Jeff's Junk

Hi Jeff,

I totally agree with you. The opener the process, the more complains you'll get.

Jeff you sound like a Microsft drone :-)

You miss the point totally. I love VS 2005, I wish I could have this tool 2 years ago.

The problem (well for me at least but surely for many) is Microsoft breaking its own rules. Being compatible with the previous version and also more important for me provide something to migrate your projects.
This apply to web projects of course, windows projects are more or less intact from the earthquake.

Yes it's seems strange to see this rumbling noise suddenly now, but you must understand that after the phase OI called the 'open the box and play', now it's time to see what we will do withj our projects the one we have to maintain every day.

And even if I am a big Microsoft advocate, there no way. It's so incredibly complex to have one single real .Net 1.1 project running under VS 2005. And this is where a lot of people like me are different from the usual Beta crowd. We are working with massive or even modest web projects, and my boss don't give a damn about VS 2005 or 2010. If I convince hime to do the jump (and this is not an issue for me) I bet you he will kill me straight if I show him the days I spend on the bloody thing. And no the Beta argument doesn't apply here, not only becuase of the Go live licence, but because the tool has bugs and missing features in a way it's really doubtful to have something ready on the first week of November. Believe me, to have Scott Guthrie sending me an email in the middle of the night to say they reckon something is wrong with the migration tool and immediately sending me a set of new assemblies, for me this is a strong proof something goes wrong. And don't be ridiculous to associate this with some VB6 petition. This has nothing in common, we are talking here wbout web applications and the lack of important features like multi project building. I can't bear the idea to live with VS 2005 for 3 to 4 years without any serious promiose of updates by Microsoft.

Of course, if someone has a different opinion, they miss the point, didn't read your post, don't know any better, etc. I've heard all of that before.

I agree with both of you !

I don't think v2 is going to be a failure, nor do I think there should be a Beta 3. But the reason I don't think there should be a Beta 3 is because I know how big a deal a beta is and I want v2 as soon as possible -- because I will be gladly using it.

But then why do I make my "negative" posts and agree with the opposite view too? Because the whole point of the open process is to get our feedback, and I'm giving it -- sometimes loudly. Why? Because I want the RTM to be as good as it can be, because I don't plan on waiting until a SP.

Will I survive without a project file? Most assuredly YES. I can make multiple server forms possible, and make web apps stay alive, and do just about anything -- given the time. And that's the rub -- why should it take lots of my time to do something that is very easy today.

I agree with Paschal. The sky is certainly not falling and VS 2005 is going to be awesome. I cannot wait for it. I do have some concerns about the "breaking changes" to the project system I've found with Beta 2. Quite frankly, I never had an issue with how ASP.NET projects were set up in the first 2 versions of VS.NET. In my opinion, the project model in the first two releases made development of ASP.NET and Winforms apps very similar. Now, it seems that everything else remains consistent except ASP.NET apps. Why?

Even migration of a small product-based ASP.NET web application I built took a couple of hours to work through and around. Why? I can deal with correcting warnings, but these were not merely warnings. The thing wouldn't compile a project built with Option Stict On and Option Explicit On unless many changes were made. And the idea of introducing a new Refresh file seems like such a hack to me.

More importantly, as Paschal mentions, is the fact that scrambling going on at this point on the part of MS is worrisome, although not un-appreciated. We hear, "These things will be fixed", but isn't the idea of a Beta 2 to be feature-locked? Some of the changes being offered aren't bug fixes, but IMHO new features. Will we get a chance to fully test and report back on the changes? Without a Beta 3 or extended RC, no.

Sorry for the long comment, and Microsoft, we love the openess you've shown on this version. I hope we never go back to the old-ways. Along those lines of openess, hear we are, offering our input.

Beta3? No, that will only slow down the process
Fix more bugs? Oh yes. We all know Microsoft won't release a service pack for vs.net. They never did. So any bug that's postponed has to be fixed NOW. And I'm not talking about silly things, but things like operator restrictions on generics for example and after reading a lot about it: the webproject crap.

personally I think the current webproject is a real PITA, so I use class library projects, but I'm unsure if I can still use class library projects in vs.net 2005 for websites.

So because they never released a SP before means they never will. That's an example of one of those irrational things I was talking about.

"So because they never released a SP before means they never will. That's an example of one of those irrational things I was talking about."
I can only conclude that they never will release a service pack. What evidence is there that they WILL release a service pack? It's 2.5 years AFTER vs.net 2003 was released! Didn't they promise a service pack for vs.net 2002 right after 2003 was released?

