31, 2009 5:38 AM:

То есть это не конец, но внимания ему, видимо, будет уделяться совсем немного.

# Phil Karras said on February 3, 2009 9:55 AM:

What part of,

"We are listening to customers regarding LINQ to SQL and will continue to evolve the product based on

feedback we receive from the community as well."

are we not reading, seeing, understanding? It looks to me that while they are working on "Entity Framework" they will continue to work on LINQ to SQL, though perhaps not as "important" to them as EF.

-pk

# Mike said on February 4, 2009 4:39 PM:

@pk (and anyone else who does not understand coporate speak):

Read the book Weasel Words or Death Sentences, books about communication in corporations. ALL 'official' posts on this issue are filled with these weasel words. The bottom line is: Microsoft WILL NOT DEVELOP LINQ TO SQL.

Sure, they are "fully committed", "listening to feedback", "evaluating the best fit", but NOWHERE does it say that they are going to make many-to-many mapping possible in LINQ to SQL, or opening the LINQ to SQL provider. NOWHERE!

It's not happening folks. So now what? We need to put pressure on Microsoft. They have two choices:

A - start an agile and independent team to develop LINQ to SQL in the same fashion as the ASP.NET MVC project

B - Open source the whole thing on Codeplex, we'll do it ourselves.

But don't count on Microsoft, they decided to f*ck their customers. Thanks for nothing Tim Mallalieu!

# Tom said on February 7, 2009 12:46 AM:

A whole book on corporate-speak?!  Read the reviews on Amazon, and you'll see that the book belongs to the category of: "Here are 3 pages that would make a pretty good magazine article; now let me pad it out to 200 pages so I can put a book on my resume."

If anyone still have trouble believing that LINQ to SQL is dead, I urge you to compare the statement above to the statement that came out when Flight Simulator (Microsoft's longest-running product) was cancelled two weeks ago.  It's seriously uncanny; they even use the word "fully committed."  Laying off the whole team, that's a funny way of showing your commitment to the product ...

# Ike said on February 12, 2009 10:19 AM:

Just wanted to cast my vote for support of table-valued functions in EF v2.  That's the only thing I used LINQ to SQL for and, even today, I would rather that functionality be in EF.  If LINQ to SQL is on it's way out.  I can only assume/hope that MS will move function support to EF.

Thanks in advance...

# Rich said on February 17, 2009 6:22 PM:

At the very least, keep LINQ to SQL in VS2010...maintained or not.  Are you saying that it will be excluded from VS2010?  Or that it will not be further enhanced in VS2010?

KEEP LINQ to SQL!

# Scott said on February 18, 2009 3:16 PM:

When I first started using LINQ to SQL, I thought "FINALLY, Microsoft has built a data access model that does what I need it to do."

And now you want me to use EF instead?  WHY?   EF is harder to work with, a heck of a lot heavier, and it's slower.  And it requires I write code with embedded strings in it?  Are you kidding?  EF is clearly inferior to L2S.  So how is it that EF is now the "preferred model"?

You guys are really, REALLY pissing me off.

# Kevin Pang said on March 21, 2009 6:08 PM:

What ASP.NET MVC Can Learn From Ruby on Rails

# ken said on March 24, 2009 5:06 PM:

MS needs to develop a flexible, living, cross-platform, quick-to-develop Data Access layer.  L2S has it's limitations, and I never found a quick easy way to update it when the data model changed, which it does frequently in a development environment.  Additionally, I find myself using MySQL more and more for web site development and L2MySql would be nice.

Whatever they come up with, they had better do it quick before too many more developers get used to L2S and then feel abandoned when it goes the way of Foxpro, BOB and playsforsure.

# X Chen said on March 25, 2009 9:35 PM:

Everybody likes linq to sql to some degree. If MS fix some of the problems, linq to sql could be the best data access component from microsoft.

Don't let it go. I am not confident EF could be as good.

# Janus007 said on March 31, 2009 3:10 PM:

What exactly has ADO.NET team contributed with in the long run?

It seems like everything they try to invent they invent as complicated and heavy-weight as possible.

Linq to Datasets - what is that? Writing Linq to some legacy old style DataSets - give me one reason? Just keep the DataSets as is to support backward compability. What happened with Linq to Objects instead? Ohh... I forgot, Linq to DataSets should be capable of doing some strange things that was never intended.

