TIOBE index for scala

[Arnab Biswas]

Hi,

TIOBE index for 2013 (released on Jan 2014) has been released. Scala is at 33. I am not able to remember Scala's ranking at the beginning of 2013 or mid (most probably it was almost the same?). It looks like TIOBE delete the old indices (at least from the web) as soon as the new index gets published. Does any one remember it? Or is there any way to get the old TIOBE indices?

Thanks,

Arnab 

[Alec Zorab]

Going from this blog post [1] it lookks like it was #36 last year

[1] http://sarroutbi-fsw.blogspot.co.uk/2013/10/the-tiobe-programming-languages-ranking.html

--
You received this message because you are subscribed to the Google Groups "scala-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email to scala-user+...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Arnab Biswas]

Thanks Alec for sharing the link.

In between, I checked wikipedia. In Dec 2013, for scala it was #31 (slight dip to #33!).

http://en.wikipedia.org/wiki/Scala_(programming_language)

Two surprises. Groovy is >50 from #18. And golang at #36. :-)

Arnab

[Seyed H. HAERI (Hossein)]

I think what is most noteworthy for Scala is D which offers similar PL constructs but is native. In TIOBE's ranking -- which is obviously not the ultimate reliable source -- D is almost always above Scala. Unlike Scala which aims for Java interoperability, D is designed for that with C. Java is going down the drain and C has ever stood steady... This is not to say that managed PLs are all doomed to death; I take F# as a clear counterexample. But, Java was just such a bad choice to pair with, and, that's paying itself off.

--
--------------------------------------------------------------------------------------------------------------

Seyed H. HAERI (Hossein)

Research Assistant
Institute for Software Systems (STS)
Technical University of Hamburg (TUHH)
Hamburg, Germany

ACCU - Professionalism in programming - http://www.accu.org/
--------------------------------------------------------------------------------------------------------------

[Arnab Biswas]

Thanks Seyed. Being a vanilla java developer, didn't have much knowledge about the other side. Very useful information. So, where exactly is java/scala lacking w.r.t D? I thought the decline of Java is being contribute by the new set of JVM based languages. Isn't it the case?

http://www.drdobbs.com/jvm/the-rise-and-fall-of-languages-in-2013/240165192

Here is an interesting snip:

Mobile programming kept Objective-C advancing and it limited Java's decline. The latter was surely due to the language falling out of fashion, especially among hobbyists, and the pressure exerted by other JVM languages, notably Groovy, JRuby, Scala, and Clojure. 

Thanks,

Arnab

[Alexandru Nedelcu]

The only languages I care about are in this list: http://adambard.com/blog/top-github-languages-for-2013-so-far/

[Alan Burlison]

On 10/01/2014 12:46, Seyed H. HAERI (Hossein) wrote:

> I think what is most noteworthy for Scala is D which offers similar PL
> constructs but is native. In TIOBE's ranking -- which is obviously not the
> ultimate reliable source -- D is almost always above Scala.

Really?

http://www.ohloh.net/languages/compare?measure=commits&percent=true&l0=dmd&l1=scala&l2=-1&l3=-1&commit=Update

> Java is going down the drain and C has ever stood steady...

Really?

http://www.ohloh.net/languages/compare?measure=commits&percent=true&l0=c&l1=java&l2=-1&commit=Update

Making statements about the future of various programming languages
based on the various web-based indexes is akin to reading chicken
entrails - it's entirely open to interpretation, and each one you look
at gives a different answer.

--
Alan Burlison
--

[Seyed H. HAERI (Hossein)]

Hi Arnab,

In this thread, I don't really want to discuss why I think Java is the devil. My point is that -- I sincerely believe that -- the decision to pair Scala with Java was fundamentally wrong, and, it's now paying off. Java is going down the drain -- whatever the reason; Scala will also follow unless it at some point obtains a divorce. As an alternative, it might from that point onward, take D as an exemplar and goes for pairing with a good native PL. Being effectively portable everywhere, C is a good candidate. And, C++ will get portable this year too (for the first time in its history).

HTH,

--Hossein

[Seyed H. HAERI (Hossein)]

Hi Alan,

> Making statements about the future of various programming languages based on the various web-based indexes is akin to reading chicken entrails - it's entirely open to interpretation, and each one you look at gives a different answer.

Indeed. :) And, thanks for throwing a different voice into the game.
Would you mind explaining why we should take this OhOh index
seriously?

Cheers,
--Hossein

P.S. Shall I humbly ask people not get me wrong? Believe me, I'm doing
my best to refrain from a language war. Please forgive me if you think
I didn't manage. Let's keep it at the technical level and avoid the
heat... :)

[Alan Burlison]

On 12/01/2014 11:49, Seyed H. HAERI (Hossein) wrote:

> Indeed. :) And, thanks for throwing a different voice into the game.
> Would you mind explaining why we should take this OhOh index
> seriously?

The whole point is you should treat them *all* with a huge degree of
caution. I chose the ohloh one as an illustration of that, because it
shows a completely different picture to TIOBE.

