MonoRail in Action Website

[Dru Sellers]

So, what does everyone want? Just a wiki dumping ground, or something more formal? -d

--
cheers,
-d

[Ayende Rahien]

I think that something a bit more formal would be better.
More light hearted than the site, is maybe what I am thinking of.

[Dru Sellers]

Cool,

Here are my initial thoughts. Please comment:

Recipe
  Name
  Description
  CodeNuggets (OrderedList of Snippets needed to explain)
  Uses/Requires (What parts of the framework)

dru

--
cheers,
-d

[leemhenson]

Yeah that looks good. Maybe a repository for hosting sample
projects/files?

[Dru Sellers]

Sounds good. Anyone else?

[Ken Egozi]

+1 fot a "samples" site.

 

I can even contribute some hosting space (ASP.NET 2.0 + SQL SERVER 2000) for it, if it cannot be done within castleproject.org (hammet?)

I think that the next logos are applicable : Powered By ActiveRecord, Powered by MonoRail, Powered by Castle, those are all good. I know I'll proudly add it to my works, where applicable.

I'll try to find an artist that will do it for free. Doesn't have to be too fancy, but formal it must be.

 

---

[Ernst Naezer]

nice to see so many +1's :) I think I can also supply the hosting (asp 2 + sql server) so that's probably no problem. Only I suck in graphic mode :)

Maybe we can collect a list of tutorials and other posible content ppl are willing to provide so we can create a structure based on that?

Let me think, I can provide some text and code demonstrating the repository pattern that allows ppl to hook up the lucense search engine as a starter. Anyone else?

greets,

ernst.

[Craig Neuwirt]

I need to provide an example site that demonstrates the ASP.NET Views with Atlas and the AtlasTooklit.

[Kevin Williams]

Atlas? Eewww.

;)

Craig Neuwirt wrote:
> I need to provide an example site that demonstrates the ASP.NET

> <http://ASP.NET> Views with Atlas and the AtlasTooklit.
>
> On 12/7/06, *Ernst Naezer* <ernst...@gmail.com

> <mailto:ernst...@gmail.com>> wrote:
>
> nice to see so many +1's :) I think I can also supply the hosting
> (asp 2 + sql server) so that's probably no problem. Only I suck in
> graphic mode :)
>
> Maybe we can collect a list of tutorials and other posible content
> ppl are willing to provide so we can create a structure based on that?
>
> Let me think, I can provide some text and code demonstrating the
> repository pattern that allows ppl to hook up the lucense search
> engine as a starter. Anyone else?
>
> greets,
>
> ernst.
>

> On 12/7/06, *Ken Egozi* <ego...@gmail.com

[Dru Sellers]

I could convert hammett's BaseCrudController to an example. As well as Ayende's IRepository<T> stuff.

I could also do a security filter.

-d

--
cheers,
-d

[Ayende Rahien]

I second the sentiment. I don't see a lot of new stuff in Atlas' Ajax library, and the asp.net stuff is useless for MonoRail

[Craig Neuwirt]

What do you mean by useless?

On 12/7/06, Ayende Rahien <aye...@ayende.com> wrote:

[Nick Parker]

Sounds great, I am working on a blog engine running Castle windsor, monorail, activerecord.  It would be nice to share this with others if there is any interest.

Nick Parker
www.developernotes.com

[Hamilton Verissimo]

And as you know, I'm really looking forward to that :-)

On 12/7/06, Craig Neuwirt <cneu...@gmail.com> wrote:

> I need to provide an example site that demonstrates the ASP.NET Views with
> Atlas and the AtlasTooklit.

--
Cheers,
hamilton verissimo
ham...@castlestronghold.com
http://www.castlestronghold.com/

[Hamilton Verissimo]

There are companies that wont embrace Castle leaving their investment
and knowledge on WebForms. We're not producing an opinionated
software, so having MR working with WebForms (and everything that
surrounds it) is a priority - from a business perspective at least.

On 12/7/06, Ayende Rahien <aye...@ayende.com> wrote:

> I second the sentiment. I don't see a lot of new stuff in Atlas' Ajax
> library, and the asp.net stuff is useless for MonoRail

[Ayende Rahien]

By the ASP.Net stuff I meant specifically the UpdatePanel.

On 12/7/06, Craig Neuwirt <cneu...@gmail.com> wrote:

[Ayende Rahien]

I don't doubt this sentiment, and I agree that having MR working with WebForms is a Good Thing.
Personally, I don't find Atlas exciting in any way, which is what I meant.

I am sure that there are a lot of people who greatly enjoy Web Forms, unfortantely, I am not one of them :-)

[Hamilton Verissimo]

I totally agree. It's just that this very week I've heard from a
prospect "try saying to a CTO that they wont be able to leverage on
third party controls", hence my priority shifting.

