Toolbox or legacy?
-
Joergen Sigvardsson wrote:
Some software are built to last for decades, installed on many sites. It is not feasible to make such software make use of the latest and the greatest.
5 Or if it is a man rated system. Do you want to go to a hospital whose systems are solid and proven reliable or do you want to go to one whose systems are upgraded to the latest MS coding options?
It'd suck if the respirator had to take a pause just because the garbage collector had to do a sweep.. :~
-- Please rise for the Futurama theme song
-
I have read a lot of posts over the years here on CP about developers "picking the right tool for the right job" or a given technology "is just another tool in the toolbox". Often when a battle occurs in the message boards about different technologies of which is better, there is always a few of those "it is just antoher tool in the toolbox" quotes. Okay, I will flaunt my age around here and say that I have been a programmer since 1981 and have plowed over many technologies (languages, frameworks, platforms) in that time. Some were great for development in the time they were used, but when new technologies appeared that made my life easier while still providing performance and maintainability, I would work with them until comfortable and then scrap the previous technology, not just cram it in my toolbox. The only time I would use the prior technologies is when I was forced to by either an employer or legacy code I had to maintain. Currently I work in C#/.NET and do not touch any other technology from the past. If I have a new project to build it will be C#/.NET. If work would come my way requiring a past technology (such as C/C++, PHP, etc) for new development, I would not take the work. To me it is a waste of my time to work on legacy systems when there is so much work to be done in the current technology I use. Anyone else out there that just loves technology and burns their bridges to past technology to remain focused on the current technology?
Rocky <>< Latest Code Blog Post: ASP.NET HttpException - Cannot use leading "..".. Latest Tech Blog Post: Vista - Little Things
Well, I agree with you in most of the points, but I just suggest you a real-world example. Fog Creek Software[^] (Joel Spolsky[^]'s Company), for FogBugz[^] developed their own compiler[^] for a special language, which targets PHP and classic ASP. This means that they can sell their product to almost any customer, both on Windows and Linux/Unix. Isn't it a great thing? Personally, I consider classic ASP way obsolete, I don't like PHP at all, but I think they're making money because they target the customer's needs, regardless of the available modern platforms and technologies.
________________________________________________ Personal Blog [ITA] - Tech Blog [ENG] Developing ScrewTurn Wiki 1.1 (1.0.7 is out)
-
Hockey wrote:
My only comment is that I find it odd you refer to PHP as "past" technology...
It's an actual technology, but it's really too "basic" in my opinion compared to (obviously) ASP.NET. So, although it is actively developed, it's very obsolete. IMHO.
________________________________________________ Personal Blog [ITA] - Tech Blog [ENG] Developing ScrewTurn Wiki 1.1 (1.0.7 is out)
-
Chris Losinger wrote:
all those technologies are old . SQL was created in the early 70s.
Old is not the issue, "better" is the point. You do not change to a new technology just because you feel like learning and using something new. The shift is only based on the benefits you gain from the new technology, minus the learning time/curve and the time it will take to move things forward. Once a person finds themselves comfortable with new technology that obsoletes their old technology, then to me, it is time to break from the old if you can and move on. There will never be a day I write a line of assembler again. Probably never a day I will do a line of C++ unless it is a performance or interace issue. Those tools are nto near as productive to me as the ones I use today, so it would cost me to use the legacy technologies.
Rocky <>< Latest Code Blog Post: ASP.NET HttpException - Cannot use leading "..".. Latest Tech Blog Post: Replacing Vista System HD & New things learned!
Rocky Moore wrote:
Old is not the issue, "better" is the point. You do not change to a new technology just because you feel like learning and using something new. The shift is only based on the benefits you gain from the new technology, minus the learning time/curve and the time it will take to move things forward.
