article Eldar. Just when preparing a new article, thinking about the material in the same vein, he wrote Eldar. Therefore, I will not repeat myself - read the first article of his, and then my postscript ...
Managers have described the Eldar, which is often present obsession with the various methodologies. From myself I will say that the programmer that results in the search for ... Do nerds - excessive zeal to reinstall anything and everything ...
In general, the whole point then read the article I can kill at once, saying that it is often age-related. Senior colleagues just need to keep with you a long line, take care of their younger colleagues, and on time, accurate shot hurts to beat them on the fingers. Then, the person gains experience, it is reasonable and prudent in their search for the ...
sysadmin.
We nerds with this quite simple: first the minimum level of rights, then more, and so, for 7-10 years, you see, the pros will grow from the ... Usually start with the fact that charge setup client sites. The main thing that a young sysadmin learned not to rearrange the stupid OS, and find the causes of failure to locate and remove them. After all, the process often involves reinstalling while preserving user configuration, its working files and the subsequent restoration. And it might not be a quick process, and yet the user still need to work. Do not even always make sense at all to eliminate the error - it just easier to get around (eg, a USB-port printer does not print, switched on the other hand, if the print was then, and figs with him, with the first USB-port ). It is important to learn how to solve the problem of minimal gestures.
Second, what has to learn sysadmin - is. healthy conservatism. Reinstall the software makes sense only if it is really justified. Not according to the principle ...
Third, almost inaccessible to the majority rule:. user needs to know. The end user rarely knows which of his problems can be solved using a computer. That is, if you can at the right time and give the right solution, then the pass for the ... Often this is a very simple solution, it is obvious to you but not for the uneducated in IT colleagues.
And then, when these principles are in your blood, then comes the understanding that the word ... Problems will be solved easily and with minimal gestures. It will be enough to make a couple of options, the worst case, to ... The equipment will operate for years without hovering, and behaving in a very predictable. From the side, you'll look like a magician, it works by itself, you're just a couple of times on the buttons pressed, and there is something to be healed, something else is run by. And all done without fuss, without excessive swarming, accurate and calibrated. That's only if you call a great professional, a true professional.
programmers.
We let programmers more florid. Almost from the very beginning of his career should solve the following dilemma. On the one hand, well, if a novice programmer will soon learn to program properly: to know the different data processing algorithms, design patterns. On the other hand, not having enough experience, your new knowledge will be applied very lopsidedly, as in such cases we say ' yes, take your hands for such a right! '. Personally, I decided for myself that beginners like to talk about the patterns do not. It is because of the dilemma articulated I oppose patterns. Because beginners are confusing, and experienced no longer needed, and so they are in practice they have mastered. And where is the line where you can ' harmless ' to examine the patterns? . I do not know there is no recipe, I do not know. Only the experience, but its ' bumps '.
When it comes to experience, then head right, ' on shelves ' are arranged and patterns, and basic data processing algorithms. And the next stage. How to properly make a decision about which approach is used to solve the problem. Decisions tends to be more. Typically, this occurs because of insufficient understanding of the problem. My recommendation is to go into the task, taking it to the smallest component parts as long as is absolutely clear it becomes clear how to solve the problem. T. e. must learn to filter out a lot of decisions through a deeper understanding of the problem.
Unfortunately, not all that easy. Often, even after a detailed study of the problem, it is still ' on hands ' is several ways to solve problems. Then suggest to use. Occam's razor. :. choose the simplest solution.
But here, too many difficulties! . Yes, we were already a considerable chain in making correct, ' the divine ', the only true solution and, nevertheless, we have a very high probability to remain in ' square one ' (t. e. make the wrong decision ). And then, near the right decision, can not relax. I'm not exaggerating. That's a real example from my practice. We introduced an ERP system. It is inculcated with a creak, and under strong pressure from superiors to subordinates who ' otbrykivalis ' from it as they could. Because the system was inconvenient. It was a powerful, feature rich, but uncomfortable. Screen forms abounded bunch of small buttons, just the size of 4x4 pixels, otherwise they would not fit into the screen. Most of the standard buttons and hot keys have been redefined in a very peculiar. Why did this happen? . So what they did not have enough to do the right thing?. intuition. Yes, at the end of the complex ways of making decisions is not something logical, understood, and intuition. Irrational sense of beauty. The feeling that if you make it so that it will be properly. Without logic, and without explanation: ... '.
And, on top of excellence, a programmer in some overlaps with an experienced admin, because the admin - destination user needs to know that, too, relies on intuition intangible.
PS: Yes, by the way, a ' divine ' software architecture does not exist. ;).
No comments:
Post a Comment