Annual Reporting Habrapost - 2015


    The past 2015 was not an easy one. So it is from 2016; the next prime number in the annual sequence will be 2017. But, we admit, not this mathematical collision created the editions of the Intel blog the biggest problems. Our harmonious system of statistics, which was traditionally published during the New Year holidays, stumbled over reform in Habré; after its propagation by budding, it became decidedly unclear what and how to count. In shock, we even missed the previous year, but this time we decided to pull ourselves together - for that they are traditions in order to comply with them.
    So, the calculation of statistics during the transition period (when the road behaves like in that bearded anecdote and on the KDPV), and, of course, the very-most posts for 2015, as well as the New Year’s contest, will be discussed.

    “Misfortune accidentally comes upon you when you are not waiting for it at all,” - so, it seems, was sung in an old song? He lived an Intel blog, did not bother, issued posts, summed up , and suddenly, without declaring war, Habra was divided into parts, and there were more blogs, and there were fewer views, pluses, comments, and all that. And this is not so bad. Intel's blog has been artificially divided into two parts; artificially - because many posts in their content should have been on Habré and GeekTimes, but they ended up in only one feed, which, of course, also affected the ratings.

    Our first thought was to introduce a coefficient that takes into account the numerical changes in the indicators of posts before and after separation. But the trouble is, it turned out to be not so simple to calculate. And although the length of time to this day seemed already sufficient, regularities were not traced. And dealing with data that is noisy to the point with gigantic dispersion is somehow not Intel's way. Reasoning sensibly, we came to a solution that was not entirely satisfactory to us, but seemed to be the only possible one: to leave everything as it is, and somehow live with it, considering the starting point a new blogging system, but not forgetting about the old glorious times. Yes, 2013 now seems like an impossible dream ... but, but we have something to strive for.

    On this lyrical preface, let me consider it finished. Now let's move on to statistics. I remind you that the apocalypse came in October 2014. 2015 is the first full year after the reform. The results of Habrahabr and GeekTimes will not be shared - we were told that this is a single blog.



    With quantitative parameters, as you can see, everything is fine with us. The authors-employees of the company continue to write interesting posts, the number of posts has increased significantly due to transfers from Intel's English resources. But do not say that for us these translations are 'free.' Firstly, an interesting article for blog readers still needs to be found (try visiting the Intel sites for developers), then translate, report and edit.



    But with ratings, as already mentioned, everything is sad. However, we will still keep in mind: “+1” of the sample of 2013 and 2015 are completely different “pluses”.



    With community indicators, too, everything is not entirely successful. The most annoying thing is that GeekTimes is pulling Habrahabr down ... but we won’t say anything about it.
    Nevertheless, the number of subscribers to both blogs over the year has increased by about a thousand people, which, of course, we add to our assets.

    We now turn to the most-most.



    And, in conclusion, a few words about the competition held in December among readers. Firstly, the number of respondents was a pleasant surprise for us. Yes, of course, we follow our audience, we know in the face of the most active commentators, but to be so friendly ... we are very pleased, honestly. The second truth revealed to us will seem trite to someone: in any case, experience is needed. We held the competition in the recent history of the blog for the first time, there was no one to make mistakes before, and we did it ourselves.

    It turns out that the law formulated back in the days of Murphy: “if you can be misunderstood, that is how it will happen” - just about the drafters of the competition questions. For example, asking the question: “Which network cards did Intel first introduce this year?”, We weren’t ready for the majority to answer “PCI Express”. After all, we had in mind the network interfaces that appeared in 2015. But they did not explicitly spell out the word “network”, and the thought of the respondents went the other way (in the wrong direction, by the way, since Intel hasn’t come up with anything new in terms of PCI Express). And the correct answer is 40G and 100G Ethernet.

    Another problematic question: “How many times did we publish a post describing all kinds of sensors that can now be connected to Intel Edison?” Here it was necessary to emphasize that there is a post entirely devoted to sensors for Edison. Not everyone found it among other posts about the Internet of Things, including those describing the connection of individual sensors. Here he is . But the most difficult was, as expected, the question about the number of hackathons. There were only three of them, believe me: one in Nizhny Novgorod , the second in St. Petersburg and the third in Moscow .
    Note that all difficult questions were evaluated in a special way: the correct answer was considered a bold plus, and the wrong one was a small minus. We will not dwell on simple ones, everything is clear with them. All winners have already been notified by email, in January our prizes will be sent to them.

    So, wish us successful posts in the already difficult 2016 year. And to you, in turn, creative successes and professional achievements!

    Also popular now: