Argh - C# can really SUCK !!!!

Chris Maunder

A code sample would be helpful. cheers, Chris Maunder

CodeGuy

George wrote: I am not sure what are you trying to say here. But I find latest Herb's comments that MC++ and C++ Standard have the same goals very fit to my theory. ... Boy, isn't that the truth. MS's sudden presence on the C++ Standard committee along with their interest in adding garbage collection and virtual machine support (what in the world?) to C++ is just scary. Essentially they are trying to make sure C++ runs on .NET and nowhere else. In case no one has seen the CUJ article George is talking about, look here. Brandon

William E Kempf

George wrote: Since it's an internal application for rather large company it isn't a problem at all - all PC's and their setup is strictly controlled to the point that we only have to make our appplication working with a single version of IE. When I say version I mean a full version like 5.50.4807.0001. It doesn't have to work any other release! This, with other comments you've made, make me wonder why you've developed this for the web at all! The excuse "because they don't have to install anything, and upgrades are automatic" is flawed, because there are numerous ways to accomplish this with out creating an application that runs inside the browser. .NET, in fact, provides a solution for this. The nice thing about these solutions: 1) they don't lock you into any specific browser or worse (in your case) browser version, 2) they don't force you to "hack" away at complex solutions to avoid the short comings of browser based application development (which you're doing from your description), 3) they provide solutions that make loading the application MUCH faster making your app more usable. I can't think of a single good reason to go down the path you have (and honestly there's few good reasons for any kind of web application, IMNSHO). William E. Kempf

William E Kempf

CodeGuy wrote: MS's sudden presence on the C++ Standard committee along with their interest in adding garbage collection and virtual machine support (what in the world?) to C++ is just scary. Why? 1) MS has always had members involved in the C++ standards committee. The only thing new is their attitude that the standards are important instead of just a necessary evil, which is a good thing for you and I. 2) Garbage collection was actually proposed for the last standard, though it got little attention AFAIK, because of scope and time constraints. But the language was specifically defined to allow for GC from day one, and most members have had some interest in a standardized GC addition to the language. Once you get over the (vague and basically unwarranted) fears about GC, there's no logical reason not to add this one. 3) VM support shouldn't scare you either. It's just another backend to target for the language. No big deal, since C++ won't be like Java... restricted (basically) to running on this VM. And for many tasks there's a true benefit to targeting a VM. I love having this extra tool in my toolbox, even if most of my code is compiled natively. It's sounds to me like you're suffering (badly) from FUD. William E. Kempf

Daniel Turini

Chris Maunder wrote: MC++ does not (yet) work with ASP.NET. It *looks* like it works, it smells like it works, but pretty soon it will crash your ASP.NET apps. Hmmm... You seem to went through this before... :suss: "In an organization, each person rises to the level of his own incompetence." Peter's Principle

George

Tomasz Sowinski wrote: It'll work in this scenario. However, it's totally inappropriate for applications targeted at general audience. Which is why I don't feel like making any. They live a violent, unstable and filled with temporary solutions and technologies lifes while I prefer a long term stable investments. ;) Tomasz Sowinski wrote: Ok. Next time you create a interactive website using C++ let me know As I said before, I would not want to work on something you have in mind. Having said that, I seem to recall a tool that actually allowed to do just that - use C++ in similiar fashion the ASP is employed.... Tomasz Sowinski wrote: No, IE was just much worse than Netscape at the beginning. Which is why I was using Lynx at that time ;)

/* I C++, therefore I am... */

George

William E. Kempf wrote: It's sounds to me like you're suffering (badly) from FUD. It sounds to me like you are suffering from brain-washing ;P

/* I C++, therefore I am... */

George

William E. Kempf wrote: Gives what out clearly? And BTW, WinForm support is supposed to exist for MC++ in VS.NET 7.1, due out RSN, so if you want to make up facts for some conspiracy theory you'd better pick a new one "Supposed to exists"? And who is making up the facts here?

/* I C++, therefore I am... */

Tomasz Sowinski

George wrote: Which is why I was using Lynx at that time I was sure you were using telnet on port 80. Tomasz Sowinski -- http://www.shooltz.com

*** Vodka. Connecting people. ***

George

William E. Kempf wrote: This, with other comments you've made, make me wonder why you've developed this for the web at all! I have no idea, it was that way when I joined in already. ;) I think that one reason is the "hype" for web applications. Most web applications should never be made one - they are just fine as a desktop. Similiarly, nobody really needs any of the .NET features. It can all be done using an existing technologies, often better and faster. But MS has to make money somehow, so there is no stopping to make tools and technologies more "productive". Still, that is not my problem and not my responsibility to decide. I only see the nice side of things (that is C++) for most of the time, even thought I know I could make a better desktop application...

