From “Oslo” to SQL Server Modeling
As I stated in my previous post, we have been on a journey with “Oslo”. At the 2007 SOA/BP conference we announced that “Oslo” was a multiyear, multiproduct effort to simplify the application development lifecycle by enhancing .NET, Visual Studio, Biztalk and SQL Server. At PDC 2008, we announced that various pieces of “Oslo” were being spun off and shipped in the application server (“Dublin”), the cloud (.NET Services), and the .NET Framework (WF/WCF 4.0). We rechristened the ‘Oslo” name for the modeling platform pieces of the overall vision.
In the year since PDC 2008, we delivered three public CTPs and conducted many software design reviews (SDRs) with key customers, partners and analysts. We listened intently to the feedback and it helped us to shape our approach toward bring this technology to market. With PDC now one week away, we are beginning to disclose the next chapter in the journey to “Oslo”, with more to be unveiled at various keynotes and sessions at the PDC event itself.
Of the key things we observed over the last year was the real, tangible customer value in applying “Oslo” to working with SQL Server. Time after time we heard that “M” would make interacting with the database easier, provided we offered a good end to end experience with tools (VS) and frameworks (Entity Framework and Data Services) that developers use today. We heard that developers wanted to use the novel data navigation/editing approach offered by “Quadrant” to access their data in whatever SQL Server they wanted, not just the “Repository”. We heard that the notion of a “Repository” as something other than SQL Server was getting in the way of our conversations with customers.
Another thing we learned was that most of the customers that we wanted to leverage the modeling platform were already using SQL Server as their “repository”. Take an application like SharePoint. It is already model-driven. It already stores its application definition in a database. Dynamics is the same way. Windows Azure is the same way. System Center is the same way. What we didn’t have was a common language, tools or models that spanned all of these applications, although they were all leveraging the same database runtime. The simplest path to get all of these customers sharing a common modeling platform seemed obvious.
Lastly, we learned that the folks on the SQL Server team were hearing the need for additional mechanisms to make the database more approachable to developers. Developers did not want use three different languages to build their database applications (T-SQL, a .NET language and a XML mapping file). Developers wanted new tools that let them deal with the truly massive amount of data they need to handle on a daily basis. Developers wanted to radically simplify their interactions with the database, with a straightforward way of writing down data and getting an application as quickly as possible.
With all of the above in mind, we just announced (at VS Connections) the transition from “Oslo” to SQL Server Modeling. At PDC, we will release a new CTP using this name, SQL Server Modeling CTP, that will begin to demonstrate how developers will use these technologies in concert with things like T-SQL, ADO.NET, ASP.NET and other parts of the .NET Framework to build database applications.
The components of the SQL Server Modeling CTP are:
- “M” is a highly productive, developer friendly, textual language for defining schemas, queries, values, functions and DSLs
- “Quadrant” is a customizable tool for interacting with large datasets stored in SQL Server databases
- “Repository” is a SQL Server role for the the secure sharing of models between applications and systems
We will announce the official names for these components as we land them, but the key thing is that all of these components are now part of SQL Server and will ship with a future release of that product.
At PDC, we will unify the “Oslo” Developer Center and the Data Developer Center. You will be able to find the new SQL Server Modeling CTP at our new home (http://msdn.microsoft.com/data) the first day of PDC. I encourage you to download this CTP and send us your feedback.
If you are attending PDC, we have some great sessions and keynotes that will highlight the work we are doing with SQL Server Modeling. My personal favorite is “Active Directory on SQL Server Modeling” (the actual title is The ‘M’-Based System.Identity Model for Accessing Directory Services), which is going to show how a serious “ISV” is using these technologies.
Speaking for myself and the team, we are very excited about this transition. Many of us have worked on numerous “v1” products while at Microsoft. This sort of transition is exactly what successful “v1” products/technologies undergo based our collective experience. You have a vision based on customer need. You write some code. You get customer feedback. You adjust. You repeat. You find the place that maximizes your investment for customers. You focus like a laser on delivering that customer value. You ship.
Looking forward to the next chapter…
[Updated: follow-up to this post: On DSLs and a few other things...]
[Updated: another follow-up post: On “M”]
I think I’m in two minds about this announcement: on the one hand, I think it brings a lot of clarity to the data modelling concepts of Oslo; but I can’t help thinking that tying M to SQL Server in this way will cripple the potential of the DSL vision: surely that has much wider applicability than is suggested by calling it “SQL Server Modeling”? I was anticipating that PDC would bring some announcements showing how M makes it easy to implement our own mini-languages in C# – and things along those lines.
