Why the fuss over MVC?
-
:confused: Planning a new project, I spent the last couple of days watching Scott Hanselman's video's on the subject, reading a lot and playing with this a bit. While I agree that it does make it simple do build in unit testing and keeps the architecture clean, it appears you give up a hell of a lot. Apparently third party controls by and large don't work. Telerik's last response I found on their support forum was basically "there is not yet a clear recommendation from Microsoft for using third-party UI controls with MVC". Ajax doesn't work easily. Basically if I want to use MVC, it appears I go back to a bare bones, standard controls only design without the rich client side experience that's become the expectation. I'm finding the same sort of things with other stuff that's all over the developer magazines and blog press lately. Sweet and sexy for a tech presentation, falls on it's face when faced with real application requirements. Or I am just getting to be a grumpy old man? I'm honestly not looking to ignite a flame war, I just hadn't looked at any new tech in the last year and now that I have I'm afraid I'm just not getting it...
MVC is great for helping to separate your code out into logically organised, easily testable code. Yes, you have to use plain vanilla HTML at the moment, but those fantasic controls you are talking about didn't appear when ASP.NET was first released. Who knows, but there's plenty of time for company's to come up with great controls that work in MVC - so don't make the mistake of being an early adopter, but wait for others to go through the pain first.
Deja View - the feeling that you've seen this post before.
-
Do you know Marc Clifton? :)
Best wishes, Hans
[CodeProject Forum Guidelines] [How To Ask A Question] [My Articles]
LOL. And he shares a similar first name. :confused: :-D :-D
Chris Meech I am Canadian. [heard in a local bar] In theory there is no difference between theory and practice. In practice there is. [Yogi Berra] Donate to help Conquer Cancer[^]
-
:confused: Planning a new project, I spent the last couple of days watching Scott Hanselman's video's on the subject, reading a lot and playing with this a bit. While I agree that it does make it simple do build in unit testing and keeps the architecture clean, it appears you give up a hell of a lot. Apparently third party controls by and large don't work. Telerik's last response I found on their support forum was basically "there is not yet a clear recommendation from Microsoft for using third-party UI controls with MVC". Ajax doesn't work easily. Basically if I want to use MVC, it appears I go back to a bare bones, standard controls only design without the rich client side experience that's become the expectation. I'm finding the same sort of things with other stuff that's all over the developer magazines and blog press lately. Sweet and sexy for a tech presentation, falls on it's face when faced with real application requirements. Or I am just getting to be a grumpy old man? I'm honestly not looking to ignite a flame war, I just hadn't looked at any new tech in the last year and now that I have I'm afraid I'm just not getting it...
I try to follow MVC to a point when developing web applications. I think it is possible to make your site ajaxy and dynamic and still maintain a site that adheres (mostly) to MVC standards. Jeff Atwood[^] discussed MVC on Monday.
Jeff Atwood wrote:
Here's one quick way to test if your application has properly segregated itself between the Model, View, and Controller roles: is your app skinnable?
But, you are also right. Third party controls can break MVC, if they don't support css control of the look and feel. I don't feel that ajax controls fit this type, sonce all of them are configurable using CSS.
Broken Bokken You can't carry out a ninja-style assasination dressed as an astronaut. It's the luminous fabric; too visible. - Tripod The story of your fighting is a poem of two words: YOU SUCK.
-
Do you know Marc Clifton? :)
Best wishes, Hans
[CodeProject Forum Guidelines] [How To Ask A Question] [My Articles]
Hans Dietrich wrote:
Do you know Marc Clifton?
:laugh: :-O Marc
-
:confused: Planning a new project, I spent the last couple of days watching Scott Hanselman's video's on the subject, reading a lot and playing with this a bit. While I agree that it does make it simple do build in unit testing and keeps the architecture clean, it appears you give up a hell of a lot. Apparently third party controls by and large don't work. Telerik's last response I found on their support forum was basically "there is not yet a clear recommendation from Microsoft for using third-party UI controls with MVC". Ajax doesn't work easily. Basically if I want to use MVC, it appears I go back to a bare bones, standard controls only design without the rich client side experience that's become the expectation. I'm finding the same sort of things with other stuff that's all over the developer magazines and blog press lately. Sweet and sexy for a tech presentation, falls on it's face when faced with real application requirements. Or I am just getting to be a grumpy old man? I'm honestly not looking to ignite a flame war, I just hadn't looked at any new tech in the last year and now that I have I'm afraid I'm just not getting it...
