What if...Java hadn't been invented....

Would anything have popped up in its place, or would we all still be using
C++?



Mike

"Mike B" <someone@invalidaddress2.someplace> wrote in message
news:ke9n8017tdd6v0tdcqu73j17tvifq808cu@4ax.com...
> Would anything have popped up in its place, or would we all still be using
> C++?

Flash would have matured much more quickly.

On Sun, 25 Apr 2004 19:55:04 +0800, Mike B
<someone@invalidaddress2.someplace> wrote or quoted :

>Would anything have popped up in its place, or would we all still be using
>C++?

Oh heavens yet. Java has a slew of competitors. Politics is even
more important than technical merit in which one prevails.

Java's niche is "safe, rigidly multiplatform language".


Java's other niche is "dynamic loading of code".


In parallel universes we might have seen an evolution of Eiffel,
Dylan, Sather, Ada, Oberon, Scheme, C# ... to take over where Java is
now.

See http://mindprod.com/jgloss/language.html


Java succeeded partly because it originally aimed to be small and fast
enough to run a TV settop box.

--
Canadian Mind Products, Roedy Green.
Coaching, problem solving, economical contract programming.
See http://mindprod.com/jgloss/jgloss.html for The Java Glossary.

Mike B <someone@invalidaddress2.someplace> wrote in message news:<ke9n8017tdd6v0tdcqu73j17tvifq808cu@4ax.com>. ..
> Would anything have popped up in its place, or would we all still be using
> C++?


i would be busy creating genetically-engineered monsters late at night
while STILL pursuing a PhD in Molec Bio after 10 years in grad school.

On Sun, 25 Apr 2004 19:56:27 GMT, Roedy Green <see@mindprod.com.invalid>
wrote:

> On Sun, 25 Apr 2004 19:55:04 +0800, Mike B
> <someone@invalidaddress2.someplace> wrote or quoted :
>
>> Would anything have popped up in its place, or would we all still be
>> using
>> C++?
>
> Oh heavens yet. Java has a slew of competitors. Politics is even
> more important than technical merit in which one prevails.
>
> Java's niche is "safe, rigidly multiplatform language".
>
>
> Java's other niche is "dynamic loading of code".
>
>
> In parallel universes we might have seen an evolution of Eiffel,
> Dylan, Sather, Ada, Oberon, Scheme, C# ... to take over where Java is
> now.
>
> See http://mindprod.com/jgloss/language.html
>
>
> Java succeeded partly because it originally aimed to be small and fast
> enough to run a TV settop box.
>

Gee - nobody remembers Magic Cap - they had a scheme for portable
code before Java. Unfortunately they didn't have the huge cash
reserves to release it for free like Sun did so it quietly went
away.

See the post-mortem here:
http://www.pencomputing.com/magic_cap/

Bill

William Brogden <wbrogden@bga.com> wrote in message news:<opr62ednstu0i8d5@giga.realtime.net>...

> Gee - nobody remembers Magic Cap - they had a scheme for portable
> code before Java. Unfortunately they didn't have the huge cash
> reserves to release it for free like Sun did so it quietly went
> away.
>
> See the post-mortem here:
> http://www.pencomputing.com/magic_cap/
>

Funny.

But don't forget - for something to be remembered, it has first to be known.

William Brogden wrote:
[snip]
> Gee - nobody remembers Magic Cap - they had a scheme for portable
> code before Java. Unfortunately they didn't have the huge cash
> reserves to release it for free like Sun did so it quietly went
> away.
>
> See the post-mortem here:
> http://www.pencomputing.com/magic_cap/

virtual machines running portable code have a long tradition---the ucsd
p-system had this, in the late 70s, and i am sure it was not the first
(implementations of lisp, e.g., certainly used vm-like interpreters long
before that).

-- j

On Mon, 26 Apr 2004 11:09:45 -0800, "Jörn W. Janneck" <jwjanneck at yahoo dot
com> wrote:

>William Brogden wrote:
>[snip]
>> Gee - nobody remembers Magic Cap - they had a scheme for portable
>> code before Java. Unfortunately they didn't have the huge cash
>> reserves to release it for free like Sun did so it quietly went
>> away.
>>
>> See the post-mortem here:
>> http://www.pencomputing.com/magic_cap/
>
>virtual machines running portable code have a long tradition---the ucsd
>p-system had this, in the late 70s, and i am sure it was not the first
>(implementations of lisp, e.g., certainly used vm-like interpreters long
>before that).
>

So what made Java "special" compared to the others, and why is Python
suddenly flavor of the month?

