One word: Monorail

I'd like to really try and focus on the positive if we can. I'm working with the team, they're awesome guys, and yes they realize some different choices should have been made. I won't speak for them - but at the same time I will go to the mat for their intention of being open and transparent, releasing what they thought is good code.

You can help me by holding off a bit on the venom - these are not bad people.

amen to that!  I COMPLETELY agree. Horrible horrible garbage.

Saying the code is horrible doesn't mean the coders are bad people.

Adding the chorus of boos is un-needed right now is my point....

"Install Resharper and prepare to see more orange lines than you've ever seen before..."

I hope you're not suggesting that developers just code explicitly to ReSharpers suggestions? I take them as just that, suggestions.

I second Scott's opinion.  This comment also rubbed me the wrong way.  Plenty of apps were coded just fine before resharper, and many will after it's gone.  Let's stop with the cult of Resharper.

On the whole though, I didn't feel that the attacks were personal to the team members, but I definitely would hate to be in Karl's code review.

Other than that, how was the play Mrs. Lincoln?

I tend to agree with Rob here. The bulk of the effort right now (both internal and external) should be on "this is how we make it better" not "lrn2k0d n00bs". The code was released to help, not to be the uber sample. It's sad that some people have taken it that way (it was inevitable as soon as Da Boyz put "Microsoft" in the description field), but treat it as any sample application that is posted online. Suggest and help, don't burn to the ground.

I'll admit I was a little surprised opening up the code, since I thought I'd see manna from heaven but instead didn't/don't really grok the overall intent. Still, it would be very cool to see more sample apps, even from MS, rather than fewer, as I'd like to see a variety of styles to choose among and learn from rather than wait for The One Best Way blessed from on high.

What would be really great is to see the project redesigned/refactored to a really high level so we can see a process of improvement and evolved mastery to learn from, rather than only the end-product of the iterations and secret advice. If blogs like this give really specific examples of better approaches, cool.

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

Pingback from  2008 December 16 - Links for today « My (almost) Daily Links

Pingback from  Reflective Perspective - Chris Alcock  » The Morning Brew #245

Yeap... And now ladies and gentelmen we all are waiting for the best of the world asp.net mvc sample application from Karl Seguin.

@robconery

I agree its great its open, superb.  However because it's open people are going to be looking at this code right now. If you do a google search you'll realize there's a lot of talk about and I think it's great that people like Karl/Chad and others are prepared to stand out say that this open source code should not be used as an example of how to use MVC.

However I don't think anyone has said that the team are not awesome guys, and definitely not that they are bad people. No one is slighting them directly and from what I've seen the dicussion has quite rightly focussed on the product not the people. Even if we were though, surely releasing open source software at MS requires a somewhat thick skin?

In any case whether they are nice or not is somewhat irrelevant to this post and in addition I assume that most people at MS are nice guys, but I think we should still be able to comment on the content they release.

Anyway I'm sure the project will get better with time.

To much focus on "microsoft released" for me, this isn't like Microsoft .NET.. its just a little project done by some of the MS community guys (c9 etc).

dgi, you missed the point..

Quite possibly the worst code I have ever seen, in this day and age I would have expected a sample solution on howto use MVC not be a project showing worst practices.

Absolute shocker :(

When I say "I agree that it is great that the project is open"...re-reading my first sentence makes me cringe :)

I am coming at this as someone that has never coded with MVC. I would expect that the sample should be a great guideline with best practices.

I hope that the input the team gets from the community is applied to the code base.

Karl is right - I would be looking at this codebase figuring that this is how it is done. Then I would have gone right down the rabbit hole with it.

It is nice to hear that the community is contributing to the project. Hopefully we'll come out at the other end with something useful.

You've been kicked (a good thing) - Trackback from DotNetKicks.com

I agree that there are some things wrong with Oxite, but the actual idea is not bad. Just the execution of the idea fell short. I'm sure they will address those issues and the product will be better in later iterations.

I don't understand how people can say that Microsoft doesn't really have to care about the code they release.  It has their name on it, so its a reflection of them. No matter how Microsoft release code, the code is going to be taken as gospel.  Microsoft built the framework, so they are going to know how best to leverage it.