Mark Tutt wrote:
Sweet and sexy for a tech presentation, falls on it's face when faced with real application requirements.
Software is rapidly morphing into the oldest profession in the world. Marc
-
I try to follow MVC to a point when developing web applications. I think it is possible to make your site ajaxy and dynamic and still maintain a site that adheres (mostly) to MVC standards. Jeff Atwood[^] discussed MVC on Monday.
Jeff Atwood wrote:
Here's one quick way to test if your application has properly segregated itself between the Model, View, and Controller roles: is your app skinnable?
But, you are also right. Third party controls can break MVC, if they don't support css control of the look and feel. I don't feel that ajax controls fit this type, sonce all of them are configurable using CSS.
Broken Bokken You can't carry out a ninja-style assasination dressed as an astronaut. It's the luminous fabric; too visible. - Tripod The story of your fighting is a poem of two words: YOU SUCK.
ASP.NET server controls require a "server" form to be present on the page <form runat="server">. Since such server form is not present by default in a MVC ViewPage server controls won't work and in most cases will throw an exception that a server form tag is missing. You can hack it in there by manually adding a server control to a MVC View page, but there are other issues requiring a lot of manual workarounds that end up polluting the nice clean view page you were hoping for.
-
Mark Tutt wrote:
Sweet and sexy for a tech presentation, falls on it's face when faced with real application requirements.
Software is rapidly morphing into the oldest profession in the world. Marc
-
:confused: Planning a new project, I spent the last couple of days watching Scott Hanselman's video's on the subject, reading a lot and playing with this a bit. While I agree that it does make it simple do build in unit testing and keeps the architecture clean, it appears you give up a hell of a lot. Apparently third party controls by and large don't work. Telerik's last response I found on their support forum was basically "there is not yet a clear recommendation from Microsoft for using third-party UI controls with MVC". Ajax doesn't work easily. Basically if I want to use MVC, it appears I go back to a bare bones, standard controls only design without the rich client side experience that's become the expectation. I'm finding the same sort of things with other stuff that's all over the developer magazines and blog press lately. Sweet and sexy for a tech presentation, falls on it's face when faced with real application requirements. Or I am just getting to be a grumpy old man? I'm honestly not looking to ignite a flame war, I just hadn't looked at any new tech in the last year and now that I have I'm afraid I'm just not getting it...
Meh. MS tried to turn web dev into VB with WebForms. The technique does make it easy to build 3rd-party controls, and just as easy to throw a site together quickly and have it sorta work. Getting past the "sorta" is another matter. The "rich client" experience made possible by ASP.NET and ASP.NET AJAX all too often translates to bloated "slow client", viewstate- and session-dependent sites full of unlinkable URLs and out of control markup. So a bunch of people started taking the good (the language and runtime support) and leaving the bad (WebForms), writing their own systems to provide proper separation of the UI, data, and background logic. Microsoft - being Microsoft - decided to get in on the action. Again, being Microsoft they'll probably find a way to continue using the VB WebForms controls in the new system, but hosting controls isn't the point. It's about building actual web sites, with URLs and such. ;)
Citizen 20.1.01
'The question is,' said Humpty Dumpty, 'which is to be master - that's all.'
-
Meh. MS tried to turn web dev into VB with WebForms. The technique does make it easy to build 3rd-party controls, and just as easy to throw a site together quickly and have it sorta work. Getting past the "sorta" is another matter. The "rich client" experience made possible by ASP.NET and ASP.NET AJAX all too often translates to bloated "slow client", viewstate- and session-dependent sites full of unlinkable URLs and out of control markup. So a bunch of people started taking the good (the language and runtime support) and leaving the bad (WebForms), writing their own systems to provide proper separation of the UI, data, and background logic. Microsoft - being Microsoft - decided to get in on the action. Again, being Microsoft they'll probably find a way to continue using the VB WebForms controls in the new system, but hosting controls isn't the point. It's about building actual web sites, with URLs and such. ;)
Citizen 20.1.01
'The question is,' said Humpty Dumpty, 'which is to be master - that's all.'
