November 04, 2019
On Monday, 4 November 2019 at 21:09:47 UTC, Ola Fosheim Grøstad wrote:
> programming" and "dlang programming". However, for D there are false positives, so they assume 90% confidence... Although they

FWIW, you'll see hits on things like:

a) …
b) …
c) …
d) programming is a…

Or stuff like:

…ph.d. programming…

But they only look at the relevance for the first 100 hits when estimating the confidence metric, which isn't really compatible with how you can structure ranking in a search engine.

You might get less confidence the further down the ranking list you get. E.g the first 100 could give 100% confidence, but then as the ranker is running out of "good matches" the number of false positives could increase significantly...

Also... the search engine might do a cut-off or use a predetermined hit-list for popular searches...

Too many potential sources for significant bias in Tiobe (and aggregating results from many search engines does not make that better, could even make it worse).
November 04, 2019
On 11/4/2019 1:03 AM, RazvanN wrote:
> Good news, everyone!
> 
> D has entered the Tiobe top 20 ranking of programming languages [1], landing on the 18th position. I have been keeping an eye on this index, and it is for the first time that I see this happening.
> 
> Cheers,
> RazvanN
> 
> [1] https://www.tiobe.com/tiobe-index/
> 
> 

Great news!

(It's been there before, until Tiobe changed their ranking method.)
November 05, 2019
On Monday, 4 November 2019 at 09:03:04 UTC, RazvanN wrote:
> Good news, everyone!
>
> D has entered the Tiobe top 20 ranking of programming languages [1], landing on the 18th position. I have been keeping an eye on this index, and it is for the first time that I see this happening.
>
> Cheers,
> RazvanN
>
> [1] https://www.tiobe.com/tiobe-index/

I've been keeping an eye on the Tiobe index in the last couple of months and I started to notice that D had a slow but steady upwards trend. It was in the 22. and 21. place in the last two months. I'm a bit surprised that it jumped all the way to 18. place. It's a good thing that this happened because the index has a lot of visibility. Journalists like to write a lot about it for some reason.

However, this index is quite noisy and flawed in several ways. So this high rating for D  might be a fluke and next month D might be back to 21. place... I'm really hoping that D is starting to show up in the top 20 more often than not in the coming months.

If you take other indices like the more complex Redmonk index that takes its data from Stackoverflow and Github and is thus biased towards open source projects, you get totally different results. For example, the Redmonk index ranks rust pretty high. This shows how arbitrary these rankings are in the end.
November 05, 2019
On 04/11/2019 09:03, RazvanN via Digitalmars-d wrote:
> Good news, everyone!
>
> D has entered the Tiobe top 20 ranking of programming languages [1], landing on the 18th position. I have been keeping an eye on this index, and it is for the first time that I see this happening.

Not that I am advocating gaming the Tiobe ranking protocol but I note that "ratings are based on the number of skilled engineers world-wide, courses and third party vendors. Popular search engines such as Google, Bing, Yahoo!, Wikipedia, Amazon, YouTube and Baidu are used to calculate the ratings."

The above being so, a quick win for a language that wished to raise its Tiobe ranking and thus its general profile would be for its community to post lots of "course" (i.e. tutorial) videos on YouTube. It is possible that a series of videos, constituting a course, might be better still. This would potentially increase the ranking directly based upon YouTube hits. If language community blogs and forums could then link to these YouTube courses then this would, in turn, increase the language's ranking on Google, Bing, and so on. Wikipedia articles linking to the YouTube courses would also be beneficial.

Just a thought.

It is ironic that an increased Tiobe ranking (amongst others) can, in
and of itself, increase the genuine user base of a language. (As,
perhaps, with the community hype around Rust).