Samuel Jack
10 Nov 09 at 09:37
[...] DouglasP blog (emphasis [...]
Aali - a developer's view : Oslo -> SQL Server Modeling
10 Nov 09 at 11:50
Disappointed by this news.
Oslo seems to have gone from a potential new Enterprise Architect modeling platform to just a modeling tool and DSL stack for SQL Server. This ignores all those large Enterprise IT departments with heterogeneous data platforms. What about support for logical and physical models? What about using ‘M’ as a modeling tool for XML Schema or augmentation with UML use cases? What about source control integration?
It seems that your ‘key’ customers at the SDRs were also SQL Server customers. Maybe you could have had an SDR for non-Microsoft customers to broaden the reach of the tools. I’m sure these customers wouldn’t have shared the sentiment that supporting other data sources would ‘get in the way’ of the tool.
Joe Wood
10 Nov 09 at 13:18
What about MGrammar? Is there any hope of getting that as a stand alone bit?
This might be a bit tangential but the focus here seems to be on Modeling and moving away from DSLs. How do you see modeling and DSLs in general, relating to each other.
Justin Chase
10 Nov 09 at 16:46
Thanks, Doug. This post clears things up quite a bit.
David R. Albrecht
10 Nov 09 at 17:35
[...] http://www.douglaspurdy.com/2009/11/10/from-oslo-to-sql-server-modeling/ a few seconds ago from Choqok [...]
Scott Banwart (sbanwart) 's status on Tuesday, 10-Nov-09 17:52:57 UTC - Identi.ca
10 Nov 09 at 17:53
Doug, I totally drank the Kool-Aid at your PDC’08 talk. I mean, I truly understand and bought into what you were doing with M and Quadrant, but I really wish Oslo would’ve focused on DSLs and LOP. Right now it really feels like Yet Another Data Access Layer for SQL Server.
As it stands now, I’m hoping the ALT.NET community can put together some sort of solution for DSLs and LOP.
Very disappointed.
Andy S
10 Nov 09 at 18:11
>> Speaking for myself and the team, we are very excited
>> about this transition. Many of us have worked on
>> numerous “v1” products while at Microsoft. This sort
>> of transition is exactly what successful “v1”
>> products/technologies undergo based our collective
>> experience.
Doug,
do you think we swallow this? Do you have any respect for your customers and the people that invest time and time into Microsoft stories? As you pointed out, time and time again, for pretty much any product, Microsoft claims it will deliver X++ and in the end we see Y–.
Why do you even bother telling all that? We know the drill.
JJ-
Jean-Jacques Dubray
10 Nov 09 at 19:13
[...] you want to know more, I suggest you to read Doug Purdy’s blog post. Posted: Tuesday, November 10, 2009 6:34 PM by lukad Filed under: Oslo, Patterns & [...]
The world of software design paradigms, .NET and other fun stuff ... : From the Capital of Norway to SQL Server
10 Nov 09 at 19:17
Very disappointing Douglas!
Microsoft sold us OSLO to be the strong movement in the MDD direction and aligned with the great job done in the MS DSL Tools. MGrammar is a great effort in such direction.
Now OSLO will be another thing totally different: a set of tools highly tied to SQL Server. That is quite different respect to the initial selling proposition we all bought!
To me, and others: modeling means increasing the level of abstraction and to tie it with a database is exactly going in the wrong direction.
Sadly, I will have to change the tools I use.
Pedro J. Molina, PhD
10 Nov 09 at 20:35
I agree with many of the comments. I don’t want another way to get at my data. I want a language, “M”, and tooling to help me define languages and models. I don’t care much for Repository or Quadrant. However, I’m very excited about “M” (MSchema, MGrammer, MGraph).
Honestly, if Microsoft dropped Repository and Quadrant, and shipped an SDK that enabled use of “M”, I’d be very happy. I don’t want any *direct* connection to data access or SQL server. If SQL Server wants to take a dependency on “M” that’s fine – just not the other direction.
Jordan Terrell
10 Nov 09 at 22:12
Whats about the DSL functionality ?
whats about the alingment with MS modeling tools in VS ?
everything is gone and OSLO is now just tied into the SQL world ? no please no….
Bernhard Merkle
10 Nov 09 at 22:24
Any chance we could shoot for something a LOT less complicated and just get a decent ERD tool? And no, Visio doesn’t count, since I it won’t forward/reverse engineer in the stand-alone versions. And no, I DON’T want to install the Enterprise Architect Edition of VS just to make an ERD. How hard is this?
