Go languag % used by programmers

[brakaman]

how is possible language Go appear in number 32 in tiobe list ?

Go only has 0,374% use percentaje. Very little.

That you think is missing from Go to be used by most programmers?

other languages of minor relevance that Go actually (I think) : Logo,
D, SAS, Q, Scheme, F#, etc. they are better classified

Is Project Google failure?

why?

[Gabor]

Fortran is number 35 in the same list.
It has a much longer history and ages well.

Would you classify it as a failure?

[Islan Dberry]

A language can be successful even if most programmers don't use it.

It's too early to ask if "Project Google" is a failure given the following:

- It's been less than two years since the language was announced.

- There are no English language books on the language.

- Library functionality is coming along nicely, but important functionality needed by some programmers is still missing.

The language is a success for me because it helps me get my job done.

[brakaman]

On 17 jul, 21:51, Gabor <g...@szfki.hu> wrote:
> Fortran is number 35 in the same list.
> It has a much longer history and ages well.
>
> Would you classify it as a failure?
>

Yes friend.

Fortran has its history, currently I think not many people are using
Fortran, is normal

But Go is a modern language, recent supported by Google.(no
comment....)

I think it should be in the top ten in Tiobe list.Otherwise I consider
it a failure.

[Paul Borman]

Fortran is still heavily used, just not in web/gui/academic environments.  I would say little new is probably being developed, though.  I do not know if non-fortran compilers have yet caught up to fortran compilers for high speed scientific computing.

I don't think you can use this sort of ranking to determine failure, though it can imply success.  A better metric would be what language is used when you have a choice.  I have a choice of 4 languages, and Go is what I am choosing.

You should also remember that Tiobe only counts what they can see, not what they can't.  Also, did you notice Objective C?  That seems to finally be doing well many years after its introduction.  It was #38 in 2006, but now is #6.  Objective C is 25 years old now.  Go, in 3 years, out performed Objective C in 25 (Go is at 32 and Objective C was at #38).  I think it is way to early to write Go's obituary.

    -Paul

[John Asmuth]

On Sunday, July 17, 2011 10:13:59 PM UTC+2, brakaman wrote:

I think it should be in the top ten in Tiobe list.Otherwise I consider
it a failure.

Utter tripe. 

[Jessta]

On Mon, Jul 18, 2011 at 5:43 AM, brakaman <jvte...@gmail.com> wrote:
> how is possible language Go appear in number 32 in tiobe list ?
>
> Go only has 0,374% use percentaje. Very little.
>
> That you think is missing from Go to be used by most programmers?

Stability. The standard library and the language are still changing.
Many programmers and businesses will avoid Go until development slows
a bit and the packages in the standard library get more stable
interfaces.

You should notice that every language in the top 20 is a least 10
years old, most are > 20 years old.
It takes time for people to decide to learn a new language and even
longer for companies to adopt to using it.

> other languages of minor relevance that Go actually (I think) :  Logo,
> D, SAS, Q, Scheme, F#, etc.  they are better classified

These languages are older or are being actively pushed by a big
company. Google isn't really marketing Go because it's still in active
development.

> Is Project Google failure?
>
> why?

--
=====================
http://jessta.id.au

[David Roundy]

On the contrary, Fortran is still widely used for new code development in the scientific community, and I'm sure it will continue to be widely used until someone develops a comparable language, with support for multidimensional arrays, the exponentiation operator, compiling to fast code, etc such as Fortran has.

Which is all rather unfortunate, since I have steadfastly refused to learn Fortran, and have yet to find a satisfactory alternative.

David

[Geoffrey Teale]

To sumarise a long presentation I gave to non-programmers:

- There are 12 million programmers in the world
- The majority of those programmers are scarcely qualified
- Most technology decisions are made by a combination of following the crowd and a false understanding of risk.
- The high cost and failure rate in software development is no coincidence.

Remember the Stevie Wonder rule - "When you believe in something you don't understand then you suffer".   In this case that means "Perhaps making programming language decisions based on what 12 million powerless idiots are doing isn't the golden road to glory and great hacks."

Go is a genuine attempt to improve the state of systems programing language beyond the point they reached in the early 1970s.  As a result the sort of people using it are mostly that small community of people who understand and care about the concens that drive such a development. 

You're not going to catch those 12 million people unless you can market heavily enough the idea that their future income depend on jobs/contacts built around go, but that goal just draws resources and energy away from making the language better.

