Just curious what everyone's take is the RTM bugs showing up with Visual Studio. Arguably, this release has had the most beta testing and community input than any previous release, and yet it seems to have the most crazy bugs in it that weren't found. I just saw another one from Osherove. What's the deal?

Jeff...

Well, it's 45 million lines of code, so there will be bugs in the code, no matter how long you test. I think everyone's main gripe with it is that once a bug is found, you're out of luck, because you almost know for certain there won't be a patch unless you call PSS and ask for it, running the risk of paying for it as well...

It looks like they're taking a lot of heat today for this release.

http://news.yahoo.com/s/zd/20051107/tc_zd/164582 http://news.yahoo.com/s/zd/20051106/tc_zd/164578

One thing that strikes me as odd, why do they have a "Check for Updates" feature when they don't plan on releasing any updates? It seems to me like they have an easy way to deploy updates for VS2K5 and even if they didn't, most developers would be happy to download and install a service pack the old fashioned way.

But so far, knock on wood, I've had pretty decent success. I used it very hard this weekend and the IDE locked up on me once. That was the first time in quite a while, and I blame that lockup on Frans for making me realize that I've been having good luck so far with this version.

Otis wrote:

Well, it's 45 million lines of code, so there will be bugs in the code, no matter how long you test. I think everyone's main gripe with it is that once a bug is found, you're out of luck, because you almost know for certain there won't be a patch unless you call PSS and ask for it, running the risk of paying for it as well...

I'm just back from the VS.NET 2005 launchparty here in the netherlands (We were the only country outside the US which had a live streaming connection with the speach of Ballmer ). Soma, the head of the developer division of Microsoft was there too. I spoke with him about service packs/fixes etc. and they are very keen on getting fixes out, how that's to be determined, as well as info about known issues and workarounds. Overall a very good conversation and I think it's going to be a lot different from the past 2.5 years. The C# program lead also replied on my blog that they'll release a service pack in the 1st half of 2006, which is good news

Otis wrote:

The C# program lead also replied on my blog that they'll release a service pack in the 1st half of 2006, which is good news

Holding breath

You have to consider the following also...

1. Who are being vocal about the bugs? It is the vocal people.

2. Some bugs are being made more of than they really are.

3. Out of I expect over 20million deployments of millions of lines of code I've heard of 4 "big" issues that people are mentioning.

I for one would prefer to deal with the bugs in 2005 to get the features it provides over staying with 2003 and it's bugs without the new features.

BOb

It's not the bugs, it's the knowledge there won't be a fix till orcas. Thankfully, due to the huge amount of airtime in the past days, they turned around and will ship fixes earlier, will publish known issues better and workarounds.

If the 'vocal people' would have been quiet, that never would have happened.

Otis wrote:

It's not the bugs, it's the knowledge there won't be a fix till orcas. Thankfully, due to the huge amount of airtime in the past days, they turned around and will ship fixes earlier, will publish known issues better and workarounds.

If the 'vocal people' would have been quiet, that never would have happened.

Why? Had they announced that there would be no service pack for 2005? If so, I didn't hear that.

BOb

Probably referring to the fact that, as of now, there still is no service pack for 2003, only QFEs and an SP for the .NET framework. Looks like MS is realizing they need to, you know, actually release fixes to the public.

Jeff...

pilotboba wrote:

Otis wrote:

It's not the bugs, it's the knowledge there won't be a fix till orcas. Thankfully, due to the huge amount of airtime in the past days, they turned around and will ship fixes earlier, will publish known issues better and workarounds.

If the 'vocal people' would have been quiet, that never would have happened.

Why? Had they announced that there would be no service pack for 2005? If so, I didn't hear that.

Why don't you believe me? Please read the replies to this blogentry: http://weblogs.asp.net/fbouma/archive/2005/11/03/429371.aspx

Then, a guy from MS comes in and posts this reply: http://weblogs.asp.net/fbouma/archive/2005/11/03/429371.aspx#429452 See the 'next version' remark? That's orcas. And not released next year.

Now, take into that light the questions from Soma to me what they should do and what he told me they will do, and add to that what I heard from VC# core team developers, I can assure you, something drastic happened at the higher levels there. Not because of the bug I found, I don't want any credit for that, but because of the unified voice from the community that they want a clear and simple and reliable way to patch their IDEs with patches for bugs found.

As I said, it's not the fact that there are bugs, everybody knows there will be bugs. It's the knowledge that potentially we had to wait for 2007+ to get an update for this and other bugs found and yet to be found. That's now going to change and I'm very happy that it will.

I also would like to add: the info about some particular nasty bugs in RTM is in ladybug or not even known to the public. But not the first stop you'd look for 'known issues' when you run into things as a normal developer.

This is a problem. For the few people who read my blog or for the few people who read other blogs about other bugs... they now know there are particular issues and how to avoid them. Especially that last part is very important: how to avoid them. The issue I ran into is nasty because it will make you lose your work and your window setup, as you have to kill the IDE. So it's important you can avoid this. Because the average developer doesn't read blogs, s/he doesn't know this issue nor how to avoid it and will run into it (or into the VB.NET IDE Crash bug in RTM they postponed for example).

To get the known issues and workarounds (!) out in the open is also something that's of upmost importancy. The more is known about which bugs are there and how to avoid them, the better. It's unclear how they will offer this, but they will work on this to get this streamlined and I think that's great.