I've given up on them for service packs for VS.NET. Perhaps we're lucky and they find it necessary to release a service pack, but I doubt it. The main reason I think they don't release a service pack is that IF they do that, there is less need to upgrade to a future vs.net, and us upgrading to a newer version is MANDATORY to get new technology implemented by developers, at least for MS. If most developers stick with .NET 1.1, stuff for longhorn will take ages to get mainstraim, at least the code based on .NET 2.0.

What evidence is there that they won't? Things change.

We have these "certainties" in the roller coaster enthusiast world too, like Cedar Point builds a new coaster every two years. Still waiting for that one to pan out.

Enterprise developers releasing software on Beta platforms? That's a complete oxymoron. You may work for big companies and produce large scale systems, but they are not "Enterprise" systems otherwise you'd not even be talking about Go Live licenses and that sort of thing. I'm all for using VS2005 for getting a head start and laying some foundation, but in my opinion the serious amount of risk you're taking working so much with a Beta product is frightening. Until the RTM version is released your playing with fire in my opinion.

I knew someone would chime in with chest thumping. Thanks!

>>>We all know Microsoft won't release a service pack for vs.net. They never did.

Really? Then what's this link, dated 8-Mar-05?

Microsoft® Visual Studio® .NET 2002 Service Pack 1
http://www.microsoft.com/downloads/details.aspx?familyid=c41d8159-b42f-4d06-a797-e510494976ee&displaylang=en

Yeah look at the date: 3/8/2005. Sorta pointless if we have to wait that long for a VS SP.

"I guess this is the down side of having a very public and open beta process. "

Actually it is the problem of having not so open Development Process. Do not know how many times I have gone in and changed Open-Source software because of small issues like this. I know, I know M$ ain't going to do anything like this, but to do the least, should listen to the Developers, Developers, Developers, Developers, Developers, Developers, Developers...(I forgot how many times Steve Ballmer said that. But I guess you get the point).

>>>Really? Then what's this link, dated 8-Mar-05?
>>>Microsoft® Visual Studio® .NET 2002 Service Pack 1
>>>http://www.microsoft.com/downloads/details.aspx?familyid=c41d8159-b42f->>>4d06-a797-e510494976ee&displaylang=en

Waiting 3 years for a service pack isn't acceptable. Six months to a year, yes; 3 years, no. The version after VS 2005 will probably be out in 3 years time anyway.

It's not about how great the Intellisense is. THE ISSUE IS: It takes me 3 days to develop an app, 10 days to workaround your so-called "Quarks".

I am ripping my hair out over here trying to publish a web project 2005 to a remote web server. No dll, no project.

I am ripping my hair out over here trying to design a DataSet with the new designer.

I am ripping my hair out over here because objects do not dispose or do not do what they are documented as doing.

I am fed up with people that have this utopia, oh the sky is so blue over here, never had a bad dev day in their life, think they are better than everyone else, act like they made this sh%, and try to talk up this billion dollar company that's wasting my damn time when it shouldn't take me so long to do some little things!

Look, you might say, "If you don't like it, don't use it..." in fact, I say that! But you know what? MS, whether you trust them or not, has a major damn monopoly on the software development world, and software world in general. Smile if you want to but I have smiled so much my damn face hurts and now I'm pissed. I can't fake it anymore... MS is great for people that like to tinker with toys but I am trying to roll products out the door here people!!

</Gripe>

BTW... I found a solution to my web project issue! Don't use them! It works! Now that's what I call a feature MS! Tell all of your customers that they really don't need to use Frontpage Server Extensions, then when you release the project, make it so that you can't publish sites to remote servers without frontpage server extensions.

Then take out the dll from developers projects all together because you know us web developers, we dont' know what them things are anyway. We all just code inline like those nifty macromedia MX people. Heck, we should just model vs.net off of macromedia MX cause MX is super neato.

I can't copy my project or publish it to remote sites like I could in 2003. But it's better than trying to publish a project for 3 hours in vs.net 2005. So some inexperienced people take over MS's flagship product. Now get all these great new features!!! Weeee!

Sounds to me like you just don't know how to adapt and don't understand the product. I've not had any problem publishing sites via the built-in FTP. No assembly required, and so what? Actually, I take that back, usually I have a reference to a class library project, and sure enough, that project's assembly makes it into the site. You're too hung up on the business of physical files, when it's all about objects. Seriously, if it takes you three hours to copy a project, that's not Microsoft's fault.