Arguably Java also suffers from it's large community of corporate drones.   The slavish tendancy to build baroque, mausiliums of intricate classes, dense with state and dripping with verbose XML is a reflection of the unthinking insanity of the 12 million.

I'd rather a tiny community usee the language well, built succesful applications and organically grew the user base whilst establishing a clean, sane library base that might later be used to improve the lives of a wider population of programmers.

I've said more than enough, I'll trundle back to the twelve million and take my punishment now.
--
Geoff Teale

[⚛]

In my opinion, the reason Go is at place 32, instead of being at e.g.
place 9, is that there is currently little effort to rewrite existing
Linux applications and libraries. Take for example the programs "bash"
or "vim" that you can find on any Linux machine. These are the types
of applications that are best candidates to fit Go's programming
model. That is: system-level applications and libraries where you do
*not* care about extreme performance (C/C++/etc is best for this), but
you still *do* care about good-enough performance, about security
holes (such as buffer overflows) and about programmer productivity.

Additional candidates for a rewrite from C to Go also include:

- make
- the syslog daemon
- Midnight Commander
- acpid
- ntpd
- ALSA utils
- coreutils (/bin/sleep, /bin/wc, ...)

Visiting the /usr/bin, /usr/sbin and /usr/lib directories might reveal
additional applications that *people are using*, are written in C and
would benefit from a rewrite to Go. The most useful benefit is the
increased programmer productivity (thanks to garbage collection) and
minor improvements in the field of security. From performance
perspective, in many cases the end-user wouldn't even notice that the
application has been rewritten from C to Go.

The basic idea here is that people out of their free will just *stop
to use C* when fixing bugs in these applications and libraries, and
when extending them with new features. People would need to forget
about C, completely, when it comes to the question of future
development of these applications and libraries.

As some participants in this forum thread already mentioned, Go is
from time to time changing its language syntax, changing the
declarations of basic library functions and removing pointless stuff
from standard libraries - this might indeed be a problem. But, in my
opinion, this a minor issue. The consistency of Linux packages is
ensured by the various Linux distributions, and this model seems to be
working well. (Well, I know that Linus Torvalds thinks that breaking
an established programming interface is a serious crime, but let us
here simply ignore his opinion for the moment.)

On Jul 17, 9:43 pm, brakaman <jvtech...@gmail.com> wrote:

[Jan Mercl]

On Sunday, July 17, 2011 9:43:12 PM UTC+2, brakaman wrote:

Is Project Google failure?

This question can be answered in a non speculative mode somewhere around year 2021 ;-)

[Zippoxer]

I didn't see a programming language released in the last few years that is as (or even half) successful as Go.

The first time I tried Go (back in 2009), I quit really fast because I could not find support on google even for seem-to-be-common problems. I planned to go back to Go few months later but nothing really improved. But now in 2011 the "Give it time" saying worked.

So... Give it time. And you can make that time shorter by contributing to Go open source projects or by creating one (or even two).

[André Moraes]

> I think it should be in the top ten in Tiobe list.Otherwise I consider
> it a failure.

If you think it's a faliure and nobody should "give a damn" why are you here?

Since you are in this malling-list, you probablye care about the
language, so in order to not make it a failure
you could spread the language and talk about it with every programmer you known.

Also erlang was around for quite some time and only a few years (if
not months) it got some momentumn
because many "NoSQL" database was using it.

One hype generated another hype.

Maybe GO can get some hype and became number one :-D on tiobe's list.

But it's not a general failure.

--
André Moraes
http://andredevchannel.blogspot.com/

[Paulo Pinto]

No, Erlang got famous because most Ericson telephone switches
are running Erlang code.

So it showed the world how good funcional programming is actually
good for heavy millions of request per second kind of world that
switches
do.

That is the Erlangs sucess story, not the NoSQL momentum.

Microsoft already has quite a few prominent sucess stories of
big companies using F# .

Same goes for Scala, Clojure and Haskell, which have a few companies
100% using them and showing their products on conferences around the
world.

Go needs to have such one. So far the only really success story
for the outside world was that now it is possible to use Go on the
App Engine.

--
Paulo

[André Moraes]

> No, Erlang got famous because most Ericson telephone switches
> are running Erlang code.

Yes, that's why Erlang was created,
But it don't became a famous and common language until some time...

I cited Erlang because the language wasn't too famous but wasn't a failure,
similar to what is happening to Go right now

[Geoffrey Teale]

On 18 July 2011 14:45, Paulo Pinto <paulo....@gmail.com> wrote:

--- %< ----

Go needs to have such one. So far the only really success story
for the outside world was that now it is possible to use Go on the
App Engine.