I agree that there are few ASP.NET MVC examples out there, but how useful is a badly code example.  Anti-patterns have their place, but common on.

Beyond starting from scratch. I don't see how the project could be saved.  Good code doesn't happen by accident it starts at the foundation and works upwards.

I just don't see the point of this useless piece of garbage.

Microsoft, disavow all knowledge of this project...and run away.

thanks karl, I was going to take a look at the code but hadn't gotten around to it yet. I knew something must be up when I saw Nikhil talking about Oxite yesterday twitter.com/.../1058691957.

Why is it MS guys always think an code criticism is a personal attack? Bullocks! It's garbage and it should be called out as such. Make it better, or, if not, pull it off the shelves. I don't sell more product b/c I am a "nice guy" I sell product because it is a superior product. If it's not superior and it's got the MS sticker on it, then it's still garbage.

My friend Dave and I embarked on a project a few weeks ago, initially just meant to be an audio CD for

This is an excellent post.  Thanks Karl.  Well thought out and presented.

Pingback from  ASP.NET MVC Archived Blog Posts, Page 1

I love it when people attribute code quality to the entity instead of the people who create it. Obviously Microsoft doesn't have any quality gates it had to go through before being released. Most projects I have worked on Management just cares does it meet the requirements. Perfection is usually expensive and hard to manage.

I would say it looks very much like any project I have worked on. Its nice to have all these ideals. But at the end of the day they had to ship it sometime.

If you have a better idea on how to do it let see a code fork from you. Otherwise your just a zealot who has no time to live to his ideals.

Its nice that you believe you know how it should be. Its completely different to offer something of value.

@Rob

I don't think this article was about how they're "bad," people.  I think it's about how Microsoft is looked at to set good examples and all you do is lead your followers down a path of death and destruction.  How can you expect your products to be used for good when you, yourself cannot even produce something worthy of the platform you provide.

Karl, I think it would be helpful (because I keep seeing comments about it) if you distinguished between Oxite failing as a "best practices real-world ASP.NET MVC project" and its success (minus the security issues) running the MIX Online site.

For example, Josh's comment immediately above me confuses this point.

Pingback from  Dew Drop - December 16, 2008 | Alvin Ashcraft's Morning Dew

I liked the background services executor.  Wasn't aware you could do that.  Thought there was a few other cool things here and there, too.

Absolutely loathe the magic strings for the views.  This application is way too big for that kind of nonsense.

Not crazy about the large controllers, but suspect you will see this (and worse) in the wild no matter what Microsoft says.   These are refactorable and not impossible to get under test.    

At least nothing ended up in the code-behind.

"But at the end of the day they had to ship it sometime."

That's a horribly wrong attitude.  Ya know, writing good code might help them ship code more efficiently.  The pieces of the Oxite codebase that Karl is complaining about probably added time to the development effort in extra debugging and testing time.  

Karl,

Thanks for the review.

Have you had a chance to look at and analyze CodeCampServer [1] and Rob's own MvcStorefront [2] example?

How do they rate?

Do you have your own example MVC app that you'd like to showcase?  I thought your ebook that you released a while back was very interesting, well put-together and helpful.  

[1] code.google.com/.../codecampserver

[2] blog.wekeroad.com/mvc-storefront

@Karl:

Of course I question your motives. Its easy to throw rocks at a glass house. It different altogether to help build one.

I think the point of Oxite is/was great way to start a dialogue. I like to see you carry past this sucks into this is how I would do controller xyz different. I think afterall that is their goal.

Attitude: Re: Having to ship code.

I haven't been on a project where the end goal was anything but meeting the expectations of the business. Zero business people or project managers care if you use domain driven design or the lastest buzz word. Instead they want it to create value for their scenario, which usually involves adhereing to a schedule and making priorities.

not only that, the trunk is broken.  no routes are installed.  there's also no history (at least using svn bridge) to go back to a time when they worked.  after looking at the code though, i decided it was best that i not run it.

Josh:

You realize that the purpose of the latest buzzwords are to ease development and better deliver on business expectations and deliver better quality to the client right?

