[erlang-questions] languages in use? [was: Time for OTP to be Renamed?]

[Robert Virding]

I would say that it is not just the erlang ecosystem (maybe) being weak which might limit where erlang is suitable, but the language itself imposes limitations on where it is useful. For example two areas where NOT to use Erlang: 

- Raw number crunching - while can do arithmetic it is not that fast at it and was never intended for it. With the Erlang you will get tomorrows weather report next month. 
- Problems where shared data is the best fit - there are applications where sharing data is the best fit, even when running very parallel. Erlang is not good at sharing. 

That said Erlang might be a very good choice for managing these types of applications where you hand over the low-level work to other languages/specialised hardware. 

Robert 

----- Original Message -----

> From: "Zachary Kessin" <zkessin@REDACTED>
> To: erlang-questions@REDACTED
> Sent: Monday, 17 February, 2014 9:42:03 AM
> Subject: Re: [erlang-questions] languages in use? [was: Time for OTP to be
> Renamed?]

> We did a podcast a while back on when not to use erlang, which I thought was
> very interesting
> http://mostlyerlang.com/2013/05/30/005-when-not-to-use-erlang/
> There are definitely places where the Erlang Ecosystem is week, but they seem
> to be things people are working on

> --Zach
> On 2/17/14, 10:10 AM, Max Bourinov wrote:

> > Erlang is absolute genius within existing programming languages for server
> > side programming for highly loaded systems. There is nothing else on the
> > market that can replace Erlang (and there is no need for it). All those
> > languages like java/.net etc makes absolutely no sense on really loaded
> > servers.
> 

> > I strongly believe that if some people/teams/companies have difficulties
> > with
> > Erlang it only shows their misunderstanding or wrong usage or poor language
> > knowledge etc. Or all above and maybe extra...
> 

> > There is nothing else to say. All insinuations about "Where Erlang in use
> > or
> > not?" are poor and have absolutely no reason to exist unless you want to
> > know who uses Erlang and where, but it is a different question.
> 

> > Erlang is really in heavy use. And it really works amazing. Absolutely
> > amazing!!!
> 

> > Best regards for all people involved in Erlang community!
> 

> > Joe Armstrong and Robert Wirding are my heroes!
> 

> > Max
> 

> > On Mon, Feb 17, 2014 at 3:41 AM, Richard A. O'Keefe < ok@REDACTED >
> > wrote:
> 

> > > On 16/02/2014, at 2:14 AM, Miles Fidelman wrote:
> > 
> 
> > > > How about this for a larger sample:
> > 
> 
> > > > http://eandt.theiet.org/news/2013/mar/cobol-skills.cfm (lede: "Dearth
> > > > of
> > > > COBOL programmers threatens business")
> > 
> 

> > > As someone who has used COBOL, who indeed has OpenCOBOL
> > 
> 
> > > on his machine (and some CICS books I've never had a
> > 
> 
> > > chance to put into practice), I have to say that
> > 
> 

> > > 'Coughlan ... also believes a certain amount of
> > 
> 
> > > "intellectual snobbery", due to the language
> > 
> 
> > > not being mathematically based and created by
> > 
> 
> > > ... Grace Hopper rather than academics, is at
> > 
> 
> > > play."
> > 
> 

> > > is *way* off the mark. COBOL is no more and no
> > 
> 
> > > less "mathematically based" than most imperative
> > 
> 
> > > programming languages. And it's news to me that C,
> > 
> 
> > > C++, Objective C, Visual Basic.NET (we used to teach
> > 
> 
> > > that, only stopped recently; I used to do it but
> > 
> 
> > > after our then HoD took over, it was decided to stop),
> > 
> 
> > > SQL, or PHP was "created by academics".
> > 
> 

> > > Our own beloved Erlang was created _by_ an industrial
> > 
> 
> > > company _for_ their industrial purposes, and it wasn't
> > 
> 
> > > the first time they'd devised their own language either.
> > 
> 

> > > The problem with COBOL is that it is just amazingly
> > 
> 
> > > verbose, not to mention error-prone.
> > 
> 

> > > Unfortunately, COBOL-to-X translators, except for the
> > 
> 
> > > very very best, tend to produce programs in language
> > 
> 
> > > X that are even more verbose, and far less understandable.
> > 
> 
> > > (I haven't used one, but I've seen materials suggesting
> > 
> 
> > > that there are some seriously good COBOL-to-X translators
> > 
> 
> > > out there, but expect to spend serious money...)
> > 
> 

> > > As for "students only really learn one language,
> > 
> 
> > > normally Java or some Java derivative", we teach
> > 
> 
> > > Python, Java, and C, and 3rd-year students are
> > 
> 
> > > at least exposed to C++ or Objective C. Coughlan
> > 
> 
> > > may be correct that students only *learn* one
> > 
> 
> > > language, but they are *taught* more.
> > 
> 

> > > The rather whiny tone of the article is one I've heard
> > 
> 
> > > before (I wonder where). If businesses are worried
> > 
> 
> > > about there not being enough COBOL programmers, the
> > 
> 
> > > answer is for *them* to club together and *run some
> > 
> 
> > > courses*. If anyone wants to do that in Dunedin, I
> > 
> 
> > > would be happy to contribute some teaching time, if
> > 
> 
> > > adequately recompensed. Oh, and since students already
> > 
> 
> > > have scary student loans, these courses should be
> > 
> 
> > > *free*. User pays: in this case it's the companies
> > 
> 
> > > that are the users, not the students.
> > 
> 

> > > > Or Indeed.Com's salary survey:
> > 
> 
> > > > http://www.indeed.com/salary/Cobol-Developer.html
> > 
> 

> > > which shows "Java J2EE Unix Developer" as offering
> > 
> 
> > > higher average salary than most of the COBOL categories.
> > 
> 

> > > _______________________________________________
> > 
> 
> > > erlang-questions mailing list
> > 
> 
> > > erlang-questions@REDACTED
> > 
> 
> > > http://erlang.org/mailman/listinfo/erlang-questions
> > 
> 

> > _______________________________________________
> 
> > erlang-questions mailing list erlang-questions@REDACTED
> > http://erlang.org/mailman/listinfo/erlang-questions
> 

> --
> Zachary Kessin
> Mostly Erlang Podcast
> Skype: zachkessin
> Twitter: @zkessin

> _______________________________________________
> erlang-questions mailing list
> erlang-questions@REDACTED
> http://erlang.org/mailman/listinfo/erlang-questions
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://erlang.org/pipermail/erlang-questions/attachments/20140217/5ea240ee/attachment.htm>