As for how it's generated, see
http://meta.ohloh.net/faqs/what-is-ohloh-code-and-what-about-koders-com/
10+ billion LOC seems to be a reasonable sample size.

--
Alan Burlison
--

[Oliver Ruebenacker]

     Hello,

On Sun, Jan 12, 2014 at 6:21 AM, Alexandru Nedelcu <al...@bionicspirit.com> wrote:

The only languages I care about are in this list: http://adambard.com/blog/top-github-languages-for-2013-so-far/

  That would probably be extremely biased towards languages typically used for small, short-lived projects.

     Best,

     Oliver

--

Oliver Ruebenacker

Founder at Relomics Consulting

Be always grateful, but never satisfied.

[Seyed H. HAERI (Hossein)]

Hi Alan,

I do also believe in the necessity of both diversity and cautious.
But, I'm still not convinced about creditability of Ohloh. What use
the samples have is in much higher significance then the sample size.

Cheers,
--Hossein

[Juha Heljoranta]

Another metric, RedMonk rankings from June:
http://redmonk.com/sogrady/2013/07/25/language-rankings-6-13/

Cheers,
Juha

[Alan Burlison]

On 12/01/2014 11:43, Seyed H. HAERI (Hossein) wrote:

> And, C++ will get portable this year too (for the first time in its history).

I don't know what your definition of 'portable is' or even how it is
relevant to Scala, but C++ still won't have a stable ABI.

--
Alan Burlison
--

[Alexandru Nedelcu]

On Jan 12, 2014 2:16 PM, "Oliver Ruebenacker" <cur...@gmail.com> wrote:
>   That would probably be extremely biased towards languages typically used for small, short-lived projects.

That's one way of looking at it. Another perspective would be that these languages are chosen by people that work on stuff in their free time and also these languages have active open-source communities. That's much more important to me than anything else. Of course, it's a biased opinion, but you have to admit - looking at open-source output is more valuable than looking at something like Tiobe.

[Oliver Ruebenacker]

     Hello,

On Sun, Jan 12, 2014 at 7:37 AM, Alexandru Nedelcu <al...@bionicspirit.com> wrote:

On Jan 12, 2014 2:16 PM, "Oliver Ruebenacker" <cur...@gmail.com> wrote:
>   That would probably be extremely biased towards languages typically used for small, short-lived projects.

That's one way of looking at it. Another perspective would be that these languages are chosen by people that work on stuff in their free time and also these languages have active open-source communities. That's much more important to me than anything else. Of course, it's a biased opinion, but you have to admit - looking at open-source output is more valuable than looking at something like Tiobe.

  I'm more interested in long-term projects, open-sourced or not. :)

  I don't see much meaning in the TIOBE index, either. I would rather ask, how many job postings require/prefer proficiency/experience in a language.

  Regarding jobs, I find this list particularly interesting.

     Best,

     Oliver

[Alan Burlison]

On 12/01/2014 12:17, Seyed H. HAERI (Hossein) wrote:

> I do also believe in the necessity of both diversity and cautious.
> But, I'm still not convinced about creditability of Ohloh. What use
> the samples have is in much higher significance then the sample size.

I'm not quite sure how to parse that last sentence but you seem to be
completely missing the point. Attempting to define the strategy for the
future of Scala based on *ANY* survey site is ill-advised, not least
because they are lagging indicators, not leading ones. The point I was
making is that if it is possible to cherry-pick one to support any
position you care to take, they are in effect useless.

It's clear both from this thread and from previous posts of yours that
you hate Java and think C++ is brilliant. You seem to have conflated
Java and the JVM - Scala is a JVM language and that's why it has a good
degree of interoperability with Java. You could just as well make the
same argument about C and Python, because they both live on top of libc
and both use the same ABI. Personally I think that choosing the JVM was
the right choice for Scala at the time the decision was made, because it
leveraged both the JVM and the large number of Java libraries that were
available, and as a result eased and speeded up the development of
Scala. But that's just my opinion.

I'm really not sure where you are going with this - I'm not clear what
your perceived problem is, let alone what your proposed solution is. It
seems to be to rewrite Scala in either C or C++. You haven't explained
the pros, cons or costs of that approach, other than it would address
your personal distaste for Java. You also haven't explained who would
resource this effort, or even why it would be likely to succeed when the
attempt to port Scala to .NET fell by the wayside.

--
Alan Burlison
--

[Jason Zaugg]

On Sun, Jan 12, 2014 at 1:54 PM, Oliver Ruebenacker <cur...@gmail.com> wrote:

     Hello,

On Sun, Jan 12, 2014 at 7:37 AM, Alexandru Nedelcu <al...@bionicspirit.com> wrote:

On Jan 12, 2014 2:16 PM, "Oliver Ruebenacker" <cur...@gmail.com> wrote:
>   That would probably be extremely biased towards languages typically used for small, short-lived projects.

That's one way of looking at it. Another perspective would be that these languages are chosen by people that work on stuff in their free time and also these languages have active open-source communities. That's much more important to me than anything else. Of course, it's a biased opinion, but you have to admit - looking at open-source output is more valuable than looking at something like Tiobe.