--- %< ---

<sarcasm>

We should not fear for Go - I mean, it's ported to Plan9 there's a ready made market! :-)

</sarcasm> 

--

Geoff 

[Steve McCoy]

Who is Tiobe and why should I care where any language is ranked on its list?

[Uriel]

On Mon, Jul 18, 2011 at 2:45 PM, Paulo Pinto <paulo....@gmail.com> wrote:
> Go needs to have such one. So far the only really success story
> for the outside world was that now it is possible to use Go on the
> App Engine.
>

See http://go-lang.cat-v.org/organizations-using-go

[Michael Jones]

I had the same question. Google search finds:

http://www.tiobe.com/index.php/content/paperinfo/tpci/index.html

so I see what the OP was commenting on. I think it is most fascinating! Look at the design of the experiment:

http://www.tiobe.com/index.php/content/paperinfo/tpci/tpci_definition.htm

Programming Language

Before discussing how the ratings are calculated, first it needs to be clarified what counts as a programming language for the TIOBE index. There are 2 criteria that should both hold:

• The language should have an own entry on Wikipedia and it should clearly state that it concerns a programming language. This is the reason why ColdFusion, (Ruby on) Rails, Excel, Cocoa, ASP and AJAX are not considered programming languages for the index.
• The programming language should be Turing complete. As a consequence, HTML and XML are not considered programming languages. This also holds for data query language SQL. SQL is not a programming language because it is, for instance, impossible to write an infinite loop in it. On the other hand, SQL extensions PL/SQL and Transact-SQL are programming languages.

:

Ratings

The ratings are calculated by counting hits of the most popular search engines. The search query that is used is

+"<language> programming"

This search query is executed for the top 7 websites of Alexa that meet the following conditions:

• The entry page of the site contains a search facility
• The result of querying the site contains an indication of the number of page hits

Based on these criteria currently the following search engines are used:

• Google: 32%
• Blogger: 32%
• Wikipedia: 16%
• YouTube: 10%
• Yahoo!: 3%
• Bing: 3%
• Baidu: 3%

Amazing! Not in the Computer Science sense, or in a language virtue sense, or as a how much usage metric, but in the sense of "how much curiosity do people express globally" this could be a fair estimate as long as all interest is directly proportional to web searches (as opposed to netlib access, ftp downloads, lines of code, or sales of books.) And of course, there is the exestential aspect -- you are only a programming language if you have a Wikipedia article. ;-)

On Mon, Jul 18, 2011 at 7:10 AM, Steve McCoy <mcc...@gmail.com> wrote:

Who is Tiobe and why should I care where any language is ranked on its list?

--

Michael T. Jones

   Chief Technology Advocate, Google Inc.

   1600 Amphitheatre Parkway, Mountain View, California 94043

   Email: m...@google.com  Mobile: 650-335-5765  Fax: 650-649-1938

   Organizing the world's information to make it universally accessible and useful

[kar]

Go excels in server environment. especially 64-bit. probably the best
imo. different tools for different jobs.

[Pete Wilson]

I have mixed feelings responding to stuff that sounds like idiocy. But
I've had a couple of glasses o wine (Becker Provencale, and Jadot
Macon Villages)(both recommended as reasonably-priced enjoyable
tipple).

I simply do not understand why anybody could think that "Web 2.0"
popularity ratings of a language which introduces vastly important new
semantics (concurrency and message-passing) and which is being created
for the business purposes of a large and rich American company can
have anything whatsoever to do with its utility, future success, or
importance.

When a language creates new semantics (new in the sense of 'new to the
wide, wide world', not new in the sense of 'OMG, nobody thought of
that before') then one may reasonably expect a quite a long period of
consideration, learning and acceptance, especially if the new
semantics are broadly useful.

Thus the posting (I am being uncharacteristically rude)(no, really) is
an indication of either imbecility or a desire to troll.

I think in my next life, I will try to be a better person and think
better of others (no, really). Meanwhile, imbeciles and trolls, please
try to think about fundamentals. At worst, we'll get better-crafted
nonsense from you.

-- P

[Paulo Pinto]

The thing is, Go has 0 new features to offer over established
languages.

For every single Go feature I can provide a similar example in a
mainstream language.

Maybe Go's advantage is how these features are blended together, but
future
will tell how successful the language in the end is.