I'd also encourage you to read through Karls archive posts, he has addressed ASP.NET MVC best practices as well as delivering a great series of posts on basic development principles.

They probably should have used the MVC Storefront if they really wanted a flag-ship ASP.NET MVC open source product. The community watched as Rob shared with us his ideas and he really incorporated some of the feedback from the community while building this application.

Oxite, on the other hand, was just realeased to the public without anyone having any input that could have improved the overall application. So, if you ask me, I don't think you can call Oxite an "Open Source" product.

"Install Resharper and prepare to see more orange lines than you've ever seen before..."

"I hope you're not suggesting that developers just code explicitly to ReSharpers suggestions? I take them as just that, suggestions."

That is what the definition of a suggestion is.  :)

On another note though, ReSharper would clean up a TON of that stuff.  Also why is Microsoft Development staff not using the greatest and latest of tools to develop?  They should all have ReSharper anyway...

...but I must digress, I haven't even looked at the code.  Regardless of how well it is written, any code can be used to learn from.

So I'm not trying to bash anyone - just saying Microsoft should shell out some $$ for ReSharper among the development staff.

@Scott:

I read Code Better feed on a regulair basis. I fully understand before my first comment what Karl does and does not deliver. (I also read the authors)

I rather see a solution rather than someone being able to poke holes in how someone else has decided to implement it.

With that said I do not put as much stock in buzz words as you do. If I said we should be using SRP, DI and DDD for this feature. Any person on my team would blink and say can you speak english.

Being able to say project X should be using DDD because I love DDD is not saying much of anything. Except that maybe the author should just move past not liking X.

Give developers a little credit. Before I would touch Oxite I would first look at the code and see if it was useful for my situation. Very few programmers would say oooh shiney Microsoft source code it must be gospel. (real do we care about those who would :)

"If you have a better idea on how to do it let see a code fork from you. Otherwise your just a zealot who has no time to live to his ideals."

Oh dear, we don't have time to do things well and anyone who tries is a "zealot" with his "ideals". Yawn.

"Its nice that you believe you know how it should be. Its completely different to offer something of value."

Your kidding right, have you read Karl's blog (hint, this isn't his first blog entry) or even the *free* book he wrote?

Funny, how all the horrible coding practices went mostly uncommented and the resharper along it's coding standards ended up as the main weak point of Karl's post to chew on among commentators...

I agree, samples by MS get way too easily evangelized to release such crap and produce million brain-dead copies of this approach just because it is done by people from or close to MS. On the other hand, why should MS teach people how to write good code? Clearly, they are struggling themself and code thrown out to the public is most probably 99% done by the guys who spend better part of the day selling themself not their code...

(disclamer; I'm the founder of Umbraco, an open source asp.net web cms and this comment is not to focus on code quality (i'm clueless on mvc) but on why ms is getting bashed)

The major flaw in the oxite release was the attitude in which it was releases. It wasn't released with a big this is wip codebase and a very limited (not to say almost none existing) functional demo of what a mvc based blogging tool could end up being. Instead it was both in terms of functionality and code quality marketed as an ace project with a pretty cocky attitude, which is cool if you live up to the hype, but backlashes of you don't. And it didn't.

I say pull the project or call and market it as a work in progress barebone blogging engine - that's what it is.

@Bunter That's probably because most people didn't download and analyze Oxite, but we all know R#.  We bitch about things we know, I guess.

Anyway, the race to 0 orange lines always seemed wrong to me.

the problem with people is that they think of Microsoft as 1 entity... they are a bunch of groups that get funding depending on if thier project will make money or not.  all thier best guys are probably working on Azure and .NET services or whatever other stuff is coming and some MVC evangalist guy hired a couple programmers to quickly do this... they did the best they could given the constraints... if you were such a hot shot why arent you developing world class software.... no youd rather be here on your soapbox pointing fingers.. show us your code so they can hit you back.. that would be a really hot post!

This sounds like a good peer review.

A poor programmer generally takes peer reviews personally.

A good programmer learns from them and writes better code in the future.

Oxite aside - what I got from all this is that I should go and see what this ReSharper is all about. I've heard of it before but never gave it a try.

