Bill Sempf

I had a good time with the Columbus Architecture Group (ColArc) Tuesday night at the ICC conference center.  I gave the Economics of Cloud Computing talk there and it was well received.

I got some great commentary from Mark Freeman about the impact that CompuServe had on early internetworking, which is a very good point.  CompuServe was born out of TSO, with a large organization reselling unused computer time.  This is very similar to the IBM TSO concept, and what Google, Microsoft, Amazon and the other large players are doing now.

Another point was the impact of grid computing, which I need to research a little more.

One of the big impacts was security, though.  How is cloud going to interact with HIPPA?  how do you convince a CIO?  What else has to happen to prepare your application for the insecurity of the cloud?

Location is a problem too. How about a state’s requirement to keep all data inside its borders?  There are tough questions there!

Anyway, thanks for having me folks, and I hope to see you next month.

You’ve been asked to make sure that the Client Search screen will stand up under load, because it will be most used screen in the application.  You set up a test user, and then run the Web Performance Test wizard in visual Studio 2010 to record the test.  You make a new test project, save it in TFS and add a new Web Performance Test.  the browser launches, and you log in as your test user.  You do a few client searches representative of the use of the system and log off.

Next, you create a Load Test.  The wizard launches, and you prescribe a 50 user test over a half hour.  You save the test, launch it and go to lunch.

When you come back, there are 23,856 errors.

What happened?  Oh, that’s right, one user can’t log into this system more than once – it was an early requirement.  Oh!  How am I going to do this then?  Do I have to record 50 Web Performance Tests? No.  You can parameterize the login.

Making a data source

Start with a CSV file of usernames and passwords.  You can make it in NotePad or Excel.

Next, we will need a datasource that points here.  Open your webest and click on the Add Data Source icon up in the test’s button bar.

You can select CSV file as the source of the data.

Pick the excel file you created and you’ll see the sample data.

The new data source will show up as one of the data sources for this test.  Probably shoulda named it something better, huh?

Binding the fields

The next major step is to bind the data to the fiends in question.  This is insanely easy.  find the step where you enter the data.  In my case, it was the Login.aspx page.  Open the Form Post Parameters folder and find the parameters for the login and password.  In this app, they are pretty easy to find.

Then follow these steps to get to the next image:

Open the properties panel with F4

Click on the fiend in your application that has the user name.  Mine is txtUser_Name

Click on the Value parameter in the properties panel.

Click the dropdown arrow.

Open the data source in the treeview.

Open the testdata table (or whatever you named it)

Click on UserName and there you go.  Teat field is now bound to the value.

This will work for any form

Remember, this isn’t just for login.  Actually, I am making a mashup of performance testing and training, adding the training data to the system using the performance test for the New Item pages.  Load testing isn’t just for performance anymore!

I have been shooting for one talk a month, and things are looking great for that goal these days.

• In January, I gave my Economics of Cloud Computing talk at CodeMash.
• This month I gave Deviant Ollem’s Lockpicking talk,
• In March I’ll be giving the Economics of Cloud Computing talk to ColArc,
• In April I’ll be giving my new Software Modeling with ASCII talk to the IEEE and ACM.
• I have a talk lined up for May, I think.
• I have talks submitted for TechEd, DevLink and CodeStock so that should give me lots to do for the summer if anyone accepts me.

Then I just have the fall to worry about.  Pretty good year so far!

Being in the design phase of a large-scale project for the State of Ohio, I have been giving a lot of thought to the problems of Modeling and Design.

You see, I am in an interesting position on most projects – I am an actual Solutions Architect.  In today’s IT environment the term ‘architect’ has grown to mean “good coder with big mouth” and I don’t really fit that mode.  I am not the best coder in the room most of the time – although the size of my mouth could be debated.

What I excel at is the designing of solutions, which leads me to use the term ‘software designer’ to describe myself.  This isn’t terribly good either, as it leads people to believe that I produce pretty graphic design and UX, which isn’t my specialty either.

What I am, in truth, is a Software Modeler.  I am the person that figures out what the domain model looks like based on the requirements of the system and the realities of the project.  The one who stands at the end of the room and scribbles boxes on the whiteboard?  Yeah, that’s me.

Thinking about this has really helped me to hone down the definition of SQL Modeling in my head.  People ask me “Why is Oslo hooked to SQL Server?  Is it just another data language?  Don’t we have enough of these?  I thought M was a language not a database thingie!  Oh woe is me!”  Well, the key is in the difference between design and modeling, and how tightly coupled the model and the database really are.

Software modeling begins and ends with data.  The user interface is, I hate to point out, now the purview of the business analyst.  When I get a functional specification for a project, it has the UI in it right there.  I don’t have to ‘design’ that any more.  All I have to figure out is how to do it.  This effort is modeling, and it is about data.

