TIOBE has always been a joke, and only even attempt to manage language chatter, not usage. It is based off job boards and internet search trends, and is highly susceptible to homonym confusion, which is especially harmul in the case of "go".

I find it amusing that people working at Google came up with an ungoogleable name for their programming language.

Unofficially, most people writing articles about Go would try to tag them with "golang" for search-ability ("golang" is also used for all the Go e-mail lists).

At least it's better than C. And Java is also a place, Python is an serpent.

The key to Googlability for everything is to put the right context keywords. Nobody just google's "go" (or "java" for that matter). They google stuff like "go string type", "go programming", "go install", etc.

That said, the bare "go" search term gives the go website as the fourth result in Google when I try it.

Generally, using 'golang' rather than 'go' gives really good search results.

Of course you can work around it if you know how to google, but Java and Python are much easier to search. If you look up "java books" or "python books" you get links about programming language books, not about coffee/Indonesia/serpents.

That's because of adoption and less sites linking and discussing go topics at the moment.

I think if you could Google when Python was in it's infancy, you would get 100s results for pythons (the snakes) before you saw Python pages...

Google search algorithm is not so naive as to understand "go" as a generic token only. It can tell from context which pages are related to "go" the language and which merely use the verb "go". The reverse is how old engines used to work, Altavista and co, where you had tons of irrelevant exact matches.

The greatest value of Tiobe is the trending it captures over years, not the actual data point at a given moment in time. Except for spikes when they changed methodologies, the trends, especially for the most popular languages tend to be reliable. Who's in the top 50 or top 100 languages, I agree is not a useful or reliable data point.

A better index is http://LangPop.com, which I recently sold on to a new owner, who hopefully will be working hard to spruce it up some, and add some new languages like Go, although Go is always going to be problematic in search engines.

Another option is Github, which ranks Go as the 27th most popular language. https://github.com/languages/Go

That's a data source I considered adding for LangPop, but back in the day, was too much of a Ruby hangout. I think these days it's probably suitable.

Any reason you wouldn't use StackOverflow tag volume/activity as a signal?

If more questions are asked about a particular language I don't think it would mean that language is being used more in the real world. Just more people are confused about it.

Not sure if stackoverflow tags are a great indicator. TIOBE tries to use job listings and google trends. Searching job listings/search engines for 'go' probably wouldn't return much results (because of the ambiguous name) which is why its a lot lower than it should be.

> I don't think it would mean that language is being used more in the real world. Just more people are confused about it.

The two are not mutually exclusive. Rather, I'd think they're directly proportional. A very straightforward and simple yet popular language should have very few questions, which doesn't seem to be the case; conversely, brainfuck is particularly cryptic and should have many questions, yet it does not.

StackOverflow is admittedly .NET heavy, but since SV is somewhat anti-Microsoft, perhaps this will balance things out.

http://LangPop.com does not track Go

Err... I said that, didn't I?

Why is it a better index?

1. It has more data sources.

2. It lets you fiddle with the 'normalized comparison' to weight the sources differently. If you play right, you can make your favorite language win!