Writing a thesis. Instructions for use. A look on the other side of the barricade

On Habré regularly appear articles aimed at helping with writing and defending a thesis (for example, here and here ). Personally, in my time I was helped by the comic commandments of the dissertation , whom for more than a dozen years (the variant that I first saw was still handwritten, and the shabby is not worse than the ancient Russian chronicles).

And in general, writing a dissertation is sometimes useful from a professional point of view, when, for example, you are engaged in a knowledge-intensive business ( which I eventually came to ) and I want to impress customers with the depth, width and height of knowledge and regalia.

But the topic of the dissertation writing procedure itself wanted to be supplemented with a little look from the other side. It so happened that I did not just once wrote my dissertation myself and forgot about it as a terrible dream, but for several years I have been a member of dissertation councils - therefore I regularly hear and see defenses passing, and, accordingly, typical problems with them related.
')
Below I will try to briefly describe the main points that should be paid attention to when preparing and defending a thesis (first of all the candidate’s, of course, because those who write a doctoral thesis are, in theory, held scientists, and themselves with a mustache).

At the same time, I will note that all of the below is in no way the official position of academic councils or, on the contrary, the secret knowledge of the clan of professors, but simply the subjective opinion of a single person taken by me.


1. First and foremost.
You should definitely give yourself the answer to the question - why do you personally need a dissertation? We need a very, very serious motivation to start this job. This is a difficult and often unpleasant work, which will take several months very intensively, and at least for a couple of years it will hang constantly with a Damocles sword. I don’t have big statistics, but I would say that out of 10 beginners it’s good if 2 complete the job.

2. Never confuse science and dissertation.
You can do this science all your life and not have degrees. You can defend a thesis that has nothing to do with science.

3. The thesis - a scientific qualification work.
What does it mean - a graduate student must show sufficient qualifications in order to be considered a scientist. If you read about the initiation rites of the ancient tribes - writing and protection are very, very reminiscent of them. You go through artificially constructed circles of hell to prove you are worthy.

4. The choice of supervisor is important, but not critical.
For to help you in the process of work, it is not a fact of what will happen. He is a critical figure at the first and last stage - when it is determined where and where to dig, and when it is necessary to suggest how best to arrange and protect the finished work.

But if you are lucky and the leader is genuinely interested in your work - appreciate. Do not let his enthusiasm burn out, show progress regularly.

5. Measure the contribution and the result.
If you put too little soul into the work, it will be very boring to do. If you invest a lot - you can be very disappointed when you realize that there is very little work needed for you but you.

6. It is necessary to clearly understand how the dissertation is analyzed.
In the worst case, the thesis itself is read by exactly one person - you yourself. In a typical - three more, opponents and a supervisor. VAK and others look at the accompanying documents. The Council itself - speaking on the defense to evaluate the work is more than enough. But the author's abstract can already read a few dozen people. And an introduction to the abstract (i.e. first pages 5) is already mastered by a hundred.

7. A very simple rule follows from this - most of the time and effort must be devoted to what will be studied most carefully.

Namely:
a. Introduction
b. Scientific novelty
c. Provisions for protection
d. List of works of the applicant
e. Opponents and Lead Organization

8. Where to start - take one. Search threads.
This is where the help of the supervisor is needed - he has to throw ideas and tasks, one of which should interest you. When you are interested, just look at the Internet and see how much has been done on it (I highly recommend looking in both Russian and English-speaking Internet. In extreme cases, if you use the language of discord, translate.google.com or similar services - they can automatically translate).

An alternative source of the topic can be anything that interests you - an article from Habr or Techcrunch, a solution to a household problem, saving the world, whatever you want. Separately, I recommend looking at the passports of the specialties of the nearest Soviets - many ideas may appear. But in any case, then it is necessary to invented the invented topic with the supervisor - he should have an eye on the "protection" of the subject. Sometimes the topic is interesting, but there is no science in it.

But the opposite is true - if the topic is boring and uninteresting - see point 1. Boredom kills motivation.

No need to take a topic where there is practically no work, or a lot. In the first case, ogrebet all the buds of the innovator, in the second - you get tired of proving novelty.

I highly recommend, if possible, to combine the topic of scientific research with what one has to do in one way or another at the main job. Yes, working and writing a dissertation at the same time is very difficult. Yes, at the final stage you will not be able to work for 2-3 months at all due to the additional workload and activities. But in the end, all the same, this approach will save a lot of time and effort.

Yes, and the obvious, but important trifle - keep everything you find somewhere. The same Evernote + Web Clipper will save a lot of time in the following steps.

