Programming language popularity
Anton Ertl
popularity, based on the number of postings in the comp.lang.* groups.
I just took another sample, and below you can find the results. For
more results and the data, on which they are based, see
<http://www.complang.tuwien.ac.at/anton/comp.lang-statistics/>.
Feb 2007 (100d) Feb 2006 (100d) Mar 2005 (185d) Dec 2003 (160 d)
|--- news.tuwien.ac.at ----| |---- news.individual.net
posts language posts language posts language posts language
1 18624 c++ 23622 java 39240 java 46418 java
2 16974 java 19346 c++ 36763 c++ 36633 c++
3 15612 c 17502 python 31491 python 28469 python
4 14073 python 15486 c 22787 c 26673 c
5 14043 ruby 14788 ruby 22115 ruby 21106 perl
6 8022 javascript 10569 javascript 20733 perl 15889 clipper
7 7194 perl 9374 perl 15874 javascript 15245 javascript
8 6527 php 8505 lisp 14157 lisp 13752 lisp
9 6483 lisp 7963 php 13786 php 12930 basic
10 6276 labview 6093 labview 12591 clipper 12852 ruby
11 4756 tcl 5488 tcl 11150 basic 12243 clarion
12 3825 fortran 5014 basic 8099 tcl 10468 pascal
13 3517 clipper 4696 clipper 7782 cobol 9472 labview
14 3013 forth 4039 fortran 7225 fortran 9312 tcl
15 2624 basic 3621 cobol 6714 pascal 6479 ada
16 2029 ada 2768 scheme 6574 labview 6064 cobol
17 1424 cobol 2180 smalltalk 6062 forth 5618 fortran
18 1418 functional 1717 pascal 5798 ada 5593 functional
19 1349 idl-pvwave 1634 forth 4245 smalltalk 3976 scheme
20 1212 vhdl 1419 vhdl 3035 scheme 3864 smalltalk
21 1178 smalltalk 1354 ada 2349 vhdl 3823 forth
22 1145 scheme 1347 idl-pvwave 2301 asm 3072 asm
23 1104 pascal 1005 asm 2217 idl-pvwave 2236 vhdl
24 936 verilog 900 pl1 1785 rexx 1764 postscript
25 828 asm 796 postscript 1725 awk 1703 objective-c
26 648 postscript 762 awk 1605 functional 1693 idl-pvwave
27 640 awk 687 functional 1423 prolog 1078 pl1
28 536 misc 637 verilog 1390 verilog 1013 verilog
29 489 pl1 551 apl 1344 postscript 1007 awk
30 464 rexx 435 misc 1144 pl1 912 prolog
31 388 objective-c 428 prolog 1084 clarion 758 rexx
32 369 apl 316 logo 718 eiffel 755 misc
33 354 prolog 312 clarion 703 misc 724 vrml
34 332 vrml 305 rexx 644 vrml 658 eiffel
35 296 haskell 302 eiffel 603 objective-c 657 logo
36 187 clarion 204 vrml 602 apl 508 mumps
37 126 mumps 198 objective-c 465 logo 443 apl
38 117 modula2 187 mumps 371 ml 358 ml
39 104 eiffel 101 asm370 353 mumps 272 modula2
40 79 ml 93 oberon 300 modula2 215 asm370
41 71 logo 83 ml 251 pop 190 pop
42 65 xharbour 79 modula2 181 asm370 147 oberon
43 64 oberon 47 dylan 115 oberon 98 modula3
44 54 asm370 30 pop 69 icon 63 dylan
45 50 dylan 14 icon 68 dylan 39 idl
46 44 icon 11 modula3 46 modula3 29 beta
47 24 idl 11 idl 40 idl 24 icon
48 21 pop 3 clos 15 hermes 21 sather
49 20 modula3 2 beta 9 beta 19 clos
50 2 sather 1 sather 8 clos 7 clu
51 1 beta 0 visual 7 prograph 6 prograph
52 0 visual 0 prograph 6 clu 6 limbo
53 0 limbo 0 limbo 2 sather 3 visual
54 0 hermes 0 hermes 2 limbo 1 php
55 0 clos 0 clu 0 visual 1 hermes
The changes at the top are small, but there seems to be a retro wave
going on, with languages like C++, C, Fortran and Forth gaining in
relative, and in some cases absolute popularity; maybe the traffic
generated by newcomers has abated in the Java and Python groups as
these languages become more established. Pascal and Basic continue
their steady decline, this time dropping below Forth. Clipper stems
the decline of the past few years this time.