On 12/7/06, Ayende Rahien <aye...@ayende.com> wrote:
> I don't doubt this sentiment, and I agree that having MR working with
> WebForms is a Good Thing.
> Personally, I don't find Atlas exciting in any way, which is what I meant.
>
> I am sure that there are a lot of people who greatly enjoy Web Forms,
> unfortantely, I am not one of them :-)

[Craig Neuwirt]

I certainly agree that using Web Forms is a little different than the approaches used

by the other view engines.  It certainly suffers from some of the constraints imposed

by ASP.NET (Like PostBacks).  I had to use it at my current client for the exact

reason Hammet identified.  I personally thought it would be a limitation, but it seems

to work just fine.  Controllers obtain the models and control workflow, while the ASP.NET

Web Forms and Controls just present using DataBinding.  Atlas is a pretty interesting

approach to Ajax and works pretty well too. 

 

On 12/7/06, Hamilton Verissimo <ham...@castlestronghold.com> wrote:

[leemhenson]

Maybe this would be a good idea for a subject area on "The Keep":
ViewComponents. We could contribute ideas on the best way to implement
commonly used components. Might also be applicable to Wizards as well:
I've just written an account registration wizard and a password
recovery wizard which tie into the standard asp.net membership
providers.

On Dec 7, 8:35 pm, "Hamilton Verissimo" <hamm...@castlestronghold.com>
wrote:

> hamm...@castlestronghold.comhttp://www.castlestronghold.com/

[Craig Neuwirt]

I assume this wasnt using the WebForms View Engine.

On 12/7/06, leemhenson <lee.m....@gmail.com> wrote:

[Dru Sellers]

ohhhhh......Password Recovery Wizard!!! That i would like to see.

On 12/7/06, leemhenson <lee.m....@gmail.com > wrote:

[Ayende Rahien]

My response to that statement is:
http://blogs.msdn.com/ericgu/archive/2006/12/05/the-siren-song-of-reuse.aspx

And a tale from a far off land called the Here&Now:

I am working on a project now (year+, two teams, lots of complexity) where the downstream team has evaluated and purchased both Infragistics - including source license and Component Art full suites.
They were drawn to the big benefits of being able to utilize 3rd party controls, speed up their dev time, etc.
Overall, I think that they spend over 5 Man Months trying to get those controls exactly the shape that they wanted.
Eventually they abandoned most of them (there are remains here are there), and are using controls built in-house  (mostly based on the default MS ones) to do all their presentations.

Overall, the adventure with 3rd party controls cost them around 30,000$.
And their codebase has four sets of controls, Component Art, Infragistics, Microsoft, Inhouse, all of them doing roughly the same thing, all of them subtly different.

Granted, in many cases, you could do quite a bit with 3rd party controls, but I find that the minute that I need to do something extra, and I always need to do something extra, it is taking me more time to get it right than it would have taken me to write it from scratch.
This is why my current project has 5 calendar controls, each written for a specific page (need to do a lot of coloring of the data, different colors, algorithms and data per page), instead of a super complex single calendar.

In most cases, I find that I don't need all that extensability / features in a control, I usually need a very small subset (which you probably couldn't sell alone, so it gets a lot of extra functionality, to fit a lot of cases). Writing that functionality by myself wouldn't take much longer than learning a new control and figuring out how to get it to where I want.

On 12/7/06, Hamilton Verissimo <ham...@castlestronghold.com> wrote:

[leemhenson]

WebForms make Baby Jesus cry.

On Dec 7, 8:45 pm, "Craig Neuwirt" <cneuw...@gmail.com> wrote:
> I assume this wasnt using the WebForms View Engine.
>

[Craig Neuwirt]

as well as many others

On 12/7/06, leemhenson <lee.m....@gmail.com> wrote:

[Craig Neuwirt]

Just out of curiousity, what about ASP.NET forms is so disliked and limiting?

 

My main gripe is the PostBack model.
 

What is peoples preferred view engine?  I havent much experience in the others, but

would like to use a different view engine on my next project.

 

 
 

On 12/7/06, Craig Neuwirt <cneu...@gmail.com> wrote:

[Dru Sellers]

I use NVelocity mostly, I tried StringTemplate but wasn't robusto enough for me. I am now wanting the syntax of the IronPython engine but on boo, cuz its the stuff!

-d

--
cheers,
-d

[G. Richard Bellamy]

I happen to be one of those folks who has a business case for using the WebForms view engine.

 

It is important to remember WHY code gets written in the business world - for BUSINESS. There are literally hundreds of thousands of developers who have been indoctrinated into the evil world of ASP.NET, and thousands of businesses who have invested huge sums developing/deploying/supporting it.

 

As a company that wants product that can be adopted by other programmers, or uneducated executives, without having to CONVINCE this audience, we're in a position where WebForm support is critical. Our audience MATTERS. Saying they are stupid for not doing things "right" doesn't change that their money spends just as well as smart folks.