But it’s great that we have ANOTHER database query language. That’s awesome. Personally, the ten I have now just aren’t that useful.
Buck Woody
10 Nov 09 at 22:33
[...] Center. You’ll be able download new SQL Server Modeling CTP here on first day of PDC,” Microsoft. Loading google.load(’search’, ‘1′); google.setOnLoadCallback(function(){ new [...]
“Oslo” now called SQL Server Modeling; Releasing at PDC 2009
10 Nov 09 at 22:36
More tight coupling – I think that I saw that coming already a year ago, so I can’t say that I’m disappointed, but I’m not impressed either.
Another wasted opportunity…
Mark Seemann
10 Nov 09 at 22:38
I just want to second Jordan’s sentiments. While I would always welcome a more simple and novel data access and tooling approach to replace–not add on top of–the others, the real excitement for me was around DSLs…creating little languages that enabled simplified programming and data entry.
Andy Heimdahl
10 Nov 09 at 23:53
I’m disappointed as well.
The concept of DSLs and tools to support the creation and manipulation and integration of DSLs with the .Net framework with the whole dynamic languages push is just a much more compelling notion than a watered down M resting on SQL Server.
I agree, there should be more tooling around SQL – look at the the sheer number of 3rd party tools successfully selling ETLs and SQL Migration management tools and ERD toos and it’s true that SQL Management Studio should be providing more out of the box when it comes to this.
But this should be a separate effort. The bigger picture here should be the focus on MDD and DSLs and combining those two so that if I wanted to create the next great language – or even the next small language – there would be tools to support me and allow me to use all the existing power of the .Net framework.
We don’t need the Purdy working on stuff that’s already flooding the market.. we want him to be working on the cool and exciting stuff… I’m hoping I’m since this is a knee-jerk reaction to a vision with details I haven’t seen… but this seems like a huge change of direction from what looked to be a very promising destination.
Sean McLellan
11 Nov 09 at 02:48
I cannot understand this decision. Oslo was the first resonable attempt from MS to go in the right direction in real DSL modeling. It is the nature of DSLs that they are developed domain specific, but the DSL tools have to be usable for any domain. Binding DSL modeling to a database is a very bad descision. I’m wondering to which customers you talked. They all might be SQL Server users, but there are lots of users out there who wanted or did apply Oslo technology for far more. From the comments above you see what they think.
So please do again “You get customer feedback. You adjust. You repeat.”
Karsten Thoms
11 Nov 09 at 06:22
I can actually harldy think of any user base for the Quadrant tools, at least as I have seen them so far. For non-developers way to geeky, for developers way to user-friendly. Never understood what the target of that tool was.
I totally agree with everyone here. The proper move would have been to have a very simple DSL story ship as part of .Net/VS. And have the Quadrant/Repository stuff either just go or move that into SQL, if necessary.
davidacoder
11 Nov 09 at 08:57
Oh, and one final comment, after that I’ll be silent
Here is my wild-ass theory why this ended up in SQL-land. This was a lemon problem. Who wanted the Oslo stuff? This was all way too abstract, so the groups that deliver stuff that customers actually want on time all said no. That left the SQL group. They have a VERY LONG story of failed grand projects, so this fits just in. WinFS, entity framework (although they seem to slowly turn around, but remember all the delays, uncertainties etc at the beginning) and now Oslo, which actually fits PERFECTLY into that series.
I’m SURE you guys tried to make this a part of .Net, but the powers didn’t want the “big vision/we change everything/we have a schedule that runs into the next millenium” stuff.
davidacoder
11 Nov 09 at 09:01
“This sort of transition is exactly what successful “v1” products/technologies undergo based our collective experience”
First it related to SOA, problem and technical domain models, it sounded like it had something to do with replacing XML, DSLs, 10x improvement. Now its about not needing to learn 3 languages and making it easier to pump out CRUD apps. Not that we didn’t see it coming, but I’m doubting you are serious when you say you are “very excited about this transition”.
I should however say that having watched MS “transition” many promising ideas I am not in any way surprised (as JJ says Y-). Its just more framework that seems totally irrelevant to the day to day job of writing high quality enterprise software.
ASP.NET MVC is a different story of course, but MS were able to stand on the shoulders of giants with it and in any case it was handled perfectly by the people involved. Oslo, not so much.
Colin Jack
11 Nov 09 at 11:31
[...] not sure what to think of the move announced by Doug Purdy. It’s not a complete surprise since the team recently moved more [...]
