Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.programming > #1286 > unrolled thread

What makes a programming language popular?

Back to article view | Back to comp.programming

Contents

  What makes a programming language popular? Aaron W. Hsu <arcfide@sacrideo.us> - 2012-02-05 19:33 -0600
    Re: What makes a programming language popular? BGB <cr88192@hotmail.com> - 2012-02-05 22:57 -0700
      Re: What makes a programming language popular? Fritz Wuehler <fritz@spamexpire-201202.rodent.frell.theremailer.net> - 2012-02-07 18:12 +0100
        Re: What makes a programming language popular? Rui Maciel <rui.maciel@gmail.com> - 2012-02-07 18:34 +0000
          Re: What makes a programming language popular? BGB <cr88192@hotmail.com> - 2012-02-07 15:25 -0700
    Re: What makes a programming language popular? tm <thomas.mertes@gmx.at> - 2012-02-06 23:27 -0800
      Re: What makes a programming language popular? BGB <cr88192@hotmail.com> - 2012-02-07 15:53 -0700
        Re: What makes a programming language popular? gremnebulin <peterdjones@yahoo.com> - 2012-02-08 04:07 -0800
          Re: What makes a programming language popular? BGB <cr88192@hotmail.com> - 2012-02-08 17:15 -0700
            Second Best Language Aaron W. Hsu <arcfide@sacrideo.us> - 2012-02-08 20:19 -0600
              Re: Second Best Language BGB <cr88192@hotmail.com> - 2012-02-08 21:27 -0700
      Re: What makes a programming language popular? gremnebulin <peterdjones@yahoo.com> - 2012-02-08 04:05 -0800

#1286 — What makes a programming language popular?

At any given time, a programming language can usually be thought of as
currently fashionable, rising in popularity, or losing popularity. I would
like to hear some thoughts on what makes a language popular, what brings it
back into popularity, or what might be the "marketing" strategy for anyone
who cares about their language making an impact and achieving wide use.

This is, of course, disregarding whether being a popular language is a good
thing or not. :-)

-- 
Aaron W. Hsu | arcfide@sacrideo.us | http://www.sacrideo.us
Programming is just another word for the lost art of thinking.

[toc] | [next] | [standalone]

#1289

On 2/5/2012 6:33 PM, Aaron W. Hsu wrote:
> At any given time, a programming language can usually be thought of as
> currently fashionable, rising in popularity, or losing popularity. I would
> like to hear some thoughts on what makes a language popular, what brings it
> back into popularity, or what might be the "marketing" strategy for anyone
> who cares about their language making an impact and achieving wide use.
>
> This is, of course, disregarding whether being a popular language is a good
> thing or not. :-)
>

some factors:
familiarity, how much does it look like what people already know and like?
what merits does it bring versus existing options?
what niches or use cases does it do well in?
how much name-recognition or feature-recognition does it have? (does it 
have popular features X, Y, and Z? does it sound like something else 
people like? how many people have heard of it? ...)
...

so, a few possibilities:
common syntax and semantics, say C-like or Java-like, or maybe an 
ECMAScript-style syntax, likely supporting packages and OO and similar;
generally does most things moderately well (it doesn't have to be 
perfect, but if *anything* is overly difficult vs the existing options, 
this will give a bad taste);
probably aiming for a popular niche, like apps/web/mobile/... or a 
"trendy" and rising niche (harder to find, if one has already heard of 
it, it is likely already hit its relative plateau or has become 
saturated, people can also try to "cause" niches, but this is harder);
having a very long feature list, especially "trendy" features (AKA: 
buzz-words);
probably a catchy yet familiar name, which carries desirable associations;
a spiffy logo, especially if it emphasizes the name or first letter in 
some catchy way;
plastering the above all over the internet (IOW: lots of advertising);
...


ultimately, it will likely all be pointless, as even if one does gain 
some popularity, unless their is some real substance, it will just as 
quickly fade away again (and, also, it is hard to compete with the large 
corporations who make a lot of business in these ways, and can pump 
millions of dollars into such campaigns, as well as "polishing the turd" 
far more than nearly any non-commercial or smaller-budget effort can 
manage).

probably better I think (for smaller projects) is to focus more on 
having at least moderately "solid" engineering in place, and with any 
luck maybe the engineering will speak for itself, but this by no means 
one will "win" the language wars (even a well engineered piece of 
technology may still never see the light of day if it turns out to be 
more of a "solution in search of a problem").

many people who support niche languages may also tend to overlook 
glaring holes or deficiencies as "not really a big deal". but, many 
people are not nearly so devoted, and can't be asked to learn the "best" 
way to work around some particular "non-issue" (if the community adopts 
a mindset of "it's not horribly broken, it's a feature", or worse, "this 
is how everything should be, just everyone else is ignorant/doesn't see 
the light/...", maybe, just maybe, it really is broken...).

everything ultimately boils down, in large part, to cost/benefit 
trade-offs, and the ever important question: "what do I get out of 
this?". people don't really care as much about "the future" or "the 
perfect solution" so much as its immediate benefit towards whatever task 
it is they are trying to get done at the moment, and a good tool is 
"just there".


or such...