There are no languages in the top 20 that seem to explode in
popularity; even idl-pvwave (what's that?), which rose steadily, to
rank 19 this year, seems to have risen mainly in relative, not that
much in absolute popularity (~13 Postings/day this year, not much
different from the 1997 and 2000 values).
- anton
--
M. Anton Ertl http://www.complang.tuwien.ac.at/anton/home.html
comp.lang.forth FAQs: http://www.complang.tuwien.ac.at/forth/faq/toc.html
New standard: http://www.forth200x.org/forth200x.html
EuroForth 2007: http://www.complang.tuwien.ac.at/anton/euroforth2007/
Andreas Kochenburger
tiobe.
The improvement of Forth's ranking to #14 is perhaps the only achievement of
werty ;-)
"Anton Ertl" <an...@mips.complang.tuwien.ac.at> schrieb im Newsbeitrag
news:2007Feb2...@mips.complang.tuwien.ac.at...
Anton Ertl
>The improvement of Forth's ranking to #14 is perhaps the only achievement of
>werty ;-)
This time around there were ~14 postings/day more in clf than in the
previous period. I don't see wertys postings any more, but does he
really post that much?
Cesar Rabak
> "Andreas Kochenburger" <a...@nospam.org> writes:
>> The improvement of Forth's ranking to #14 is perhaps the only achievement of
>> werty ;-)
>
> This time around there were ~14 postings/day more in clf than in the
> previous period. I don't see wertys postings any more, but does he
> really post that much?
>
> - anton
messages in c.l.forth this will change the status?
werty
Of what value is the word "popular"
Hitler was 100% popular in 1938 ..
Trade unionists have control on America .
They like C/C++ ....
They want to control Forth ....
Im a Union buster .....
English is the first to go ! No more stupid text .
The game box NDS Lite at $140 is the fastest
method to advance ...
Its bios is clear and simple , and the comany intended
for you to hack this box .
GP2X is Linux , thus more time will be needed
to develope software .
There is a free file "FlashMe" you place on
a expansion card ..
.then short the SL-1 PCB "reset" contacts
thru a tiny hole , near battery ..
Now it will not check for ROM'd protection
password in all Nintendo products ..
And you will probably never again use the
smaller of the 2 slots alias "DS" slot .
Forth will reappear in a NDS Lite , game box ..
Anton Ertl
>Your comment bring another intersting question: if we put automated
>messages in c.l.forth this will change the status?
In the long run: If a significant part of the newsgroup is automated
noise, this decreases the signal/noise ration, especially for
potential new readers, who check out the newsgroup unfiltered. This
will reduce the number of new readers and consequently new posters in
the newsgroup, and also the number of old readers and posters
(although possibly not as much, thanks to kill-files). The reduction
in on-topic posters will decrease the signal-noise ratio even more,
until more or less only the automatic noise is left. Then the
newsgroup will be dead (look at alt.games.diablo for an example of
where that has happened).
The newsgroup might look good in my survey (until somebody notices
that it contains only off-topic traffic and proposes to rmgroup it),
but it will be lost as a communication medium for the Forth community.
And I think that having that medium is far more important than looking
good in my survey.
If you want to help the newsgroup and it's status in the survey, write
good on-topic articles (including interesting on-topic questions).
That will increase the number of postings as well as increasing the
signal/noise ratio. But it cannot be automated.
Cesar Rabak
> Cesar Rabak <csr...@yahoo.com.br> writes:
>> Your comment bring another intersting question: if we put automated
>> messages in c.l.forth this will change the status?
Thanks for posting. I obviously brought it as food for thought, and my
reasoning is complimentary to yours (see the bottom).