Which is exactly our point in saying the right tool for the right job. If you are required to write software that is compatible and designed for one technology, then do so. But one technology is not always compatible with another, or at least "efficient". You put it well, a shift based on benefits you gain minus the learning curve. There are some software paths, high performance simulations, detailed and fast 3D graphics, etc. that just aren't compatible with C# "efficiently". Sure I could switch over, I know how to, and I may some day. But right now there would be no benefit, there would be significant loss in efficiency. The right tool for the right job. Your job may be C#, mine is not. That doesn't make my job any less entertaining. You are welcome to challenge me on the hyper-fast frame-rate real-time 3D market using C# vs my C++. I love a good challenge. SimDIS after years of trying to ignore me are finally starting to wake up and pay attention. They have dozens of programmers chasing after my performance achievements. They might actually make it, but I doubt it. Not because they are not good enough, they are, and smart enough, but they started off on the wrong path, the wrong core technology. As they hit the augmented reality alignment issues, the accuracy and speed is just not enough and they will be rewriting to change paths. This is what I mean by the right tool for the right job. When you choose the wrong tool to solve a problem, you run smack into a wall either in performance or level of ability. Not everyone writing "desktop applications" writes an SQL user interface to a database. Some of us are doing much more. http://www.dtc.army.mil/ttr/ttr0505.pdf[^] (page 7) http://www.csc.com/industries/government/casestudies/1649.shtml[^] and that is just the pub
-
I used to develop in PHP three years ago, and I became quite good. Then I discovered ASP.NET. Never ever looked back. :-D In my opinion, there is no reason for using PHP for a new project, even the smallest one. But maybe I'm being too drastic.
________________________________________________ Personal Blog [ITA] - Tech Blog [ENG] Developing ScrewTurn Wiki 1.1 (1.0.7 is out)
-
I used to develop in PHP three years ago, and I became quite good. Then I discovered ASP.NET. Never ever looked back. :-D In my opinion, there is no reason for using PHP for a new project, even the smallest one. But maybe I'm being too drastic.
________________________________________________ Personal Blog [ITA] - Tech Blog [ENG] Developing ScrewTurn Wiki 1.1 (1.0.7 is out)
Dario Solera wrote:
there is no reason for using PHP for a new project
When you're coding for a linux server that doesn't support asp... Man, the egos are so thick around here that you can almost hurt yourself when you bump up against one.
"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 -
Dario Solera wrote:
there is no reason for using PHP for a new project
When you're coding for a linux server that doesn't support asp... Man, the egos are so thick around here that you can almost hurt yourself when you bump up against one.
"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, 10/26/2001John Simmons / outlaw programmer wrote:
coding for a linux server
:wtf: :-D
________________________________________________ Personal Blog [ITA] - Tech Blog [ENG] Developing ScrewTurn Wiki 1.1 (1.0.7 is out)
-
Jörgen, I think you, me, and John Simmonns are the last bastions of C++-ness here at CP.
Software Zen:
delete this;Gary R. Wheeler wrote:
Jörgen, I think you, me, and John Simmonns are the last bastions of C++-ness here at CP.
hardly. :) It's all of us in the real-time graphics. :)
_________________________ Asu no koto o ieba, tenjo de nezumi ga warau. Talk about things of tomorrow and the mice in the ceiling laugh. (Japanese Proverb)
-
John Simmons / outlaw programmer wrote:
coding for a linux server
:wtf: :-D
________________________________________________ Personal Blog [ITA] - Tech Blog [ENG] Developing ScrewTurn Wiki 1.1 (1.0.7 is out)
it wasn't until this year that Windows servers outsold Unix servers. there are still tons and tons of *nix boxes out there, running the web.
-
I used to develop in PHP three years ago, and I became quite good. Then I discovered ASP.NET. Never ever looked back. :-D In my opinion, there is no reason for using PHP for a new project, even the smallest one. But maybe I'm being too drastic.
________________________________________________ Personal Blog [ITA] - Tech Blog [ENG] Developing ScrewTurn Wiki 1.1 (1.0.7 is out)
Dario Solera wrote:
In my opinion, there is no reason for using PHP for a new project, even the smallest one
here's a reason: .Net is not an option on the server you are writing for
-
More for us! :-D
-- Mr. Bender's Wardrobe by ROBOTANY 500
Joergen Sigvardsson wrote:
More for us!
and more of the FUN stuff!!
_________________________ Asu no koto o ieba, tenjo de nezumi ga warau. Talk about things of tomorrow and the mice in the ceiling laugh. (Japanese Proverb)
-
it wasn't until this year that Windows servers outsold Unix servers. there are still tons and tons of *nix boxes out there, running the web.
