Vladimir Varankin

> but here's alternative tests saying there's almost no difference o_0 @tadatuta this article speaks about CSS rules performance, not about query selection performance, doesn't it?

@tadatuta plz, don't forget destruction part of i-bem__dom: https://github.com/bem/bem-core/blob/636443a5c69677dd573ab18e9390a2ae29a96164/common.blocks/i-bem/__dom/i-bem__dom.js#L980 as well as force block initialisation, for example after `Block#findBlock*()` call https://github.com/bem/bem-core/blob/636443a5c69677dd573ab18e9390a2ae29a96164/common.blocks/i-bem/__dom/i-bem__dom.js#L123 /cc @dfilatov

I don't mind if this would be fixed in minor release. But since nobody cares, one could wait for a major release without hacks :)

Actually the problem from jsbin's example is that `i-bem` initializes a new empty instance of `input` after blur event happens. As this new instance is linked to the same DOM...

@rakchaev @corpix @Andre-487 вы можете описать, про какие именно «преимущества модульной системы» идет речь, с учетом того, что: 1. объект `BEM` остается в глобальной области видимости; 2. никакие другие блоки...

> upd Пример хорош для «порезался об ножницы — нужны царапки», имхо. Я готов поспорить, что количество людей которые про это даже не задумаются (не то, чтобы попытаются понять), будет...

I've spent some time reading about etcd's data-model; indeed it's a bit more complex than I expected after looking at sqlite's driver but still doable. Will try to sketch up...

I'd like to +1 on the issue. Current API seems a bit wasteful in 100% of our cases: the aerospike-client returns `BinMap`s, that're `map[string]interface{}`, and we convert them into strictly...

I agree to relicense my contributions.

:+1: