C Beats Java As Number One Language According To TIOBE Index 535
mikejuk writes "Every January it is traditional to compare the state of the languages as indicated by the TIOBE index. So what's up and what's down this year? There have been headlines that C# is the language of the year, but this is based on a new language index. What the TIOBE index shows is that Java is no longer number one as it has been beaten by C — yes C not C++ or even Objective C."
definition (Score:5, Informative)
thx, bye.
Re:Dying gasps (Score:5, Informative)
You would be surprised how many mission critical embedded systems - are still being written in C
The other one (Score:5, Informative)
Is called PYPL (PopularitY of Programming Languages), and it ranked C# as #1 and C down in #5 based on a different methadology. Honestly, they both sound pretty silly to me.
https://sites.google.com/site/pydatalog/pypl/PyPL-PopularitY-of-Programming-Language [google.com]
Re:...Bash? (Score:5, Informative)
Yep, they use frequency of search on the internet for the language to estimate. Which means confusing, and easily broken languages like C, and infrequently used(and thus easily forgotten) languages like bash get a huge leg-up.
Not surprising (Score:4, Informative)
Re:definition (Score:1, Informative)
So what this really means is that coding in C requires more searches than Java in order to remember how to use the language.
Yes, unfortunately TIOBE is bollocks. (Score:5, Informative)
Seriously, for the last fucking time, can we stop posting on Slashdot random shit picked up from TIOBE? The TIOBE index is so completely and utterly full of fail that I can't believe people are STILL clinging onto it as evidence of anything whatsoever.
It shouldn't be traditional to do anything with TIOBE, except perhaps laugh at it or set it on fire.
So once last time, one final fucking time I'll try and explain to the 'tards who think it has any merit whatsoever why it absolutely does not.
We start here, with the TIOBE index definition, the horses mouth explanation of how they cludge together this table of bollocks they call and "index":
http://www.tiobe.com/index.php/content/paperinfo/tpci/tpci_definition.htm [tiobe.com]
First, there is their definition of programming language. They require two criteria, these are:
1) That the language have an entry on Wikipedia
2) That the language be Turing complete
This means that if I go and delete the Wikipedia entry on C, right this moment, it is no longer a programming language, and hence no longer beating anything. Apparently.
The next step, is to scroll past the big list of languages, to the ratings section, where we see that they state they take the top 9 sites on Alexa that have a search option, and they execute the search:
+" programming"
Then weight the results as follows:
Google: 30%
Blogger: 30%
Wikipedia: 15%
YouTube: 9%
Baidu: 6%
Yahoo!: 3%
Bing: 3%
Amazon: 3%
The first problem here is with search engines like Google, I run this query against C++ and note the following:
"About 21,500,000 results"
In other words, Google's figure is hardly anything like a reasonable estimate because a) Most these results are fucking bollocks, and b) The number is at best a ballpark - this accounts for 30% of the weighting.
The next problem is that Blogger, Wikipedia, and YouTube account for 54% of the weighting. These are all sites that have user generated content, as such you could literally, right now, pick one of the lowest languages on the list, and go create a bunch of fake accounts, talking about it, and turn it into the fastest growing language of the moment quite trivially.
To cite an example, I just ran their query on English Wikipedia for the PILOT programming language and got one result. A few fake or modified Wikipedia entries later and tada, suddenly PILOT has grown massively in popularity.
The next point is the following:
"Possible false positives for a query are already filtered out in the definition of "hits(PL,SE)". This is done by using a manually determined confidence factor per query."
In other words yes, they apply an utterly arbitrary decision to each language about what does and doesn't count. Or to put it simply, they apply a completely arbitrary factor in which you can have no confidence of being of any actual worth. I say this because further down they have a list of terms they filter out manually, they have a list of the confidence factors they use, and it takes little more than a second to realise massive gaps and failings in these confidence factors.
For example, they have 100% confidence in the language "Scheme" with the exceptions "tv", and "channel" - I mean really? the word Scheme wouldn't possibly used for anything else? Seriously?
So can we finally put to bed the idea that TIOBE tells us anything of any value whatsoever? As I've pointed out before a far better methodology would at least taken into account important programming sites like Stack Overflow, but ideally you'd simply refer to job advert listings on job sites across the globe - these will tell you far more about what languages are sought after, what languages are being used, and what languages are growing in popularity than any of this shit.
Finally I do recall last year stumbling across a competitor to TIOBE that was at least slightly better but still not ap
Re:a bit of latency (Score:5, Informative)
Here they have the most popular iOS game development library ported for programming on android in C++.
Re:TIOBE algorithms (Score:2, Informative)
Jets suck -- About 4,770,000 results
Yankees suck -- About 1,430,000 results
Knicks suck -- About 1,370,000 results
Krypton sucks -- About 166,000 results
Re:C? (Score:5, Informative)
Encapsulation - the ability to hide functions inside classes is a far bigger feature of C++ than any of the above.
And how do you hide functions? You put them behind a "private" or "protected" access specifier, but you still have to show them in the class definition in the header file. That's not hiding. That's saying "look at all my nifty functions, none of which you can use, neener neener neener".
In C you prefix those functions with a "static" keyword, and they aren't visible anywhere outside the original source file. Once you compile them into a .o file it's as if they never existed. That is hiding.
Re:C used in your favorite programming language (Score:4, Informative)
It should be noted that for most programming languages, it is highly likely that the compiler and other code used for most if not all programming languages are written in C.
And the C compilers are written in C++. LLVM has been C++ from the beginning, GCC is transitioning to C++ and is now being built with G++, not GCC, and allows C++ constructs in the code. I think many JVMs are written in C++, too.