Author: Rauf Aliev

I thought it was in America where the portions are big. And there was an extra large option on the menu, but that’s not the one I chose—I went for a smaller one. Find the glasses)

Please provide the text you need translated.

Please provide the text you need translated.

Getting ready for takeoff. Right in front of me on the mat, a Muslim is praying. Polite, apologized before and after. By the way, do they carry a compass with them?

Funny, the word apparatchiks in English doesn’t need decryption

I recently read a Facebook post saying that if you break a thermometer, you must call the ghostbusters who will carry out the demercurization of the area and save your life. Crawling around with a piece of paper and gathering the beads is just foolery, apparently. It suggests that you can’t scrape it all up yourself, and only professional help can save the space and its inhabitants.

I decided to see what it’s like here. Everyone seems to be obsessed with such problems. Each similar incident is blown up to the scale of a cosmic problem and there is always a company offering professional help with the consequences. But somehow, mercury removal services are hard to find.

Just for fun, try to find at least one American organization that performs demercurization of premises. There is Poison Control, but as I understand it, they focus on giving the right advice and helping the caller if necessary. If you manage to find an organization (surely there must be one in the States), try to find their prices.

Yesterday we watched four episodes of Stranger Things in one go.

Interestingly, Netflix decided it was best to release all 10 episodes of Stranger Things 2 at once, rather than releasing them one per week. Apparently, subscription statistics show that the latter does not bring in more money than the former. And that’s quite fascinating. Think about it: watching all 10 episodes costs nothing (they give you a free month of subscription), or $7-14 per month if you’re paying. Releasing episodes weekly would have accustomed users to returning to the service, and likely would have pushed them beyond the free month with the series unfinished, which would have motivated them not to cancel the subscription. Meanwhile, there would have been heated discussions at water coolers and during kitchen breaks, which could have generated more buzz for the series. But the owners believe it is right to release all 10 at once. It’s interesting how much this decision is deliberate and calculated.

I released the sixth part of my presentation at SAP Moscow. This time I will talk about two topics: page fragment caching and PDF generation on the server using templates. The first experiment is explored in three variations: standard, using Varnish, and built on Nginx/Memcached. The second experiment also demonstrates several ways to solve the template-based PDF generation task. I discuss the pros and cons of these solutions, as well as reveal the implementation details.

Despite it ostensibly being about Hybris, the topics mentioned are not related to Hybris at all. They can be effectively used in any web system.

Welcome)

Yesterday we watched “Your Name” by Makoto Shinkai. A very beautiful animation, I recommend it. Romance, beautiful Japanese scenery, just as usual with Shinkai. The plot is a bit stretched and raises a bunch of questions, but that’s not very important, the film sets both the mood and the atmosphere well.

I believe that in a few years a new business will emerge, similar to today’s cryptocurrency mining, but focused on recognizing data from real-world sensors, such as video images or audio clips. Some Google or Yandex will be able to perform primary classification of objects, using their vast cluster with a neural network to determine that a picture shows a crocodile and a face. Third-party “contractors” will be able to focus on recognizing types of crocodiles or facial expressions, earning money from this specialization. Of course, for the first example, the neural network will need to be trained on crocodiles, and nothing else. Since there are very many object classes, a single, even large network owned by a giant like Google or Yandex will not be capable of handling everything, but it can “outsource” correctly: to another company that has a narrow specialization, and together they will provide the client with the best result.

Theoretically, recognition tasks can be assigned to more than one “contractor” and the results can be merged. For example, face recognition might be performed on different networks, trained on different races and nationalities. If the prediction of the contractor’s network turns out to be incorrect, their rating decreases, if it is confirmed – it increases.

Most likely, the system will not be structured as a tree with Google or Yandex at the top, and smaller companies below, but also as a network, where one company enriches the input stream with new knowledge, which helps the neighboring network to perform its specialized work better. For instance, understanding a crocodile’s breed might narrow down the recognition of the face of the aboriginal standing next to it, since breeds are tied to localities.

What do you think?