Re: IDE for C++: KDevelop vs Emacs vs Other

• Previous message: Daniel James: "Re: IDE for C++: KDevelop vs Emacs vs Other"
• In reply to: LEE Sau Dan: "Re: IDE for C++: KDevelop vs Emacs vs Other"
• Next in thread: LEE Sau Dan: "Re: IDE for C++: KDevelop vs Emacs vs Other"
• Reply: LEE Sau Dan: "Re: IDE for C++: KDevelop vs Emacs vs Other"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Date: Tue, 30 Mar 2004 12:05:51 +0100

In article news:<m3wu55anrt.fsf@mika.informatik.uni-freiburg.de>, LEE
Sau Dan wrote:
> Emacs has provided a "last-resort" method to handle this case:
>
> M-x function-name
>
> Of course, you need to know the function name to use this ...

Doesn't help much then, does it? When you get to the stage of wanting
to use an operation for which you can't remember thee keystrokes (and
possibly can't remember the exact name either) it becomes *much*
easier to have some sort of menu of commands.

> Daniel> not because it's the *best* way to do these things, but
> Daniel> simply because it *can* do them ... and it's just easier
> Daniel> to use the tool you know.
>
> But it's often better, if not the best. Think about using RCS.
> Those confusing 'ci' and 'co'. I started learning and using RCS
> via Emacs's simpler interface. Only after a few years, upon
> watchihng VI-using colleagues ...

<aside>
Your pseudo-proportionally-spaeced NG postings are a pain to quote and
edit! I suppose you do that because you're using Gnus/Emacs to edit
your posts and so you can, without regard for whether it's a good
idea?
</aside>

That's not the point. You're using Emacs as an IDE, and comparing it's
extra utility as an IDE with the utility afforded by vi and
non-integrated tools. *Of course* an IDE is going to be more powerful
than standalone tools, that's not at issue. The question is whether
emacs *used as an editor* is as easy to learn/use as other editors;
there is also the question of whether emacs *used as an IDE* is as
easy to learn/use as other IDEs. The two questions don't necessarily
have the same answer.

> Think about M-x compile. Does GCC bring me to the offending lines
> automatically?

No, of course not. gcc is a compiler not an IDE.

> Emacs augments GCC (or other compilers) with that feature.

Yes. So do other IDEs. That's what they do. Please compare like with
like.

> Daniel> ... it may be a better editor than the ones I use, but
> Daniel> it's easier to use the tools I already know.
>
> Emacs often makes those tools even easier. And sometimes adds
> more value to those tools.

You misunderstand. I was talking about alternative editors that I
might use *instead of* Emacs, not tools I might use *with* Emacs.

Of course, many of those editors are *just* editors, while Emacs is
capable of being so much more. The question one has to ask is whether
it is better to have an integrated set of tools with a perplexingly
illogical interface or a mixture of non-integrated tools -- some of
which may have excellent interfaces and some of which may not. In fact
the most productive interface for most people usually turns out to be
a compromise -- a set of mostly-integrates tools with an interface
that's easy to use but not especially powerful. I'd put most of the
better Visual IDEs (KDE, Anjuta, Kylix, Visual Studio, etc.) into this
category. In terms of sheer power Emacs puts them to shame, but
they're much easier to work with and a hell of a lot of people do do a
hell of a lot of work with them.

This is what makes me so *cross* about Emacs - I do appreciate its
power, but I think it a crying shame that such a great tool should be
so hampered by a stupidly and unnecessarily arcane interface.

> I write both Prolog and LISP programs. I can't understand why it's
> so difficult to adapt from one to another.

[We're getting off-topic here ...]

I've written both. I learnt LISP first but had no difficulty switching
to Prolog. In fact, Prolog *always* seemed easier and more, well,
logical. IME LISP is unreadable and virtually unmaintainable. I would
never use LISP in a commercial environment (mind you, I'm not sure
that I'd use Prolog, either). Whoever named it Lots of Infuriating
Single Parentheses may have been joking, but they had a point.

> Prolog is nice in theory, but too ugly in practice (when
> cuts come into picture).

Prolog is nice for solving problems that lend themselves to expression
in Prolog. Unfortunately real-world problems often don't, and the cut
is a pragmatic way of imposing real-world constraints on the pure
logic of Prolog. I agree that it destroys the elegance of the language
.. and perhaps shows that such a language isn't really suitable for
real-world problems (which I'd say of LISP, too).

Cheers,
 Daniel.
 

• Previous message: Daniel James: "Re: IDE for C++: KDevelop vs Emacs vs Other"
• In reply to: LEE Sau Dan: "Re: IDE for C++: KDevelop vs Emacs vs Other"
• Next in thread: LEE Sau Dan: "Re: IDE for C++: KDevelop vs Emacs vs Other"
• Reply: LEE Sau Dan: "Re: IDE for C++: KDevelop vs Emacs vs Other"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]