Alexander Mikhailian

Although the popular description of the Open-Domain QA centers on the fact that the questions are not limited by a specific domain, this is not totally correct unless the term domain is used in the philosophical sense of a sphere of knowledge, influence, or activity, which is listed as 'sense #4' in the Merriam-Webster's dictionnary.

However, the term domain may also have a mathematically sounding meaning of a ...set of elements to which a mathematical or logical variable is limited; specifically : the set on which a function is defined. See 'sense #5' of the same dictionary. In this case, it can be said to be equivalent to the CS term data.

As evaluation conferences like TREC teach us, there is always a fixed amount of input data for a QA system. Which means that every QA system participating in TREC is to be called a closed-domain system if the term domain is used in its 5th sense.

Выдали мне не так давно на работе компьютер dualboot: Windows XP и Ubuntu.

И действительно, оказалось полезно, из Ubuntu я пару раз запускал Windows XP через kvm. В первый раз она просила её активировать. Ну дай думаю активирую -- я ведь раньше этого не делал, может положено так. Через несколько месяцев вдруг понадобилось в Windows XP опять зайти. Сделал ребут, запускаю виндовс -- а она говорит, что надо опять активироваться. Ну ладно думаю. Давай. И тут она отвечает, что количество активаций истекло и через 3 дня она типа самоуничтожится.

Так и остался я у разбитого корыта с одним Ubuntu на машине. Windows теперь запускать страшно, вдруг она не дай бог на меня настучит в местный офис Microsoft.

Everyone and his friends are afraid of calling the langue and parole concepts of Saussure by their English names.

After giving in some thought and trying different translations in the discussions, I came out with language and discourse and the optimal translations. Feel free to criticize.

More on Chaos control.

The most dangerous outcome of the underplanning is employee frustration. Without clear day-to-day targets, employees loose interest in the work. Some become anxious, not willing to take on responsibilities beyond what they consider their assignment.

The danger of the overplanning is that detailed plans are hard to change.

The rigidness of the plan is proportional to the time spent on it, as the only reason not to change plans is often sheer inertia of those that invested already a lot of time to write it up, in the first place.

An ideal planning phase is short and concise, and the planing granularity is barely sufficient to distribute the day-to-day assignments to the employees.

A project manager plays with two major forces: Chaos and Freedom. When a project starts, nothing is clear, the previous stages like business analysis and user requirements collection have likely been a complete failure and collected input won't help much the team and the manager. At this point, the manager has to control the Chaos and reduce it to the minimum, by creating an environment where everyone know exactly what to do, how to do it and how much time would it take.

So he starts to plan, as the work moves on. The days roll out, and he slowly realizes that there is not much left of the initial planning, and the team is swirling out in different directions. This is where the Freedom control takes precedence. The manager has to give everyone enough Freedom so that every morning employees felt the urge to scratch an itch, while still working in the planned direction.

He slowly realizes that the Chaos/Freedom ratio is unique for each employee, and he has to work out a personal approach to everyone if the wants the team to work at full throttle.

By the time everything is crystal clear in his mind, and all the errors have been consumed and learned from, the project ends, and the manager moves on the next one or bails out.