 

In other words, I would like to be able to ride the coattails of the Microsoft Marketing Juggernaut, and this has been the only sticking point I've had when discussing the Castle Stack at every level of the business process.

 

And the fact that this seems to be, consistently, the ONLY business argument AGAINST the Castle Stack (aside from the Open Source vs Private Sector argument), it would make sense for us to adopt an approach that will build a bridge between the objection and adoption.

 

For what it's worth.

 

- Richard

---

From: castle-pro...@googlegroups.com [mailto:castle-pro...@googlegroups.com] On Behalf Of Craig Neuwirt
Sent: Thursday, December 07, 2006 12:39 PM

To: castle-pro...@googlegroups.com
Subject: Re: MonoRail in Action Website

[leemhenson]

- The amount of cruft you have to write to do simple things.
- ViewState
- PostBack
- The overall WebForms model just doesn't sit right with me: it's too
geared towards making websites act like desktop apps.
- Solely using WebForms means you lose sight of what's actually
happening in terms of http - I'd forgotten so many of the basics.
- I like having control over every line of code that arrives on the
browser, right down to ids. WebForms likes to munge everything up.
- By having more control over my exact output, I feel more capable of
integrating other libraries like Prototype.

On Dec 7, 9:13 pm, "Craig Neuwirt" <cneuw...@gmail.com> wrote:
> Just out of curiousity, what about ASP.NET forms is so disliked and
> limiting?
>
> My main gripe is the PostBack model.
>
> What is peoples preferred view engine? I havent much experience in the
> others, but
> would like to use a different view engine on my next project.
>

> On 12/7/06, Craig Neuwirt <cneuw...@gmail.com> wrote:
>
>
>
> > as well as many others
>

[Ayende Rahien]

It is a leaky abstract at best.
The page lifecycle sounds nice in theory, but try to figure out what happens when in a page where you are doing "complex" stuff like adding controls dynamically, where the control relies on something happening in the OnLoad, but actually was loaded afterward, etc.

For a long time I had the page lifecycle diagram in my office, I used it as a target practice.

We have an inspirational message on the walls here: "If something goes wrong, check the view state. If it isn't this, check the view state again."

I got a line that goes like this:

ctrl.Name = "WTF?!?!?!?!?!";// DON'T DARE TOUCH THIS LINE
// the above line set the name of the control to a stable name, otherwise ASP.Net will generate the name for it
// and it will fail every third postback. I am serious, DO NOT TOUCH THIS!

The line above cost me two days and was escelated to MS before we solved it (without their help).

The full story is here: http://www.ayende.com/Blog/2005/11/23/ItsTheNameStupid.aspx

[Craig Neuwirt]

These are all great points and very true observations.  Thanks for sharing them with me.

 

[Ayende Rahien]

One of the advantages of having an active blog is that I can refer people to it.
Check this post for my reasons to dislike ASP.Net Web Forms model.
http://www.ayende.com/Blog/LeakyAbstractionsASPNetAndChoices.aspx

[Craig Neuwirt]

Yup, that is a great entry as are many of your others.

[leemhenson]

+1

On Dec 7, 9:50 pm, "Craig Neuwirt" <cneuw...@gmail.com> wrote:
> Yup, that is a great entry as are many of your others.
>
> On 12/7/06, Ayende Rahien <aye...@ayende.com> wrote:
>
>
>
> > One of the advantages of having an active blog is that I can refer people
> > to it.
> > Check this post for my reasons to dislike ASP.Net Web Forms model.
> >http://www.ayende.com/Blog/LeakyAbstractionsASPNetAndChoices.aspx
>

> > On 12/7/06, Craig Neuwirt <cneuw...@gmail.com > wrote:
>
> > > Just out of curiousity, what about ASP.NET <http://asp.net/> forms is so

> > > disliked and limiting?
>
> > > My main gripe is the PostBack model.
>
> > > What is peoples preferred view engine? I havent much experience in the
> > > others, but
> > > would like to use a different view engine on my next project.
>

> > > On 12/7/06, Craig Neuwirt <cneuw...@gmail.com > wrote:
>
> > > > as well as many others
>

[Ayende Rahien]

LOL,
Thanks guys.

On 12/7/06, leemhenson <lee.m....@gmail.com> wrote:

[Kevin Williams]

I have seen more 2000-5000 line *.aspx.cs files than I can possibly
remember. I even wrote one, but I've learned my lesson. The tight
coupling between presentation code and code-behind classes creates the
trap of building the whole app in code-behind. How do you test that?
Browse, click and pray. That doesn't sit well with me at all.

That said, our company is adopting Monorail across the board (public
company, tons of web code) but webforms is the rule for now, primarily
for the lower barrier to learning for our offshore folks and the need to
support third-party controls. I've been pushing for using the
CompositeViewEngine and only use aspx for the pages with 3rd party
controls, but some people look at NVelocity like it came from Mars. My
next battle is IoC, which may come down to using Spring.Net instead of
Windsor, which is better than not using IoC at all. Oh, and ORM may be
iBatis over ActiveRecord/NHibernate due to the stronger stored procedure
support, but using ORM may be a long way off. Arrgh!

Craig Neuwirt wrote:
> Just out of curiousity, what about ASP.NET <http://ASP.NET> forms is so

> disliked and limiting?
>
> My main gripe is the PostBack model.
>
> What is peoples preferred view engine? I havent much experience in the
> others, but
> would like to use a different view engine on my next project.
>
>
>

> On 12/7/06, *Craig Neuwirt* <cneu...@gmail.com

> <mailto:cneu...@gmail.com>> wrote:
>
> as well as many others
>
>

> On 12/7/06, *leemhenson* <lee.m....@gmail.com

> <mailto:lee.m....@gmail.com>> wrote:
>
>
> WebForms make Baby Jesus cry.
>
> On Dec 7, 8:45 pm, "Craig Neuwirt" < cneuw...@gmail.com

> <mailto:cneuw...@gmail.com>> wrote:
> > I assume this wasnt using the WebForms View Engine.
> >
> > On 12/7/06, leemhenson < lee.m.hen...@gmail.com

> <mailto:lee.m.hen...@gmail.com>> wrote:
> >
> >
> >
> > > Maybe this would be a good idea for a subject area on "The Keep":
> > > ViewComponents. We could contribute ideas on the best way to
> implement
> > > commonly used components. Might also be applicable to Wizards
> as well:
> > > I've just written an account registration wizard and a password
> > > recovery wizard which tie into the standard asp.net

> <http://asp.net/> membership

> > > providers.
> >
> > > On Dec 7, 8:35 pm, "Hamilton Verissimo" <

> hamm...@castlestronghold.com <mailto:hamm...@castlestronghold.com>>

> > > wrote:
> > > > I totally agree. It's just that this very week I've heard
> from a
> > > > prospect "try saying to a CTO that they wont be able to
> leverage on
> > > > third party controls", hence my priority shifting.
> >
> > > > On 12/7/06, Ayende Rahien < aye...@ayende.com

[leemhenson]

All that, *and* having to deal with pirates? Rather you than me! >:]

On Dec 7, 11:19 pm, Kevin Williams <k...@bantamtech.com> wrote:
> I have seen more 2000-5000 line *.aspx.cs files than I can possibly
> remember. I even wrote one, but I've learned my lesson. The tight
> coupling between presentation code and code-behind classes creates the
> trap of building the whole app in code-behind. How do you test that?
> Browse, click and pray. That doesn't sit well with me at all.
>
> That said, our company is adopting Monorail across the board (public
> company, tons of web code) but webforms is the rule for now, primarily
> for the lower barrier to learning for our offshore folks and the need to
> support third-party controls. I've been pushing for using the
> CompositeViewEngine and only use aspx for the pages with 3rd party
> controls, but some people look at NVelocity like it came from Mars. My
> next battle is IoC, which may come down to using Spring.Net instead of
> Windsor, which is better than not using IoC at all. Oh, and ORM may be
> iBatis over ActiveRecord/NHibernate due to the stronger stored procedure
> support, but using ORM may be a long way off. Arrgh!
>
> Craig Neuwirt wrote:
> > Just out of curiousity, what about ASP.NET <http://ASP.NET> forms is so
> > disliked and limiting?
>
> > My main gripe is the PostBack model.
>
> > What is peoples preferred view engine? I havent much experience in the
> > others, but
> > would like to use a different view engine on my next project.
>

> > On 12/7/06, *Craig Neuwirt* <cneuw...@gmail.com

> > <mailto:cneuw...@gmail.com>> wrote:
>
> > as well as many others
>

> > On 12/7/06, *leemhenson* <lee.m.hen...@gmail.com

[Ayende Rahien]

I may be missing something, but what pirates?

[Jonathon Rossi]

I agree with those big pages. I was building my current project with web
forms (not the view engine) and I was getting really complex pages. I had
one page about 1500 lines that I need to keep doing more work and I spent so
long working out where I had to put things to make them work because I
hadn't worked on that page in a couple of weeks. Now my actions are only a
few lines just to get what is needed done.

I really like being able to have complete control over the HTML now using
NVelocity; I find I am so much more productive writing pages even though I
am writing the HTML for everything and not using any web form controls. But
I can understand why businesses need them. I also like the cleanness of
monorail, you don't need to hack around with the asp.net web controls to get
them to do what you want when you want it and the likes.