Who has ever used DataSets with the designer? Who uses DataSets at all?

Show me which objects aren't disposed.

And for that matter, if you're "pulling your hair out," then maybe you should not use a beta product and bitch about how much you can't stand it.

Beta 2 to RTM? Where's beta 3, RC, RC 2, RC Gold??? Maybe I don't understand the point of these web projects but that's the whole point, at least I understood the other kinds of projects. As far as understanding the product, I'm an MCSD/MCAD .net AND I had to live through understanding vs.net 2002.

It is microsoft's fault that it takes three hours HTF is it my fault that they promise you don't need FP server extensions for remote publishing but you really do? I don't want to run an FTP server on my box.

Who uses DataSets? I use DataSets. Lots of people use DataSets with the designer. Why rinvent the wheel? Especially with the new binary serialization formatter and the major performance improvements. But then they go and break the designer. Besides, you do realize that DataSets and XSD are the future, don't you? WinFS uses DataSets and XSD, SQL 2005 uses DataSets and XSD, and web services now support binary serialization for DataSets and XSD.

You also think it's a good idea to publish your source code when going to production? I love the fact that you compile web projects. But now those MX users are even happy to break the code when they can. What's this App_Code and App_Data too? MS talks about how it's so important to use nTier architecture and then throws their Data access layer in their default web project. Give me a break.

Create a new instance of an IsolatedStorageFileStream and use it in one section of your code:
DataSet myDataSet = new DataSet();
using(IsolatedStorageFileStream isolatedStorageFileStream = new IsolatedStorageFileStream("MyXmlFile", FileMode.OpenOrCreate)
{
myDataSet.WriteXml(isolatedStorageFileStream);
}

now in a different class... try to open and read the isolated file... the stream remains open. The GC is supposed to dispose/close resources that are required by other objects. Instead you get a FileAccessException because the file is in use.

You also need to take a look at some of these responses from Microsoft. Bugs that make certain 'features' unusable:
http://lab.msdn.microsoft.com/productfeedback/viewfeedback.aspx?feedbackid=c2f7d047-7eda-47fa-bc8a-5801b9e0d271
http://lab.msdn.microsoft.com/productfeedback/viewfeedback.aspx?feedbackid=4a93e304-4ae7-4519-96e1-70f3a0202b6a
http://lab.msdn.microsoft.com/productfeedback/viewfeedback.aspx?feedbackid=963554f0-cca4-45e9-8431-b0b70da5251f

The DataSet designer was almost awesome in 2003. Now it's unusable. That's my point!

BTW, that IsolatedStorageFileStream issue and many other bugs are not just in vs.net 2005, they're still in vs.nte 2003!!!

Did I mention it was beta software? I'm sorry, did you pay for VS 2005? That's about what I thought.

>I'm an MCSD/MCAD .net
Good for you. That doesn't mean anything other than you can memorize stuff.

>HTF is it my fault that they promise you don't need FP server extensions for remote publishing but you really do? I don't want to run an FTP server on my box.

So then how do you expect to get the files on the server? They give you every possible option. They all work.

>Lots of people use DataSets with the designer.

DataSets blow. They're overkill for 90% of what anyone does in Web projects on a daily basis.

>You also think it's a good idea to publish your source code when going to production?

I didn't say it was a good idea, and I don't. What does that have to do with anything? You don't have to publish the code.

OK, so you found a couple of bugs, and they're in a beta product. It must suX0rz teh big one!!!111

Seriously... you're painting a much darker picture than what's really there.

I just have to say, a lot of you people are whiny little bitches. You have no value to add to this subject, you just want to be heard and you don't even know why, you probably spend a half hour editing your post and still it comes out sounding as though it were written by a 15 year old.

A couple points. Backward Compatibility for DEVELOPERS is a pipe-dream. If it worked for you in 1.0, but has changed in 2.0, tough shit. That's called advancment. Backward Compatibility for your users is infinitely more valuable, software that breaks because of technological advancment is software that doesn't sell, therefore Microsoft has stood by their word on backward compatibility. V1 apps will run on a system with a V2 framework, V1 crutches non-inclusive.

