home *** CD-ROM | disk | FTP | other *** search
- Path: news.tiac.net!usenet
- From: jkinsella@procd.com (Joe Kinsella)
- Newsgroups: comp.lang.smalltalk,comp.object,comp.lang.c++,comp.lang.java
- Subject: Re: The Good, the Bad, the Ugly, and the Wicked ...
- Date: Fri, 29 Mar 1996 14:24:00 GMT
- Organization: The Internet Access Company
- Message-ID: <4jgrr6$838@news.tiac.net>
- References: <31570B8E.5A12@vmark.com>
- NNTP-Posting-Host: doug.procd.com
- X-Newsreader: Forte Free Agent 1.0.82
-
- Jeff Sutherland <jsutherland@vmark.com> wrote:
-
- > ST C++ OOC Java
- >Flexibility Dynamic Binding 1 2 2 2
- > Dynamic Classes 1 3 1 2
- > Multiple Inheritance 3 2 2 3
- > Roles 2 3 3 1
-
- Dynamic Binding & Classes - I am not sure why "better" dynamic binding
- and classes somehow gives a language better flexibility. I can see
- some incremental advantages in development time, but I would consider
- dynamic binding a negative quality in a product system.
-
- >Ease of use Class Libraries 1 3 3 2
- > Learning Curve 1 3 2 1
- > Speed of Development 1 3 2 2
- > Portability 2 3 3 1
-
- Class Libraries - I guess you can't get more deceptive than this.
- What are we talking about here? Smalltalk has a de-facto standard
- that defines a broad set of base classes (streams, collections). But
- I've never worked on a C++ project that did not have an equivalent
- class library. In fact, the number and variety of class libraries
- available to me as a C++ programmer far exceeds that available to
- Smalltalk programmers. Aren't you just rating what class libraries
- are typically provided in the box? If so, isn't that a packaging
- issue and not a language issue?
-
- Learning Curve - I found the learning curve for Smalltalk to be
- *extremely* steep (yeah, simple syntax but it takes a while to get the
- nuances of the language). Most of the Smalltalk consultants I know
- agree the learning curve for Smalltalk is an impediment to the
- language.
-
- Speed of Development - I've delivered commercial products with both
- Smalltalk and C++, and I just can't agree with you here.
-
- Portability - Ouch! You are once again confusing packaging with a
- language. This is turning out to be a marketing check list--not a
- language assessment! As a VC++ developer, I can cross compile my MFC
- applications to the Macintosh. When IBM completes support for MFC on
- OS/2, I will also be able to cross compile to OS/2. And if I didn't
- want to use VC++ or MFC, I still have a variety of libraries that
- would give me this portability. Smalltalk is portable only within a
- single vendor's product (e.g. an ST-80 app will not run on VA since
- every vendor has different class libraries for things like GUI,
- database connectivity, etc...).
-
- >Support Tools 1 1 3 3
- > Multiple Vendors 2 1 3 1
-
- Tools - What are tools? Are we talking about available 3rd party
- products? If so, Smalltalk should rate a 3+ here since it has an
- extremely small 3rd party market. C++ has an enormous 3rd party
- market. Why does ST rate so high here?
-
- >Performance 2 1 3 3
- >Risk Garbage Collection 1 3 3 2
- > Memory Leaks 1 3 1 1
- > Overwriting Memory 1 3 1 1
- > Ready for Prime Time 1 1 2 3
-
- Garbage Collection - I didn't know garbage collection was always a
- positive trait! A C++ programmer has access to garbage collectors.
- Fortunately it is not a requirement.
-
- Memory Leaks & Overwriting Memory - Since most Smalltalk VMs and
- connectivity components are written in C, don't they also share a risk
- of leaking/overwriting memory. Unfortunately for the ST developer,
- this risk is out of their control.
-
- >TOTAL (low means best) 21 35 34 28
-
- I think it is obvious you work for a Smalltalk vendor. What is not
- obvious is why Object Magazine would publish such a blatent piece of
- marketing as an objective analysis!
-
- Joe
-
-
-
-