Guillaume Roques Response - What he really meant...

First off to everyone that has linked to this blog and even tweeted... yes traffic came from 2 tweets :) . . . I appreciate it. Today 205 readers checked out DOT BLOAT alone and over 1300 views since I have started blogging again, which is pretty cool. Is there a possibility momentum is building for the truth about Visual Studio? By the way, if you are linking to this blog and want a link back let me know and I will add it to the side bar.



A bit about Kinnect
Soma in his most recent blog mentioned Kinnect. Ok so my grandson has an xbox360, which is CTP 7 / RC3 / SERVICE PACK 210 TO BETA RTM 6 of the XBOX 160 which (in case you forgot) contained only pong and pac-man (just kidding). Anyway the kid is asking for this kinnect thing and more then likely I am going to get it for him. But anyway in that spirit here is a brief infomerical about kinnect before getting into the "bloat" of what I have to say. (Yes it is a repost but still funny)



Soma's Reply
I seriously appreciate the fact, Soma posted a reply and my comments. Below is what he wrote.


My blog isn’t moderated and comments should show up within a couple minutes of posting. Sometimes MSDN’s automatic spam filter triggers on blog posts, as it seemed to have with yours, so I’m posting it here:

.... My comments were posted here in his blog ...

Mark, I do want to tell you that I hear your concerns. You are passionate about Visual FoxPro, which is great. We will have to respectfully agree to disagree about Visual Studio and .NET. I hope we can lay this topic to rest.


While there is not much in his reply, I am glad he read my concerns unfortunately he didn't address any of them. I would like to clarify one point. I am "passionate" about "productive" development tools. I am passionate about the FoxPro paradigm and the ease and speed in which applications can be written. If Visual Studio was at least equal to or more efficient then VFP, I wouldn't have anything to complain about. I truly feel the community would be best served if developers voiced their opinion about the bugs and problematic paradigms of Visual Studio in an attempt to improve the tools and frameworks Microsoft is releasing. Our job is hard enough without tools and bloat getting in our way and slowing us down.

Now What I want to do is break down Guillaume Roques's response.

If you are a bit confused, check out my prior post before proceeding.

But first, lets pull out the questions in Tom comments.

1) Why was Ajax Toolkit discouraged and why did you start touting jscript all of sudden? Just because of open source pressure or that they out-witted you guys?

2) Why are you guys promoting MVC. What is the problem with regular way of doing ASP.NET sites?

3) Do you have scientific or statistical proof that MVC is better than the original approach?

4) How many .NET developers actually use MVC?

5) Why did you guys discourage LINQ-To-SQL and started encouraging Entity Framework?


Guillaume's reply to Tom - What he really meant.


Tom:

When we think developers that are consuming Ajax it breaks into two segments:

a) Developers that want to write some client side script in their pages.

[Ok I have to ask this, what is the hang up with developers who want to write code? This makes absolutely zero business sense! Read that again ZERO "BUSINESS SENSE" either your client is paying for your play time or your company is. Either way it is a "waste" of time in most cases.

This is so simple, each line of code we have to write takes time and secondly no matter how simple the code is, each line of code is possible point of failure in your application. Folks, misspelled variables names, logic errors, unclosed brackets etc ... whatever the case is HAPPENS! Unless I'm the only asshole programmer that makes these type of mistakes and "the get close to the metal VS guys" are freaking perfect? In terms of html, jquery and javascript, in Visual Studio 2008, it is even worse since the compiler won't even catch the majority of the errors nor is intellisense fully implemented! There is a simple saying "time is money" and that is true for a reason. If someone has a trick to writing more lines of code by hand in less time then reusing a UI class for example please fill me in.

Let me make this clear: Programmers are hired to deliver a solution. Programmers are not paid to get an erection and masturbate over a link list implementation they wrote or the 500,000 lines of MVC code using 20 different technologies you created for a 20 pages website. Most client's could give a shit about your algorithm artwork. The primary reason for coding standards and writing clean code is for maintainability NOT to address some twisted technology fetish you have. If you want to write code use C++, C or better yet assembler trust me you can write a ton of code in those languages, I can speak from experience I started coding on a PDP11.

Moreover you guys are using .NET OS WRAPPER CLASSES and JQUERY to do most of the work for you, what metal are you getting close too? Cool you can type a variables or object and you know how to write some bullshit plumbing logic or call some bloated bastardized os wrapper class framework method and your head swells. Not all programmers are bad, but I am tired of the arrogance that seems to be running rampant in certain circles and with some individuals. First the VS guys looked down on VFP and now MVC dudes are looking now on webforms - isn't the solution what matters? Yeah I am just venting I will get back on track.]


b) Developers that want their applications to automatically take advantage of Ajax functionality.