Perhaps the Python phenomenon is partly explained by the fact that an average
PC has 100 times the processing power and memory compared to those from the
early 1990's. Essentially the disk I/O bottleneck, which was a major problem
is now, in many cases, the only problem.

Mike

Mike B wrote:
> On Mon, 26 Apr 2004 11:09:45 -0800, "Jörn W. Janneck" <jwjanneck at yahoo
dot
> com> wrote:
>
>> William Brogden wrote:
>> [snip]
>>> Gee - nobody remembers Magic Cap - they had a scheme for portable
>>> code before Java. Unfortunately they didn't have the huge cash
>>> reserves to release it for free like Sun did so it quietly went
>>> away.
>>>
>>> See the post-mortem here:
>>> http://www.pencomputing.com/magic_cap/
>>
>> virtual machines running portable code have a long tradition---the ucsd
>> p-system had this, in the late 70s, and i am sure it was not the first
>> (implementations of lisp, e.g., certainly used vm-like interpreters long
>> before that).
>>
>
> So what made Java "special" compared to the others,

hard to say, but possible explanations might be: it was object-oriented,
garbage-collected, looked a lot like c/c++, came with a useful (by the
standards of the day) library, and a "cool" (albeit perhaps somewhat
synthetic) application scenario (applets) that sat on top of the
internet-wave.

sure, other languages/systems had some of the same features, but maybe it
was the mix, the timing, and cunning marketing that made it happen for java.

> and why is Python
> suddenly flavor of the month?
>
> Perhaps the Python phenomenon is partly explained by the fact that an
average
> PC has 100 times the processing power and memory compared to those from
the
> early 1990's. Essentially the disk I/O bottleneck, which was a major
problem
> is now, in many cases, the only problem.

yes, that sounds reasonable. the availability of a lot of processing power
certainly changes the relative importance of language characteristics.

-- j

On Tue, 27 Apr 2004 10:30:26 -0800, "Jörn W. Janneck" <jwjanneck at
yahoo dot com> wrote or quoted :

>looked a lot like c/c++,

I think it was the superficial resemblance to C++. It made people feel
comfortable.

Yet the things I like least about Java are kludgy things inherited
from C, like switch, the () {} forests, the mishmash of prefix/postfix
notation, and the for loop that can do anything but eat if you don't
get every keystroke bang on.

--
Canadian Mind Products, Roedy Green.
Coaching, problem solving, economical contract programming.
See http://mindprod.com/jgloss/jgloss.html for The Java Glossary.

Roedy Green wrote:
> On Tue, 27 Apr 2004 10:30:26 -0800, "Jörn W. Janneck" <jwjanneck at
> yahoo dot com> wrote or quoted :
>
>> looked a lot like c/c++,
>
> I think it was the superficial resemblance to C++. It made people feel
> comfortable.

perhaps, but this still begs the question why so many of the recently
popular languages look so very unlike c/c++. personally, i think that this
issue (i.e. superficial similarity to the currently popular language) is
similarly overrated as the issue of "native" look and feel. of course,
programmers do not want to needlessly learn a new language that does pretty
much the same thing as a language they already know---they do resist
gratuitous differences, and quite rightly so.

but if a new language solves a real problem better than their current tools,
it is my experience that people will happily learn its syntax in order to be
able to use it. afaict, the myth of the conservative programmer is the
result of either (a) frustration on the part of language designers who were
unsuccessful in convincing their target audience of the value of their
proposition, or (b) calcification in management layers who used to be
technical once, but do not want to get involved with the new-fangled stuff
and thus project their own reluctance onto an assumed clientele.

real technical people, who have real problems to solve, will use whatever
tool it takes to solve them, with or without curly braces, and they do not
shy away from learning new stuff if they see the value of it.

that said, i agree that it was probably part of java's success to express
those constructs that it has in common with other popular languages in an
established syntax, even if that syntax is rather kludgy. otoh, it should
not be overlooked that java also broke with many deeply-engrained c
traditions---no include files, for instance, no top-level functions, no
pointers, automatic gc, etc. for years, poor souls would haunt java
newsgroups asking how they can do destructors, or complaining that they
can't, or trying to come up with proposals for extending the language to
allow them.