ADO.NET DataServices - what is that? Deploy dataservices over the Internet, don't make me laugh *LOL*

What is up with that team?

And now they are fighting Linq To SQL and trying to convince people that L2E is the preferred way - yeah yeah.. still laughing.

Wonder how many have tried the new "“Table Splitting”: Mapping multiple entity types to the same table."

Again I must laugh, why the heck is that such a big issue? It shouldn't be anything to even think about, it should just be there intuitively. Why even think about how the data should be modelled from the relational model to object model, why not just drop all the strange rules and just model the object entities and let the mapping do some magic. I can continue, what about updating and inserting etc. etc. uuuhhh you can't insert an object in a disconnected environment because bla bla... because what? Useless information if you ask me, it should just be something implicit instead of all those strange rules. Hint: If you come back with an object into a disconnected datacontext, then the datacontext should reconize the object - how hard would that be to create?

But nono, instead they try to build strange technologies that very few people need.

And the worst thing, they try to fight L2S before even L2E is mature enough for some real projects.

NB: I could be wrong, I just googled self-tracking entities, they could cover what I'm talking about, but nevertheless it's the whole concept that irritates me *LOL*

# Kenny said on April 10, 2009 3:07 PM:

So, the people pissed that the time and money spent learning L2S -- you people honestly believe 100% of that is gone? Even 75%? Wow. I find it difficult to believe you guys are that ignorant.

You learned about anonymous types. You learned lambda's. You learned how to handle things. Your *code* shouldn't require that much more effort to go to L2E.

Get your panties out of the bunch and calm down. Getting all emotional only makes you look like a kid who can't handle the real world.

L2E is fantastic. I started learning it over L2S. At the moment I /only/ use L2S if I'm writing a throw-away application -- otherwise it's L2E. The skills I learn go back and forth. That simple.

<Arnold> Stop whining </Arnold>

# Kenny said on April 10, 2009 3:07 PM:

So, the people pissed that the time and money spent learning L2S -- you people honestly believe 100% of that is gone? Even 75%? Wow. I find it difficult to believe you guys are that ignorant.

You learned about anonymous types. You learned lambda's. You learned how to handle things. Your *code* shouldn't require that much more effort to go to L2E.

Get your panties out of the bunch and calm down. Getting all emotional only makes you look like a kid who can't handle the real world.

L2E is fantastic. I started learning it over L2S. At the moment I /only/ use L2S if I'm writing a throw-away application -- otherwise it's L2E. The skills I learn go back and forth. That simple.

<Arnold> Stop whining </Arnold>

# Articles said on April 11, 2009 1:22 AM:

Microsoft Kills LINQ to SQL? For the last couple of months, I&#39;ve been hearing all kinds of complains

# Articles said on April 11, 2009 4:27 AM:

For the last couple of months, I&#39;ve been hearing all kinds of complains, cries, nervous breakdowns

# Petr Antos said on May 13, 2009 6:19 AM:

dont worry ... L2S was only lightweight wrapper around relational database approach, allowing only simple automatic persistence of simple inheritance trees; simply because of LINQ - as this has nothing to do particulary with databases only. And L2S has its flaws for multitier usage, and is not intended to be enhanced (as it is impossible to manage for VERY HIGHER level "REAL" ORM). So Entity Framework WILL integrate good things from L2S, reuse its querying capabilities inside (although its query language originaly was not powered by LINQ?) and as making "really good" ORM (allowing to forgot for SQL completelly) is nobel-price task IMHO, it is simply very, very, very much complex to do it well, than simple L2S wrapper - so be patient, and *believe* ...

# VistaDB.Net Blog said on May 16, 2009 2:28 AM:

Linq2SQL's 'big brother' - Entity Framework

# VistaDB.Net Blog said on May 16, 2009 2:34 AM:

Microsoft killing LINQ to SQL?

# medyum said on May 20, 2009 1:25 PM:

Hibernate and EF are only good for bloated development environments who have no competative advantage in the market. Personally i think productivity in the industry has dived in the last 5-6 years . LINQ2SQL was a nice solution that fitted 90% of the apps that we need to write not the huge monsters that need massive development and support teams for little business gain except to make developers/architects good that they have made the "perfect product".