The vs2002 Service Pack released was released shortly after vs2003, not 2005, the web link itself was updated. Anyone who thinks they have the upper hand by declaring that it was a late service pack doesn't know their head from their ass, and even if it was a late service pack 99.9% of you whiners don't write commercial software, at BEST you create a website and MAYBE even an intranet portal. You haven't actually experienced any of the 'real' bugs that exist in the framework and those that have probably thought it was THEIR OWN FAULT IN THE FIRST PLACE.(really, I'm laughing inside, because it's funny).

And Tom,. I don't know what your problem is because I really haven't had any problem publishing web projects to my web servers, nor building, I still see my DLLS goign out and I still use codebehind as it is approriate. I'd like to sit here and pretend I'm a wh0re but I'm not, and the fact is this is a beta release, that when released, was a 3 MONTH OLD BUILD, that's right.. it was already outdated before it was shipped. The majority of problems peopel are having have already been fixed. You shouldn't be using whidbey for anything your bread and butter relies on anyway, are you insane?

I'm never hiring an MCAD/MCSE, but then everyone I know personally knows that about me.

The DataSet designer was always a packacge of shit, and it's only just now getting better. Most peopels problem is they really don't understand what's going on internally. If you had any idea, you'd be lauding the new designer as the future direction of code-generation and data modelling.

But no. You complain it's unusable, even go so far as to claim the original peice of shit they called a designer was usable. No, it's not the friendliest thing on the planet, yes it still needs some keyboard improvements, and I have no doubt they will be resolved before shipping, I of course use PDC2000->V1.0 as a template for this shit, Microsoft generally ships a usable, stable product that serves the exact purpose it was intended to. The path to 2.0 isn't a clear one, stop talking as if it were.

Delete the beta. None of you are worthy. Some of you obviously fail to understand what beta means, and in your correspondence with Microsoft you're totally inept at comprehending exactly what they have told you. You've become doomsayers because you WANT to.

Bleh, I have better shit to be doing.

Suck it up.

And oh, you don't need FP server extensions to publish, as I still don't use them at the office. A fileshare does help though?

Try 2 minutes to write down my ideas and 3 to revise...

You do need frontpage server extensions to publish to a remote server, if you are not using FTP. If you read anything below the third paragraph in my last post you would see that the issues I am having are not just in vs 2005 or .net 2.0, they're also in .net 1.1. Did I pay for MS Software? No, but who does and who should?

About that SP1, did you ever see that check for updates feature work for you in 2002 or 2003?

DataSet designer: add 20 tables to one schema and try to use it. I thought it was great too, until I tried. The performance is worse than before. It takes over a minute to refresh the schema diagram because it generates the code every time. Are you sure that you know how everything works under the hood? Navigating anything over 20 entities in the schema designer becomes difficult. Create a new XSD Schema in 2003, add two elements and move one out of the view. Go to View>XML Schema Overview. Did you know that was there? It isn't in 2005. There's no zoom in zoom out tool or a high level navigation view, as in the previous version, Look at the code generated by the DataTableAdapter. They're using string concatination to build the SQL statements. Why that's bad? You probably wouldn't know.

The class designer is great and model driven development is where it's at. But it's like 4 steps forward, two steps back. We are making progress...

Where exactly do you find the DLL for your web project? Where is the option to publish a web project without the code? There isn't one.

What do you know about the software I have or have not shipped? Would writing software for the largest retailer in the US count as experience in your book? Would developing software for the worlds 3rd largest software company and shipping to the worlds largest retailer count? Would working with the developers that created Amazon.com, Walmart.com, to name a couple count? Would presenting to over 10 of the world's fortune 100 companies CEOs, CFOs and CIOs count? How about working directly with Macromedia, Charles Schwab, the entire Taiwan Technical Deligation? Don't make yourself such an ass you little Microsoft Koolaid, cowboy, think you know it all cause your mom pays for your gas.


Would developing cutting edge vector based UI's integrated with web service back ends, deployed to millions of users count? Look who's talking about shipping software LOL... MS has slipped on their dates over and over and over again. So don't use them as your model, princess. You sit here and talk all this hype about vs.net but you are just a little bitch. Core .net bugs or general OS infrastructure problems have forced us to cancel huge projects. You don't know the millions and millions of dollars, 500 million to be exact, that I've seen fly out the window because of issues like these... Huge teams put out of work because even Microsoft tech support couldn't help figure out why it didn't work... Microsoft built labs for me to help figure out some of these issues and they spent weeks trying to fix them... yet they couldn't seem to fix it.