Regards, Jonathon Rossi

> -----Original Message-----
> From: castle-pro...@googlegroups.com [mailto:castle-project-
> de...@googlegroups.com] On Behalf Of Kevin Williams
> Sent: Friday, 8 December 2006 9:19 AM
> To: castle-pro...@googlegroups.com
> Subject: Re: MonoRail in Action Website
>
>

[leemhenson]

Arrrgh!
http://en.wikipedia.org/wiki/Pirates_vs._Ninjas

Can I get a +1 for pirates?

Sorry.....I'm working very long hours at the moment. I think madness is
setting in.

On Dec 7, 11:36 pm, "Ayende Rahien" <aye...@ayende.com> wrote:
> I may be missing something, but what pirates?
>

[Adam Mills]

Hell yeah you can!
Funnily enough all candidates who apply here are asked to fill out a
quick get to know the candidate survey... favourite books, blogs and
whether pirates are cooler than ninjas.

-----Original Message-----
From: castle-pro...@googlegroups.com
[mailto:castle-pro...@googlegroups.com] On Behalf Of leemhenson
Sent: Friday, 8 December 2006 10:42 AM
To: Castle Project Development List
Subject: Re: MonoRail in Action Website

[leemhenson]

Amen brother.

On Dec 8, 12:06 am, "Adam Mills" <adam.mi...@fortytwo.com.au> wrote:
> Hell yeah you can!
> Funnily enough all candidates who apply here are asked to fill out a
> quick get to know the candidate survey... favourite books, blogs and
> whether pirates are cooler than ninjas.
>
> -----Original Message-----
> From: castle-pro...@googlegroups.com
>
> [mailto:castle-pro...@googlegroups.com] On Behalf Of leemhenson
> Sent: Friday, 8 December 2006 10:42 AM
> To: Castle Project Development List
> Subject: Re: MonoRail in Action Website
>

> Arrrgh!http://en.wikipedia.org/wiki/Pirates_vs._Ninjas

[Andrew Hallock]

I stopped using Web forms about a year ago, and never looked back.
Recently, a client of mine (soon to be someone else's client)
discovered DotNetNuke...it's like a bad dream, really it is...

Ayende: "...you could do quite a bit with 3rd party controls, but I

find that the minute that I need to do something extra, and I always
need to do something extra, it is taking me more time to get it right
than it would have taken me to write it from scratch."

So true. Those shiny, third-party controls are *usually* bloated,
hard to customize, and have that magic demo feel: "Hey, look what I
can set up in 10 minutes to impress everyone!"

I usually opt for lightweight JS components that can be hooked
together, like command-line tools in Linux/Unix. Best part is, the
controllers remain mostly untouched, callable via XMLHttpRequest or a
regular request.

The only things I remember about "Atlas" were hours of configuration
and too much abstraction. I'm sure it's wonderful in the hands of
people who know it and care about it, but I don't have the patience.

Web controls make the easy things easier and the hard things
impossible (can't remember the origin of that quote, but it's a great
pretentious and glib statement about MS products :-)).

On 12/7/06, leemhenson <lee.m....@gmail.com > wrote:
>

[Hamilton Verissimo]

On 12/7/06, G. Richard Bellamy <rbel...@pteradigm.com> wrote:
> As a company that wants product that can be adopted by other programmers, or
> uneducated executives, without having to CONVINCE this audience, we're in a
> position where WebForm support is critical. Our audience MATTERS. Saying
> they are stupid for not doing things "right" doesn't change that their money
> spends just as well as smart folks.

Richard, I dont think MonoRail's positioning is "the right way", or
"for smart people". I've tried to expose it as a "different way", a
"simpler way" to web development.

Giving a step back, what moves Castle, what's our big goal? The
simplification. Let's make life bearable for .Net programmers. Let's
present an alternative to application development for those not
completely satisfied with the "one Microsoft way". Let's try to create
some boundaries so in the end you end up with a better system...

For some reason, most developers on MS land dont buy it. Be it for FUD
or something else. And that's fine with me, we will never reach a full
adoption. If Castle was conceived on the Java side of the fence I'm
positive it would have a larger user base. Those folks knows what IoC
is since 1999. They know web frameworks and what makes a good one.
Best of all, they know how to judge the maturity of a project based on
its code and the community behind it. But I happen to like .net, so
here we are...

However I feel that people who gets involved with castle, are above
average, have a higher IQ. That's enough to make me happy :-)

[Jonathon Rossi]

I started working with "atlas" a while back and it seemed good when I
watched a screencast. When you get into it, it seems a little strange
because it is based on the web forms model of trying to be like winforms. I
can see heaps of people abusing the UpdatePanel and not understanding
anything about what is happening and making slow AJAX requests. The other
concern is when you use an UpdatePanel it re-renders the whole page and
sends back the portion that was inside the UpdatePanel. So instead of making
it look like the client is calling the server using XMLHttpRequest and atlas
hides all that and allows you to hook up triggers on events like a client
side button click which ends up being server-side. After you add atlas into
the mix I believe it makes your pages so much harder to debug compared to
the elegant way monorail implements AJAX.