Today I needed to estimate an inventory system with a few quirks.  Without thinking, I got out my notepad, sketched an ER diagram, and worked from that.  How many POCOs?  How many adapters?  How big is the entity model?  Yes, eventually I had to estimate how long it would take me to build the screens, but the bulk of the work was in the domain model, where the work is done, as it should be.

Imagine if I could have instead described my model in text, without all of the brittleness that the ER diagram entails, and have that text description emit both the databases and the POCOs for me?  And then, if something changes, I could change my scratch model and my other code changes?  What if it didn’t break my changes or additions?  What if it didn’t create scores of stored procedures that are impossible to manage? 

Wouldn’t that be nice.

This, though, is the goal of SQL Modeling.  It is clear as day exactly where it fits in the product lifecycle – right there overtop of my fancy notebook scrawls.  Those boxes and chicken feet have a place somewhere, but it’s time for a modeling solution that actually is a solution.

As part of a project I am working on, I need to create an appointment calendar with appointments that can be tagged with and one (or more) of four tag groups.  The appointment viewer can then be filtered to see only appointments tagged with selected properties.

The tags are stored in a generic pot with all of the tags, and then are themselves tagged by Type. 
(Can amyone say MetaInformation.  Yeah, there’s an app for that.)  The tags of each type should have their own checkboxlist. The UI looks like this:

There are a number of ways that I could implement the collection that I bind to the listbox, but I am using Telerik controls and they would really like a generic list.  OK, I can handle that, right?  I can make a generic List of the particular POCO, and just pass it through.  But … how do I filter?  In the old days I would write a sproc, but these days, we have LINQ, and this is a job for LINQ to Objects (really the only LINQ that matters …)

protected List<Tag> Clients 

{ 

    get 

    { 

        var clients = from x in Tags 

                      where x.Type == "Clients" 
                      select x;         return clients.ToList();  
    } 
} 
  
protected List<Tag> Resources 

{ 

    get 

    { 

        var resource = from x in Tags 

                    where x.Type == "Resource" 
                    select x; 

        return resource.ToList(); 

    } 

} 
  
protected List<Tag> Staff 

{ 

    get 

    { 

        var staff = from x in Tags 

                      where x.Type == "Staff" 

                      select x; 

        return staff.ToList(); 

    } 

} 

protected List<Tag> TextTags 
{ 

    get 

    {         var tag = from x in Tags 

                    where x.Type == "Tag" 

                    select x; 

        return tag.ToList(); 

    } 

} 

Now I can work from the Tags collection in the entity model, and easily get just the values I need.  I can get rid of the magic values with an enumerator, but I thought it was a good use of Linq.

In M you can describe values and use MGraph to infer a data model.  for instance, I can tell it a little data, like this:

module TheStates
{
  States  => { "Ohio", "Indiana", "Kentucky" };
}

… and get back the SQL code for my inferred table:

set xact_abort on;
go

begin transaction;
go

set ansi_nulls on;
go

if not exists 
(
    select *
    from [sys].[schemas]
    where [name] = N'TheStates'
)
    execute [sp_executesql] N'create schema [TheStates]';
go

create table [TheStates].[States]
(
    [Item] nvarchar(max) not null
);
go

insert into [TheStates].[States] ([Item])
    values (N'Ohio'),
        (N'Indiana'),
        (N'Kentucky');
go

commit transaction;
go

The problem is with a real world model, say a collection of states that represents a route.  if you try to model that with an extent, like this:

module Proto
{
  Representative =>
  {
    Id => "4";
    Name => "Bill";
    States  => { "Ohio", "Indiana", "Kentucky" };
  } where identity Id;
}

Well, at least it is a Not Yet Implemented error.

untitled3  4,3-8,22    Error   M3999   Not yet implemented: Initializing a collection with an expression that yields a collection (Microsoft.M.SemanticGraph.FromExpressionSymbol)

Still, without something that straightforward implemented, it is tough to call this a modeling language for data.  Hopefully it is in the works soon.

Well, if you went to PDC and did the thing I always do, which is go to every talk that lists Don Box as a speaker, then you probably heard about SQ!L Modeling.  I don’t care about the SQL part – SQL is a database as far as I am concerned – it is an implementation detail.  As a software architect, though, I am very interested in Modeling.

So I get the bits and install them.  Great.  I have Quadrant as a Start Menu item, but Quadrant is Access.  Not interested (for this anyway).  I want to tell Visual Studio how I need my software built.  I need M.

The New Project Dialog has a Oslo Library that might be interesting.  I thought they had nixed Oslo though?  What’s up with that?  Anyway, the description says “A project for creating Oslo Flavored CSharp Library” whatever that means.  I just want to design software, people.  I don’t make ice cream.