/* I C++, therefore I am... */

CodeGuy

William E. Kempf wrote: 2) Garbage collection was actually proposed for the last standard, though it got little attention AFAIK, because of scope and time constraints. But the language was specifically defined to allow for GC from day one, and most members have had some interest in a standardized GC addition to the language. Once you get over the (vague and basically unwarranted) fears about GC, there's no logical reason not to add this one. Except unlike threading or networking support (which occurs across many different platforms), GC and VM support doesn't. These requirements are not coming out of the blue, or postponed from a prior meeting -- Herb makes it pretty clear in his article MS is interested in making C++ target .NET. This is FUD?? As you mentioned in your point #1, MS has shown zero interest in the past in the Standard. Essentially, they sat on their hands for 6 years or so while we had a crap version of STL and no worthwhile template support. Now they have a new version of C++ that is rooted in .NET with GC & VM support, but still will not compile many third-party libraries according to the current Standard. Again, it's pretty clear that MS is attempting to change the Standard to match MC++, while the current Standard hasn't even been complied with! Who are they kidding? I'm not scared by the addition of GC & VM features to the Standard*. I am scared by the presence of a 900-pound gorilla on the Committee who will attempt to bully a Standard through that suits them and no one else. Brandon * Although I do think the Standard is already weighed down by its own complexity. I think more attention should be given toward simplification of existing features rather than new libraries.

George

Tomasz Sowinski wrote: I was sure you were using telnet on port 80. LOL!

/* I C++, therefore I am... */

Tomasz Sowinski

CodeGuy wrote: MS has shown zero interest in the past in the Standard. ... which seems to worry you. CodeGuy wrote: I am scared by the presence of a 900-pound gorilla on the Committee ... which seems to worry you as well. Is there anything M$ should do about their presence in this body which could make you less scared? Tomasz Sowinski -- http://www.shooltz.com

*** Vodka. Connecting people. ***

Tomasz Sowinski

George wrote: Similiarly, nobody really needs any of the .NET features. This is very strong statement. I think you should rephrase it to 'I (George) don't need any of the .NET features'. Tomasz Sowinski -- http://www.shooltz.com

*** Vodka. Connecting people. ***

George

Tomasz Sowinski wrote: This is very strong statement. I think you should rephrase it to 'I (George) don't need any of the .NET features'. I'm sorry but the best I can do for you is to rephrase to: I think that nobody really needs any of the .NET features. ;P

/* I C++, therefore I am... */

CodeGuy

They could 1) match the Standard in VC++ (i.e. bring the product up to where it can compile 3rd party libraries like Boost), and then 2) contribute to Standard C++ in an open fashion. What I mean by #2 is that MS right now has the .NET VM, which includes GC and other features. Theoretically, this could be ported to other platforms, but right now, it only exists on Windows. Once -- or if ever -- .NET or other truly portable VMs exist for lots of other operating systems, then it makes sense to say, OK, let's add VM and GC support to the Standard. Brandon

William E Kempf

Is happening. VC++ 7.1 is rumored (by people who would know) to be extremely compliant, and it's been developed using Boost as a test bed. 2) MS has ported it, as read in the article, to other platforms. They've also paid money to others so they could do the same. There's also numerous efforts to do this by folks with no ties to MS. The whole thing has also been _STANDARDISED_ by the ECMA, and, as the article stated, may be standardised by ISO as well. I don't get why you're so stuck on all of this? William E. Kempf

CodeGuy

"is rumored" ... MS has already quietly missed the deadline for VC++ 7.1 (summer). Again, from past history and current statements, I have to believe that MS is more interested in making MC++ the Standard rather than complying with the existing one. We'll see. The proof is in the pudding. 2) "maybe be standardized by ISO" ... Again, why are we talking about standardizing the next version of ISO (not ECMA) C++ on a platform that 1) hasn't been standardized by the ISO yet and 2) isn't widely available? Even MS is having a hard time getting the CLI available because it's rolled up in a 20MB .NET framework download. I think that you and I have a VERY different outlook on what C++ should evolve into.

Mike Nordell

William E. Kempf wrote: 1) MS has always had members involved in the C++ standards committee Is that really true? AFAIK they had one person on the committee. One person that often was absent during committee meetings. Calling that "involved" isn't really the truth I think. Search and you shall find. 2) Garbage collection was actually proposed for the last standard And AFAIK completely ignored (for good reasons). 3) VM support shouldn't scare you either. It's just another backend to target for the language. Exactly. Another back-end. I.e. it would just be another target CPU/environment that should be nothing more than that back-end (compiler-) switch. Now look at what Micros~1 instead (true to their habits) tries to do (hints: a boy born in Austria and named Adolf tried it once in a place with the country code DE)...

Christian Graus

I thought I read somewhere that C# was whizbang because it initialised numbers to 0, unlike C++. But I agree that the ref behaviour is fair enough, it's the 'out' behaviour that is frustrating. Christian Hey, at least Logo had, at it's inception, a mechanical turtle. VB has always lacked even that... - Shog9 04-09-2002 During last 10 years, with invention of VB and similar programming environments, every ill-educated moron became able to develop software. - Alex E. - 12-Sept-2002