  I'm more interested in long-term projects, open-sourced or not. :)

  I don't see much meaning in the TIOBE index, either. I would rather ask, how many job postings require/prefer proficiency/experience in a language.

  Regarding jobs, I find this list particularly interesting.

Could I suggest to move discussion over PL popularity indices to scala-debate? There are also plenty of threads in the archives of our mailing lists that delve into Tiobe, its methodology, how Scala fares for those interested.

scala-user should remain focussed on Q&A and discussions about the Scala programming language itself.

-jason

[Seyed H. HAERI (Hossein)]

Hi Juha,

That was a very interesting study. Thanks! :) I can agree that GitHub
and StackOverflow are both becoming increasingly important to the
daily practice of programming. TIOBE, however, seems to have
difficulties mining StackOverflow and hasn't still considered GitHub:

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

Yet, I humbly think a study based on simply these two hubs is too
narrow to be reliable as an alternative to TIOBE.

Cheers,
--Hossein

[Seyed H. HAERI (Hossein)]

Hi Alan,

>> And, C++ will get portable this year too (for the first time in its history).
>
> I don't know what your definition of 'portable is'

platform-independent

> or even how it is relevant to Scala,

Java was born portable and, by being married to it, so is Scala.

> but C++ still won't have a stable ABI.

Hmmm... Would people excuse me for a small deviation here to ask for
what ABI stands for?

Cheers,
--Hossein

[Seyed H. HAERI (Hossein)]

Dear Alan,

You have raised valid points that I wish I was welcome on this list to
address. I don't, however, like to miss Jason's polite invitation to
cut this discussion here. I am not, at the moment, ready to port this
thread to scala-debate. But, would be happy to take it off-list if
you're still up for it. :)

Cheers,
--Hossein

[Nils Kilden-Pedersen]

Application Brogramming Interface?

 

Cheers,
--Hossein

[Alan Burlison]

On 12/01/2014 15:58, Seyed H. HAERI (Hossein) wrote:

> Hmmm... Would people excuse me for a small deviation here to ask for
> what ABI stands for?

http://en.wikipedia.org/wiki/Application_binary_interface

--
Alan Burlison
--

[Alexandru Nedelcu]

On Sun, Jan 12, 2014 at 2:54 PM, Oliver Ruebenacker <cur...@gmail.com> wrote:

I don't see much meaning in the TIOBE index, either. I would rather ask, how many job postings require/prefer proficiency/experience in a language.

Out of curiosity, have you ever had problems finding a job for Scala?

--
Alexandru Nedelcu
www.bionicspirit.com

PGP Public Key:
https://bionicspirit.com/key.aexpk

[Oliver Ruebenacker]

     Hello,

On Sun, Jan 12, 2014 at 5:04 PM, Alexandru Nedelcu <al...@bionicspirit.com> wrote:

On Sun, Jan 12, 2014 at 2:54 PM, Oliver Ruebenacker <cur...@gmail.com> wrote:

I don't see much meaning in the TIOBE index, either. I would rather ask, how many job postings require/prefer proficiency/experience in a language.

Out of curiosity, have you ever had problems finding a job for Scala?

  Can't tell, as I started using Scala on my current job. My guess would be that if you don't care what kind of software you are developing, the possibilities are endless.

  My preferred job is developing software to analyze biological pathways and genomes, and the programming language is a secondary concern. My preference would be first Scala and second Java, but I'm open to anything if the scientific background is exciting. If you look at computational biology and bioinformatics positions, most ask for either R or Python, or they ask for some Java framework like Hadoop, Spring or Hibernate.

  This is one reason not to trust TIOBE or any index that is counting explicit mention of a language: very often, Java is implied by the framework or technology used (e.g. Tomcat).

  That said, I have seen a few mention of Scala within my focus area.

     Best,

     Oliver

[Oliver Plohmann]

Am Freitag, 10. Januar 2014 13:46:12 UTC+1 schrieb Hossein:

I think what is most noteworthy for Scala is D which offers similar PL constructs but is native.

 

Good observation. Agree.

 

In TIOBE's ranking -- which is obviously not the ultimate reliable source -- D is almost always above Scala. Unlike Scala which aims for Java interoperability, D is designed for that with C. Java is going down the drain and C has ever stood steady... This is not to say that managed PLs are all doomed to death; I take F# as a clear counterexample. But, Java was just such a bad choice to pair with, and, that's paying itself off.

 

Facebook lately moved its first program written in D in production. The D people were very proud about it. It is in large part probably because Andrei Alexandrescu (one of the main D compiler developers) works for Facebook. Other than that there are very few companies known to using D (I only know sociomantic.com). There are much more companies listed on Akka's homepage to use Akka than companies that on the D homepage are mentioned to using D. So I really wonder why Scala and D are that close on the Tiobe index.

 

To "measure" how much a language is used inside companies counting job ads for languages might be a suitable approach. If I do this for my country (Germany) I find very very few job ads for D and very few ads for Scala. What Tiobe says at least doesn't apply much to the place I live.