The Rise And Fall of Languages in 2010 | Hacker News Reader

The Rise And Fall of Languages in 2010

drdobbs.com

41 pointsfogus24 comments

"Over the years, the index has been both praised and vilified. The latter by language adherents who are unhappy over the decline of their favorite idiom."

Really? That's the best defense that can be brought for the methodology - a textbook ad hominem attack?

Criticism merely reinforces the uncomfortable truths that the programming elite have silenced for so long! Anyone who asks for reproducibility in science is part of the conspiracy too!

I thought TIOBE was vilified because it does a really bad job of measuring what it claims to measure.

This paints a very different picture: https://github.com/languages

edit: I realize this is measuring something different, but it suggests that the reports of the decline of Ruby and JavaScript could be wrong.

There are two major problems with this:

Any project which has html documentation, which contains javascript, gets counted as javascript.

git and github in particular seem very popular in some communities (ruby is a good example) and less so in others (C++ comes to mind, as one I am familar with).

Github tries to alleviate some of this, but they're not great at it yet. They've said in the past that they discount the existence of well-known JS frameworks like jquery.

BitBucket seems to have a lot of Python projects hosted there which might take a chunk out of Python's share on Github.

Alternative opinion: http://langpop.com/

The langpop link is interesting because it does an analysis by source. If you are familiar with Craigslist, you can see some of the problems with the methodologies. The SF Bay Area is heavily into web development, so PHP is no surprise. But C and C++ are second and third and essentiallt tied. That much C development going on? Not really. The majority of listing have C\C++ some where in the acronym stew, just to indicate that they want real programmers. So what were companies hiring a year ago, when last I looked? PHP, Java, C#, Ruby, and C++. All the rest was HR noise. But it's more interesting than nothing.

Allright, who has been playing with the random number generator?

One thing you'll have to do before looking at the data presented here (if you want to give it a second look) is to take into account that there is a lot more to programming than just web programming.

But even with that in mind I have a hard time seeing Assembly on the rise and JavaScript dropping and named in one breath with delphi.

And I'm not a fan of any of those (ok, maybe assembly), so that can't be it.

If you want to skip the article to go right to the data it's here:

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

Ada, RPG on OS/400 and Transact SQL? Really?

Anyone risks a hypothesis on why this happened in 2010?

Lot defense work, people not replacing OS/400 machines and still needing new features, and MS sold a lot of databases this year[1].

[1] If you go through your corporate approval process to replace mysql because your mad at Oracle, don't be surprised if SQL Server walks in the door.

Why would I want to replace MySQL. It's GPL. Larry may scream all he wants and business will continue as usual. The worst that can happen is to use a forked version.

Drizzle is very interesting, BTW.

Welcome to the wonderful world of corporate America, a good Microsoft salesman, and listening to the tech press. :)

It never ceases to amaze me how resilient corporations are, given so many decisions are delegated to the people least qualified to make them.

The only reason I read mainstream tech publications is to know which silly ideas I will have to demolish in the near future because some clueless PHB decided it would be the next great thing.

Yes and if you believe their data, Pascal and ADA are also on the ascendancy.

I can see some rationales for both modern Pascal dialects and Ada as application languages. Foot-shooting is less readily achieved than in C or C++, and the compilers and libraries are mature, so you can do some pretty heavy lifting. As well, Oracle is scaring people away from the JVM, so where the answer for a lot of apps used to be "Java, period," now we're probably seeing more investigation of alternatives.

I can't imagine anyone being chased away from Java by Oracle's IP claims would run to Java. They are such different languages, so much so that this assertion is as audacious as claiming people who recently quit smoking are becoming addicted to cabbage and other leafy vegetables.

Ugh, Tiobe. I wish we could come up with a better measure for this. I don't really feel that Tiobe's index is worth anything at all.

JavaScript with a double-arrow drop in 2010?

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

"Tiobe programming community index is an ordered list of programming languages, sorted by the frequency of web search using the name of the language as the keyword"

Maybe people are searching more for specific frameworks and libraries instead of the language. I work mostly in Python and it's been a long time since I last googled for "python". These days I google for "django" or "appengine".

Because obviously the most popular languages must be the best ones.

... and obviously it's trivial to filter out false cognates when analyzing search term volume.

... and obviously it's statistically meaningful to compare year over year trends without accounting for changes in global search volume.

... and obviously you can trust an index even if it doesn't know the difference between a web browser and a language implementation.

I fear that attempting to name the "best" languages is just as fruitless as trying to name to the most "popular" languages.

Reply on news.ycombinator.com