“Oslo” renames to SQL Server Modeling - Gerben van Loon
11 Nov 09 at 12:06
[...] Conference (PDC) at http://msdn.microsoft.com/data. Douglas Purdy, a Microsoft software engineer announced the good news on his blog. There are three important functions of SQL Server Modeling, and Purdy provides us a glimpse of [...]
DatabaseProNews » Blog Archive » Microsoft Will Demonstrate SQL Server Modeling At PDC 2009
11 Nov 09 at 13:33
For anyone that was looking to Oslo as a new set of tools for Model Driven Development (like me), I can only recommend that you give Irony (http://irony.codeplex.com) a go. It’s my favourite new DSL-builder for .net. Then there’s always boo if you want the advanced stuff: http://www.manning.com/rahien/
Maybe in 5 years we’ll have built enough awesomeness with these OSS tools for Microsoft to have some giant shoulders to stand on!
Rob Fonseca-Ensor
11 Nov 09 at 13:47
Blogged: “M” and Oslo’s Future – http://blog.jordanterrell.com/post/m-and-oslos-future.aspx
Jordan Terrell
11 Nov 09 at 17:38
Blogged: “M” and Oslo’s Future – http://blog.jordanterrell.com/post/m-and-oslos-future.aspx
Jordan Terrell
11 Nov 09 at 17:39
So, the response to this decision is almost universally negative. The question is… does Microsoft actually care what anyone has to say? Complain all you want, but I doubt it will make one iota of difference with what happens. The bottom line seems to be that the team failed to deliver something compelling enough to sell on it’s own, so the work is being plundered by the SQL team in an attempt to sell more SQL licenses.
Jesse Ezell
11 Nov 09 at 18:53
if you have ever tried to ship a big v1 at a big software company, you know what this transition is and what it is like — and that this is a _very_ positive step for customers and the team.
in retrospect, were trying hyper-transparent with this project from the beginning (even before folks like myself, Don, Anderson, Sells, etc. joined it) and i certainly learned the downside of that approach.
we do welcome the feedback and i appreciate everyone that took the time to let their thoughts be known.
i know that a number of comments wonder how we think about DSLs now and i’ll post something about that soon. short story is that it is still there; safe and sound.
that all said, we are _very_ excited for this technology to land and ship as part of SQL Server.
i was just talking to don about it — and i haven’t been happier about our real ability to impact customers lives so directly in a long time.
The great irony to all these comments is that all we did was change the name from “Oslo” to a SQL Server Modeling and now we get the #fail tag.
If we had called it Windows Modeling or .NET Modeling would it have been #success?
douglasp
11 Nov 09 at 21:32
“in retrospect, were trying hyper-transparent with this project from the beginning (even before folks like myself, Don, Anderson, Sells, etc. joined it) and i certainly learned the downside of that approach.”
I wondered how long till someone said that and I disagree.
Look at the MVC guys, transparency in action. Thing is they didn’t hype it as much, they just got on with solving a problem for people, delivering regular improvements and listening to feedback.
Put it this way, if you’d said nothing last PDC and announced SQL Server Modeling now I’m pretty sure most people wouldn’t have even been interested enough to comment on it. However having announced it in such a blaze of glory, making big big promises about your work in a really interesting area and having everyone commenting on it right up to Fowler himself, its no surprise that your getting a bit more of a reaction.
I thus don’t think its that you shouldn’t be transparent, instead I think its that you should try to avoid the hype and should try to really ensure you get accross the message about what problems you are trying to solve.
“The great irony to all these comments is that all we did was change the name from “Oslo” to a SQL Server Modeling and now we get the #fail tag.”
I think in general the reaction isn’t to the rename itself, if SQL Server Modeling fits then its as good a name as any. Whether it should fit is the real question though.
Interested to hear your views on how the DSL side still fits in though, I’m hoping the discussion covers supporting interesting behavior (rules/policy/workflow) not just on the data side of it though.
Colin Jack
11 Nov 09 at 22:37
Doug,
do you even read what you write? Is that changing the “name”?
>> Time after time we heard that “M” would make
>> interacting with the database easier, provided we
>> offered a good end to end experience with tools
>> (VS) and frameworks (Entity Framework and Data
>> Services) that developers use today. We heard that
>> developers wanted to use the novel data
>> navigation/editing approach offered by “Quadrant”
>> to access their data in whatever SQL Server they
>> wanted, not just the “Repository”.
I would call that changing the direction, but well, it’s clear you simply have no clue about this whole thing, maybe for you it’s all the same, as long as it ships, it does not matter what it is.
Jean-Jacques Dubray
11 Nov 09 at 23:23
I think the reason behind of the transition is really Microsoft internal issues. “Oslo” is a good conceptual and innovative product, but hardly generate any direct revenue for a while. As a result, there is no other product groups really want it, then it ends up in SQL server dumpster.