Regards, Jonathon Rossi

> -----Original Message-----
> From: castle-pro...@googlegroups.com [mailto:castle-project-

> de...@googlegroups.com] On Behalf Of Andrew Hallock
> Sent: Friday, 8 December 2006 10:18 AM
> To: castle-pro...@googlegroups.com
> Subject: Re: MonoRail in Action Website
>
>

[goodwill]

Hehe havent you check my post in MR Usage? I have asp.net membership
implemented in MR, with password recovery already :)

[goodwill]

Well I guess Hammett always tell us- you are the idiot, check the doc!
Ok thats just a joke, and thanks hammett for always doing that. I think
I am not being trained well enough on open source project either, thats
why stupid questions always comes in.

Ok, wanna add my 2 cents on this topic. I feel that its a fake
impression you need a lot of controls and truly board support to make
things work the way you want. All too often M$ programmers are stuck
with vendor's way, and keep crying when vendor doesn't provide things
the way they want. Let me put ORM for an example. We have been writing
our ORM since 2003, and yet how many of us in .net world really have
knowledge and the momentum to write apps with an ORM framework? I can
say at least 90% people are using ADO.NET with DataSet and Tables,
that, IMHO, sucks when you want your program to be really OO. The
problem is if you want to use the component in the way they behave, you
have to write it the way M$ expect you to do so, thats why a lot of
people stuck in the not good approach. The actual fact is, even if I am
writing my ORM and the binding engine myself, I think afterall I have
better productivity then stuck with M$ solution, especially when you
have a medium scale project which maintainability plays a very
important role. So if companies really feel they cant switch to MR
because of the existing investment in ASP.NET, I think its just they
haven't analyze the problem in depth enough. I am not disagreeing with
others who think MR + WebForms bridging is important, but my experience
is sometimes move bravely to a new direction might give you a lot more
benefit than try to stuck with something broken.

[Kevin Williams]

Subtle, but funny. +1 :)

[Kevin Williams]

Andrew Hal lock wrote:
> I stopped using Web forms about a year ago, and never looked back.
> Recently, a client of mine (soon to be someone else's client)
> discovered DotNetNuke...it's like a bad dream, really it is...

Yeah! I found nuke's to be horrific, and even the idea of doing that
style of code is asp.net is horrifying.

> So true. Those shiny, third-party controls are *usually* bloated,
> hard to customize, and have that magic demo feel: "Hey, look what I
> can set up in 10 minutes to impress everyone!"

The trick is getting the Pointy Haired Bosses to see past that
impressive demo. It is for me, at least.

[Hamilton Verissimo]

On 12/8/06, Kevin Williams <ke...@bantamtech.com> wrote:
> The trick is getting the Pointy Haired Bosses to see past that
> impressive demo. It is for me, at least.

And how can we, as a community, try to change this perception?

[Kevin Williams]

This is an excellent summary of my thoughts of Castle as well.

I'm a big ran of Ruby on Rails - I've used it publicly, contributed to
it, and done presentations on it. Since I make a living working with
.NET, Castle is an angel in the darkness of asp.net development. I love
the design patterns and the simplicity and flexibility they bring to
.NET development. I haven't seen anything else like it, and I'm proud to
be a part of it, however small.

[Bart Reyserhove]

And how can we, as a community, try to change this perception?

It might be interesting to create some sort of list with advantages and disadvantages (or benefits and concerns...) for both ASP.NET and MR. If everyone contributes a little we will quickly get a very usable list in discussions with project managers/ team leaders, and others.
 

[Kevin Williams]

Hamilton Verissimo wrote:
> On 12/8/06, Kevin Williams <ke...@bantamtech.com> wrote:
>> The trick is getting the Pointy Haired Bosses to see past that
>> impressive demo. It is for me, at least.
>
> And how can we, as a community, try to change this perception?
>
>

Perhaps some screencasts to shed more light on Castle's benefits. How
simple it is to develop with. How the testability will ensure a better
product. How the speed of development and easy app maintenance can
improve the company profits.

Perhaps comparing the use of bad practices compares to using Castle
could put some FUD to rest. For example, how Separation of Concerns
using NVelocity, view components, and ajax creates a more performant and
stable app with fewer lines of code than a traditional aspx page with
third-party controls. Cheaper, too.

[Hamilton Verissimo]

We have something along these lines
http://wiki.castleproject.org/index.php/MonoRail:View_Engines_Comparisson

Kinda outdated, though and only covers the view engines.

[Jonathon Rossi]

That is really for the developers, most managers wouldn't understand that.