[There is nothing automatic about the the Ajax Toolkit, it is merely the lesser of the two evils.]

Our investments into jQuery

[What was Microsoft's investment in JQuery? Some cash to buy licensing? Adding intellisense to Visual Notepad 2010 or copying the library functions, JQuery is an open source toolkit.]


are targeting people in the first batch while the Ajax Control Toolkit and the rich server controls offered by third party vendors target people in the later case.

[Wait a second, offered by third party vendors? WTF - Microsoft wrote the Ajax toolkit and the default set of server controls. I know Microsoft has A LOT of technology, more then most developers want, but does Guillaume even know what products Microsoft offers?]


We think there are developers in both of these segments and are trying to target them both just like below in the MVC section you can see we are trying to target multiple types of developers.
[What happened to DRY? Maybe Guillaume didn't get Scott "THE DOT" Gu's email or read his blog. (Come on did you think I wouldn't mention the GU!). VFP and VB developers got targeted alright, with a freaking bullseye.]

This isn’t about choosing one Ajax strategy over the other you can use one or the other or both together.
[This is a crock, first he never answered the question and second can someone tell me when Microsoft released an Ajax toolkit update? If Microsoft is no longer updating or supporting a toolkit how can Microsoft advocate using it?]

With regards to MVC, we have had in the past one framework for developing Web Applications and that is ASP.NET Web Forms. This framework was designed to make the transition from developing desktop applications to developing server-based web applications very easy it also gives developers great benefits in productivity
[Guillaume should of said, this tool has the "POTENTIAL" to give developers great benefits in productivity. How many releases do we have to wait until the webform designer gets fixed? That designer is a bug ridden mess. Moreover the html generated by the rich set of server controls and the form designer doesn't even render pages consistently across different version of Intercrap Explorer without hacks and work arounds. Most developers end up writing the html by hand in Visual Notepad I meant Studio, how is that productive? Moreover why can't we visually subclass UI controls in Visual Studio's non-functioning class browser so we can reuse them?

Let's proceed I think you get my point.]

by letting them leverage rich web controls that have lots of built in functionality


[True they do have lots of functionality unfortunately visual studio doesn't support all of the functionality. For example, the listview rich server control which by the way do we really need 6 different data grids, Microsoft simply omitted the UI and made it read only so we are coding by hand to implement it. That is sloppy crap development plain and simple. How many of us could sell excel as a custom solution to a client and expect the client to key in the formulas? Maybe we can use Microsoft's Marketing spin and explain to the client how they are getting close to the metal and this is a feature? This is exactly what Microsoft is doing to their development community. No sense beating this to death Guillaume thinks webform controls were written by a 3rd party vendor.]


While trying to grow the types of developers that can use ASP.NET we also identified the MVC pattern as a common pattern used by many other web frameworks such as Ruby on Rails and CakePHP and as a result we created ASP.NET MVC to provide a bridge to developers that like that pattern to be able to easily adopt ASP.NET and the power of the .NET framework and more generally our Web Platform. We are focusing on providing choices to various types of customers. We will continue to invest in all areas of ASP.NET (Web Forms and MVC).

[Ok here is what I think is happening: Microsoft is scared shitless of the open source community, Google's endorsement carries a lot of weight and showcase sites like facebook give it creditability, more importantly there is NO way they can buy out their competition this time like they did with Foxpro.

Microsoft is simply trying to duplicate open source technology as close as possible. This is a terrible strategy for one .BLOAT is not required in open Source and open source developers are not going to change teams most hate Microsoft. Even if these hurdles didn't exists there is still one other problem - COST! What Microsoft did accomplish with MVC is split their own community, great move.

The only way you compete with open source is by building a "BETTER" and more productive environment. This concept works, Microsoft did it with word going against Word Perfect and Excel went up against Lotus. Microsoft needs to scrap the API wrappers known as the .BLOAT framework, build complete development tools that work well and start changing their image, this is NOT a difficult formula. Until they do they are going to struggle in the web platform arena, if they haven't already lost that battle.

Let me throw this fun fact out there. MVC was invented at Xerox Parc in the 70's this is not NEW. If Microsoft was interested in this paradigm they could have moved in this direction from the start with interdev. Microsoft's movement towards MVC is a hail mary pass to try and compete with open source frameworks. Microsoft flat out doesn't have the confidence in their software development toolkits that they can stand on their own merit and attract developers.]