This singals a far away path for Microsoft to be really support DSL for production. Fortunately, I don’t put much effort to look into it. And from now on, my technology pick up for the DSL direction will have to be Scala and Ruby based lanaguages.
Xin Zhang
11 Nov 09 at 23:46
@jdubray:
I appreciate the professionalism of your comment.
As much as I would love to engage in an ad hominem discussion, I’ll refrain.
I was making an observation about what we heard from customers.
Each of the things I mentioned were already parts of “Oslo” — we had a VS story for M and Quadrant worked over SQL databases, etc.
We heard that customers wanted to use these particular features to help them with real world problems that they faced building applications today.
Modulo some new features that expand the capabilities of “Oslo” based on this feedback, the CTP is largely a rename.
douglasp
12 Nov 09 at 00:09
@colinjack
I don’t think MVC is even close to a similar comp, but I hear you.
Many of us working on the project get very excited at the possibilities (and still are).
We think that landing this project in SQL Server is _exactly_ the right place for it.
I think that there is a serious mismatch on how developers think about SQL Server that we need to overcome, but from a customer value and investment standpoint — this is exactly the right place for any of this to see the light of day.
Lastly and as I said in a previous comment, modulo some improvements that we made based on customer feedback, the CTP is largely a rename.
douglasp
12 Nov 09 at 00:18
So, after two years of hype and soul searching, Oslo finally found its identity (and the problem it is solving) – yet another object-relational mapping.
Great effort and a true success story !
G.Mladenov
12 Nov 09 at 00:25
[...] thing happened after Douglas Purdy blogged about renaming From “Oslo” to SQL Server Modeling. Comments and insight began streaming through twitter. Later that evening David Ing suggested [...]
Microsoft Renaming Oslo Flows Through the Social Networks
12 Nov 09 at 00:55
Everyone says it is now fail..
Everything is suggesting instead of Creating DSL which are stored in a repsoitry and can be compiled to binary/CIL and used from a development environment ( which is what people are interested in ) to another LINQ/SQL data lib ( which is like ho hum no matter how good) .
Is it just a name change ? We are all confused ,to this end the name change is a failure. We need more explanations.
The good thing about an open plan is you get community feedback now, if this was the release you would have few customers. Maybe a lesson for Midori.
Ben
12 Nov 09 at 01:01
Oslo had a very high potential for being used in non-data applications (see behavioral DSLs). Making it dependent on SQL Server kills that potential.
I was very happy in previous CTPs to see that M was usable fully independent of SQL Server, and I hope it stays that way.
Application behavior source code does not belong in a database, and requiring a “deploy to DB” to run an application would be a major annoyance.
Christos Karras
12 Nov 09 at 01:55
[...] initial post on the retiring of the “Oslo” project name drew a lot of attention and [...]
On DSLs and a few other things… at Douglas Purdy
12 Nov 09 at 06:14
@douglasp
Not saying they’re equiv, just saying that maybe less hype is the key to making transparency work.
Really glad you guys are happy with the way its going, from the outside it does seem like politics/internal organization is stopping MS from delivering some great software but I hope you guys do manage to persevere and produce something great inside the your new home.
Colin Jack
12 Nov 09 at 09:13
@davidacoder what software are we not delivering? all the components that we released as part of “Oslo” are still there. is this about the fact that it is called “SQL Server”?
douglasp
12 Nov 09 at 09:36
This is really lame. The original vision of Oslo was the right one – it was bold and visionary and could have been going far – especially since it can really play well with .Net framework.
Now this is just a lame modeling tool for SQL Server.
Dody Gunawinata
12 Nov 09 at 09:42
It seemed to me that the Oslo team wants to build Oslo together with the community. I think someone should have asked the community first what they think about such major change, instead of enthusiastically announcing it after a year.
Ales Sturala
12 Nov 09 at 10:08
We certainly want to build the “M” language with the community. In addition, we want to move the idea of modeling forward with the community. We think this is positive step for these set of technologies toward those goals.
douglasp
12 Nov 09 at 10:15
I see, I’m late on this…
I’m also disappointed. The “SQL” in it makes a large part of the community, namely developers, turn away.
And that is what the many have told you all the way. I don’t know who you listened to (apart from internal customers as system center, axapta, e.g.). This is the opposite of what the community interested in modelling was hoping for.
— Reading your follow-up now.