Regards, Jonathon Rossi

> -----Original Message-----
> From: castle-pro...@googlegroups.com [mailto:castle-project-
> de...@googlegroups.com] On Behalf Of Hamilton Verissimo
> Sent: Saturday, 9 December 2006 1:22 AM
> To: castle-pro...@googlegroups.com
> Subject: Re: MonoRail in Action Website
>
>

[Hamilton Verissimo]

We are a project for developers.

On 12/8/06, Jonathon Rossi <jo...@jonorossi.com> wrote:
>
> That is really for the developers, most managers wouldn't understand that.

[Jonathon Rossi]

I know, but if you need to convince the managers to let the developers use
MR.

Regards, Jonathon Rossi

> -----Original Message-----
> From: castle-pro...@googlegroups.com [mailto:castle-project-
> de...@googlegroups.com] On Behalf Of Hamilton Verissimo
> Sent: Saturday, 9 December 2006 1:45 AM
> To: castle-pro...@googlegroups.com
> Subject: Re: MonoRail in Action Website
>
>

[Bart Reyserhove]

I would hope that managers that have to take that kind of decisions know what they are talking about...

btw, it does not only concern managers, but maybe even more important fellow developers. It might be interesting convincing your colleagues first.

[Hamilton Verissimo]

In that case I think a white paper would be more appropriate as it
targets both audiences.

Screencast would be too hands-on, only useful for developers getting
up to speed with it.

On 12/8/06, Bart Reyserhove <bart.re...@gmail.com> wrote:

> I would hope that managers that have to take that kind of decisions know
> what they are talking about...
>
> btw, it does not only concern managers, but maybe even more important fellow
> developers. It might be interesting convincing your colleagues first.

[Jonathon Rossi]

I can see your point. If your whole team was really interested in using MR then you would have a much better chance in getting permission from above anyway.

 

I can see a few wow type screencasts helping to show how easy it is to start building apps with MR. Could be the simple build a blog example but I don’t think that would show too much of what you can do with MR. however you would need to limit it to about 15mins. The screencast wouldn’t be to teach you how to use MR but just to show you what can be done and how streamlined the experience is. I believe that many developers would give you 15mins of their time to watch you work your magic while they can leisurely watch you breeze through it. This would get many developers excited because I know I got excited and ready to start working with it when I watched some RoR screencasts. Does anyone else think this would be a good idea?

 

Regards, Jonathon Rossi

---

[Jonathon Rossi]

I don't work at a web development company so I can't speak from experience
but I would expect that it would be easier to convince developers into
wanting to use MR and then they can lobby their manager to use it. We would
still need some sort of whitepaper as you said but I'm not sure if many
developers would sit there and read a big document. I know as a developer I
like to see the system in action with an example project.

Regards, Jonathon Rossi

> -----Original Message-----
> From: castle-pro...@googlegroups.com [mailto:castle-project-
> de...@googlegroups.com] On Behalf Of Hamilton Verissimo
> Sent: Saturday, 9 December 2006 1:57 AM
> To: castle-pro...@googlegroups.com
> Subject: Re: MonoRail in Action Website
>
>

[Kevin Williams]

I think both are valid because people think and learn differently. Some
need to read it, some need to see or hear it, some need to touch it and
do it. We need to be able to reach all learning styles if we want wider
acceptance.

[Tanner Burson]

On 12/8/06, Hamilton Verissimo <ham...@castlestronghold.com> wrote:

On 12/8/06, Kevin Williams <ke...@bantamtech.com> wrote:
> The trick is getting the Pointy Haired Bosses to see past that
> impressive demo. It is for me, at least.

And how can we, as a community, try to change this perception?

Use Castle's strengths, and not hide it's weaknesses.  From my experiences managers are much more likely to get on board with a "product" (everything is a product to some of them) if both the advantages and disadvantages are clearly spelled out. 

I work more or less as a single developer shop, so I don't have to do a lot of pitches on my choice of tools.  But when I have been questioned about the reasoning behind not using the builtin database and web frameworks, the answer has always been simple.  Castle allows me to create an application faster, with less code, and more "safely" than the alternatives provided in .NET.  The safety comes from being able to Unit Test heavily, built in Web Testing (even though I've switched to Watir for this), much simpler build and deployment procedure, and an active community to help resolve tough issues, and continually grow the "product". 

--
Cheers,
hamilton verissimo
ham...@castlestronghold.com
http://www.castlestronghold.com/

http://www.tannerburson.com

[goodwill]

I think code length comparison is one of the good way. I have a feeling
after I use MR + NVelocity code length significantly reduced, and run
time performance too- I feel MR is faster than ASP.NET as well. Can
have some factual list on that part?