>
> In the long run: If a significant part of the newsgroup is automated
> noise, this decreases the signal/noise ration, especially for
> potential new readers, who check out the newsgroup unfiltered. This
> will reduce the number of new readers and consequently new posters in
> the newsgroup, and also the number of old readers and posters
> (although possibly not as much, thanks to kill-files). The reduction
> in on-topic posters will decrease the signal-noise ratio even more,
> until more or less only the automatic noise is left. Then the
> newsgroup will be dead (look at alt.games.diablo for an example of
> where that has happened).
This seems a sound prediction to me, and a fate we should avoid.
>
> The newsgroup might look good in my survey (until somebody notices
> that it contains only off-topic traffic and proposes to rmgroup it),
> but it will be lost as a communication medium for the Forth community.
> And I think that having that medium is far more important than looking
> good in my survey.
Yes. The only value of a NG nowadays is the (e-)gathering for people
interested in the subject (OT here Forth).
>
> If you want to help the newsgroup and it's status in the survey, write
> good on-topic articles (including interesting on-topic questions).
Seems a nice suggestion for the whole usenet (while it lasts).
> That will increase the number of postings as well as increasing the
> signal/noise ratio. But it cannot be automated.
Agree.
Now about my additional thinking. I believe the problem of surveying
just the number of messages can somewhat skew some stats as:
1) some groups have an active community and stay on topic, sometimes
even being harsh;
2) some groups have an active community but have some lenghty threads
that wander messages (sometimes in the hundreds);
3) some groups have low activity and spammers and other non-topic
posters can 'increase' the traffic in noticeable ways percent wise.
It is not an exhaustive list, but to show the issues with only message
counting.
Regards,
--
Cesar Rabak
sp...@controlq.com
> Your comment bring another intersting question: if we put automated messages
> in c.l.forth this will change the status?
>
----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! >100,000 Newsgroups
---= East/West-Coast Server Farms - Total Privacy via Encryption =---
Anton Ertl
>Now about my additional thinking. I believe the problem of surveying
>just the number of messages can somewhat skew some stats as:
>
>1) some groups have an active community and stay on topic, sometimes
>even being harsh;
An extreme example is comp.lang.c, where every posting that is on
something other than ANSI C is flamed; if you are lucky, a different
newsgroup is suggested. Still, comp.lang.c is the comp.lang newsgroup
with the highest traffic this time around (other languages are higher
in the language ranking by having the traffic distributed in more
groups).
>It is not an exhaustive list, but to show the issues with only message
>counting.
Sure, it's not perfect (but what would be?) and has its weaknesses,
but at least it's simple.
sp...@controlq.com
>> It is not an exhaustive list, but to show the issues with only message
>> counting.
>
> Sure, it's not perfect (but what would be?) and has its weaknesses,
> but at least it's simple.
>
> - anton
>
Actually, I like the simplicity of it, and I'm aware that one must draw
one's own conclusions, but I'm not afraid of that. Thanks Anton, for
posting this most illuminating message (the OP), whatever it means.
Cheers,
Rob Sciuk
----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----
Cesar Rabak
> Cesar Rabak <csr...@yahoo.com.br> writes:
>> Now about my additional thinking. I believe the problem of surveying
>> just the number of messages can somewhat skew some stats as:
>>
>> 1) some groups have an active community and stay on topic, sometimes
>> even being harsh;
>
> An extreme example is comp.lang.c, where every posting that is on
> something other than ANSI C is flamed; if you are lucky, a different
> newsgroup is suggested. Still, comp.lang.c is the comp.lang newsgroup
> with the highest traffic this time around (other languages are higher
> in the language ranking by having the traffic distributed in more
> groups).
Yes.
>
>> It is not an exhaustive list, but to show the issues with only message
>> counting.
>
> Sure, it's not perfect (but what would be?) and has its weaknesses,
> but at least it's simple.
>
I agree and believe it is a good indicator we can use as is. Any attempt
to 'improve' would require too much energy versus the possible increase
in accuracy that I don't think it is worth of.
[]s
--
Cesar Rabak