> Yet the things I like least about Java are kludgy things inherited
> from C, like switch, the () {} forests, the mishmash of prefix/postfix
> notation, and the for loop that can do anything but eat if you don't
> get every keystroke bang on.

apart from these superficial issues, there is also a whole lot of more
structural nonsense in the java design, otherwise one could just slap a new
parser onto it and have a great language. alas...

-- j

On Tue, 27 Apr 2004 12:07:12 -0800, Jörn W. Janneck <> wrote:
> Roedy Green wrote:
>> On Tue, 27 Apr 2004 10:30:26 -0800, "Jörn W. Janneck" <jwjanneck at
>> yahoo dot com> wrote or quoted :
>>
>>> looked a lot like c/c++,
>>
>> I think it was the superficial resemblance to C++. It made people feel
>> comfortable.
>
> perhaps, but this still begs the question why so many of the recently
> popular languages look so very unlike c/c++. personally, i think that this
> issue (i.e. superficial similarity to the currently popular language) is
> similarly overrated as the issue of "native" look and feel. of course,
> programmers do not want to needlessly learn a new language that does pretty
> much the same thing as a language they already know---they do resist
> gratuitous differences, and quite rightly so.
>
> but if a new language solves a real problem better than their current tools,
> it is my experience that people will happily learn its syntax in order to be
> able to use it. afaict, the myth of the conservative programmer is the
> result of either (a) frustration on the part of language designers who were
> unsuccessful in convincing their target audience of the value of their
> proposition, or (b) calcification in management layers who used to be
> technical once, but do not want to get involved with the new-fangled stuff
> and thus project their own reluctance onto an assumed clientele.

No, it is management which assumes that "nice syntax" automatically
means "academic impractical language".

On Tue, 27 Apr 2004 12:07:12 -0800, "Jörn W. Janneck" <jwjanneck at
yahoo dot com> wrote or quoted :

>apart from these superficial issues, there is also a whole lot of more
>structural nonsense in the java design, otherwise one could just slap a new
>parser onto it and have a great language.

That was my idea, to create a language almost identical to Java in
functionality but with cleaner syntax and a large dollop of sugar to
make the common idioms terse.

The various ideas I had are at http://mindprod.com/jgloss/bali.html

Some are gradually making their way into mainstream Java.

If JavaC were open source would be probably thousands of such
languages. Programmers who work alone might decide they could finally
have things THEIR way.

--
Canadian Mind Products, Roedy Green.
Coaching, problem solving, economical contract programming.
See http://mindprod.com/jgloss/jgloss.html for The Java Glossary.

On Tue, 27 Apr 2004 20:49:12 GMT, Roedy Green <see@mindprod.com.invalid> wrote:
> On Tue, 27 Apr 2004 12:07:12 -0800, "Jörn W. Janneck" <jwjanneck at
> yahoo dot com> wrote or quoted :
>
>>apart from these superficial issues, there is also a whole lot of more
>>structural nonsense in the java design, otherwise one could just slap a new
>>parser onto it and have a great language.
>
> That was my idea, to create a language almost identical to Java in
> functionality but with cleaner syntax and a large dollop of sugar to
> make the common idioms terse.

Been there, done that.

Sather or Dylan.

Dr Chaos wrote:
> On Tue, 27 Apr 2004 12:07:12 -0800, Jörn W. Janneck <> wrote:
[snip]
>> but if a new language solves a real problem better than their current
tools,
>> it is my experience that people will happily learn its syntax in order to
be
>> able to use it. afaict, the myth of the conservative programmer is the
>> result of either (a) frustration on the part of language designers who
were
>> unsuccessful in convincing their target audience of the value of their
>> proposition, or (b) calcification in management layers who used to be
>> technical once, but do not want to get involved with the new-fangled
stuff
>> and thus project their own reluctance onto an assumed clientele.
>
> No, it is management which assumes that "nice syntax" automatically
> means "academic impractical language".

afaict, "nice" does not even enter the equation, unless it is defined as
"similar to c". management do not care about esthetics (and neither should
they, arguably). their perception is that many people "want" c/c++, for
various (and varying) reasons, mostly associated with some form of legacy,
either as code, or as user qualification or even just user
preference/prejudice.

to some extent, that is even true, as people's imagination is influenced by
their past experience (the traditional problem in letting users tell you
what they want). to some extent, their perception comes from talking to
other managers. and the third ingredient is their own prejudice, from either
the time when they did technical work, or when they at least bothered to
scan technical publications.

-- j