Chris Losinger wrote:
there are still tons and tons of *nix boxes out there, running the web.
I know, it was a joke. :)
________________________________________________ Personal Blog [ITA] - Tech Blog [ENG] Developing ScrewTurn Wiki 1.1 (1.0.7 is out)
-
Dario Solera wrote:
In my opinion, there is no reason for using PHP for a new project, even the smallest one
here's a reason: .Net is not an option on the server you are writing for
Chris Losinger wrote:
here's a reason: .Net is not an option on the server you are writing for
Of course, but since I'm better with .NET than with PHP, surely no one is going to hire me to code PHP. It was an opinion from my point of view.
________________________________________________ Personal Blog [ITA] - Tech Blog [ENG] Developing ScrewTurn Wiki 1.1 (1.0.7 is out)
-
It'd suck if the respirator had to take a pause just because the garbage collector had to do a sweep.. :~
-- Please rise for the Futurama theme song
-
I have read a lot of posts over the years here on CP about developers "picking the right tool for the right job" or a given technology "is just another tool in the toolbox". Often when a battle occurs in the message boards about different technologies of which is better, there is always a few of those "it is just antoher tool in the toolbox" quotes. Okay, I will flaunt my age around here and say that I have been a programmer since 1981 and have plowed over many technologies (languages, frameworks, platforms) in that time. Some were great for development in the time they were used, but when new technologies appeared that made my life easier while still providing performance and maintainability, I would work with them until comfortable and then scrap the previous technology, not just cram it in my toolbox. The only time I would use the prior technologies is when I was forced to by either an employer or legacy code I had to maintain. Currently I work in C#/.NET and do not touch any other technology from the past. If I have a new project to build it will be C#/.NET. If work would come my way requiring a past technology (such as C/C++, PHP, etc) for new development, I would not take the work. To me it is a waste of my time to work on legacy systems when there is so much work to be done in the current technology I use. Anyone else out there that just loves technology and burns their bridges to past technology to remain focused on the current technology?
Rocky <>< Latest Code Blog Post: ASP.NET HttpException - Cannot use leading "..".. Latest Tech Blog Post: Vista - Little Things
I am surprised that some people call C++ legacy. COBOL, FORTRAN or PASCAL would have been fine, but C++ is certainly not legacy. I gave a few interviews earlier this year, and everyone expected me to know C++, and sometimes Java as well, but no one talked about C#/.NET. But then I had applied in 'big' companies only.
-
I have read a lot of posts over the years here on CP about developers "picking the right tool for the right job" or a given technology "is just another tool in the toolbox". Often when a battle occurs in the message boards about different technologies of which is better, there is always a few of those "it is just antoher tool in the toolbox" quotes. Okay, I will flaunt my age around here and say that I have been a programmer since 1981 and have plowed over many technologies (languages, frameworks, platforms) in that time. Some were great for development in the time they were used, but when new technologies appeared that made my life easier while still providing performance and maintainability, I would work with them until comfortable and then scrap the previous technology, not just cram it in my toolbox. The only time I would use the prior technologies is when I was forced to by either an employer or legacy code I had to maintain. Currently I work in C#/.NET and do not touch any other technology from the past. If I have a new project to build it will be C#/.NET. If work would come my way requiring a past technology (such as C/C++, PHP, etc) for new development, I would not take the work. To me it is a waste of my time to work on legacy systems when there is so much work to be done in the current technology I use. Anyone else out there that just loves technology and burns their bridges to past technology to remain focused on the current technology?
Rocky <>< Latest Code Blog Post: ASP.NET HttpException - Cannot use leading "..".. Latest Tech Blog Post: Vista - Little Things
To me, code packages written in C# for the same task are simpler than those written C++. So sometimes I write them in C# and then use another C# program already written to translate it into C++ to take advantage of C++ that several poster in this thread have already listed so that the complexities of C++, which is abstracted away by C#, can be handled or generated consistently. In C++, you have complete control over memory and low level issues so that you do not have to be bound to a particular memory management pattern--a shoe that may not fit all feet. Also you can extend your already existing systems, legacy or not, to take advantage of your new packages in both of you code bases, namely C++ and C# ones. Regards, S. Ying CryptoGateway
-
You must have missed the part about "if it's the most appropriate". Of course "appropriate" is a relative term. For me, .Net is appropriate only if the task is a web app, or if the requirements state that the task must be implemented with .net. Even then, I will still select C++ over anything else, again, if given the choice. C# has no benefit over C++ where desktop apps are concerned. At work, I have to support VC6 apps, VS2005 C++/MFC apps, and I'm working on a ASP/C# web app. At home, I'll always use C++/MFC because *I'm* dictating the requirements for those applications. I'll probably always use PHP for my web stuff because ASP/C# has left a really bad taste in my mouth (of course, it might be because the IDE is a completely unstable piece of CRAP, but that's a topic for another thread).
