|
I agree each language has its own unique characteristics. Many have a unique purpose. We should not all move to a general-purpose language that would be good for scientific and business and other purposes.
cheers,
Donsw
My Recent Article : Ajax Calendar Control
|
|
|
|
|
Right now, I am desiging a computer language called Tree. In this functional/imperative/declarative language, tree data structure is the first class citizen. All data structures are implemented, using trees, for example, array is a single level tree and a single-linked list is a tree where each parent node has only one child node. This Tree language works really well and optimized if your problem domain has something to do with tree data structures, example HTML DOM or XML DOM. And this Tree language of mine is going to trump every conceivable computer languages in this world, and they include C# and Haskell!!
World Domination for me!! *Evil laughs* Hahahaha.........
|
|
|
|
|
|
Presumably your program self destructs if you try to print it of print from it.
|
|
|
|
|
Ever had to solve the same problem in 2 or more languages? For the people who pay our salaries, this could be perceived as duplication in effort. Overall productivity improves if we reduce the number of languages, simply because we can solve the same problem fewer times.
|
|
|
|
|
I agree, legacy systems shouldn't be updated in newer languages. If it's written in COBOL, leave it in COBOL.
|
|
|
|
|
There will be a minimum of two languages possible #1 Whatever Microsoft promotes #2 Everyone else's
Not that I am particularly anti-Microsoft, just recognizing that they will always do their best to maintain a proprietary offering.
I really think we should reduce the number of languages. Being a pretty old programmer, I have seen many come and go. While there is always some reason behind the next new one, what really happens is that our whole industry just goes through a re-learning and re-development exercise primarily for our own benefit, whenever the next new "latest and greatest" comes along.
|
|
|
|
|
Generally one of the toughest things is switching development team to another language. Since than, studing the Java from developer's point of view looks more promising, cause thereafter you can program practically whole range of applications(desktop,web,mobile, etc...). I'm not talking about writing compilers, drivers, ... where the Assembly language is required.
|
|
|
|
|
While the goal of a single language would be worthwhile, it is difficult to see how it would work in practice.
For example, there are major differences in language requirements between Teaching langauges, Desktop languages, and Embedded langauages. While C# or VB are pretty good for the first two, I shudder to think of garbage collection during interrupt! C++ can be used for Embedded, and Desktop, but as a Teaching language it allows too many horrors. Assembler is quick - good for embedded sometimes - and would certainly teach people a thing or two, but as a Desktop langauge? Hmm.
I think the problem is that we need a major change in the way we program before we can consider a single langauge.
|
|
|
|
|
Indeed. "Horses for courses" is the saying I like to use.
The low-end embedded work I've been doing over the last 20+ years uses mostly C (not C++, except in some of the higher-end systems). C is still a better choice for the low-end processors.
However, there are many languages that fit other jobs better. My current employer, for instance, uses LabVIEW to run automated testing. We also use Python for automating some of the development tasks.
I have no objection to C# or VB for a Windows-based platform. Or Java, or C++ or whatever gets the job done efficiently. Sometimes, you have to adjust your expectations to the developer(s) as well.
As you say, though, it's the way things are done rather than the language in which they're done that needs changing. Too many people jump in with no thought about design, testability or maintainability.
It's possible to write garbage in any language.
Professional Geek,
Amateur Stage-Levelling Gauge
|
|
|
|
|
There are way too many tasks in different domains waiting for new solutions (and new languages). But new languages need to be bring radically new concepts (such as object or functional programming).
|
|
|
|
|
I dont agree with you. We don't need more languages than the few hundreds we already have today. But we need an active evolution on the existing ones. Then there will always be a good language to chose for in the future
Sidenote: What we don't need at all is this copy-cat language behaviour of some languages. They usually don't offer a big advantage and often realize with time that they really do need to implement all those hateful features they bragged about not needing in the beginning ... as seen with Java + C# who now copy the features from C++ which they loathed so much.
|
|
|
|
|
...Everything eventually works off of Binary. Just think, if we could just use that we would only need two buttons on our keyboard.
|
|
|
|
|
We might need more than two - back when you started computers by keying in binary on the front panel, you needed "address/data" toggles, plus "store" and "execute" before the damn thing would start. (Still more reliable than windows, tho.)
|
|
|
|
|
Why would you say windows is not relia
Sorry my windows locked
I complain about windows (can we say bloat) but it does pay some of the bills.
It also has improved over the years. I remember a 95 seminary where they said 95 has a half life so you WILL be reinstalling.
|
|
|
|
|
djj55 wrote: I remember a 95 seminary where the said ...
I didn't know one could get ordained in Windows, I guess it's how one becomes a Windows Evangelist.
|
|
|
|
|
as soon as i ascend to my rightful place as God Of Programming, i will forbid the use of any language other than Haskell.
compiler writers will be permitted to use Assembler appropriate to the machine. but only for the duration of the compiler project.
|
|
|
|
|
|
A true puritan would demand that the hardware consume Haskell directly.
Steve
|
|
|
|
|
Sorry, but Puritans would have the hardware assassinate Haskell, not consume it.
That's because Haskell's practitioners never stop proclaiming it to be the King of Languages.
Puritans use C it's a spartan language free of decoration, and why is it called C - short for Cromwell.
|
|
|
|
|
As long as this site still gets new or revised articles in pure C++ (no .NET stuff) and maybe even ongoing support for MFC (which I still need to use, alas) then I'm happy
Thanks a lot for keeping this site up and running guys!
PS: It would be nice to add Qt articles, it's a good library with honest C++ code in it! http://www.qtsoftware.com - Nokias Qt
|
|
|
|
|
I agree! Why did you put the "Joke" sign in your post?
|
|
|
|
|
I am in the opinion that keeping it simple makes good sense. Too many languages makes life too complicated trying to keep up with the newest thing that only lasted a few years. At some point, the average Joe might need a PHD just to code the simplest projct in the future and to make matters worse, by the time he is done getting that diploma, the language is of no value anymore. This hinders progress in my opinion. I love new enhancements to languages for sure, but don't go out of our way to make 5 new languages when one would suffice to accomplish the end means. Niche markets have their rewards for pay, but are a killer to those who choose them in a market like this one where the niche becomes a death sentance because no one wants it right now . Languages take time to learn, just like learning any other language. Visual basic was a good start in this direction and I think it will continue. C# is another good start, but personally I like to declare MyNewThing as anewwidget rather than anewwidget MyNewThing. Its a lot like which came first, the chicken or the egg. In my book, definately the chicken. I also realize that not all of you would agree with me. I am just saying, lets not waste time learning languages that might only last a short time and killing extra brain cells I might need later along the way.
|
|
|
|
|
I remember a "simple" language, that has been modified to be more than it ever was intended. Yes Visual Basic .NET can do a lot of things but look at the size. If people wanted more feature why not just go to another language? You can not argue that you would need to learn more as the jump from VB6 to .NET has a learning curve.
A language designed for a specific niche is not a bad thing as you can stream line it and do away with overhead. With all the GUI interfaces now a language could be as DiscoJimmy states translated on the fly.
Hey it is early here, so this may not make sense when I am fully awake.
djj
|
|
|
|
|
To keep academics employed, whether that be gainfully or painfully is for you to decide.
|
|
|
|