I understand that some people say: "Why all the fuss about a couple of bugs?". Though that's not the point: I didn't post the bug to start a war, I posted the bug to see if others ran into it as well and to get it validated on ladybug plus to see if the community could get MS in gear to release a fix IF it's a bug. The more fuss, the more chance they will do something about it.

I'm not sure how you think of this, but I for one want bugs to be patched and patches to be released in the open, so the people who WANT TO download the patch can do so. And also, known issues and workarounds to be posted if no patch is available/possible. A good example of that last part is the VS.NET 2003 HTML editor. MS confirmed in 2004 that the editor was broken and couldn't be fixed. Though they did that in a thread in a vs.net newsgroup. How many people have read that? A small group compared to the group who ran into the issues the HTML editor has/had in VS.NET 2003. If they would have posted the info, and possible workarounds (there were a couple), it would be way better, because customers could then find the known issues, read teh work arounds and move on.

The community, and that's also you, can make a difference. If you're not happy with something, say so: here, or elsewhere. The thing is that if more and more people express their frustration, the more chance you have it will get fixed/updated. One of the examples is Response.TransferFile(), an undocumented addition to ASP.NET's Response object in .NET 1.1 SP1, to overcome the misery of BinaryWrite and WriteFile.

Though I have to agree with the fact that people should be realistic as well. I therefore tried to express the bug I found as 'hmm, is this a bug or is it me?' instead of '(&@#@#@! bl00dy VS.NET $#(@($@ ~!!!', as I've seen a couple of others do. It's not realistic, and frankly I would be a hypocrite if I did so, because my own work isn't bugfree either, and neither will your work be, no matter how hard we try. But as I said: it's not that I ran into a bug, it's the knowledge that a bugfix would be hard to get if not impossible. Well, not for me, but for OTHER people who don't have the benefit to pull a string and get certain things easier/earlier from MS. Things had to change, MS had to realize that bugfixes are things you should release to your customers a.s.a.p.

Frans,

First, I want to go on record that I wasn't complaining that you published info about the bug you found. I think this needs to be done. I came from the Visual FoxPro community and was an alpha/beta tester on the last three releases. Many times people would complain about bugs in the public beta and I would mention it on the beta forum and MS didn't know about it. Although usually one of the beta testers would officially report bugs that had been found by non 'official' testers.

Secondly, I guess I was giving MS the benefit of the doubt that "deferred to next version" didn't HAVE to mean next major release. Once again, I guess I am tempering my expectations from the Visual FoxPro group which has never made a firm decision whether to have or not have a service pack but has made the decision based on issues that are found and how serious they are. For example, when they released Visual FoxPro 9 recently they said "there are no current plans for a service pack." But, there is already an SP 1 public beta available with plans to release it by year end.

What I was really complaining about was the doom sayers that lable the product as "buggy" or "crap" or whatever because they find a single bug. Not the fact that they found or publish info about the bug.

Sorry if I hit a sore spot. No worries!

BOb

pilotboba wrote:

First, I want to go on record that I wasn't complaining that you published info about the bug you found. I think this needs to be done. I came from the Visual FoxPro community and was an alpha/beta tester on the last three releases. Many times people would complain about bugs in the public beta and I would mention it on the beta forum and MS didn't know about it. Although usually one of the beta testers would officially report bugs that had been found by non 'official' testers.

No problem I perhaps didn't wrote it down that clear, but I'm with you when complaining is easy and should be avoided if there's nothing really to complain about, just hot air (as we've seen in the past week alot as well indeed, with people crying foul about vs.net being crap/buggy etc. ). Though their uninstaller of the beta seems to be buggy in one area: with VB.NET beta2 installed, you're out of luck in some cases when you're working in VB.NET 2005 RTM installed on the same machine after removing beta2. Most crashes in VB.NET RTM seem to originate from the fact that some old beta2 goo is still there.

Secondly, I guess I was giving MS the benefit of the doubt that "deferred to next version" didn't HAVE to mean next major release. Once again, I guess I am tempering my expectations from the Visual FoxPro group which has never made a firm decision whether to have or not have a service pack but has made the decision based on issues that are found and how serious they are. For example, when they released Visual FoxPro 9 recently they said "there are no current plans for a service pack." But, there is already an SP 1 public beta available with plans to release it by year end.

Heh Well, I think the groups are different. In the past 2, 3 years, the upper management layers in the dev division weren't that concerned about service packs or fixes. Now they are, while the developers who actually wrote the stuff were already concerned about service packs for a couple of years. I think it depends on who calls the shots in which group and if they decide / have the funds to release a SP.

What I was really complaining about was the doom sayers that lable the product as "buggy" or "crap" or whatever because they find a single bug. Not the fact that they found or publish info about the bug.

Oh, 100% agreed. If someone would call vs.net 2005 'buggy' I don't know what they'll call their own code, but it won't be above 'miserable goo'.

Sorry if I hit a sore spot. No worries!

No sore spots were hit . I thought I should explain a little about in which corner I was standing, for the situation where a reader of this thread would think I'm in the corner of the 'crap!'-sayers

I suppose you guys have heard about the pre-Orcas tool - MQ...

http://blogs.msdn.com/eric/archive/2005/11/04/489108.aspx http://blogs.msdn.com/somasegar/archive/2005/11/08/490694.aspx

MQ won't be a release of a product, if that's what you're thinking. I was thinking that too, but MS told me it's a milestone for a change in how they work. So no 'in-between release'