From a data perspective, ASP.NET works with any data solution in .NET but Entity Framework has been getting better and better with each release

[Given how bad EF is, they are not exactly setting a very high bar!]

and with the release of EF Feature CTP 5 we think that it is very compelling for web data scenarios and are working to provide more guidance on how to use it.

[Guillaume wrote "working to provide more guidance", am I the only one that had a WTF moment? Is Guillaume stating Microsoft really doesn't know how to use EF? If so then how in the hell can he say it is compelling to use, when they are still trying to provide guidance on how to use it?

What is so f#$ked up here is Microsoft, since the birth of Visual Studio, has been trying to implement a data solution and FAILS over and over again! Yet they have Visual FoxPro laying there and it F#$King works, hell most of the SQL Server indexing internals are based on VFP'S rushmore technology. Does it seriously require a brain surgeon to say hmmm maybe we should put a data centric extension language into Visual Studio.

The real issue is, Microsoft is not looking to make developers productive, their data goal is to build some type of data access strategy that the community will accept which will lock web and LOB applications into Sql Server. Move over this is CTP5 which means they released 5 versions of this thing and still can't get it right! This CTP/RC/RTM notion is BULLSHIT this is "RELEASE AND PRAY" software development. Microsoft is "using" the community as a group of "FREE" BETA testers throwing crap out there until something sticks. By labeling software in this fashion they can put this stuff out there without having to worry about quality, analysis or if it even works right. GU can always write on this blog well this is CTP6 no worries we will have it fixed when we "GO LIVE" sometime in 2020 and CTP7 will be out in a couple weeks. When FoxPro went Visual/OOP in version 3.0 there was only 6 total releases and couple service packs for each release. EF is only a data technology folks it is not even a complete development tool and they are on their 6th attempt. While they know their goal, SQL Server LockIn, they are clueless how to get there or better stated what the community will tolerate.

This is important so let me state it another way, as long as "connection string based" data access is around and implemented using standard Structured Query Language and stored procedures, portability exists and Microsoft can not accomplish the goal of SQL Server Lock In! I believe the syntax structure change in LINQ was introduced to force developers into a new paradigm and out of SPROCS to raise the barrier of entry into MySql. The bottom line is SQL Server is Microsoft's cash cow in the development area this is an expensive package and MySql works and it's free!


Guillaume

Director of Product Management - Microsoft

Now let's return to the 5 questions Tom asked and see how many were answered.

1) Why was Ajax Toolkit discouraged and why did you start touting jscript all of sudden? Just because of open source pressure or that they out-witted you guys?

Guillaume didn't even know Microsoft wrote the Ajax toolkit.

2) Why are you guys promoting MVC. What is the problem with regular way of doing ASP.NET sites?

Guillaume tried to spin an answer. It short we can't beat them so let's join them!

3) Do you have scientific or statistical proof that MVC is better than the original approach?

Guillaume ignored this one.

4) How many .NET developers actually use MVC?

Guillaume ignored this one as well. Many people have been asking this question and Microsoft has been silent. If the MVC numbers were high, I think it is safe to assume Soma would be blogging about it. Other then the hype of something new and the cheerleaders spinning MVC, my opinion is not many companies have jumped on this bandwagon.

5) Why did you guys discourage LINQ-To-SQL and started encouraging Entity Framework?

I think Guillaume will get back to you Tom once Microsoft figures out how to use EF. IMHO To lock developers into SQL Server.

I know this was long - there was even more there then I initially thought. If you made it this far, I hope you enjoyed reading this post as much as I enjoyed writing it. The one good thing about blogging about VS there is never a lack of material.

Until Next Time.

.Mark!

MVC AND RAZOR BIZSPARK AND SCOTT GU AKA DOT GLUE

This is my first blog post in a while, I took some time away from programming. To bring you up to speed, I stopped using Visual Studio and .BLOAT and I made the switch to PHP, MYSQL and playing with Ruby a bit. Seems odd after 20+ years of programming and strictly with MS Tools since VFP 2.0 and interdev but it is worth it. Granted the open source tools may not be anything to get excited over and the syntax a little cryptic but it is nice to know you can actually learn a language or technology and it will be around a month from now.

On a side note: many of you may not be aware of this but I was a Microsoft cheerleader for VS 6.0, Sql Server and VFP 9. Now I dread even touching their software. My main pc setup I run linux and open office and another PC running XP used for Visual Studio and testing the sites on I.E. so I can put all the hacks in required to render a page correctly.