9. What to start with - double 2. When there is a theme, but nothing else.
Personally, I recommend the snowflake method (writers love to use it). In this case, it is applied as follows - it is necessary to compile the content of the thesis (take on the Internet / library any publicly available theses in the specialty and see how it usually looks).

The content greatly clarifies the picture - because you still do not know what specifically you will write in the review or, say, the system architecture, but you already understand that there should be such parts in the work. It can be redone many times until you have a feeling that on each point you theoretically have something to write or you know how to approach it. Well and, accordingly, that if all the items are filled, you get a coherent story about solving some problems.

My own snowflake, in articles and business intelligence, most often looks like this: Chapter 1. The urgency of the problem. Review of available methods. Advantages and disadvantages. Perspectives. Chapter 2. The theoretical part proposed by me. Methods Algorithms. Comparison and niche. Chapter 3. The practical part offered by me. Architecture. Performance Research Comparison with competitors. Chapter 4. Practical application. Task - problem, solution, results.

10. Quite often, a graduate student (especially in high school) is perceived as a free labor force .
Avoid. Or bargain - because everything you do should bring you closer to your goal - a dissertation. It is not necessary that this was directly the text of the dissertation. An article, a program or just public speaking skills, acquaintances and good relationships are also needed.

11. Defended in the “native” Council is immeasurably easier.
In this regard, sometimes it makes sense to ask, and which specialties and topics are relevant and generally can be considered at the Council. Each Council has its own set of specialties. Definition and passport specialties can be found on the Internet.

12. No need to postpone writing articles for the last moment. Now at least 10 publications are required for the candidate’s, and two or three of them (for different specialties in different ways) should be on the VAK list (detailed requirements here . Thanks to the user of DreamWalker for the link).

If there is a ready more or less complete piece - describe and publish. If you want to modify - write in the article about the plans for improvement. And how to finalize - another article. Of course, it is not necessary to describe every sneeze, but to polish to a mirror shine is senseless.

Article writing is a skill that trains. I wrote my first scientific article for more than three months. Now on topics that I understand well, I can write an article in 4 hours (though I don’t have to do that, the article is still a serious matter).

13. Participation in conferences , as well as studying other abstracts or simply attending the pre-defense / department meeting is also useful - the more others examples are seen, the clearer the requirements and expectations become.

14. Any questions or ideas that you have in the course of writing the text, write in a separate file.
If the topic of the thesis is changing - keep the old versions in drafts. Any comments that you give in the course of work, keep, as well as the answers to them. In addition to being just a useful source of your own ideas, this is all very useful for you at the final stage - the preparation of additional materials.
Of course, all my friends and acquaintances of my friends are crystally honest people, but I heard that there are times when the applicant has to write the manager’s comment himself. Reviews of opponents. Ten reviews on the abstract from different universities from all over the country. Acts of implementation. Conclusions And many other equally interesting documents. Let's just say that the situation is not excluded that you will have to do this too.

15.Type errors and problems - the lack of scientific novelty.
A very big problem inherent in many works. People are not able to formulate what is specifically made new (and sometimes do not think about it in the process of writing - because in the diploma, for example, it is rarely required). But the PhD thesis in complexity is about 10 degrees.

For each point of scientific novelty, I recommend the triad formulation:
The proposed <defensible idea> , which differs <specifics of the proposed approach> , which allows <pragmatic benefit> .

16. Typical errors and problems - the lack of comparison with competitors.
According to 9 dissertations out of 10, I could write the same question as an opponent's remark, without even looking at the work. They say that insufficient analysis and comparison of the proposed method with respect to existing analogues was carried out.
The problem is both in the theoretical part of the comparison, and in the practical. And, interestingly, there is a review of known methods. Their shortcomings also exist. And then comes the offer of his beloved, fostered, blood and sweat. But almost never compare yours with the well-known, and under honest conditions.
As a result, the person describes the proposed method, draws beautiful flowcharts, demonstrates the program screens, and to the question of what, in fact, you are better than Method X, which also solves this problem, cannot answer anything.

17. Typical errors and problems - limitations of the proposed method.
It is always necessary to clearly formulate in which conditions the method is applicable and gives good results in comparison with analogs, and where it should be applied carefully or in combination with others or not at all.
Do not be afraid to say that the method can not something. It is necessary to be afraid of unreasonable statements for wider applicability than in reality.
Candidate work is always a search for your niche. We must be able, on the one hand, not to become impudent and not to claim that you invented perpetual motion, and on the other hand, to defend your modest gains to death.