"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, 10/26/2001I can not speak for the web side of things, as I have never done any of that stuff (except for some SMTP email capabilities). But for desktop apps I will definitely be sticking to C++, with or without MFC. When I first joined CP the debate was between MFC and WTL. Many of the 'progressive' people were claiming that WTL was better because it did not require the 1MB runtime that came with MFC. Smaller, lighter footprint and all that. Now those same people (for the most part) are claiming that C#, with it's 21MB runtime is the better way to go. Come on guys, make up your minds, which is better? C++ with a relatively small footprint, that *may* take slightly longer to develop, or C# with it's huge runtime?
You may be right
I may be crazy
-- Billy Joel --Within you lies the power for good, use it!!!
-
I can not speak for the web side of things, as I have never done any of that stuff (except for some SMTP email capabilities). But for desktop apps I will definitely be sticking to C++, with or without MFC. When I first joined CP the debate was between MFC and WTL. Many of the 'progressive' people were claiming that WTL was better because it did not require the 1MB runtime that came with MFC. Smaller, lighter footprint and all that. Now those same people (for the most part) are claiming that C#, with it's 21MB runtime is the better way to go. Come on guys, make up your minds, which is better? C++ with a relatively small footprint, that *may* take slightly longer to develop, or C# with it's huge runtime?
You may be right
I may be crazy
-- Billy Joel --Within you lies the power for good, use it!!!
-
I can not speak for the web side of things, as I have never done any of that stuff (except for some SMTP email capabilities). But for desktop apps I will definitely be sticking to C++, with or without MFC. When I first joined CP the debate was between MFC and WTL. Many of the 'progressive' people were claiming that WTL was better because it did not require the 1MB runtime that came with MFC. Smaller, lighter footprint and all that. Now those same people (for the most part) are claiming that C#, with it's 21MB runtime is the better way to go. Come on guys, make up your minds, which is better? C++ with a relatively small footprint, that *may* take slightly longer to develop, or C# with it's huge runtime?
You may be right
I may be crazy
-- Billy Joel --Within you lies the power for good, use it!!!
PJ Arends wrote:
C++ with a relatively small footprint, that *may* take slightly longer to develop, or C# with it's huge runtime?
The size of the runtime isn't really the issue for me (or my particular set of clients). Marc
People are just notoriously impossible. --DavidCrow
There's NO excuse for not commenting your code. -- John Simmons / outlaw programmer
People who say that they will refactor their code later to make it "good" don't understand refactoring, nor the art and craft of programming. -- Josh Smith -
I can not speak for the web side of things, as I have never done any of that stuff (except for some SMTP email capabilities). But for desktop apps I will definitely be sticking to C++, with or without MFC. When I first joined CP the debate was between MFC and WTL. Many of the 'progressive' people were claiming that WTL was better because it did not require the 1MB runtime that came with MFC. Smaller, lighter footprint and all that. Now those same people (for the most part) are claiming that C#, with it's 21MB runtime is the better way to go. Come on guys, make up your minds, which is better? C++ with a relatively small footprint, that *may* take slightly longer to develop, or C# with it's huge runtime?
You may be right
I may be crazy
-- Billy Joel --Within you lies the power for good, use it!!!
PJ Arends wrote:
Smaller, lighter footprint and all that. Now those same people (for the most part) are claiming that C#, with it's 21MB runtime is the better way to go.
Yes that is the funniest part. I am not only talking about CP but various other places too.
Debugging is twice as hard as writing the code in the first place. Therefore, if you write the code as cleverly as possible, you are, by definition, not smart enough to debug it. -Brian Kernighan