I honestly not convinced we will ever make the evil asp.net web forms
work well with MR enough. Itself is just a burden, watch this :)

<asp:GridView ID="mygrid" DataKeyNames="ID" DataSourceID="">...
<templatecolumn>
<header>Code</header>
<editTemplate>...
<itemTemplate>...
</templateColumn>
...
...
</asp:GridView>
<asp:ObjectDataSource ID="MyObjectDataSource">
<selectcommand>
<parameter>
<parameter>...
</selectcommand>

not to mention when you have to access the column in a grid....
RowDataBound(object sender, ....)
Label myDamnLabel=((GridView) sender).FindControl("yuck...");
myDamnLabel.Text=((MyBOClass)

I think thats enough for the joke :) But, very true fact. In MR I think
within 10 statement you got entire thing up and running nicely.

So, anyone can do that whitepaper? I am very interest to know.

[Ayende Rahien]

There is an MVP week in Israel soon, and Justin (a pal of mine) is going to cover that with AR vs. DAL, at least.
I think that the main problem of Castle is that it help solves the problems that you have after you finished 30% of the project. That is when you start discoverring the limits of the stuff that MS is preaching. It takes prior experiance with this pain to make the trade off visible.

[Ayende Rahien]

What you are thinking about is team lead / architect position. By amazing coincidence, I was ask to prepare such a thing recently. It is a PPT for a presentation to pitch Active Record to team leads.

http://www.ayende.com/presentations.aspx

On 12/8/06, Bart Reyserhove <bart.re...@gmail.com> wrote:

[Ayende Rahien]

Will someone please make RowDataBound stop hurting me?

[Bart Reyserhove]

Ayende,

Very useful ppt. I am thinking of doing some sort of knowledge sharing session in my company (mainly to other developers/analysts but also project managers) and the ppt shows in a couple of slides what ActiveRecord can do, how easy it is to use it, and -very important- how it puts developers on the right track of writing good maintainable code...

A similar ppt on MR would be very interesting too. If I find the time I'll give it a try.

[Kevin Williams]

Ayende Rahien wrote:
> I think that the main problem of Castle is that it help solves the
> problems that you have after you finished 30% of the project. That is
> when you start discoverring the limits of the stuff that MS is
> preaching. It takes prior experiance with this pain to make the trade
> off visible.

I totally agree.

It would be great if we found a way to communicate these benefits and
pains to those who have never experienced that pain first hand.

[Tanner Burson]

On 12/8/06, Kevin Williams <ke...@bantamtech.com> wrote:

I'm actually "new" to .NET development, having only started in the last couple of years.  My first large web application in .NET was also in MR.  I chose this route, because of my experiences in other languages (PHP, Ruby, and to a lesser extent Java).  I haven't worked in a language yet, where  my productivity didn't skyrocket by using third party tools.  I was quickly frustrated by ASP.NET's postback, and viewstate issues, just in simple applications.  This led me to look for MVC frameworks, and I found MR.

I have never really experienced the "tough" issues in .NET development, because I skipped it and came straight here.  I think Castle is in a great position to target other developers like myself, those switching from other languages, looking for a more comfortable environment than ASP.NET.  Things that have really appealed to me, AR, MR, Marc-Andre's generator, the Brail view engine, and the emphasis on testing and automated builds. 

Things I've found lacking?  Documentation regarding common mistakes (I spend a lot of time hunting through the forums to find easy solutions to problems I run into).  Class heirarchy / integration diagrams for the public classes of AR and to a lesser extent MR.  There are lots of things I've found were easily solved just by changing the parent class of my objects (ARValidationBase for example).  And the biggest thing I've been missing is code.  The proposed site would be a huge help to fellow noobs like myself.  Docs are good for a lot, but a concise code sample can be a lot more enlightening than a page of text.

I look forward to being able to contribute some snippets I've worked out on what are likely simple problems, but took me longer than I thought they should.  Those working on starting this site, please keep the list informed of the progress!

http://www.tannerburson.com

[evarlast]

I just finished reading this thread. I've been slowly learning and
using Active Record and Mono Rail for almost a year now. I love it. A
couple months ago I was asked by an MVP what I didn't like about
DataSet and DataTable and I my initial reply was to stand there
gapejawed. Not everyone even realizes the problem. Most people don't
see that DataSets are a problem. Most people think that Typed DataSets
are typed, so they must be good OO now. I realized then that Castle
needs the advocacy that you all have discussed in this thread.

I have the pleasure of attending codemash (www.codemash.org) in January
and bringing castle to as my .net programmers as possible is my plan.
I'll be looking over this thread, the docs, and Oren's presentations to
build an arsenel of firepower trying to gain converts.

I think one easy approach to convincing people is pointing out MS's
double talk. MS recommends n-tiered applications and building good OO
models and having a great business object layer, yet they give us tools
like ASP.NET and its DataGridView which really only binds to
SqlObjectDataSource or whatever it is called. Am I the only one who
sees these two models don't match? I hope not.

I just want to thank you all for your castle contributions. I only
wish I could contribute half as much as most of you.