[toc] | [prev] | [next] | [standalone]

#1294

> On 2/5/2012 6:33 PM, Aaron W. Hsu wrote:
> > At any given time, a programming language can usually be thought of as
> > currently fashionable, rising in popularity, or losing popularity. I would
> > like to hear some thoughts on what makes a language popular, what brings it
> > back into popularity, or what might be the "marketing" strategy for anyone
> > who cares about their language making an impact and achieving wide use.

The most obvious factor is what is the OS and its interface written in? If
not for UNIX being written in C with an interface defined in C header files,
C wouldn't have seen the light of day outside Bell Labs.

> some factors:
> familiarity, how much does it look like what people already know and like?

Historically irrelevant, now probably very relevant.

> what merits does it bring versus existing options?

Irrelevant. What do you need to use to get the job done in environment X?
That is what you will choose whether the language itself is good or bad.

The answer for UNIX-eunichs is always C

[toc] | [prev] | [next] | [standalone]

#1295

Fritz Wuehler wrote:

> The most obvious factor is what is the OS and its interface written in? 

This is irrelevant.  If this was the case then no interpreted language would 
ever have succeeded. Yet, Java is one of the most popular programming 
languages, along with perl, PHP and python.


> If
> not for UNIX being written in C with an interface defined in C header
> files, C wouldn't have seen the light of day outside Bell Labs.

This is stupid.  A group of smart people took the time to design and develop 
a new programming language because at that time no other programming 
language successfully scratched the itch they needed scratching.  As the 
world was packed with people who also wanted to scratch that particular 
itch, the C programming language quickly took off.  This has absolutely 
nothing to do with UNIX or "C header files".


>> some factors:
>> familiarity, how much does it look like what people already know and
>> like?
> 
> Historically irrelevant, now probably very relevant.

What do you mean by "historically"?  One of the reasons behind the success 
of programming languages such as C++, Java, C#, javascript and others is how 
they managed to become familiar by copying the syntax of the C programming 
language.


>> what merits does it bring versus existing options?
> 
> Irrelevant. What do you need to use to get the job done in environment X?
> That is what you will choose whether the language itself is good or bad.

The main reason that defines if a particular programming language is good or 
bad is if gets the job done and how well it does the job when compared with 
alternatives.  Therefore, why can you claim it is irrelevant?


> The answer for UNIX-eunichs is always C

How do you explain the multitude of software developed to run on UNIX and 
unix-like OSs that is actually developed in languages other than C?


Rui Maciel

[toc] | [prev] | [next] | [standalone]

#1296

On 2/7/2012 11:34 AM, Rui Maciel wrote:
> Fritz Wuehler wrote:
>
>> The most obvious factor is what is the OS and its interface written in?
>
> This is irrelevant.  If this was the case then no interpreted language would
> ever have succeeded. Yet, Java is one of the most popular programming
> languages, along with perl, PHP and python.
>

yeah.

it is not what the OS is written in, but what can most easily access the 
relevant functionality.

for many tasks, this means C or C++, but it is not always the case, and 
a "not entirely terrible" FFI can go a long ways to helping in this case 
(IOW: can I access the relevant APIs without writing a crapload of ugly 
cruft and boilerplate?...).


>
>> If
>> not for UNIX being written in C with an interface defined in C header
>> files, C wouldn't have seen the light of day outside Bell Labs.
>
> This is stupid.  A group of smart people took the time to design and develop
> a new programming language because at that time no other programming
> language successfully scratched the itch they needed scratching.  As the
> world was packed with people who also wanted to scratch that particular
> itch, the C programming language quickly took off.  This has absolutely
> nothing to do with UNIX or "C header files".
>

yeah...

otherwise, no one would have bothered to port C compilers to other, 
non-Unix, targets (such as MS-DOS).