Lars Corneliussen
12 Nov 09 at 12:13
Maybe your customers just don’t use twitter, blogs, the web, …
Can’t see anybody who’s happy over this move there, at least: http://twitter.com/#search?q=sql%20server%20modelling
Lars Corneliussen
12 Nov 09 at 13:11
Doug,
I’d like to place your “renaming” announcement in context of the track record of the CSD for the last decade.
I came to know the CSD back in 1999 when BizTalk was announced. Let’s look at what you guys shipped since then and let’s discuss who is “unprofessional”:
- BizTalk: Announced 1999 -> could not be used for anything until v2004 (5 years)
- Jupiter: killed
- EDRA: killed
- SOAP Toolkit, ASMX, WSE, : announced as short lived, transition technologies to WCF
- WCF: announced in 2003 -> shipped in 2007, in 2009 customers claim “no gain, just pain” over ASMX
- WF: supposed to revolutionize BPM, nowhere to be found
- Oslo: announced in fall 2007, 2 years later, no bits shipped yet, many “remodeling” and “renaming” activity, who knows what ships and what doesn’t, who knows which direction each piece is taking
All these “products” are heavily patched by P&P and MCS because they can’t do much on their own. There is no ecosystem that grew on top of them (how could it grow?)
In the standards area, not much to report either. WS-* was launched to support BizTalk, you guys “had to kill” ebXML in order to gain marketshare. We have seen how successful that has been. The CSD couldn’t even deliver a WS-Addressing spec that made sense. It took 8 embarrassing years to complete the WS-*, just because your products couldn’t follow.
In the MDE space, DSL Tools and Software Factories have been a painful embarrassment as well.
In 2009, the CSD was folded in the SQL Server division, after such a track record, this should not be a surprise to anyone.
I’ll let your readers find “professional” qualifiers.
Congratulations on a Job Well Done. Mission Accomplished !
Jean-Jacques Dubray
12 Nov 09 at 16:11
@jdubray: i’ll approve the post, but i want to be very clear to readers that you have very bad data and it is clear (at least to me) that there is some deeper motivation at work here.
douglasp
12 Nov 09 at 19:23
I have no motivation. I am just trying to play back objectively what an outsider can observe coming out of the CSD. Countless time I had to deal with Microsoft sales people trying to sell me the future of X, Y or Z. I am just reflecting as a customer what I see.
I’d be happy to be corrected.
Jean-Jacques Dubray
12 Nov 09 at 19:41
I could spend the entire day correcting your timeline.
To start, there was no CSD in 1999.
None of the people that ran CSD were even remotely close to Biztalk in 1999.
You fail to mention that the people than ran CSD delivered VB, VBA, J++, C#, VS.NET, .NET Framework, COM, COM+, MTS, the list goes on and on…
Your timeline goes downhill from there.
I understand frustration about the difference between technology vision and delivery, but that rarely has anything to do with orgs/people within company being incompetent (as you imply).
I am still waiting on Smalltalk to have an impact (I have been trying to build it for 10 years, so I am at least trying to address this problem).
I still have my Newton.
But this is a general problem of technology and the nature of the business.
douglasp
12 Nov 09 at 19:53
I had not asserted there was such a rift between the BTS team and the “CSD”, that’s interesting.
Look, Doug, at the end of the day, I think you get the point, all I am trying to say is that announcing things 2-4 years ahead and delivering little or changing direction mid-flight is not necessarily something that works well with customers. Maybe you guys could give just a tiny bit of consideration to that aspect of your “go-to-market” strategy.
I know of 2 companies that could have used Oslo that delayed their MDE effort to see what will come out from your team. How many more have been in that situation? how many more did you stand up? Do you realize how much damage you inflict on them in terms of time to market whether it is for SOA, BPM or now MDE? (I don’t think you do because of the kind of trumpeting post that you write which is in sharp contrast with how high you hang your customer base)
Jean-Jacques Dubray
12 Nov 09 at 20:33
I’ll reply to myself.
I’ll add that I do think that we could learn to be much more pragmatic in how we talk about these things to customers.
The number of stars that have to align in get the full breadth of a vision implemented in v1 rarely happens.
That doesn’t mean we should stop, but it does mean that we should be very, very clear that it is a vision and there are going to be many bumps in the road, disappointments, revectoring, etc.
That all said, and to unwind completely, the retiring on the “Oslo” name is good thing from my standpoint and it represents us getting real about delivering this technology to customers.
douglasp
12 Nov 09 at 20:38
@jdubray
This feedback I completely get.
We need to do better.
Could you send me a note offline about the 2 companies you refer to, I can make sure we do a call down to let them know what we are doing (douglasp@microsoft.com)?