I’ll go ahead and model SHARP, my event system, because I know that system well and have been modeling it for years.  To start, I name the Oslo Library SharpLibrary.  The resultant code is unusual, but it is a new language after all.

module SharpLibrary
{
    type Model
    {
        Id : Integer32 => AutoNumber();
        
        Field : Integer32;
        
    }
    
    Modelsamples : {Model*} where identity Id;
    
}

Don’t let the ‘module’ statement fool you – this isn’t a RAD application development language that will remain nameless.  M is a language that will model your middle tier.  Let’s model.  The goal is to build the simple SHARP model, which I used in VB for Dummies and the C# All In One.

I will start with the Conference entity, because it is sorta the middle of the system.  ID is as the example, but I want to add a Title and Description.  First thing is that M needs intellisense.  What’s a string again? String?  Varchar?  Oh, no of course it is Text.  Of course.

More later!

I am working on the table of contents for my new book on designing software with SQL Server Modeling, or TheModelingSystemFormerlyKnownAsOslo.  There is a lot to be shared, but TOCs are kinda boring.  I thought I would use a post to fill out any random thoughts.

I think I am going to break the book into three parts.  First, some Microsoft-centric bits about modeling in general.  Second, a detailed look at M.  Finally, we are gonna build an app.  I want to make this thing functional.  If I can't build real, usable software with it and make my life easier, then I don't want it, and I don't want you to use it.  Proves how much I think of the SQL Modeling team if I am writing a book on it, huh?

Anyway.  The tl;dr on M is that is eats examples and poops your database AND your POCOs.  If they eventually get to the point where complex examples are handles gracefully, then we might have something here.

I am still not sold on Quadrant.  They need to stop showing SQL Server table level examples.  That's.  Not.  The.  Point.  If I can't gain visibility into my AD and my file structure and my partner's accounting system, I'll just use Crystal Reports or Query Analyzer.

Most people don't understand that we have been given unprecedented access into a technology that probably won't be ready for prime time for a couple of years.  M is rough.  Quadrant is rough.  There is a lot of SCOPE work to do, not to mention a little coding.  This concept is just starting to find itself.  There are technologies that SSM depends on that haven't been built yet.  Have a little patience, people.

Now, this  Linq to M idea is a very interesting look into the thinking of the SSM team. We describe something in M.  We upload to the repository.  We can 'query' the example in C# with Linq.  Great.  Are we ever going to use M to move REAL data, rather than example data?  No?  Then how is this different from Linq to SQL?  It isn't?  Oh, OK.

Generally, though, this is something I have been saying since my "Web Design With The End in Mind" article back in 1999.  The data is the application.  If you understand the behavior of the data - which you somehow must store and business rules (or metadata) then your application is done. Skin it and go get a beer.

Wow, Shawn Wildermuth's article on Textual DSLs is really good.  Why haven't I seen this?  I thought he was just a UI geek.  Color me impressed.

Sorry this is so stream-of-consciousness.  It is late and this cough is keeping me from sleeping.

I think that SQL Server Modeling has the potential to dramatically change how Microsoft architects design software for clients.  I think it has the potential to improve design too, by making good, principled design part of the fabric rather than a documented rule.  I really, really wish they hadn't pinned it to SQL Server, but I guess I get it.  The data is the application.  Tha'ts why I have called it the Data Driven Web for fifteen years.

A recent reader emailed to wax poignant about VB6, and ask about dynamic typing – among other things – in VB.NET.  although I feel that static typing has a more firmly established place than ever in software development, I commiserated with him in this response:

A lot of people agree with you.  There is even a well-established petition at http://classicvb.org/ that states these and a lot of other problems with VB.NET, and the .NET Framework in general.

However, you have to remember that VB.NET is a language, and VB6 is a program, language and framework all in one.  As I pointed out in Chapter 1, they are designed to do different things.  A lot of the things that you could do in VB6 are considered bad practice now, due to scalability and other issues, although many are coming back into vogue. 

For instance, you point out the Variant issue.  What you are saying here is that VB6 is dynamically typed and VB.NET is strongly typed.  Coding for a strongly typed system is a lot slower than for a dynamically typed system.  You will, however, have a lot fewer errors with a statically typed system because any typing issues will usually be caught at compile time.

I have always found it interesting that many community members belittle VB6 for being a 'hack' out of one side of their mouth, and then extol the speed and ease of Ruby out of the other side of their mouth, never realizing that the dynamic typing system is a common defining characteristic of both 'languages'. 

I should add that dynamic typing is coming back in VB10 (as it is in C# 4.0) with the dynamic language runtime.  Hopefully that will return some of the flexibility to the language that you are looking for.