Thanks :)

Oxite? MVC Best Practices?

Denny, try it. After couple of weeks of using, visual studio without r# seems seems like a notepad with couple of extra menu commands...

Good lord, those guys just wanted to release something....Instead of applauding them for having the courage to go hey look at this, instead of maybe contacting them and offering to help (Big props to Rob for that) you harpoon them and by extension used it to bash MS. That's pretty harsh and doesn't help anybody, and that's especially disappointing considering the effort that you've gone to make the community a smarter place.

Critisism is great when it targets the work and the men behind.

But I would be glad if you could add suggestions for each problem you spot :

- When you talk about the lack of "Model". What kind of model code did you expect from a "simple blog engine" ?

- About the use of a base controller with 400 lines, what is useless, what should be moved, where shoud it be moved, according to you ?

- Could you talk about "a right way" for making localization ?

- About testability, What should you test ? and how ?

...

Hmm, off course, not in the same post, but what about a second post with those suggestions ?

Karl, of course you are right in pointing out the problems and yes you are right in your views on what the problems are but somehow still this post just seems wrong?

I feel the urge to critise you for something but of course you are right--maybe thats what is bugging me, or maybe it's your photo...really man you should be carefull 'what you put out for everyone to see' :)

Pingback from  ASP.NET MVC, Oxite - Golden Shower « Tribulations of a giant dwarf

Pingback from  ASP.NET MVC (Beta) - FUBAR « Tribulations of a giant dwarf

@Rob,

Hrmm Rob, remember this post where you went all out on the netTiers framework.  Let's not act as if you've been all gung ho with open source communities this whole time.

blog.wekeroad.com/.../Zero-Code-DAL-The-.NET-ActionPack

@Frank - yesss! Conery likes to fling mud when he wants to, and act like the voice of reason when it suits him. Anyone that reads his crap carefully will see this. Self-serving and two-faced. -1 CodeBetter for having him cross post here.

Be a critical reader!

I don't really understand the "Hey, at least they put something out there" train of thought.  This is obviously a quite flawed piece of software put out there with a label stating that it's a Best Practice.

Putting something out there just for the sake of putting something out there, even it if it's of horrible quality is not a good approach to take.  In this case, it actually causes more harm than good.

Quote from Oxite page - http://www.codeplex.com/oxite

What this isn't

This isn't a best-practices example of ASP.NET MVC design guidelines. For that, check out Rob Conery's MVC Storefront application (there are others, start here). We are working to make it a better example of these guidelines, but it is most definitely a work-in-progress.

This isn't a finished product. It has no install, it has no documentation. It is our code, exactly as we are using it. If you'd like something that is farther along and has all the support, documentation and handy installation features you'd expect from a finished product, then you might want to check out BlogEngine.NET, SubText or dasBlog ... we are fans of all of these blogging engines... none of which use MVC at the moment but are far more 'finished'.

This isn't for non-developers. This follows somewhat from the previous point, but to be clear... if you aren't familar with a web.config and with some database poking, then this isn't the project for you. You might have more luck with the other blog engines mentioned in the previous point, but for a non-dev I'd probably suggest you check out Graffiti.

@ Conker - bullocks! That's a marketing patch! Here is how it showed up:

The team behind Channel 9, Channel 8, Channel 10, TechNet Edge, and Mix Online have released an open source blogging / basic content management platform called Oxite.

In addition to the core blogging features you’d expect, Oxite is a full working example of ASP.NET MVC in action!

And it’s from the real world:  Oxite powers the MIX Online site. Oxite was a greenfield (built from scratch) project but the dev team borrowed functionality, code, and lessons learned from the Channel 9 codebase.

Oxite is an open source, standards compliant, and highly extensible content management platform that can run anything from blogs to big web sites. We know this because it runs MIX Online.

http://visitmix.com/lab/oxite/

-------------------------------------------------

They were looking for something much bigger, got their peter sanded and now they are trying some smoke and mirrors. Don't be so naive.

Well, Rob has apparantly taken some of the criticisms in this blog here: blog.wekeroad.com/.../some-thoughts-on-oxite

Karl, you put the final NAIL  on the cfn.