Shog9 wrote:
It's about building actual web sites, with URLs and such.
Shhh. They'll hear you.
Deja View - the feeling that you've seen this post before.
-
Meh. MS tried to turn web dev into VB with WebForms. The technique does make it easy to build 3rd-party controls, and just as easy to throw a site together quickly and have it sorta work. Getting past the "sorta" is another matter. The "rich client" experience made possible by ASP.NET and ASP.NET AJAX all too often translates to bloated "slow client", viewstate- and session-dependent sites full of unlinkable URLs and out of control markup. So a bunch of people started taking the good (the language and runtime support) and leaving the bad (WebForms), writing their own systems to provide proper separation of the UI, data, and background logic. Microsoft - being Microsoft - decided to get in on the action. Again, being Microsoft they'll probably find a way to continue using the VB WebForms controls in the new system, but hosting controls isn't the point. It's about building actual web sites, with URLs and such. ;)
Citizen 20.1.01
'The question is,' said Humpty Dumpty, 'which is to be master - that's all.'
Agreed. WebForms does make it easy to throw something together quickly and then try to make it work. There was a reason I went looking at the shiny new toys, hoping they made it better. However, with all the hand waving over the new stuff, I'd expected a better solution to the controls issue, i.e. it's still ASP.NET. I was sooo jazzed watching Hanselman show lambda expressions for creating links so that a change that would have broken a link before will break the build now and the helper methods that handled all the databinding plumbing. Then the shoe dropped and I felt like the hottie I was looking at turned out to be a guy! Like it or not, customer expectations for web apps have just been getting higher. We're not going to convince customers that in the interest of clean seperation they can't have web apps like Gmail, Google Docs, Google Calendar, etc. they've seen it's possible. But they're not going to pay to have them written from scratch either. I predict MS will have to solve the controls issue before MVC makes it, at least in the world I work in... I need a beer, is it 5 yet? :((
-
Agreed. WebForms does make it easy to throw something together quickly and then try to make it work. There was a reason I went looking at the shiny new toys, hoping they made it better. However, with all the hand waving over the new stuff, I'd expected a better solution to the controls issue, i.e. it's still ASP.NET. I was sooo jazzed watching Hanselman show lambda expressions for creating links so that a change that would have broken a link before will break the build now and the helper methods that handled all the databinding plumbing. Then the shoe dropped and I felt like the hottie I was looking at turned out to be a guy! Like it or not, customer expectations for web apps have just been getting higher. We're not going to convince customers that in the interest of clean seperation they can't have web apps like Gmail, Google Docs, Google Calendar, etc. they've seen it's possible. But they're not going to pay to have them written from scratch either. I predict MS will have to solve the controls issue before MVC makes it, at least in the world I work in... I need a beer, is it 5 yet? :((
Mark Tutt wrote:
We're not going to convince customers that in the interest of clean seperation they can't have web apps like Gmail, Google Docs, Google Calendar, etc. they've seen it's possible. But they're not going to pay to have them written from scratch either.
:shrug: Hence the title of my reply. It's like the old cliche, "good, fast, cheap - pick two". I've no doubt that some vendors will eventually roll out MVC-friendly controls, but it'll take a bit more time and thought from 'em than the WebForms verisons did.
Citizen 20.1.01
'The question is,' said Humpty Dumpty, 'which is to be master - that's all.'
-
Mark Tutt wrote:
We're not going to convince customers that in the interest of clean seperation they can't have web apps like Gmail, Google Docs, Google Calendar, etc. they've seen it's possible. But they're not going to pay to have them written from scratch either.
:shrug: Hence the title of my reply. It's like the old cliche, "good, fast, cheap - pick two". I've no doubt that some vendors will eventually roll out MVC-friendly controls, but it'll take a bit more time and thought from 'em than the WebForms verisons did.