Thanks.
douglasp
12 Nov 09 at 20:43
I think this is a good move – it’s time to turn the ideas of Oslo into something real. SQL Server seems like a good home.
Maybe I’m biased, because I only develop large workflow driven information systems. For me, data modelling almost always leads to storage in SQL Server.
Right now I have to build my own tool chain. Which is fun but keeps me away from building actual customer workflows. So I’m behind any effort to make a better story for us developers.
Callan Hodgskin
13 Nov 09 at 03:51
>> I’ll add that I do think that we could learn to
>> be much more pragmatic in how we talk about these
>> things to customers.
Go back and watch the videos of you talking at PDC 2008 and ask yourself… what message should I have been trying to send.
digory
13 Nov 09 at 04:06
[...] From "Oslo" to SQL Server Modeling – Douglas Purdy talks about where Oslo came from, and what its future is looking like for the project, in the form of SQL Server Modeling, and that they will now ship with SQL Server [...]
Reflective Perspective - Chris Alcock » The Morning Brew #476
13 Nov 09 at 08:33
SQL Server Modeling
Hmm. . . SSM acronym is already well known for
Same Sex Marrige
Any chance we can roll back to Oslo? Windows 7 kept the development name, why not do the same?
Alex
13 Nov 09 at 10:50
Doug – you’ve mentioned a few times how you think SQL Server is misunderstood and isn’t just a database engine. Could you provide some examples, or maybe point to some resources that explain what you mean?
For example, how can I use M to model SOA and synchronize the model with code in a source repository – if the model *has* to be in the database. Are you assuming the checkout is directly to a local SQL instance, or even SQL CE?
If this is out of scope for V1 in SQL Modeling then what is the plan going forward? Would you consider a separate sub-product to address the concerns of the commentors on this post? Maybe use the community in the same way as MVC has done? I would be interested in a FrontLine engagement if you feel there are still misunderstandings of the requirements or the products.
Joe Wood
13 Nov 09 at 21:25
@douglasp
>retiring on the “Oslo” name is good thing from my
I don’t think that many developers here care that much about renaming only (although I think – Server Modeling or .NET Modeling without SQL would be better)
I think most of them (I too) care about:
a) Bundeling – if I use M-technology do I have to
ship SQL Server with my software ?
b) The announced CTP has as feature:
M – textual language ….. for SQL server
-> does that include or exclude the usage
of M as DSL tool for other areas but SQL ?
c) I think most of the developers assumed
Oslo technology would be decoupled from
SQL generally, but would also ship with
a layer which would allow a tight coupling
with SQL server. Why directly bind the
whole thing to SQL server directly.
To sum it up – I think (thought?) Oslo has (had?) a huge potential to leverage DSL/MDD technology in many areas of software development – including SQL server.
Now its potential seems to be restricted to SQL server only. Please say I’m totally wrong thinking that !
Andre
13 Nov 09 at 21:45
Rather than put everything under the SQL server nest, or BizTalk (later Dublin), would Microsoft consider creating a few products to accomplish all sorts of aspects in Enterprise SOA, ESB, BPM, complex event processing etc.
IBM and Tibco both have many products that specialized in something and can be used together to achieve a complicated enterprise solution.
Nowadays, Microsoft seems to be trying to create the SILVER bullet solution under SQL server for all. Very likely then, it may ends up do nothing well.
Xin Zhang
14 Nov 09 at 02:12
[...] couple of things related to “M” and respond to some questions that were posed in response to my From “Oslo” to SQL Server Modeling and On DSLs and a few other things… [...]
On “M” at Douglas Purdy
14 Nov 09 at 02:13
[...] http://www.douglaspurdy.com/2009/11/10/from-oslo-to-sql-server-modeling/ [...]
“M” & OSLO diventano SQL Server Modeling - Impedance Mismatch
16 Nov 09 at 17:11
[...] More information:http://www.douglaspurdy.com/2009/11/10/from-oslo-to-sql-server-modeling/http://social.msdn.microsoft.com/Forums/en-US/oslo/threads/ [...]
SQL Server Modeling - the new name for Oslo - Johan Åhlén
17 Nov 09 at 15:07
The obvious conclusion is that Oslo will be tied to SQL Server.
For those interested in a model-driven aproach that would be more loosely coupled with implementation layers, one can have a look at CodeFluent.
By the way, we will support the different Oslo pieces on their applicable layer once they are available on the market.
Regards,
Daniel COHEN-ZARDI
Daniel COHEN-ZARDI
18 Nov 09 at 15:28
[...] Doug Purdy’s blog (owner of Modeling) and On DSLs and a few other [...]