>
>>> some factors:
>>> familiarity, how much does it look like what people already know and
>>> like?
>>
>> Historically irrelevant, now probably very relevant.
>
> What do you mean by "historically"?  One of the reasons behind the success
> of programming languages such as C++, Java, C#, javascript and others is how
> they managed to become familiar by copying the syntax of the C programming
> language.
>

agreed.

actually, in many cases each has made a fair amount of inroads by 
copying the syntax/semantics/... of preceding popular languages.

sort of know C++ or Java?... you should have little problem learning C#, ...

many languages which have looked new or unfamiliar have often been 
received coldly, and the less familiar it looks, the harder time it will 
have.

not that it has to dogmatically clone the preceding languages (and 
especially not their built-in limitations), but being close enough that 
people who know the old language can figure out the new language without 
too much pain or a steep learning curve seems fairly relevant.


>
>>> what merits does it bring versus existing options?
>>
>> Irrelevant. What do you need to use to get the job done in environment X?
>> That is what you will choose whether the language itself is good or bad.
>
> The main reason that defines if a particular programming language is good or
> bad is if gets the job done and how well it does the job when compared with
> alternatives.  Therefore, why can you claim it is irrelevant?
>

yep.

"getting the job done" is a big factor in what defines "merit".

it one is left to choose "well, I could use language X or I could use 
Perl?". if Perl can perform the task easily, but X would make it a 
horrible pain or have some other crippling drawback, this sort of counts 
against it having all that many "merits" (these are, in-fact, "demerits").

if the reverse is true:
the task would be horribly painful in Perl, but trivial in X, and X 
lacks any obvious drawbacks, well then, X has "merit" with regards to 
the task at hand.


merit is not built from "magic" or "what should be", but rather is built 
(much like nearly everything else) from cost/benefit tradeoffs and the 
pursuit of self-interest (not necessarily money, but pursuit of money is 
a potential form pursuit of self-interest may take).

so, the merit of something is thus how well it aligns with ones' 
self-interests (within the confines of bounded-rationality), which may 
in turn be defined by various other factors (both internal and external).


>
>> The answer for UNIX-eunichs is always C
>
> How do you explain the multitude of software developed to run on UNIX and
> unix-like OSs that is actually developed in languages other than C?
>

yep.

many use Java, and some use C# (even if, IMHO, Mono is sort of crap, but 
whatever...).

likewise, the world is not Unix, and I personally mostly develop on 
Windows (actually, if everything were about Unix, then probably the 
world wouldn't have become dominated by Windows systems...).


Windows in-fact had a fair amount of merit from the POV of most 
consumers and end-users:
was user friendly;
came by default on most new PCs;
wasn't absurdly expensive;
...

now, it had drawbacks early on:
was buggy, prone to crashing, ...
but this was outweighed by its other merits (and, MS has since greatly 
improved stability as well).

[toc] | [prev] | [next] | [standalone]

#1293

On 6 Feb., 02:33, Aaron W. Hsu <arcf...@sacrideo.us> wrote:
> At any given time, a programming language can usually be thought of as
> currently fashionable, rising in popularity, or losing popularity. I would
> like to hear some thoughts on what makes a language popular, ...

Marketing done by a big company or organization.

Consider languages like Java, C#, Objective-C, VisualBasic, Delphi,
PL/SQL and Go. All backed more or less by big companies, which
do a lot of marketing (directly or indirectly) for these languages.


Greetings Thomas Mertes

--
Seed7 Homepage:  http://seed7.sourceforge.net
Seed7 - The extensible programming language: User defined statements
and operators, abstract data types, templates without special
syntax, OO with interfaces and multiple dispatch, statically typed,
interpreted or compiled, portable, runs under linux/unix/windows.

[toc] | [prev] | [next] | [standalone]

#1297

On 2/7/2012 12:27 AM, tm wrote:
> On 6 Feb., 02:33, Aaron W. Hsu<arcf...@sacrideo.us>  wrote:
>> At any given time, a programming language can usually be thought of as
>> currently fashionable, rising in popularity, or losing popularity. I would
>> like to hear some thoughts on what makes a language popular, ...
>
> Marketing done by a big company or organization.
>
> Consider languages like Java, C#, Objective-C, VisualBasic, Delphi,
> PL/SQL and Go. All backed more or less by big companies, which
> do a lot of marketing (directly or indirectly) for these languages.
>

marketing can help, but it is not what ultimately "makes" a language.

even if well advertised, if the language is "obviously crap" (IOW: has 
major or obvious drawbacks), then any such popularity will soon fade 
away as people loose interest in favor of "better" options.

similarly, many languages have gained what positions they have mostly by 
rising from the bottom up (and spreading via word-of-mouth and similar).


granted, popularity is an aggregate factor (as determined by what most 
people "actually" use), rather than necessarily the opinions of a 
small-but-vocal minority.

sometimes people confuse this (thinking this vocal minority is in-fact 
the majority opinion), which would likely seem to result in a situation 
where society seems to be saying one thing but doing another (as it more 
follows along with the position of the silent majority).