18. Typical mistakes and problems - Inconsistency passport specialty.
One of the worst problems is because when found automatically means the cancellation of the thesis. It is recommended to study the passport of a specialty, indicate which items of the passport the work corresponds to, and even include certain words and formulations of these items in the goals and results.

19. Make friends with the Scientific Secretary of the Council.
For you, this is the first experience, he / she sees dozens of works every year and knows all the requirements. Listen carefully. Follow the generally accepted rules. See item 3 - this is a qualifying work. And, unfortunately or fortunately, not the best place to show your individuality.

20. Do not consider others as idiots.
When you come to the defense, then in your subject, with a considerable degree of probability, you will understand better than most, if not all members of the dissertation council. Just because it is a narrow niche that you personally dug up the last three years. And they may have completely different scientific interests.

But this does not mean at all that in a very short time they will not be able to assess what you have done and find all the shortcomings in what you have forgotten or failed to do. And here respectful attitude towards others will help a lot.

A separate part of this rule - do not consider your opponents to be idiots. Therefore, do not engage in plagiarism and copy-paste (including translations of foreign articles). There are almost no people who cannot use the Internet. Remained except those who are reluctant to grab your hand.

Also, do not consider other enemies - almost never, and no one has a purpose specifically to harm you, to overwhelm or catch. They just often play in another team - they need, for example, not that you defend yourself, but that your defense does not set up a council or department (in case the dissertation turns out to be so bad or scandalous that the punitive gaze of VAK turns in their direction) .

21. The most important thing to always remember the main rule is the salvation of the drowning hands of the drowning themselves. There are many good people in life who are capable and willing to help. But to flinty anyway you.

Bonus game. Answers to questions posed by beta testers (in the sense of the first readers)

1. And how, good work goes, what level?
I will try to give a total assessment, from the experience of not only my own advice, but also when I am present at the invitation in other cities / universities on the defenses. To be extremely honest, out of 10 candidates that I see, 3 really deserve a degree without question. Another 4 - so-so, are on the belly. And with the rest of the trouble. And these are the ones that successfully passed all the preliminary filters, i.e. Initially, the sample is even worse.
For doctors, the situation is better, but there and filtering is much more serious. Here I would say that out of 10 jobs, about 5 are worthy, another 4 are at the level of a very good candidate, and the last one is bad at best.

Unfortunately, while the Soviets are evaluated by the number of candidates, and not by the quality of work, the situation will not improve. And all the reforms of the RAS are not focused on improving the quality of science, but simply a classic redistribution of spheres of influence. But this is a completely different topic of discussion.

2. But with all your requirements, how would you rate your own dissertation?
Good question. Even now I am not ashamed of my doctoral thesis; everything that is written there has not only a theoretical basis, but has been repeatedly proved in practice. But the candidate, to be honest, I was damp, at C grade.

3. And what does a degree generally provide for further work and business?
Depends on the type of work. In my opinion,
- if it is necessary to work in state structures, then it is respected;
- If in educational - definitely needed, because many posts informally require its availability.
- If in business, it depends on the role you play in business. Say, if as a developer (from technical to CTO), then rather yes. If in a business role - probably not. Moreover, if Russia is still perceived normally, and quite often there are people who successfully combine scientific degrees and business acumen, then in the West, most often, a clear separation is implied. Therefore, there is a chance that we will have to break through the stereotype of a mad scientist who is able to come up with rocket science, but not very familiar with commercial and business realities - that is, contracting skills, team building and management, etc.

Therefore, I personally have several different business cards, with different beautiful words - under the situation.

4. Does she help the candidate for hiring?
Depends on work, of course. In my opinion, much more often - it interferes. Especially if the person in the process of obtaining a dissertation worked for several years exclusively at the institute / university.

For in this case (not always, but often) a very interesting combination is obtained - the newly created PhD has ambitions and a sense of superiority, and all his habits and work style are not very suitable for business and team development.

If you managed to defend your thesis on a topic that correlates with the main work, and ideally without leaving it, or if you work in high-tech industries - then why not.

5. Is it possible to earn money on science?
It is not necessary to go to science in order to earn money - its task is to move the world forward, discovering the unknown and coming up with something new. Another question is that a scientist should be able to do science and not die of hunger.

6. And what will happen to science in Russia now and on?
But this is the main question. And I deliberately do not want to give a detailed answer to it - for everyone interested must ask it himself and find his answer. My such - science in Russia has not died, will not die, and will be only stronger, but not thanks, but in spite of.

So I wish everyone (even if it was said on a completely different occasion) - a cold head, clean hands and a warm heart.