Ranking the Programming Languages
As you probably know Jelastic is an international company, American-Ukrainian-Russian to be exact. Recently I found very interesting post by Ruslan Shevchenko on one of the most popular Ukrainian tech blogs DOU. It’s very detailed stats on programming languages popularity among Ukrainian developers according to different parameters. So, let’s dive into this great post!
Brief summary: Java is on the top, a generation not acquainted with C++ comes to the development, Apex has appeared on the radar.
The ranking is based on a survey lasting from April 7 to May 5 2012, it included 2,758 participants (2472 of them are from Ukraine). 41% of them took part in the previous surveys, and 59% – filled the form for the first time.
The chart below demonstrates the stats on programming languages popularity according to the first written program.
It’s obvious that the number of developers, who started programming with C++ decreased comparing with the previous year. It’s quite interesting that a new generation of developers chooses Pascal and Basic for education.
And what is going on now?
This chart is based on the answers of users, who managed to pick out one programming language as the basic.
So, you can see that the biggest part of our audience codes on Java – 26% (9 months ago it was C# with 21% result), the top five is closed by Python (6%). 1C is on the border of marginality with 1% frequency of use. It should be pointed out that Apex has appeared on the radar (as I understand this is Salesforce embedded language).
Let’s see the trends.
Now let’s see what our audience thinks about future trends:
In this case Python would be faster than PHP, and Scala would be the last in the second group.
Let’s build the language attachment index as a percentage of users, who would choose it for their new commercial project in their specific area:
The most comfortable languages for their users are Python (0,91) and C# (0,88), and the most uncomfortable is 1С and PL/SQL (only 0,33 for each one).
Migration patterns for the upcoming projects:
- PHP developers targeted at Python (16%) and Ruby (9%), then at Java (5%) and C# (4%);
- From Java – at Scala (10%), then C# (55), Python (3%) and Groovy (3%).
Note that no new wave language (Kotlin, Ceylon) has been even mentioned.
Let’s see now what languages are used as additional. By the way, 13% of the audience prefers not to mark out their main language.
Now let’s consider the distribution of languages in developers’ projects (77% of respondents have their own projects):
Let’s examine the demographics now.
Here we have a little surprise: the audience of our survey has changed – there are many young developers and the number of developers with more than 5-year experience has decreased.
It seems DOU is losing the audience of experienced developers for several reasons: starting with another wave of emigration and reducing social activity with age and ending with relationship problems between HRs and programmers (resume writing and coding of startup in two days is still more interesting for young programmers).
Here are the current numbers:
Programmers with more than 10-year experience have 13% of a share (comparing with 18% of 2011), programmers with 1-year or less experience have17% (last year it was 14%).
The distribution by the age has not changed:
The most of developers are from 20 to 30 years old. It’s quite interesting, that in this year’s survey 13 guys older than 50 have taken part and 8 indicated their main programming language: Java for 3 of them and Objective-C, Delphi, PHP, PL/SQL and Scala for the rest.
The age group from 40 to 50 also includes Java programmers (15) most of all, they are followed by C # (6), and only then C + + developers (4).
The ratio between the total experience and work experience with specific language:
Also it’s interesting if we have any difference between using programming languages in our country and in diaspora?
Yes, we have:
It seems that the prevalence of Java is a property of outsourcing industry focus, we also work more with PHP, than with Python, and in the diaspora everything is vice versa. Also the use of C ++ is much higher abroad.
Another source of information on the distribution of languages is the relative frequency of vacancies.
We see that the leaders are the same, but the number of vacancies on C# is significantly higher than the number of vacancies on Java – it seems that the use of Java has grown faster simply because of people selection easiness.
Features of the markets are visible and obvious:
DOU is more focused on outsourcing and it.rabota.ua – on the domestic market, so on it.rabota.ua the PHP share is much greater, and on DOU the same is with Java and C #. There are no 1C vacancies in DOU, but at the same time they constitute a significant part in the ads on it.rabota.ua. The situation with Oracle PL/SQL is reverse.
And finally here’s the table with data sorted by market share among the DOU audience:
So now you can compare this data with statistics in your country. Any differences? I’ll be glad to discuss this in comments below.