Citizen 20.1.01
'The question is,' said Humpty Dumpty, 'which is to be master - that's all.'
Oooh. Marketing opportunity there. Must make note to talk to the team tomorrow - there's a window of opportunity just opened.
Deja View - the feeling that you've seen this post before.
-
Agreed. WebForms does make it easy to throw something together quickly and then try to make it work. There was a reason I went looking at the shiny new toys, hoping they made it better. However, with all the hand waving over the new stuff, I'd expected a better solution to the controls issue, i.e. it's still ASP.NET. I was sooo jazzed watching Hanselman show lambda expressions for creating links so that a change that would have broken a link before will break the build now and the helper methods that handled all the databinding plumbing. Then the shoe dropped and I felt like the hottie I was looking at turned out to be a guy! Like it or not, customer expectations for web apps have just been getting higher. We're not going to convince customers that in the interest of clean seperation they can't have web apps like Gmail, Google Docs, Google Calendar, etc. they've seen it's possible. But they're not going to pay to have them written from scratch either. I predict MS will have to solve the controls issue before MVC makes it, at least in the world I work in... I need a beer, is it 5 yet? :((
It is refreshing to see people reading between lines on all those podcasts and blogs touting extremelly poor clones of Google pushing MS to reshape likely yet another failed web effort. To think it all began as fight against Java Server Pages + Apache tools bloat... Perhaps MVC and MVP are more problematic than just controls (wait and see what all those casts and reflection and events and datasets and serialisation do, otherwise why did they bother inventing Dependency junk to speed it up and make it more flexible for WPF). Same for classic ASP.NET forms with or without view-state laughing stock. In short, for me if you don't have a good and OS-licence-free API to your fast and feature-rich service, it doesn't really matter how pretty the Web Form is. And Telerik UIs really put me off. Bundling WCF+JSON+ATOM with AJAX.ASP and Silverlight is their attempt to take it further but the infrastructure behind all their server bloat (what makes it 'Web x.0') can simply not even approach Google (data and analytics) and Amazon (on-demand + ship an OS image) ideas. I want the service not the flashy bits that make the process so damn slow. Users want to work without delays of oh-so-obvious powered-by-ASP.NET server-side. Don't see them ever do it in all honesty..and the train is gone for many companies thinking they'll just produce another demo with 3rd party controls and sell it when all those service vendors give it virtually for free and so you can style it all you want.. even the Ruby toy gets more investment than all those ASP.NET houses put together.
-
It is refreshing to see people reading between lines on all those podcasts and blogs touting extremelly poor clones of Google pushing MS to reshape likely yet another failed web effort. To think it all began as fight against Java Server Pages + Apache tools bloat... Perhaps MVC and MVP are more problematic than just controls (wait and see what all those casts and reflection and events and datasets and serialisation do, otherwise why did they bother inventing Dependency junk to speed it up and make it more flexible for WPF). Same for classic ASP.NET forms with or without view-state laughing stock. In short, for me if you don't have a good and OS-licence-free API to your fast and feature-rich service, it doesn't really matter how pretty the Web Form is. And Telerik UIs really put me off. Bundling WCF+JSON+ATOM with AJAX.ASP and Silverlight is their attempt to take it further but the infrastructure behind all their server bloat (what makes it 'Web x.0') can simply not even approach Google (data and analytics) and Amazon (on-demand + ship an OS image) ideas. I want the service not the flashy bits that make the process so damn slow. Users want to work without delays of oh-so-obvious powered-by-ASP.NET server-side. Don't see them ever do it in all honesty..and the train is gone for many companies thinking they'll just produce another demo with 3rd party controls and sell it when all those service vendors give it virtually for free and so you can style it all you want.. even the Ruby toy gets more investment than all those ASP.NET houses put together.
Interesting rant. For me it's all about being pragmatic. This isn't for a service, it's an intranet application. I could care less about the politics and motivations behind the tech, I simply want to get the project done and move onto the next. With a respite from pressing deadlines for once, I took the time to look at new tech. One of the requirements of this contract is a resource scheduling feature. The Customer has 1000's of locations and wants a web app with zero client deployment requirements. Wants drag and drop event scheduling, inline editing, "Like google calendar". Telerik RadScheduler can give me the majority of that for near free in terms of development hours if I use WebForms, or I can spend days/weeks in javascript hell. While I might have liked to have made use of "better" tech, it's not worth the cost to me or the customer.