I just read the blog postings on the Internet about silver light, now Microsoft is back pedaling with silver light after they slipped up and let the cat out of the bag that this is a dying platform once HTML 5 is released. Yeah so what netflix uses silverlight I wonder how much Bill Gates paid them.

Here are some other tool sets or frameworks (I use that term loosely related to Microsoft) if you still think Microsoft has any plan for it's development product line, pop fly is in the grave after all the hype my Friend Soma wrote on his blog. From what I been hearing from a reliable source the Linq to Sql team has been dissolved and the ADO.NET team is taking it over, which they have already stated they are no longer enhancing the product. So the BLOATED slow and unstable entity framework with LINQ will be your tool of choice for data access, have fun! Remind me again when was the last time the MS Ajax toolkit has been updated?

Now all my new development is using open source development tools, I love knowing I am no longer feeding the Microsoft machine or supporting the economy of India. I am so glad I am outside the webforms mvc drama, hats off Microsoft for splitting your own development community. I still get the occasional client for which I have to painfully open Visual Studio. This one last week I had was blog worthy.

I had to implement several changes to an MVC website, I open up the source code and it is over 2 freaking million lines of code using the BETA version of MVC which is not compatible with the production version, WTF! Folks it took 2 hours to configure a computer just to maintain this code. Where is the productivity in that mess.

Trust me THE GLUE and the MVC lovers are full of shit about MVC. This company has 15 programmers supporting this single website and still required my assistance in implementing these changes because it turned into such a bastardize amount of spaghetti code and the performance was subpar. How can you expect anything else when the following array of technologies are used:

1) .BLOAT Wrapper classes
2) Jquery
3) Javascript
4) Blend
5) C#
6) CSS
7) HTML
8) Visual Studio
9) MVC
10) Webservices
11) SQL Server Stored Procedures
12) Entity Framework
13) LINQ
14) Dataset - to fill in the holes, like reporting, where LINQ and EF don't work
16) AJAX
17) ASPX - for somes pages that the MVC framework was to time consuming to use.

Folks this is what DOT GLUE doesn't tell you about the learning curve and technology overhead when you get into Microsoft MVC and blindly implement stuff after reading the blogs written by cheerleaders. Seriously MVC is 1970's programming writing htmlhelper a million times and alligator syntax it is seriously a step back closer to classic ASP.

Sure there are so many things wrong with this scenario as personally I wouldn't recommend using any Microsoft tools until service pack two is released (but the problem is Microsoft tools never make it into production anymore with any of their development tools they just call everything a RC, CTP) and definitely would NOT use beta software on a production website, but besides the point, 2 million lines of code what the hell are people thinking. Is that worth getting "CLOSE TO THE CHROME or METAL" however microsoft and the cheerleaders are spinning Scott Gu's and the Visual Studios Team's inability to build an innvoative productive development paradigm, write contols, class browsers (that works) and build designers. Given they can't write anything worthwhile on their own they copied Rails or at least tried to in their own BLOATED fashion. Disagree it is bloat refer to the above list.

Look point blank, Microsoft's MVC has zero RAD tools, no controls and for the most part you are on technology overload. Are developers that in love with .BLOAT that they pay to use an over bloated set of Operating System Wrapper Classes in controllers? If you want the MVC model use Ruby it seems to work well and the best part is it is not based on Microsoft's .CRAP wrapper classes.

What I find amusing is how Microsoft's is scrambling now to release the new MVC Razor view engine in hopes it fixes all the problems with their current view engine for MVC.

Then I read Scott GU'S blog and there are all these want-a-bees suck ups praising DOT GLUE for MVC release candidate 78,233 while DOT GLUE is posting comments about all the hacks and bugs in that framework and it has only been out a couple days. Didn't MSFT test any of this software? given DOT GLUE only writes elementary school tutorials on his blog that the want-a-bees cream over you know he won't find a bug in that framework, what a freaking joke.

Yes you read that correct DOT GLUE is stating in short YES WE F'D UP ANOTHER RELEASE and still the MSFT cheerleaders and want-a-bees are thanking him. This is directly off the DOT GLUE'S blog. "... If you installed the async CTP and the new MVC RC3 .... You will likely experience problems in VS with debugging (potentially other things)... I didn't paraphrase the (potentially other things) DOT GLUE wrote it - so what he is saying is he doesn't even know what the hell will go wrong.