The truth is: "developer will flock and consider Oxite as the Hooly Grail"

That is usually the cause and that is why MSFT Only developers for the most part develop bad software.

I started up as a Unix developer (C/C++), then Java before .NET. I followed only the principles from good software development. However, in an attempt to make stuff [super] easy, MSFT always generates a culture of "BAD" developers by labeling certain "OPEN" projects as the perfect example of "best practices".

Not to be all "Anti MSFT" but, they have the responsibility of "NOT MISLEADING" the community with horrible software. The Unity application blocks are good but they don't promote them, instead we get CRAP like this as the "BEST" CMS ASP.NET MVC package.

One thing I like about C/J2EE projects is: They might have a learning curve and horrible IDEs, but they filter out a lot of phony developers

Oxite, in my opinion should be totally re-written.

Pingback from  Der perfekte Zeitpunkt zum Source Code veröffentlichen? | Code-Inside Blog

The problem is that there are NOT enough good sample open source .NET applications out there for people to review, discuss/critique and learn from.

Tobin, i dont see any part of your quote suggesting that Oxite developers claim its a 'best'  MVC application of some sort:

"Oxite is a full working example of ASP.NET MVC in action!" - its true

"And it’s from the real world:  Oxite powers the MIX Online site." - true as well.

So where is the problem exactly ?

All they need to do is to clean up and refactor their code with some automatic tool of their choice , but  its hardly a problem.

LINQ to SQL is a part of the framework and i dont see how it have "questionable" future any more than any other class inside Framework. LINQ is particularly good for exactly those kind of projects .

@conkin:

I would say that the release of Oxite to the world is akin to a programmer saying, "well, it compiled", or, "it works on my computer", etc., and saying to everyone else, "it's good to go!" Opening Oxite's code as open-source basically revealed it to the ultimate code review session. The heading on their website, especially with the proclamations "we're using it in production", implied certain things about the code that maybe aren't so true...

Karl pointed out some of the issues, both wtih the initial (naive?) branding of it, how some programmers will look at it down the road and the implications of that, as well as actual technical issues. It was a criticism, and he's not necessarily required to point out alternatives.

Rob Conery addresses many of the technical issues on his blog (not sure which came first, Rob's post or Karl's, or if they're ships passing in the night), and hopefully they'll be fixed soon.

The Oxite web page needs to be modified, however, to indicate that the code is a work in progress and that we're ALL learning from it, both the downloaders and the developers. Whether it's being used "in production" or not is simply a red herring.

I know I d/l'd it to look at (along with Kigg and MVC Storefront apps), to see how/if the developers obfuscate the database access layer, and integrate it with L2S. From the cricicisms about it, I'm not going to be looking at it further.

Karl,

I think that you need to take a look at the way you write and accept some of the responsibility for the tone of your responses. When your lead paragraph closes with this sentence:

<blockquote>One would expect Microsoft to realize this and seize the opportunity to showcase above normal coding prowess and leadership. (You might also expect someone at CodeBetter to piss all over it...)</blockquote>

YOU have made it personal. Pissing on something is not a gift, it's an insult and you made it clear at the beginning of the piece that you were out to get them. Using adjectives like "horrible", "mediocre", etc, is not an attempt to add value to a conversation, it's an attempt to be a movie critic.

And saying that the MVC group should have adopted the recommendations of your friends is more than a bit arrogant.

You have done a good job of identifying problems, that a good place to start. Now you have to help fix them or you should keep your opinions to yourself (or do a better job at voicing them politely).

"YOU have made it personal."

:)

"Using adjectives like "horrible", "mediocre", etc, is not an attempt to add value to a conversation"

:))

And the best, the numero uno:

"You have done a good job of identifying problems, that a good place to start."

:DDD

Be easy evangelists, be easy...and try to code.

@Karl

You obviously know what you are talking about.  Please consider helping 15,000+ people by introducing some code patches to the MVC demo.

Looking at other MVC storefront, it doesn't seem to be compatible with the new MVC builds.

Pingback from  The perfect time to release source code? | Code-Inside Blog International

Developer preference.