-
Meh. MS tried to turn web dev into VB with WebForms. The technique does make it easy to build 3rd-party controls, and just as easy to throw a site together quickly and have it sorta work. Getting past the "sorta" is another matter. The "rich client" experience made possible by ASP.NET and ASP.NET AJAX all too often translates to bloated "slow client", viewstate- and session-dependent sites full of unlinkable URLs and out of control markup. So a bunch of people started taking the good (the language and runtime support) and leaving the bad (WebForms), writing their own systems to provide proper separation of the UI, data, and background logic. Microsoft - being Microsoft - decided to get in on the action. Again, being Microsoft they'll probably find a way to continue using the VB WebForms controls in the new system, but hosting controls isn't the point. It's about building actual web sites, with URLs and such. ;)
Citizen 20.1.01
'The question is,' said Humpty Dumpty, 'which is to be master - that's all.'
Shog9 wrote:
with URLs and such
Don't try to confuse me with your techno-babble! I know when you're trying to bullshit me! And that rotating thingie should be mauve.
We are a big screwed up dysfunctional psychotic happy family - some more screwed up, others more happy, but everybody's psychotic joint venture definition of CP
blog: TDD - the Aha! | Linkify!| FoldWithUs! | sighist -
:confused: Planning a new project, I spent the last couple of days watching Scott Hanselman's video's on the subject, reading a lot and playing with this a bit. While I agree that it does make it simple do build in unit testing and keeps the architecture clean, it appears you give up a hell of a lot. Apparently third party controls by and large don't work. Telerik's last response I found on their support forum was basically "there is not yet a clear recommendation from Microsoft for using third-party UI controls with MVC". Ajax doesn't work easily. Basically if I want to use MVC, it appears I go back to a bare bones, standard controls only design without the rich client side experience that's become the expectation. I'm finding the same sort of things with other stuff that's all over the developer magazines and blog press lately. Sweet and sexy for a tech presentation, falls on it's face when faced with real application requirements. Or I am just getting to be a grumpy old man? I'm honestly not looking to ignite a flame war, I just hadn't looked at any new tech in the last year and now that I have I'm afraid I'm just not getting it...
Mark Tutt wrote:
it does make it simple do build in unit testing and keeps the architecture clean, it appears you give up a hell of a lot.
How about this analogy: I want a big car because I have to haul equipment around and it makes it easier, but I want it to do 0 to 60 in less that 5 seconds because I like to drive sporty cars, and I want the fuel economy to be 50+MPG because fuel is so expensive now.
Upcoming FREE developer events: * Developer Day Scotland Recent blog posts: * Introduction to LINQ to XML (Part 1) - (Part 2) My website | Blog
-
ASP.NET server controls require a "server" form to be present on the page <form runat="server">. Since such server form is not present by default in a MVC ViewPage server controls won't work and in most cases will throw an exception that a server form tag is missing. You can hack it in there by manually adding a server control to a MVC View page, but there are other issues requiring a lot of manual workarounds that end up polluting the nice clean view page you were hoping for.
I think you might be talking about 2 different things. MVC the pattern is way of separating model from presentation. OTOH Microsoft MVC is a specific implementation the MVC pattern. The MVC pattern can be implemented in different ways on different platforms. Some MVC frameworks do provide for 3rd party ASP.Net controls (but may have other disadvantages over the Microsoft effort). Personally, I rather like Microsoft's MVC - I a bit of a purist at heart, and it is rather a sleek model. Unfortunately, in the real world, I really couldn't go without all the fancy UI stuff - AJAX and the like - so will have to wait for the technology to mature.