-- 
Mark Rousell






November 05, 2019
On 11/4/19 2:17 PM, JN wrote:
> On Monday, 4 November 2019 at 09:36:33 UTC, RazvanN wrote:
>>
>> I think that all this Rust hype is artificial and people with large codebases that started looking into it are not that pleased with it.
> 
> It's not artificial at all. Many big companies start their projects in Rust now. Rust became the go-to language for new projects that would use C/C++ normally. They definitely have the hype behind it and the hype helps build its ecosystem, which will be helpful in maintaining the popularity after the hype dies over.
Yeah. Also some companies try to drop the current language for Rust and then after some time of Rust experience switch back. Considering huge hype about Rust and comparing it to Go it's obviously that something is wrong with Rust adopting.
I totally agree to Andrei about Rust and I think it is a good but niche language and with abnormal hype.
November 05, 2019
On Tuesday, 5 November 2019 at 04:00:02 UTC, Mark Rousell wrote:
> The above being so, a quick win for a language that wished to raise its Tiobe ranking and thus its general profile would be for its community to post lots of "course" (i.e. tutorial) videos on YouTube.

That depends on the ranker. Seems like Google is trying to reduce the number of hits from the same website. So, with that in mind you would be better off getting bloggers on various blogging sites to write about it in different languages. See:

https://www.tiobe.com/tiobe-index/programming-languages-definition/

If you look at the description of Tiobe it seems like they aggregate Google hits from various languages. So the best strategy seems to be to game Google's ranker and do it in all the languages Tiobe are tracking:

Google.com: 7.69%
Google.com.hk: 5.85%
Google.co.in: 5.23%
Google.co.jp: 4.00%
Google.com.br: 3.69%
Google.de: 3.38%
Google.ru: 2.77%
Google.fr: 2.15%
Google.it: 1.85%
Google.es: 1.54%
Google.cn: 1.23%
Google.com.tw: 0.92%
Google.com.mx: 0.31%

November 05, 2019
On Tuesday, 5 November 2019 at 02:08:28 UTC, Gregor Mückl wrote:
> However, this index is quite noisy and flawed in several ways.

Right, the big dips in the graph are from Google adjusting their ranker... AFAIK.
November 05, 2019
On Tuesday, 5 November 2019 at 02:08:28 UTC, Gregor Mückl wrote:
> If you take other indices like the more complex Redmonk index that takes its data from Stackoverflow and Github and is thus biased towards open source projects, you get totally different results. For example, the Redmonk index ranks rust pretty high. This shows how arbitrary these rankings are in the end.

Redmonk is flawed as well, of course. A language with worse documentation, is more difficult to use or appeal more to beginners (like PhP) will have a stronger presence on Stackoverflow...

Github is probably the better metric, but counting number of projects is silly. Some languages have a culture of miniature repositories. Others have lots of forking.  But yeah, if you did static analysis on all the source code on Github then that could measure something reasonable.

Also the counting on Github seems to be approximate. I get ±10% in repeated searches their web interface, e.g. sometimes 7000 hits, sometimes 8000 hits…

Searches on github:

2k language:nim
8k language:d
10k language:julia
23k language:go
43k language:rust
759k language:c
880k language:c#
890k language:c++
3400k language:java


November 05, 2019
Ok, so here is another github metric that is more interesting. Doing a "language:d" search and then look at the number of _users_:

Zig: 10
Pony: 25
Nim:  400
Crystal: 500
Ada:  600
Vala: 800
D:     2000
Pascal: 12000
Dart: 13000
Rust: 16000
Swift: 24000
Go:   128000
C++: 706000

Clearly, this metric favours older languages. It would have been better to look at new repos only, but I don't know if that is possible with the github interface.

November 05, 2019
On Tue, Nov 5, 2019 at 11:50 AM Ola Fosheim Grøstad via Digitalmars-d <digitalmars-d@puremagic.com> wrote:
>
>

I am using openhub interface:

https://www.openhub.net/languages/compare?utf8=✓&measure=commits&language_name%5B%5D=erlang&language_name%5B%5D=dmd&language_name%5B%5D=-1&language_name%5B%5D=-1&language_name%5B%5D=-1&commit=Update