--
Paulo

[Paul Borman]

Of course, you mean you can provide a single mainstream language that has all the features, right?

[bflm]

On Jul 19, 7:59 am, Paulo Pinto <paulo.jpi...@gmail.com> wrote:
> The thing is, Go has 0 new features to offer over established
> languages.
>
> For every single Go feature I can provide a similar example in a
> mainstream language.

I don't think you can. Does e.g. standard C knows split stacks so it
can handle 100k goroutines in about half GB of memory?

[Paul Borman]

The Cray 2 had stack segments in the mid to late 80's.  Part of every language (C, Fortran, Pascal).  It was a necessity as the memory management hardware was a single base address/limit address pair.  However, none of these languages also attempted to provide a large number of "threads" (but you didn't have to pre-allocate stack for threads).

Paulo is right that there is very little new out there, but I think he misses the point as the issue is if the features exist together in one language (I tend to only use a single programming language per source file...)

[bflm]

On Jul 19, 8:41 am, Paul Borman <bor...@google.com> wrote:
> > I don't think you can. Does e.g. standard C knows split stacks so it
> > can handle 100k goroutines in about half GB of memory?
>
> The Cray 2 had stack segments in the mid to late 80's.  Part of every
> language (C, Fortran, Pascal).

A specific machine is not relevant. I meant if split stacks support in
the language is part of the C standard (any of them), so you can use
them with any standard conforming C compiler. I think that's not
possible. AFAIK the same for ISO Pascal. Have no idea how/if Fortran,
though.

> Paulo is right that there is very little new out there, but I think he
> misses the point as the issue is if the features exist together in one
> language (I tend to only use a single programming language per source
> file...)

Signed.

[Geoffrey Teale]

This thread is starting to look like troll food.  Everyone on this list is already interested enough in Go to see past theee kinds of non-constuctive criticisms.

Perhaps the best attitude is that adopted by fans of Everton football club:  those who understand need no explanations, those who don't just don't matter.

Regards.
Geoff Teale

[Paulo Pinto]

No I did not miss the point.

Quoting myself:
"Maybe Go's advantage is how these features are blended together..."

On Jul 19, 8:41 am, Paul Borman <bor...@google.com> wrote:

[Paulo Pinto]

Everyone that attended a compiler development course, knows
that your remark is an implementation issue, not a language
feature.

[Gustavo Niemeyer]

> That you think is missing from Go to be used by most programmers?

We're missing *you*.

(pause for thinking)

--
Gustavo Niemeyer
http://niemeyer.net
http://niemeyer.net/plus
http://niemeyer.net/twitter
http://niemeyer.net/blog

-- I never filed a patent.

[⚛]

On Jul 19, 5:41 am, Pete Wilson <p...@kivadesigngroupe.com> wrote:

> I have mixed feelings responding to stuff that sounds like idiocy. But
> I've had a couple of glasses o wine (Becker Provencale, and Jadot
> Macon Villages)(both recommended as reasonably-priced enjoyable
> tipple).
>
> I simply do not understand why anybody could think that "Web 2.0"
> popularity ratings of a language which introduces vastly important new
> semantics (concurrency and message-passing) and which is being created
> for the business purposes of a large and rich American company can
> have anything whatsoever to do with its utility, future success, or
> importance.
>
> When a language creates new semantics (new in the sense of 'new to the
> wide, wide world', not new in the sense of 'OMG, nobody thought of
> that before') then one may reasonably expect a quite a long period of
> consideration, learning and acceptance, especially if the new
> semantics are broadly useful.
>
> Thus the posting (I am being uncharacteristically rude)(no, really) is
> an indication of either imbecility or a desire to troll.

No. The poster of the original question asked the question because he
strives for understanding, wants to know how things work, and wants
other people to find contradictions in his current understanding of
the world - so that in the end he gains a better understanding of the
world and himself.

[Ryan Petersen]

Go was Tiobe's language of the year in 2009- they must not see it as a failure...

On Sun, Jul 17, 2011 at 3:13 PM, brakaman <jvte...@gmail.com> wrote:

On 17 jul, 21:51, Gabor <g...@szfki.hu> wrote:
> Fortran is number 35 in the same list.
> It has a much longer history and ages well.
>
> Would you classify it as a failure?
>

Yes friend.

Fortran has its history, currently I think not many people are using
Fortran, is normal

But Go is a modern language, recent supported by Google.(no
comment....)

