Hacker News Reader

Comment by throwaway0255 | Hacker News Reader

Comparisons in Google Trends are extremely easy to manipulate (or just interpret wrong). The ".js" you added to Vue is killing its results, and the generic term "react" is propping that one up.

https://trends.google.com/trends/explore?date=today%205-y&ge...

or how about

https://trends.google.com/trends/explore?date=today%205-y&ge...

Maybe alexa matters?

https://www.alexa.com/siteinfo/vuejs.org

https://www.alexa.com/siteinfo/reactjs.org

Or maybe one community uses their respective primary domain more than the other for one reason or another.

React also has more than 4x as many dependent packages in npm, which could be propping up its install count.

View on HN

> Comparisons in Google Trends are extremely easy to manipulate (or just interpret wrong). The ".js" you added to Vue is killing its results, and the generic term "react" is propping that one up.

The search I linked to is not a keyword search but rather extracted topics. Your criticism is just based on misunderstanding how Google trends work. Not saying it is accurate but in both of your "corrections" to my link you are getting less accurate Google Trend results.

You typed vue.js so technically your results are less accurate. :/

Regardless of the fact Google trends is not even an accurate way to gauge usage of vue or react to begin with.

Vue.js category that I used is the official Google aggregate category. You do not understand how Google Trends work. There are keywords you can use or you can pick Google aggregate categories. I pick the aggregate category, which generally should be the most accurate.

I am not responding to any more because it seems that people refuse to understand it.

Quite likely Google is underrating Vue, because Vue has heavy presence in China and Google does not.

Sounds like you’re upset vue is more popular than you wished.

I think it's interesting that much like on the global trends on Google [1], Alexa surfaces that Vue seems to have some traction in China:

    Country         Percent of Visitors   Rank in Country
    China           59.1%                 631
    United States   8.7%                  7,245
    Japan           3.7%                  6,109
    Iran            2.8%                  2,373
    India           2.2%                  8,422

    Country         Percent of Visitors   Rank in Country
    United States   33.3%                 3,312
    China           9.8%                  6,694
    India           8.5%                  4,016
    Japan           4.9%                  7,585
    Iran            3.6%                  2,632

The top Sites linking into React (according to those Alexa links):

upenn.edu (US) hateblo.jp (Japan) facebook.github.io (US) yiibai.com (China) infoq.com (China)

And to Vue:

sina.com.cn (China) csdn.net (China) wikia.com (US) cnblogs.com (China) blog.jp (Japan)

[1]: https://trends.google.com/trends/explore?date=today%205-y&q=...

My experience with Alexa incoming links is really inaccurate. The numbers are fairly decent for popularity though.

> React also has more than 4x as many dependent packages in npm, which could be propping up its install count.

I don't think this would do anything. If you have three packages that depend on React, you still only install it once. If anything, the number of packages that depend on React is an indication of its popularity.

Similarly to number of stackoverflow questions not being a valid gauge for popularity, I don't think that this is necessarily representative of popularity. What if react is simply more mature and doesn't need to market itself as much, so more people are using it than need to find out about it?

If you look further in Alexa by country I believe it says that React's rank is higher than Vue's site, but that in China Vue is insanely hot.

Reply on news.ycombinator.com