How egregious is my crime to consider unmanaged code?
-
I guess it depends on the quality of your codebase. I wouldn't suggest, anyway, to build GUIs in pure C++ for business applications. I think that a native backend can have its advantages, such as speed, although fast hardware is a commodity nowadays. It also depend, IMO, how many modifications you plan to do.
If you truly believe you need to pick a mobile phone that "says something" about your personality, don't bother. You don't have a personality. A mental illness, maybe - but not a personality. - Charlie Brooker My Photos/CP Flickr Group - ScrewTurn Wiki
Dario Solera wrote:
although fast hardware is a commodity nowadays.
That should never be a consideration when designing your code.
A human being should be able to change a diaper, plan an invasion, butcher a hog, conn a ship, design a building, write a sonnet, balance accounts, build a wall, set a bone, comfort the dying, take orders, give orders, cooperate, act alone, solve equations, analyze a new problem, pitch manure, program a computer, cook a tasty meal, fight efficiently, die gallantly. Specialization is for insects. - -Lazarus Long
-
We have a ton of existing C++ code I can reuse to build an n-tier enterprise managment solution. Does the lounge consider it a crime to consider using unmanaged C++ for a new project?
Do we weigh less at high tide?
Why would it matter to you want *we* think. If you have code that can be leveraged, by all means use it. If you're more comfortable using native C++, and if management hasn't dictated otherwise, by all means use it. And consider yourself lucky that you have a job that allows the development of unmanaged code.
"Why don't you tie a kerosene-soaked rag around your ankles so the ants won't climb up and eat your candy ass..." - Dale Earnhardt, 1997
-----
"...the staggering layers of obscenity in your statement make it a work of art on so many levels." - Jason Jystad, 10/26/2001 -
Chris Austin wrote:
Has there been a C++0x draft standard compiler implementation?
As far as I know only Intel. MS gave a nod a few months back, it got a pretty mad ranting in the lounge with those complaining that MS was making a mistake keeping C++ alive. GNU has an early draft like draft 1 which can be enabled. I'll be loading the latest fedora today so I will see if they made it through to the final draft that was sent for approval. As far as I know only Intel has attempted to keep up with the drafts, every release of the compiler brings the draft support up to date.
_________________________ Asu no koto o ieba, tenjo de nezumi ga warau. Talk about things of tomorrow and the mice in the ceiling laugh. (Japanese Proverb)
El Corazon wrote:
As far as I know only Intel.
El Corazon wrote:
I'll be loading the latest fedora today so I will see if they made it through to the final draft that was sent for approval.
Thanks, I'll have to give it a look.
A human being should be able to change a diaper, plan an invasion, butcher a hog, conn a ship, design a building, write a sonnet, balance accounts, build a wall, set a bone, comfort the dying, take orders, give orders, cooperate, act alone, solve equations, analyze a new problem, pitch manure, program a computer, cook a tasty meal, fight efficiently, die gallantly. Specialization is for insects. - -Lazarus Long
-
Chris Austin wrote:
Has there been a C++0x draft standard compiler implementation?
As far as I know only Intel. MS gave a nod a few months back, it got a pretty mad ranting in the lounge with those complaining that MS was making a mistake keeping C++ alive. GNU has an early draft like draft 1 which can be enabled. I'll be loading the latest fedora today so I will see if they made it through to the final draft that was sent for approval. As far as I know only Intel has attempted to keep up with the drafts, every release of the compiler brings the draft support up to date.
_________________________ Asu no koto o ieba, tenjo de nezumi ga warau. Talk about things of tomorrow and the mice in the ceiling laugh. (Japanese Proverb)
I wonder who those poor deluded souls are, the ones that think Microsoft is making a mistake by keeping C++ alive. Maybe they're the ones that would like Microsoft to rewrite Windows, Office and Visual Studio from scratch in C# or Ruby on Rails.
-
I certainly don't consider it a crime. I've developed n-tier enterprise solutions in C++ and VB with COM and DCOM. The weak link in the chain back them was IIS. These days you could go for Apache instead.
Nothing is exactly what it seems but everything with seems can be unpicked.
Matthew Faithfull wrote:
The weak link in the chain back them was IIS.
What veersion of IIS was that? I was involved in development of a machine translation web farm and IIS 5 worked like a charm - we were not using ASP, but ISAPI extensions though.
Matthew Faithfull wrote:
These days you could go for Apache instead.
These days IIS (6+) is faster and more secure than Apache.
-
chris ruff wrote:
Does the lounge consider it a crime to consider using unmanaged C++ for a new project?
Nope. These newfangled languages are poorly architected, the supporting frameworks are kludgy and buggy, and the designers appear to be syntactical sugared and eye candied, to say the least. I'm dead serious. C# is cool and thoroughly enjoy the n-tier architecture that I build with it, but I'm depressed by what they're doing to the language and the framework. I continually encounter areas of the framework that don't come up to snuff performance-wise when working in an n-tier environment or that are so dumbed down as to be unusable, either case requiring a replacement of what .NET provides. I'm disappointed with the language enhancements, feeling that there is no roadmap other than "screw everybody else's ideas, but don't admit they even have ideas because we're going to hijack them anyways." The changes to C++ that someone posted about a week or so ago, that the Intel compiler supports, that's stuff that gets me wishing I'd developed Interacx in C++. Seriously. Marc
Marc Clifton wrote:
The changes to C++ that someone posted about a week or so ago, that the Intel compiler supports, that's stuff that gets me wishing I'd developed Interacx in C++. Seriously.
:-O What people forget is that all technology changes. Some good, some bad. It's like marketing evolution in a bizarre way. Folks develop somethings with an eye for goodies, we call them creaping featurisms, good ideas that seemed good at the time, by some guy in a back room who never actually tried to field anything but add features to his product... he never actually "uses it." When it is REALLY bad, I named them after a reference in a book, feaping creaturisms, when the creaping features actually inhibit your work. A compiler is a general purpose tool, but applications are specific. Things that work good for one product may actually inhibit productivity for another purpose. Some of the directions of C# and C++ have specific reasons, for good and bad, each have their benefits, and each are dividing those bases of application support. I am very pleased at the new features of C++, some of the features of C# make me glad too, and some are disappointing. This is because each of us has an application direction, what one person writes doesn't mean that everything will always fit that way. C++ has also had a revision in 99 and 03 that went somewhat unknown, it has continued to expand and increase in capability, and as near as I can tell will continue to do so. In some ways it has not competition, in others it does. Even in areas where it does have competition with multi-year releases in changes of language specifications, people forget about asynchronous update releases between competitors. What you see is C# increasing, and then C++, and then C# and then C++, etc. This is highly normal, it doesn't mean that one will disappear, either one. It means that both are growing in technology over time, which is good for all of us. :-D
_________________________ Asu no koto o ieba, tenjo de nezumi ga warau. Talk about things of tomorrow and the mice in the ceiling laugh. (Japanese Proverb)
-
Why would it matter to you want *we* think. If you have code that can be leveraged, by all means use it. If you're more comfortable using native C++, and if management hasn't dictated otherwise, by all means use it. And consider yourself lucky that you have a job that allows the development of unmanaged code.
"Why don't you tie a kerosene-soaked rag around your ankles so the ants won't climb up and eat your candy ass..." - Dale Earnhardt, 1997
-----
"...the staggering layers of obscenity in your statement make it a work of art on so many levels." - Jason Jystad, 10/26/2001Well, first of all I *am* interested in what the lounge thinks. There is a ton of programming horsepower perusing these pages. I am constantly bombarded with _opinions_ from all levels and I want to know, myself, that I am not screwing my customer(s). Put differently, I think it would be *bad* to create bigapp+ with Turbo Pascal, Power Builder, Fortran, unmanaged VB, etc. But I am not sure that it is bad to use VC++ unmanaged. I know it isn't bad for me! I also have had the maddening experience of having 'frameworks' and code-families-for-everybody let me down to the point that I have to include native DLLs, OCX (active-x, anyone?) solutions and etc. just to be able to do something that is SIMPLE in C, C++, ASM. So that is why I am looking for other's opinions. thnx for listening...
Do we weigh less at high tide?
-
I do. C++ is best when you know what you're doing. I'd recommend it whole-heartedly if you were setting up, say, a 2-tier, 3-tier, or 1111-tier enterprise management solution. But since you need a variable number of tiers, and presumably they're gonna change, all the time, while the software is running, i'd go with something more flexible. I hear there's a new version of Perl in the pipes... :rolleyes:
But who is the king of all of these folks?
right. you need something with JS's "exec" or equivalent, because you're going to find yourself needing to execute code that gets pulled out of a database somewhere, which can be changed any time the marketing guys change their mind about which strategic vision they need to execute on. embedding a Lisp interpreter might prove handy, too.
-
Marc Clifton wrote:
The changes to C++ that someone posted about a week or so ago, that the Intel compiler supports, that's stuff that gets me wishing I'd developed Interacx in C++. Seriously.
Too bad platform ports are so time consuming.
A human being should be able to change a diaper, plan an invasion, butcher a hog, conn a ship, design a building, write a sonnet, balance accounts, build a wall, set a bone, comfort the dying, take orders, give orders, cooperate, act alone, solve equations, analyze a new problem, pitch manure, program a computer, cook a tasty meal, fight efficiently, die gallantly. Specialization is for insects. - -Lazarus Long
Chris Austin wrote:
Too bad platform ports are so time consuming.
how so? much of choices in multi-platform support are changing. From gaming code to others, changes in architecture are enabling greater choice in multi-platform capability. The problem is we started with two completely different architectures that were never designed to compete. MS killed all the competition on the PC arena for a long time, and Unix was strictly big-iron and MS never attempted to fight the big-iron control. Mac was the lonely guy in the corner that does some great things, but people forget is there half the time. We have been going through a growing level of effort that is bringing these architectures together, and it is getting easier. A lot of it is the choices we make that make a difference. With openGL defining its own shader compiler in its standard it opened up a lot of cross-compatibility between graphics cards that was difficult even under DirectX, which was windows only. Compilers are taking up threading as part of the feature sets, which until multi-core, was avoided like the plague and I kept telling people... it is coming, prepare for it, learn it, it is only to our benefit. But a lone voice is not worth much. Times change, things grow, and cross-platform is getting easier. :-D
_________________________ Asu no koto o ieba, tenjo de nezumi ga warau. Talk about things of tomorrow and the mice in the ceiling laugh. (Japanese Proverb)
-
Chris Austin wrote:
Too bad platform ports are so time consuming.
how so? much of choices in multi-platform support are changing. From gaming code to others, changes in architecture are enabling greater choice in multi-platform capability. The problem is we started with two completely different architectures that were never designed to compete. MS killed all the competition on the PC arena for a long time, and Unix was strictly big-iron and MS never attempted to fight the big-iron control. Mac was the lonely guy in the corner that does some great things, but people forget is there half the time. We have been going through a growing level of effort that is bringing these architectures together, and it is getting easier. A lot of it is the choices we make that make a difference. With openGL defining its own shader compiler in its standard it opened up a lot of cross-compatibility between graphics cards that was difficult even under DirectX, which was windows only. Compilers are taking up threading as part of the feature sets, which until multi-core, was avoided like the plague and I kept telling people... it is coming, prepare for it, learn it, it is only to our benefit. But a lone voice is not worth much. Times change, things grow, and cross-platform is getting easier. :-D
_________________________ Asu no koto o ieba, tenjo de nezumi ga warau. Talk about things of tomorrow and the mice in the ceiling laugh. (Japanese Proverb)
El Corazon wrote:
how so?
Sorry I wasn't clear. What I meant and should have said was that framework/development platform ports are time consuming.
El Corazon wrote:
With openGL defining its own shader compiler in its standard it opened up a lot of cross-compatibility between graphics cards that was difficult even under DirectX,
OpenGL 3 is getting pretty excited also. And, I agree with 100% of what you said.
A human being should be able to change a diaper, plan an invasion, butcher a hog, conn a ship, design a building, write a sonnet, balance accounts, build a wall, set a bone, comfort the dying, take orders, give orders, cooperate, act alone, solve equations, analyze a new problem, pitch manure, program a computer, cook a tasty meal, fight efficiently, die gallantly. Specialization is for insects. - -Lazarus Long
-
El Corazon wrote:there are those who believe C++ is dead, and those who do not. You are about to find out who. Read that as "there are those who bought the .NET marketing and advertising hype, and those who did not."
to_be_defined wrote:
Read that as "there are those who bought the .NET marketing and advertising hype, and those who did not."
No, I don't even think it is that. And I think it is a poor representation of those who believe that C++ is dead. Oh sure I think .NET was rather optimistically presented as the falling in love and fairies coming out your @ss that was mentioned elsewhere.... at least until a few people felt what that really was like. Some liked the feeling, others did not, hey that is life. What ever floats your boat, go for it. Just don't tell me that it is an enjoyable experience.... :laugh: It really comes down to issues of support and use. There is still a lot of use left in C++. and there is a lot in C#. You can take either concept too far until the fairies come out to play, generally I avoid that. I like to keep both real. There are ways to accomplish activities in both, there is some overlap, and there are areas where each shine rather spectacularly. I just feel we need to keep our eyes open that both are good for different things, with some overlap. Live and let live. :-D
_________________________ Asu no koto o ieba, tenjo de nezumi ga warau. Talk about things of tomorrow and the mice in the ceiling laugh. (Japanese Proverb)
-
Well, first of all I *am* interested in what the lounge thinks. There is a ton of programming horsepower perusing these pages. I am constantly bombarded with _opinions_ from all levels and I want to know, myself, that I am not screwing my customer(s). Put differently, I think it would be *bad* to create bigapp+ with Turbo Pascal, Power Builder, Fortran, unmanaged VB, etc. But I am not sure that it is bad to use VC++ unmanaged. I know it isn't bad for me! I also have had the maddening experience of having 'frameworks' and code-families-for-everybody let me down to the point that I have to include native DLLs, OCX (active-x, anyone?) solutions and etc. just to be able to do something that is SIMPLE in C, C++, ASM. So that is why I am looking for other's opinions. thnx for listening...
Do we weigh less at high tide?
We are using a mixture of old C and C++ and C# now in our project. Some of the C source is well over 20 years old now. Some of the C++ parts are over 10 years old now too. We decided to use C# for its support of XML and some date/time stuff that doesn't exist in the CRT. It works pretty well. We also use managed C++ as a bridge between the old C code and some new C# code also. The C code calling the C# code. Speed of the execution is not really an issue with our software, so this works for us. You might consider using your C++ in managed C++ if you want to reuse your old code and get some of the .NET goodies also.
-
Matthew Faithfull wrote:
The weak link in the chain back them was IIS.
What veersion of IIS was that? I was involved in development of a machine translation web farm and IIS 5 worked like a charm - we were not using ASP, but ISAPI extensions though.
Matthew Faithfull wrote:
These days you could go for Apache instead.
These days IIS (6+) is faster and more secure than Apache.
That was the dreaded bug fest known as IIS 3 and we were using a hand full of trivial ASP pages. The ASP engine was fine, the VB presentation layer was as good as bullet proof after my grumpy scottish colleague spent some weeks beating it sticks and error handlers, my C++ DCOM Session handling request routing middleware was better than anything MS came up with for years after I wrote it ( I'm still not sure it's been bettered ) and the back end database was 20+ years stable and many times more efficient than Oracle of SQL Server in its application domain. Unfortunately getting the system to stay up long enough to demo it involved some very fast clicking, fast talking, a deal of praying and a couple of goes at it. If Apache had supported ASP, even a little, at the time it would have saved me 3 months out of my life.
Nothing is exactly what it seems but everything with seems can be unpicked.
-
right. you need something with JS's "exec" or equivalent, because you're going to find yourself needing to execute code that gets pulled out of a database somewhere, which can be changed any time the marketing guys change their mind about which strategic vision they need to execute on. embedding a Lisp interpreter might prove handy, too.
Chris Losinger wrote:
embedding a Lisp interpreter might prove handy, too.
For sure, although you'll need to add a SOAP layer that pulls in XML-encoded XML data from a distributed database server first, and then uses Java to compile it to Lisp so that it can be interpreted properly. It just wouldn't be enterprisy enough otherwise.
But who is the king of all of these folks?
-
to_be_defined wrote:
Read that as "there are those who bought the .NET marketing and advertising hype, and those who did not."
No, I don't even think it is that. And I think it is a poor representation of those who believe that C++ is dead. Oh sure I think .NET was rather optimistically presented as the falling in love and fairies coming out your @ss that was mentioned elsewhere.... at least until a few people felt what that really was like. Some liked the feeling, others did not, hey that is life. What ever floats your boat, go for it. Just don't tell me that it is an enjoyable experience.... :laugh: It really comes down to issues of support and use. There is still a lot of use left in C++. and there is a lot in C#. You can take either concept too far until the fairies come out to play, generally I avoid that. I like to keep both real. There are ways to accomplish activities in both, there is some overlap, and there are areas where each shine rather spectacularly. I just feel we need to keep our eyes open that both are good for different things, with some overlap. Live and let live. :-D
_________________________ Asu no koto o ieba, tenjo de nezumi ga warau. Talk about things of tomorrow and the mice in the ceiling laugh. (Japanese Proverb)
El Corazon wrote:
falling in love and fairies coming out your @ss
:omg:
Software Zen:
delete this; -
We have a ton of existing C++ code I can reuse to build an n-tier enterprise managment solution. Does the lounge consider it a crime to consider using unmanaged C++ for a new project?
Do we weigh less at high tide?
If you write code for money then there is only one crime: spending too much money to make too little money. I wouldn't start any new project in unmanaged code but if I had a huge investment in an existing library I'd use it as is, there's no profit in re-inventing the wheel.
When everyone is a hero no one is a hero.
-
chris ruff wrote:
Does the lounge consider it a crime to consider using unmanaged C++ for a new project?
Nope. These newfangled languages are poorly architected, the supporting frameworks are kludgy and buggy, and the designers appear to be syntactical sugared and eye candied, to say the least. I'm dead serious. C# is cool and thoroughly enjoy the n-tier architecture that I build with it, but I'm depressed by what they're doing to the language and the framework. I continually encounter areas of the framework that don't come up to snuff performance-wise when working in an n-tier environment or that are so dumbed down as to be unusable, either case requiring a replacement of what .NET provides. I'm disappointed with the language enhancements, feeling that there is no roadmap other than "screw everybody else's ideas, but don't admit they even have ideas because we're going to hijack them anyways." The changes to C++ that someone posted about a week or so ago, that the Intel compiler supports, that's stuff that gets me wishing I'd developed Interacx in C++. Seriously. Marc
Marc Clifton wrote:
don't come up to snuff performance-wise when working in an n-tier environment
Ok, I seriously don't want to argue the merits of managed code in any way, but managed n-tier development is an area I'm deeply involved in and I can't fathom what you're saying here. Performance has never been an issue for me, it scales beautifully, where do you get this from?
When everyone is a hero no one is a hero.
-
Dario Solera wrote:
although fast hardware is a commodity nowadays.
That should never be a consideration when designing your code.
A human being should be able to change a diaper, plan an invasion, butcher a hog, conn a ship, design a building, write a sonnet, balance accounts, build a wall, set a bone, comfort the dying, take orders, give orders, cooperate, act alone, solve equations, analyze a new problem, pitch manure, program a computer, cook a tasty meal, fight efficiently, die gallantly. Specialization is for insects. - -Lazarus Long
That should *always* be a consideration when designing your code. Designing software isn't about dogmatic adherence to rules and regulations. You have a need, you take *everything* into consideration when designing a solution, the best design is the best design. I could quickly go out of business spending an inordinate amount of time fiddling around trying eke out the last millisecond of performance while my end users languished waiting for an update. People who have a serious need for software will gladly supply any hardware required because it *is* in fact a commodity these days.
When everyone is a hero no one is a hero.
-
El Corazon wrote:there are those who believe C++ is dead, and those who do not. You are about to find out who. Read that as "there are those who bought the .NET marketing and advertising hype, and those who did not."
-
That should *always* be a consideration when designing your code. Designing software isn't about dogmatic adherence to rules and regulations. You have a need, you take *everything* into consideration when designing a solution, the best design is the best design. I could quickly go out of business spending an inordinate amount of time fiddling around trying eke out the last millisecond of performance while my end users languished waiting for an update. People who have a serious need for software will gladly supply any hardware required because it *is* in fact a commodity these days.
When everyone is a hero no one is a hero.
and people with a need for performance that can't be solved by just throwing hardware at it will pay extra for C++ with hand assembled hotloops.
Otherwise [Microsoft is] toast in the long term no matter how much money they've got. They would be already if the Linux community didn't have it's head so firmly up it's own command line buffer that it looks like taking 15 years to find the desktop. -- Matthew Faithfull