Don't take offense to the fact that I am pointing out flaws. I am a developer and I understand that it's not easy to ship bug free software. You might be right about most of the bugs being fixed in the final release and I wouldn't know yet because I don't have it. But what's really scary is that we have had to deal with issues from vs2002 all the way up to today with service packs coming AFTER a entirely new release of the product, with new bugs...

I am really looking forward to 2005 and .net 2.0 but I would rather work with a product that I can count on 100 percent.

I'm not taking offense to anything... nor am I the one chest thumping certifications and my resume. I'm still not impressed either.

You only need FP extensions if that's how you want to put your code on the server. You can also use FTP, UNC, or burn a CD and take it to the server. What else do you want?

I see plenty of compiled assemblies, and one step better, no .aspx markup at all when I use Build -> Publish. No previous version did that. Even then, if I was doing a straight copy, the uncompiled code on the server in code-behind or in the pages is little more than glue to tie the UI to the class library I write in a separate project. That's hardly a revolutionary way to build an ASP.NET site. Things like app_code and the like are for the benefit of hobbyists, people building small apps, or people that just don't need to a hundred tiers to an app to say "hello world."

And if you pissed away half a billion dollars on an IT project, then that's your fault, not Microsoft's. There isn't anything in the product that would ever merit that kind of loss.

I have to agree with much of what Tom is saying. If the RTM is even close to what Beta 2 represents now in terms of outstanding bugs and performance issues (and there's no reason to think it won't be with the amount of postponed issues at the Lab) then I'm not sure Visual Studio 2005 will be the best bet for my company and my personal development. I think Microsoft should've focused on improving what was lacking in VS.NET 2003 and maybe add some core features such as generics, GridView, TransactionScope, etc.

plex, DataSets and DataTables are used extensively in practically every project I've worked on. I'm also familiar with the custom business objects model as well, and even in my custom business object model I use untyped DataTables to retrieve data from the database.

In VS 2005 I have a single DataSet schema that contains all my database tables (without relationships--I have to manually delete relationships automatically created for me every time I configure a TableAdapter in the designer, it's a real PITA). I then derive each of the data tables in my business tier and add business logic. It's a very good model and it removes a lot of the plumbing code I had to write in VS.NET 2003. I love being able to manage the data objects (DataTable) and data components (TableAdapter) in one place. It lets me worry about the BUSINESS side (the important side) of my application rather than worrying about select statement parameters. The DataSet designer in Beta 2 is buggy and pathetically slow. Anything larger than the smallest pet project renders the DataSet designer useless. Just take the Northwind or similar database, import it into a DataSet and try and use the designer. It's a joke.

I think Tom's concern is (and I know mine is) that Microsoft is changing what were otherwise good designs between VS.NET 2003 and VS 2005. Eliminating Web projects (I know they said they were reimplementing after Beta 2 but I'll believe that when I see it, and in what form I see it). Removing key DataSet designer features like zooming really hurt the product. You should remember that just because you may not use DataSets doesn't mean no one else does.

Also, Tom's comment about tiered architecture is correct. Strangely, Microsoft seems to push people *away* from tiered architectures by introducing such concepts as SqlDataSource controls. I couldn't believe it when I first saw an example of an asp:SqlDataSource control.....complete with inline SQL statement. Are these the kinds of precedents Microsoft wants to set? We need to move toward better development practices and features such as these hurt that effort.

True, Beta 2 is not far from being probably the most kickass development tool ever created. Combined with [now delayed until 2006] Team Foundation Server it will definitely improve a lot of the bad things developers must deal with currently when coding with Microsoft tools (source control being the primary example here).

It would be great to see a Beta 3. We've all been patient, we can be patient a bit more. The majority of production code does not ride on Microsoft's release date for VS 2005. I hope Microsoft clears up some of the biggest issues (performance and scalability) before RTM.

You and everyone else act as if the bugs you find are being ignored, and they're not. Beta 2 isn't going to be anything like the RTM version.

Of course they are not being ignored. All the bugs I'm referring to have had Microsoft responses. Unfortunately all the responses are Postponed and as far as I know that means postponed until Orcas. Additionally Microsoft has stated numerous times in many of these Lab issues that Beta 2 is feature-complete and they are only working on the most critical of bugs. How is it that RTM won't look anything like Beta 2?

There's a difference between a *bug* and a *feature* that's missing.

Take a look at the first thread in this link:
http://netdialect.com/2005/tabid/103/ItemIndex/-1/Default.aspx

As mentioned above, the feature missing in the article is not the only postponed bug.