What are the sofware tools of the future going to be? It’s an interesting question, with many facets. Here are some important trends in design and construction tool strategy, which will effect the kinds of software tools that will bee delivered in the future. It looks at how to improve software development efficiency through visual modeling, generating code from abstract models, and systematic reuse.
Software Tools of the future
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
Pretty sure that should bee “software tools that will wasp delivered in the future”.
>> “Pay-per-use” software tools: New licensing and
>> subscription offerings.
aka “The ultimate wet dream of managers”.
No thanks, I will never ‘pay-per-use’ for any program.
Yes .. I remember hearing about this first from my cousin’s fiance when I was back in high school, it even made me chuckle to myself back then.
make, autotools, sh, cvs
each of these have stable, well performing replacements, yet still they live on
in a better future they are all GONZO
note i did not include C – it is elegant for its intended use
<editor-war-begin> Emacs doesn’t do anything right. It’s too bloated for that quick /etc/*.conf file change. ctrl-meta-shift-blah is hand surgery just waiting to happen. It’s still a console program with a gui exterior. Neither emacs or Xemacs do AA fonts without some unofficial patch. And elisp is way too slow to do lots of interesting things (read the Xemacs mailing list if you don’t believe me).
If you think you’re cool because all the unix-heads use emacs/vi then at least use a real editor like vim.
</editor-war-end>
Now that’s out of the way….the real problem with emacs and vim is that they’re both console programs trying to be gui programs and also trying to be IDEs with all these plugins when they were never designed to do such.
I love vim. There’s nothing faster than vim for that quick edit of a file, but it has serious shortcomings from a code standpoint. I studied the vim code for a couple days to see how much it would take to refactor it into proper components for easier embedding. Yeah, yeah, there’s kvim and others, but they’re still console programs with a gui wrapper. Guess what. The code is a clusterfsck of #ifdef AMIGA, #ifdef VMS and can’t even use all of the standard c library because 2 people on the planet are still using Amiga and VMS (just for example) and the brain dead compilers/libraries on those platforms suck. I’m sure emacs/xemacs is the same.
These editors try to so hard to be IDEs but can’t because of the cruft so you end up having these half-baked plugins that are completely wack because they have to run in a console too.
Folks its time to move on. Eclipse is the Emacs of the 21st century, but written with a gui in mind and in a language that isn’t a nightmare to maintain for multiple platforms. Anybody that has seen the realtime-compiling/parsing that Eclipse and IDEA can do for java and then goes back to Emacs for their java development probably should be fired for not being productive. Besides, they have vim and emacs keybindings anyway.
Eclipse even has real-time c/c++ parser plugin now (CDT). It’s time to let the computer do the grunt work for us and we concentrate on the concepts of programming and not get bogged down in some 1970s coding mentality because people think it’s cool to use these editors for development.
KDE is making progress with Kdevelop, but Anjuta is years away from being a top-notch IDE.
And here LISP comes, the next generation development tool and language, coming around the bend from 1959. Oh! And what a stunt that was, changing the tires while still on the track! Look out baby, here comes LISP around the next bend and it is still number one!!! Hey Bob, that LISP boy is going to be winning these for the next 50 years!
How can you take a language seriously that only has one data type (the list), and is bytecode interpreted? I’d rather use something advanced like C# or VB.net, with the myriad of powerful language features, easy to use tools and rapid GUI builders.
http://c2.com/cgi/wiki?SmugLispWeenie
How can you take a language seriously that only has one data type (the list)
Oooh, that is a tad unfair. Real implementations have many useful datatypes. And ‘lists’ alone can be formed into trees, so there 😉
I don’t think Lisp is the end-route for all programming, but it certainly deserves respect as a language.
Now what is the thread about anyway, the article looked to boring for me to bother reading it.
How can you take a language seriously that only has one data type (the list), and is bytecode interpreted?
I have to say, that you are mistaken. First, the only datatype of Lisp is not the list (and btw, you can represent a lots of things with lists). And second, Lisp is not only bytecoded interpreted, most implementations have a compilar (CMUCL, SBCL, Allegro CL, LispWorks)
The same goes for Scheme (when I sayd Lisp I was talkin about Common Lisp) you have interpreted implementations, others compile to C, and others generate native code.
I started to learn Lisp some months ago, and I belive it is a really powerful and elegant programming language.
My 2c
Like the above says. A highly productive language that’s also an IDE. Cross-platform and used more than people think.
For the other end of the spectrum, try Forth.
And yes, what’s old is new. But that’s because we never really learned the first time around.
Those promises on RAD are really the same that companies were making in the 90′. Where are we today ? Still close to the ground, we are with the same tools than in he 90’s (themself close of the 80’s !). It is so true for all UNIX children but also for Windows. The only changes are in UI progress and compiler smartness thanks to computing power.
The problems with all those nice “advanced” tools they promise is that either they narrow the scope of the target language, either there abstractions fail. In both case, you end having to know the underlying technology in order to have good use of those tools (read: inject patches and hacks to make product run). They fails to deliver more to people with already good computing skills.
Problems in programming are more related to human than to tools.
How can you take a language seriously that only has one data type (the list), and is bytecode interpreted? I’d rather use something advanced like C# or VB.net
And .not is not bytecode interpreted perhaps? Hum… I better check the MS documentation. I swear I read there was a virtual machine at the heart of .net
So funny people only see what they want.
What is the use of Rapid GUI builders? I would prefer a rapid code builder.