[toc] | [prev] | [next] | [standalone]

#1302

On Feb 7, 10:53 pm, BGB <cr88...@hotmail.com> wrote:
> On 2/7/2012 12:27 AM, tm wrote:
>
> > On 6 Feb., 02:33, Aaron W. Hsu<arcf...@sacrideo.us>  wrote:
> >> At any given time, a programming language can usually be thought of as
> >> currently fashionable, rising in popularity, or losing popularity. I would
> >> like to hear some thoughts on what makes a language popular, ...
>
> > Marketing done by a big company or organization.
>
> > Consider languages like Java, C#, Objective-C, VisualBasic, Delphi,
> > PL/SQL and Go. All backed more or less by big companies, which
> > do a lot of marketing (directly or indirectly) for these languages.
>
> marketing can help, but it is not what ultimately "makes" a language.
>
> even if well advertised, if the language is "obviously crap" (IOW: has
> major or obvious drawbacks), then any such popularity will soon fade
> away as people loose interest in favor of "better" options.

eg. PASCAL, hyped by being used in college courses but still fatally
flawed.

[toc] | [prev] | [next] | [standalone]

#1304

On 2/8/2012 5:07 AM, gremnebulin wrote:
> On Feb 7, 10:53 pm, BGB<cr88...@hotmail.com>  wrote:
>> On 2/7/2012 12:27 AM, tm wrote:
>>
>>> On 6 Feb., 02:33, Aaron W. Hsu<arcf...@sacrideo.us>    wrote:
>>>> At any given time, a programming language can usually be thought of as
>>>> currently fashionable, rising in popularity, or losing popularity. I would
>>>> like to hear some thoughts on what makes a language popular, ...
>>
>>> Marketing done by a big company or organization.
>>
>>> Consider languages like Java, C#, Objective-C, VisualBasic, Delphi,
>>> PL/SQL and Go. All backed more or less by big companies, which
>>> do a lot of marketing (directly or indirectly) for these languages.
>>
>> marketing can help, but it is not what ultimately "makes" a language.
>>
>> even if well advertised, if the language is "obviously crap" (IOW: has
>> major or obvious drawbacks), then any such popularity will soon fade
>> away as people loose interest in favor of "better" options.
>
> eg. PASCAL, hyped by being used in college courses but still fatally
> flawed.

yeah. it wasn't like no one heard about Pascal, just no one particularly 
wants to use it either...

so, marketing and hype can't "make" a language, it also has to be 
something people will willingly want to use...

[toc] | [prev] | [next] | [standalone]

#1306 — Second Best Language

BGB <cr88192@hotmail.com> writes:

>yeah. it wasn't like no one heard about Pascal, just no one particularly 
>wants to use it either...

Knuth used Pascal in TeX, and I believe he said that he did so not because
he liked the language, but because it was everyone's second best language. I
wonder what the equivalent is these days? C?


-- 
Aaron W. Hsu | arcfide@sacrideo.us | http://www.sacrideo.us
Programming is just another word for the lost art of thinking.

[toc] | [prev] | [next] | [standalone]

#1309 — Re: Second Best Language

On 2/8/2012 7:19 PM, Aaron W. Hsu wrote:
> BGB<cr88192@hotmail.com>  writes:
>
>> yeah. it wasn't like no one heard about Pascal, just no one particularly
>> wants to use it either...
>
> Knuth used Pascal in TeX, and I believe he said that he did so not because
> he liked the language, but because it was everyone's second best language. I
> wonder what the equivalent is these days? C?
>

possibly C, or maybe Java...

[toc] | [prev] | [next] | [standalone]

#1301

On Feb 7, 7:27 am, tm <thomas.mer...@gmx.at> wrote:
> On 6 Feb., 02:33, Aaron W. Hsu <arcf...@sacrideo.us> wrote:
>
> > At any given time, a programming language can usually be thought of as
> > currently fashionable, rising in popularity, or losing popularity. I would
> > like to hear some thoughts on what makes a language popular, ...
>
> Marketing done by a big company or organization.
>
> Consider languages like Java, C#, Objective-C, VisualBasic, Delphi,
> PL/SQL and Go. All backed more or less by big companies, which
> do a lot of marketing (directly or indirectly) for these languages.

Counteraxamples: C, python, ruby

[toc] | [prev] | [standalone]

Back to top | Article view | comp.programming

csiph-web