[bflm]

On Jul 19, 9:27 am, Paulo Pinto <paulo.jpi...@gmail.com> wrote:
> Everyone that attended a compiler development course, knows
> that your remark is an implementation issue, not a language
> feature.

So any language which has/may have/will have such feature mentioned/
prescribed/defined in the specs invalidates the assertion, right? ;-)

[Steven Blenkinsop]

It would be an implementation guarantee, not a language feature. You
can write a correct program in a language without caring about
implementation details, but you can't guess how it will perform, which
includes the possibility of it exceeding available resources.
Implementation guarantees give you the information needed for this.

[pavolstartrek]

This reaction merits to be written in to golden book of human wisdom.

On Jul 18, 7:35 am, Geoffrey Teale <tea...@gmail.com> wrote:
> To sumarise a long presentation I gave to non-programmers:
>
> - There are 12 million programmers in the world
> - The majority of those programmers are scarcely qualified
> - Most technology decisions are made by a combination of following the crowd
> and a false understanding of risk.
> - The high cost and failure rate in software development is no coincidence.
>
> Remember the Stevie Wonder rule - "When you believe in something you don't
> understand then you suffer".   In this case that means "Perhaps making
> programming language decisions based on what 12 million powerless idiots are
> doing isn't the golden road to glory and great hacks."
>
> Go is a genuine attempt to improve the state of systems programing language
> beyond the point they reached in the early 1970s.  As a result the sort of
> people using it are mostly that small community of people who understand and
> care about the concens that drive such a development.
>
> You're not going to catch those 12 million people unless you can market
> heavily enough the idea that their future income depend on jobs/contacts
> built around go, but that goal just draws resources and energy away from
> making the language better.
>
> Arguably Java also suffers from it's large community of corporate drones.
> The slavish tendancy to build baroque, mausiliums of intricate classes,
> dense with state and dripping with verbose XML is a reflection of the
> unthinking insanity of the 12 million.
>
> I'd rather a tiny community usee the language well, built succesful
> applications and organically grew the user base whilst establishing a clean,
> sane library base that might later be used to improve the lives of a wider
> population of programmers.
>
> I've said more than enough, I'll trundle back to the twelve million and take
> my punishment now.
> --
> Geoff Teale
> On 18 Jul 2011 06:53, "David Roundy" <daverou...@gmail.com> wrote:
>
>
>
>
>
>
>
> > On the contrary, Fortran is still widely used for new code development in
> > the scientific community, and I'm sure it will continue to be widely used
> > until someone develops a comparable language, with support for
> > multidimensional arrays, the exponentiation operator, compiling to fast
> > code, etc such as Fortran has.
>
> > Which is all rather unfortunate, since I have steadfastly refused to learn
> > Fortran, and have yet to find a satisfactory alternative.
>
> > David
> > On Jul 17, 2011 9:30 PM, "Paul Borman" <bor...@google.com> wrote:
> >> Fortran is still heavily used, just not in web/gui/academic environments.
> > I
> >> would say little new is probably being developed, though. I do not know
> if
> >> non-fortran compilers have yet caught up to fortran compilers for high
> > speed
> >> scientific computing.
>
> >> I don't think you can use this sort of ranking to determine failure,
> > though
> >> it can imply success. A better metric would be what language is used when
> >> you have a choice. I have a choice of 4 languages, and Go is what I am
> >> choosing.
>
> >> You should also remember that Tiobe only counts what they can see, not
> > what
> >> they can't. Also, did you notice Objective C? That seems to finally be
> >> doing well many years after its introduction. It was #38 in 2006, but now
> >> is #6. Objective C is 25 years old now. Go, in 3 years, out performed
> >> Objective C in 25 (Go is at 32 and Objective C was at #38). I think it is
> >> way to early to write Go's obituary.
>
> >> -Paul

[Joshua]

Its actually a good thing that Go is not in the top ten list. It is a new language and needs time to mature. If it gains popularity too soon, people look at it and then stop using it. Clojure is sort of in this boat right now so its become just a niche language. If it takes too long, people never look at it. Sometimes there is a "killer app" that is released that creates buzz and pulls in many users. Ruby on Rails is an example of this kind of scenario.

I predict that you will see Go gain popularity when the first books are released. It will continue to gain popularity if it provides a simple way for Professors to teach Computer Science subjects. This is dependent on maturity of the language and the libraries readily available.