There are no silver bullets.

One man's trash is another man's treasure.

The grass is always greener.

History repeats itself.

Need I go on?

@Conker

Check the history on the page, I believe they changed the page to have that bit about it not being best-practice (probably as a result of feedback). So the fact that message is now what it is shows the positive power that the community can have.

"So the fact that message is now what it is shows the positive power that the community can have."

Hm...imho positive power is if they change/rewrite/refactor the code.

Changing the page _instead of_ the code is just PR.

Pingback from  Triple Shot Links # 11 | Caffeinated Coder

@Conker,

Did you read http://visitmix.com/Lab/Oxite ? Especially the part about 'built for testability and hence most likely to be voted "hottest in class" by TDD fans' ?

You know, we shouldn't complain. There's a lot of money to be made by good developers in fixing bad developers' code.

Don't believe me? Raise your hand if you were handed a project with 0% code coverage, where core business logic was implemented in a private method of a System.Web.UI.Page and were told to implement new functionality. That's how you rack up those billable hours!

@João P. Bragança,

There are easier ways to make a living than that.  I'm sick of explaining to clients that the work is going to take longer than they want because I'm fighting with bad code.

Thanks Karl for this eye-opener for all the MS loving zealots, even if they don't care to listen.

And to all those MS loving zealots / defenders, here's my question to you (since I don't understand marketing, different teams, who owns what etc.) -

Who owns MSDN???? And please show me the code samples / examples from within MSDN which show this usage -

Interface x = new Class();

e.g. IList<int> x = new List<int>();

instead of, Class x = new Class();

i.e. List<int> x = new List<int>();

And if you can't come up with too many, please stop defending yourself.

Thanks again Karl.

Best regards,

Manoj.

#.think.in infoDose #12 (15th Dec - 19th Dec)

I bet all those who criticized Karl (Josh et all) are not coders or used to be coders some time when a code implementation like oxite was something.

Don't worry about your language and tone karl. such code needs to be criticized loud and clear because before you know it hundred enterprise applications will be build using oxite as an example just like hundreds were build using the asp.net storefront which used horrible design techniques that took me years to recover from.

OXITE SUCKS.. if you have a problem with that let me know.

@DaRage:

I hope you can read this with your head so far up Karl's you know what.

Open Source code teaches you something regardless of whether it is good or bad. (Of course there is always something good amid the not so good for any nontrivial project) If you have something better maybe you should release it to the world so we can critique it.

To sit there with zero contribution and fire shots at a team who is trying to contribute something is counter productive and adds zero value. Thats been my point all along.

Why has is it under Mix Labs?

Because they thought it would be cool to show what they are running their site on. Full stop.

To think anything from any source is gospel always is well rediculious. This is what your friend Karl has been using as a the main point of his diatribe. Microsoft should of course release the magic bullet using all the lastest glitzy buzzwords and in a timely fashion. Anything else is well unacceptable because it will confuse the new people... Don't you think this thesis is a bit well silly.

"If you have something better maybe you should release it to the world so we can critique it."

I can't lay egg but I know if it's addle...

"If you have something better maybe you should release it to the world so we can critique it."

I'm sorry, but if I've seen a horrible movie, I have every write to say (or write, or blog) it's horrible without having written, directed, produced and starred in my own film.

@Josh

Actually it is your argument that is silly. neither I or Karl or anybody has to write any open code in order to criticize oxite or any other project. And contrary to what you think, it is constructive and important to criticize loudly and harshly projects like oxite because of its association with Microsoft will likely grant it  publicity unlike hundreds of well coded open source projects found source forge or codeplex like Arch# (have you heard of it? i bet not).

The fact it's an open source project is what makes it open to criticism and not vise versa because if it was a closed source there would be nothing to criticize.

lastly, DDD, DI, SRP, etc are not buzzwords as you like to describe them. They are important scientific terms in the field called software engineering if you haven't heard of it and understanding them is crucial to anybody working in this field which I'm sure you're not a part of. Judging by what you wrote, you're somebody who doesn't write code for a living and thus doesn't have the passion to defend quality code, so just go away or i'll write some code to kick your you know what.