3 Screens and The Cloud – PDC09
18 Nov 09 at 17:09
Wow, didn’t see that one coming. I thought it was a general purpose model language. My take away from previous sessions was that you could use it anywhere. Tying it into SQL Server seems like a very limiting scope to put all that power.
Jason Short
19 Nov 09 at 15:52
As a former MS partner, now customer (startup CEO), I am disappointed. Aside from substance of the move (which I let more technically inclined folks comment on), the name change sends the wrong message to MS customers, especially Enterprise customers.
Working with the CSD folks over the years, I really thought Oslo was the ONE for the Enterprise market and get it right. By right I mean really go after the platform, lock in the stack, have a compelling, integreated value proposition, be a leader and leverage all the great assets already in place (yes, I think BizTalk Server is great – great product, great value).
If Microsoft loses its foothold in the Enterprise over the years, we may well look back at this moment and say, ‘oh, this is when MS really missed the boat. This it could have been Microsoft’s big bet for the Enterprise, finally stepping out of the shadows of IBM, Oracle and SAP, putting it together beyond Windows, Office and SQL Server.’
Yeah, I know, it’s not 1995 anymore, but MS has to bet on something big for the Enterprise sooner or later. I’m sure the investment numbers are impressive in a nubmer of areas, including SSM. But what’s gone now is the inspiration of the Oslo vision. It made me belive that MS really gets it: how work is changing, how business is changing and how technology enabling these trends is evolving. I saw Oslo as all that. SSM just doesn’t do it for me. Sorry folks!
But then again, the underlying issue probably needs to addressed way beyond the great folks in the SQL Server and Oslo teams – this is steveb deciding between pleasing Wall Street next quarter or betting the company on something tha could be great. I’m afraid its the former. As a long term investor, I wish he’d pick the latter.
Michael Klotz
20 Nov 09 at 04:53
[...] project was put together by a friend of mine Dana Kaufman, who works closely with the M/Oslo (now SQL Server Modeling) group. One of the black eyes of the Routing Service that I’ll get into at some point [...]
PDC Traffic | PHP Scribe
20 Nov 09 at 20:35
this seems like a “you can’t handle the truth” situation. THe main point is that Microsoft – trhough CSD or whatever name it had before is full of pseudo gurus, “architects”, etc telling customer that they know what is better for you and what innovation looks like, when in reality they have no clue whatsoever. They try to look like they know what they are doing (usualy projecting the usual microsoft arrogant attitude), but they are really lost trying to “innovate” and convince people MS is still a “great” technology company. When you “squeeze” what is really comming out of these “innovations” you will find little to nothing to get “excited”, but they have to justify their big salaries, so relax and drink the coolaid.
same goes for entity framework, mobile fiascoS(emphasis on plural), WF “re-write”, etc.
Customers are the ultimate victims. I – for one, am sick and tired of these crappy, “upbeat” PR pieces. Get real.
skeptical
28 Nov 09 at 03:42
[...] Conference (PDC) at http://msdn.microsoft.com/data. Douglas Purdy, a Microsoft software engineer announced the good news on his blog. There are three important functions of SQL Server Modeling, and Purdy provides us a glimpse of [...]
Microsoft Will Demonstrate SQL Server Modeling At PDC 2009 | DatabaseProNews
17 Dec 09 at 19:55
I predict that Oslo or SQL Server Modelling will be dropped within 2 years.
As for this being a tool to unify SQL, BizTalk, etc I’ve seen no evidence of this. Many people don’t even see a practical use of the tool.
It was a silly idea in the first place and just gave Don Box something to play with. Why don’t let just let Don go out to passure, COM+ was frankly your best work.. just leave it now.
Bob
8 Jan 10 at 15:39
[...] engineering vakgebied. Hoewel Microsoft het hiervoor ontwikkelde platform Oslo min of meer in de ijskast heeft geplaatst, lijkt de Java community met Eclipse serieuze stappen te nemen om een volwassen [...]
Duchess | Women in Java Technology
17 Jan 10 at 14:09
[...] SQL Server Modeling, and the language M is also available to define other data structures. See Doug Purdy’s blog post for more [...]
Heikniemi Hardcoded » “Denali” to finally ship Oslo?
15 Feb 10 at 15:12
[...] have missed it. Oslo , the codename for Microsoft’s new modeling platform, has been renamed to SQL Server Modeling. Confirms this that the focus will shift to SQL Server instead of creating a modeling platform [...]
Antoine Savelkoul |
26 Aug 10 at 10:20