-
Agreed. WebForms does make it easy to throw something together quickly and then try to make it work. There was a reason I went looking at the shiny new toys, hoping they made it better. However, with all the hand waving over the new stuff, I'd expected a better solution to the controls issue, i.e. it's still ASP.NET. I was sooo jazzed watching Hanselman show lambda expressions for creating links so that a change that would have broken a link before will break the build now and the helper methods that handled all the databinding plumbing. Then the shoe dropped and I felt like the hottie I was looking at turned out to be a guy! Like it or not, customer expectations for web apps have just been getting higher. We're not going to convince customers that in the interest of clean seperation they can't have web apps like Gmail, Google Docs, Google Calendar, etc. they've seen it's possible. But they're not going to pay to have them written from scratch either. I predict MS will have to solve the controls issue before MVC makes it, at least in the world I work in... I need a beer, is it 5 yet? :((
I'm not disagreeing with what's been said and this is sort of a reply to the tread in general. It's only a CTP; each version that has been released so far has had some sweeping changes; Microsoft are encouraging you to get involved and raise concerns. I;m not banging the drum for Microsoft, but these comments are based on soemthing that isn't final (nor is it close if I understand correctly), and there are plenty of opportunites to get involved. Just a read of Scott Guthrie's blog and the his responses to issues raised there would indicate that they are taking feedback seriously. In fact, I would go as far as to say there has been a quantum shift at MS recently, in particular in the way they address community concerns. -- Slips into fire protective suit :) --
The only thing unpredictable about me is just how predictable I'm going to be.
-
:confused: Planning a new project, I spent the last couple of days watching Scott Hanselman's video's on the subject, reading a lot and playing with this a bit. While I agree that it does make it simple do build in unit testing and keeps the architecture clean, it appears you give up a hell of a lot. Apparently third party controls by and large don't work. Telerik's last response I found on their support forum was basically "there is not yet a clear recommendation from Microsoft for using third-party UI controls with MVC". Ajax doesn't work easily. Basically if I want to use MVC, it appears I go back to a bare bones, standard controls only design without the rich client side experience that's become the expectation. I'm finding the same sort of things with other stuff that's all over the developer magazines and blog press lately. Sweet and sexy for a tech presentation, falls on it's face when faced with real application requirements. Or I am just getting to be a grumpy old man? I'm honestly not looking to ignite a flame war, I just hadn't looked at any new tech in the last year and now that I have I'm afraid I'm just not getting it...
Having worked with ASP.NET MVC, I will say yes, it is a little rough around the edges at the moment, but so was ASP.NET at the beginning. AJAX works, other controls also, I've had no problems providing rich UI experiences. MVC is just another tool. Just as a hammer isn't the best choice for digging a hole, you have to choice the the right tool for development. Not knowing the exact problem you are facing with third party controls it seems like a typical response from them. It's not their fault, it must be Microsoft. I've had this response from others, including Telerik, Infragistics, DevExpress.
only two letters away from being an asset
-
:confused: Planning a new project, I spent the last couple of days watching Scott Hanselman's video's on the subject, reading a lot and playing with this a bit. While I agree that it does make it simple do build in unit testing and keeps the architecture clean, it appears you give up a hell of a lot. Apparently third party controls by and large don't work. Telerik's last response I found on their support forum was basically "there is not yet a clear recommendation from Microsoft for using third-party UI controls with MVC". Ajax doesn't work easily. Basically if I want to use MVC, it appears I go back to a bare bones, standard controls only design without the rich client side experience that's become the expectation. I'm finding the same sort of things with other stuff that's all over the developer magazines and blog press lately. Sweet and sexy for a tech presentation, falls on it's face when faced with real application requirements. Or I am just getting to be a grumpy old man? I'm honestly not looking to ignite a flame war, I just hadn't looked at any new tech in the last year and now that I have I'm afraid I'm just not getting it...
Just a comment on the third party controls aspect of your post. Check out componentart. They don't have full support yet, but they're taking steps, and will soon have a release with MVC capabilities. Also, i would argue that building rich client-side experiences becomes easier when you step away from the often unnecessarily-complicated web forms. When you are building the UI, all you need to focus on is building the UI. It is true that you need to do a few more things yourself, but there are more and more helper classes available all of the time which abstract away some of that stuff. Keep an eye on Rob Conery's blog, as he's often releasing helpful things in this area. Anyway, my vote is for MVC. .. :)