Joshua

[⚛]

I read your post yesterday and today, but unfortunately I am unable to
make sense out of it. Below are my notes about your ideas, maybe we
can somehow sort it out and arrive at a mutual agreement.

On Jul 20, 7:46 pm, Joshua <jhaw...@gmail.com> wrote:
> Its actually a good thing that Go is not in the top ten list.

I think Go is not in the top 10 list because it is impossible to be in
the top 10 in such a short time. All languages in top 10 have a
momentum, top 10 is crowded. It is physically impossible for a new
language to be in top 10. It has little to do with it "being a good
thing" or "being a bad thing".

> It is a new language and needs time to mature. If it gains popularity too soon,
> people look at it and then stop using it.

In my opinion, innovations in programming languages always happen for
a particular reason. To solve a particular set of issues. If a
programming language solves widespread issues, people will not stop
using it. Most certainly, if a new programming language solves issues
that other languages do not solve - and if it is a practical
programming language (which means: I am excluding academic programming
languages here) - then fears of the language being killed because "it
is gaining popularity too soon" seem irrational to me.

Well, if young unexperienced people create a new language, the
language might get killed because of popularity. That is because young
people might panic or might do something stupid. But I do not think
this is Go's case - the people who created it seem experienced-enough
to me.

> Clojure is sort of in this boat
> right now so its become just a niche language. If it takes too long, people
> never look at it. Sometimes there is a "killer app" that is released that
> creates buzz and pulls in many users. Ruby on Rails is an example of this
> kind of scenario.

I don't know Ruby on Rails well, so I am unable to make a comment
here.

> I predict that you will see Go gain popularity when the first books are
> released.

Maybe. But there are counter-examples to this. Books about Haskell are
old by comparison with the age of Go. Despite that, Haskell has not
taken over the world. And that's despite the fact that some
universities are using it to teach students about programming.
Arguably, some people truly believe that Haskell and functional
programming will eventually take over the world in the long term - it
is a form of religion, I guess.

In my opinion, to make a prediction means to uncover/describe a
necessity. Because stuff always happens within a particular system of
rules, any prediction should also describe (or make a reference to)
the rules of the system and should show that those rules correspond to
reality. One cannot pull predictions out of "thin air".

> It will continue to gain popularity if it provides a simple way
> for Professors to teach Computer Science subjects.

Many parts of Computer Science (CS) are independent from any
programming language.

Some parts of CS are using a particular programming language, because
that language is the best tool to make the point (example: Lisp) - it
is unlikely for any new language to invade the space. The existing
programming language used in the field is already well suited for the
task (or even best suited for the task), and there is no real benefit
to be gained by changing to any other language. Besides that, people
would have to rewrite all CS books in the field or create a dictionary
for translating old books, articles and blogs into the new notation
(both of which seems unlikely).

In many CS areas, there is already competition for Go. For example,
communication sequential processes are in Lua and in Erlang.

[Paulo Pinto]

Well their slogan is "avoid success at all cost". :)

But actually many companies are now doing Haskell stuff, even Intel
cared
to convert one of their frameworks to Haskell.
http://software.intel.com/en-us/blogs/2010/05/27/announcing-intel-concurrent-collections-for-haskell-01/

And Microsoft is still pumping money into its development, although
they tend to
be more busy with F#.

I remember that in FOSDEM 2008, it was very hard to get Haskell or
Erlang books, every time the
book selling stands got refiled, the books were quickly gone.

--
Paulo

[⚛]

On Jul 21, 8:02 pm, ⚛ <0xe2.0x9a.0...@gmail.com> wrote:

> In many CS areas, there is already competition for Go. For example,
> communication sequential processes are in Lua and in Erlang.

On second thought, I should have written "concurrency" instead of
"communicating sequential processes" in the above sentence.

[bflm]

On Jul 19, 8:24 pm, Steven Blenkinsop <steven...@gmail.com> wrote:
> It would be an implementation guarantee, not a language feature. You
> can write a correct program in a language without caring about
> implementation details, but you can't guess how it will perform, which
> includes the possibility of it exceeding available resources.
> Implementation guarantees give you the information needed for this.

True but not complete. E.g. an "implementation guarantee" in language
X which says goroutines(fibers, ...) do not stick exclusively to their
creating thread do change the semantics of the language in a way, that
some programs become correct/incorrect with this "implementation
guarantee" in effect or not (subject to other details/conditions I'm
ignoring for simplicity).

It doesn't matter if you artificially rename a concept to fabricate an
argument - the semantics of the language judges it all.