# Medyum said on May 22, 2009 6:36 AM:

You learned about anonymous types. You learned lambda's. You learned how to handle things. Your *code* shouldn't require that much more effort to go to L2E.

Get your panties out of the bunch and calm down. Getting all emotional only makes you look like a kid who can't handle the real world.

# Quinn said on May 22, 2009 11:36 AM:

L2S is a great solution for our needs.  Why not evolve something that works so well and is as popular as L2S?  

# Lioncall said on May 26, 2009 11:52 PM:

I suggest implementing a Repository pattern to simplify transition between NHibernate,L2S and L2E. See http://www.codeplex.com/backgroundmotion    

# PJB said on May 27, 2009 2:13 PM:

This reminds me of the DAO, RDO, ADO debates. Look like MSFT never learns, and we continue to dive head first into all of their offerings.

# XУЛИГAH said on May 30, 2009 9:14 PM:

Информация полезная. Спасибо. Хотелось бы только обновлений почаще

# Вадим said on June 11, 2009 6:13 PM:

У меня очень скоро в bookmarks место закончится, но я буду рад добавлять с вашего блога и далее ссылочки на интересные темы!

# 出会い said on June 11, 2009 9:48 PM:

ヒマだょ…誰かかまってぉ…会って遊んだりできる人募集!とりあえずメール下さい☆ uau-love@docomo.ne.jp

# 小向美奈子 said on June 12, 2009 9:35 PM:

話題の小向美奈子ストリップを隠し撮り!入念なボディチェックをすり抜けて超小型カメラで撮影した神動画がアップ中!期間限定配信の衝撃的映像を見逃すな

# 小向美奈子 said on June 12, 2009 10:04 PM:

話題の小向美奈子ストリップを隠し撮り!入念なボディチェックをすり抜けて超小型カメラで撮影した神動画がアップ中!期間限定配信の衝撃的映像を見逃すな

# 家出掲示板 said on June 13, 2009 9:46 PM:

カワイイ子ほど家出してみたくなるようです。家出掲示板でそのような子と出会ってみませんか?彼女たちは夕食をおごってあげるだけでお礼にHなご奉仕をしてくれちゃったりします

# 右脳左脳 said on June 14, 2009 9:58 PM:

あなたは右脳派?もしくは左脳派?隠されたあなたの性格分析が3分で出来ちゃう診断サイトの決定版!合コンや話のネタにも使える右脳左脳チェッカーを試してみよう

# Bill Sempf said on June 15, 2009 4:04 PM:

My Visual Studio 2010 presentation for ICC

# セレブラブ said on June 15, 2009 9:13 PM:

セレブラブでは性欲のある男性を募集しています。セフレパートナーを探している20代・30代の女性たちが多数登録されています。セレブと遊びたい、Hがしたいという方は無料登録からどうぞ

# 逆援助 said on June 16, 2009 9:16 PM:

セレブ達は一般の人達とは接する機会もなく、その出会う唯一の場所が「逆援助倶楽部」です。 男性はお金、女性はSEXを要求する場合が多いようです。これは女性に圧倒的な財力があるから成り立つことの出来る関係ではないでしょうか?

# 救援部 said on June 17, 2009 9:10 PM:

貴方のオ○ニーライフのお手伝い、救援部でHな見せたがり女性からエロ写メ、ムービーをゲットしよう!近所の女の子なら実際に合ってHな事ができちゃうかも!?夏に向けて開放的になっている女の子と遊んじゃおう

# 出会い said on June 18, 2009 9:48 PM:

まったぁ〜りしたデートがしたいです☆結構いつでもヒマしてます♪ m-g-j@docomo.ne.jp 年齢と名前くらいは入れてくれるとメール返信しやすいかも…

# VS2010学习 said on June 21, 2009 2:42 PM:

A lot of people are confused about whether or not they should use LINQ to SQL , because the word on the

# Денис said on June 23, 2009 10:09 PM:

Можно и поспорить по этому вопросу, ведь только в споре проявляется истина. :)

# sohbeti said on July 1, 2009 4:34 PM:

perfect this blog, kameralı sesli sohbet chat girişi.

# hikaye said on July 4, 2009 3:16 PM:

L2S is a great solution for our needs.  Why not evolve something that works so well and is as popular as L2S?

Leave a Comment

(required) 
(optional)
(required) 
Page view tracker