Every person that posts thanks scott or cool, great job, I have a question for you what the hell is wrong with you? Look I will spell it out for you THE SOFTWARE DOESN'T WORK OR INSTALL CORRECTLY without jumping through hoops and Microsoft is even losing control of their technology base. Let me ask you this when was the last time a client told you good job for giving them software that is unstable? Moreover if you bought a car and the door wouldn't open and you were told just crawl through the trunk would you be happy. Why do so many developers HOLD MICROSOFT and DOT GLUE to such a pathetic level of competance?

The funniest post was some dude wrote: (and I am paraphrasing here) I opened up the MVC release candidate and I started to cry because this that and the other thing didn't work.... and DOT GLUE wrote some crap back like don't cry blah blah blah. I literally became physically ill reading his blog. Folks there is a reason this tool is ONLY being used by a small number of websites, regardless of how much spin DOT GLUE puts on it, its a trainwreck!

After being involved in the open source community for months now, I think there is such a negative impression and distrust towards Microsoft, and rightfully so, there is not a development tool they can release (and that includes Webmatrix - which is like an unfinished version of "access" for web development and still .CRAP based) that would gain back the developers that have left their camp. Soma is blogging about giving .BLOAT away under the Biz Spark umbrella to trick users into using it. Look, Microsoft doesn't care about your success, just ask any VFP or VB programmer, this is a marketing gimmick - don't buy into in. Microsoft is fully aware of the cost to change platforms so they give you their tools for free for 3 years (which is how long its take to write 100 million lines of code and hire a tribe of indian developers to use .BLOAT) knowing there is a high cost barrier to change platforms.

Unless Microsoft changes course, gets their head out of their asses and fixes this mess in 5 years the only folks still using .BLOAT and Visual Studio will be intranet corporate enterprise accounts, it will be a non-existant player on the web - sort of like BING (But it's Not Google search engine) and Live (is that even still around?).

Don't expect me to rip open source EVER, the software is free and it is the lesser of the two evils. Moreover the open source community didn't issue an end of life on every single development platform they had that worked, like Microsoft did. I find it completely ironic there is MORE stability in open source then with Microsoft tools.

Enough for now take care everyone.

Your Friend
.Mark

By The Way, If you read my Bing blog post when it first was released, I predicted it's failure and sure enough it happened, so what does Microsoft do, it gets balled up with yahoo and buys Bing's way into facebook and still can't make any inroads. Bing was written using Visual Studio and Google and Facebook weren't, need I say more.

Who is really to blame for Visual Studio?

Scott's Gu posted a chapter from the new MVC book on his blog, which I finished reading a few minutes ago. I was amazed by the backward progression of "modern" development tools and programming techniques. Scott Gu, Microsoft and the cheerleaders act as if this debacle known as Visual Studio was created outside the sweep of history, or that they cannot possibly glean wisdom from yesterday's technologies. Is Scott Gu and Microsoft really so bright, so brilliant, so ingenious, that they need not consult with the wisdom of prior development tools used in the 6.0 days? Yes, the old fashioned musty things - can be our salvation - if only Microsoft would install them and realize how productive these "legacy" technologies are in comparison to .NET. Instead of learning from experience thereby building on the "legacy" knowledge base in order to guide future development tools, Microsoft and Scott Gu in a clear lapse of thought insist on reinventing the wheel. Their exercise in stupidity comes at great cost to developer productivity.

If we could provide only Lotus 1-2-3, as the next revision to our complex business systems that our client possess - then merely instruct them how to manually enter data into cells and create formulas while cheerleading the control they have over their data and complex algorithms they "get" to happily create spinning the extra work as a benefit -while we know know full well in their ignorance we would receive payment along with praise in the process for introducing them to this "great new user definable flexible technology" would we? If we answer yes, then we can't necessarily blame Scott Gu, Soma and Microsoft for the current state of Visual Studio and .bloat as we became the ignorant end users.

Perhaps we should blame the people who offer praise for Microsoft, pathetically thanking them for providing "new" technology and development tools that mandates hand-coding a user interface - employing coding techniques predating FoxPro 2.0 - which on a productivity level equate closer to Mainframe COBOL. The cheerleaders as much as Soma, Scott Gu and Microsoft bear the responsibility for our current state of affairs. Have we grown so accustom to horrific software by Microsoft, that lack of Visual Designers, debuggers and nonsensical glue code we now find praise-worthy instead of appalling?

If we don't expend the energy and time to speak up, we can only blame ourselves, when Visual Studio 2012 forces us to use technologies that resemble assembler.

.Mark