# Skimming through a math paper with a group

I am an M.Sc. student in mathematics. I was recently invited by some Ph.D. students and Post-Docs (a group of 5 people, including myself) to join their study group. We are reading a specific text, which should get us ready to read some more advanced work. This work is relevant to the research of some of the other people in the group, but my main goal is to experience this learning methodology with the added advantage of getting to know a few aesthetic results in mathematics.

We met around 6 times, and it is not what I'm used to from courses in the sense that we don't completely understand all the details. Nevertheless, we go on reading.

For example:

1. We encounter a definition and we can't find out its exact meaning. In this case we usually know of an example of a mathematical object satisfying this definition (because it's mentioned in the text) and we just try to see how the propositions in the text apply to the specific example.

2. A proof is given with very few details - we manage to fill in some of the gaps, but not all of them, so we just take an example again and simply accept the statement of the theorem so we can use it later.

3. An excercise is given in the text and we only solve part of it.

We allow ourselves to skip some details because this text is only meant to get us ready for some more advanced, but more specific, material. My question is how we can find out whether or not we are gaining anything, and how we can gain more given the fact that we are all busy and don't want to invest much more time in this specific reading (we have a 3 hours meeting every week).

I have a feeling that I'm "getting used" to some ideas and facts while reading this text (in contrast with "completely understanding"), but I'm not sure if I'm really gaining anything or whether it's just an illusion and I'm not sure how to test my gain of knowledge. The exercises in the text allow us to test our understanding of the details, but not of the general ideas.

EDIT: I will clarify what the question is, in response to aeismail's comment:

As Charles and Nunoxic say, the question of whether shallow-reading is useful is separate from the fact that we are studying in a group. So, the 2 separate questions are:

1. When reading without understanding all the details, how can I find out whether or not I'm gaining anything?

2. How can we make the process of studying in a group for 3 hours a week most efficient?

These 2 may have better been asked as 2 separate questions, but I did not notice that (in my mind they were related because the group study was the first time I encountered shallow-reading). To summarize the answers I got so far:

1. It is possible, for some people, to gain knowledge from shallow-reading and one way to test it is to see if you understand why each topic is being developed and why the text is structured the way it is.

2. When studying in a group, one should test his ability to work out the details himself after the group sessions.

I think the answer I got for (1) is excellent and the answer for (2) is somewhat lacking so far.

• Do you read together or is the reading an assignment and the 3 hours are meant for discussions?
– user107
Commented Mar 10, 2012 at 9:30
• @Nunoxic: Each of us goes over the planned reading once before the meeting, and then we read it together again. Sometimes we give an assignment to complete an exercise we started solving together or to find a definition in the literature. Anyway, the work we do out of our 3 hours meeting is kept rather minimal. Commented Mar 10, 2012 at 9:40
• user302099: Welcome to Academia.SE. At the moment, I have to vote to close this question, but only because I don't know what you're specifically asking for. Do you want tips on improving the process? Do you want to know what other approaches there are? You need to provide a specific question to be answered. Commented Mar 10, 2012 at 14:47
• @aeismail: I thought I was specific in the sentence: "My question is how we...". There I asked for 2 things: 1) a way to test whether or not this is working. 2) a way to gain more in this process given the time constraints. I will try to think how to make it more specific and if you can explain why this sentence does not make a specific question that would help too. Commented Mar 10, 2012 at 15:59
• @aeismail: Additionally, Charles answered my question partly by saying "it works for some people" and Nunoxic gave a way to test my understanding by asking for the reason something is being developed. Both of them separate the question of studying in a group from the question of shallow-reading, which is an important observation. Commented Mar 10, 2012 at 16:16

My answer may not be completely relevant but I still thought it was worth putting in. I have had a few such sessions and I realized a few things. Not all of them could be true in general and I might have been a bit extreme with what I treat as knowledge. Here goes:

Your level of understanding is directly proportional to a few things:

• Your ability to frame and ask grammatically correct sentences as an individual. This also encompasses communication of ideas/questions to experts of the field. For instance, suppose you are learning Linear Algebra in a group. You have a few gaps which are filled by others in the group. However, unless you can form sentences using Linear Algebra "handles", its unlikely you'll get far in research. Literature is way to dense in keywords. Unless you can talk in terms of Column Space, Rank and Eigenvectors (Rather than Linear Combo of all column or the vector which only scales) you are far from knowledgeable.

• Your ability to participate in discussions. Its not difficult to lag and be left behind in a group of impatient, overachieving academics. Further, it can be a bit demotivating at times when the senior students who read the same content (owing to their heightened intuition) seem to grasp more. As junior students, it is often necessary to substitute the lack of intuition by more work.

• Your ability to appreciate the nuances of the field. I cannot stress this enough from my experience. If you cannot appreciate the subtleties yourself, you didn't learn much. It doesn't take much to learn Elementary Fluid Mechanics per se. But, IMO, you actually "learn" when you go OMG when you see the transport equation and play with it till you are satisfied. An extension of this is motivation. When some concept is developed by an author, he doesn't write in random order. One of the most important aspects is to be able to understand why something is being developed. For instance, most Aerodynamics books start off with Euler Angles and then move on to Quaternions. Its simple to understand what Quaternions do and how to solve equations based on them. However, unless you know that they are used to prevent gimbal locks, there is no point of knowing about them.

• The usual: Your ability to write alternate proofs, write codes (if possible), interpret the results of these codes and the usual other-than-textbook stuff.

One ending comment : Your level of knowledge is a function of how good you are in the group and how good you are without them. If you are able to develop proofs in those 3 hours together but aren't able to get started on your own later, you need to investigate whats wrong.

If you want to know how good these group sessions are, find out how good you are getting at that field as an individual

There are two positive points of group reading and shallow reading that I feel were not stressed enough by the existing answers: acclimatization and motivation.

By acclimatization I mean gradually learning as to what is interesting and what to pay attention to in a particular field. When you read by yourself (shallow or deep) you only have your own knowledge and intuition to guide you. As a junior student, you might not know what is considered interesting in a given field. If you read by yourself, it is very difficult to spot what is new and what is interesting, especially for some of the less-than-stellar papers that often comprise the bulk of your reading (sure, if you only read the greats they might make things clear, but usually if you can just stick to the greats you are probably reading something old). This was mentioned in @Nunoxic's answer, by attending the group are you learning how to ask questions? How to use the lingo of the field?

By motivation I mean having the extra commitment that helps you to read more. As graduate students we are highly self-motivated, but that doesn't mean we can't benefit from external motivation. By committing to a group, you force yourself to keep up with your reading and work. Some people can replace group meetings by an equal (or even greater) amount of individual work, but I doubt those are the norm. I always schedule a certain number of group meetings and projects to keep myself committed. This makes sure that even on slow weeks where every proof I try fails, and every idea I have is derivative, I still have something to motivate me: the commitment to the group. Further, when I am the junior student in a group, I usually feel the extra pressure to not "be dumb" and tend to invest more time and effort into understanding the material better. The pressure helps me, but it is definitely not for everyone and you should see what works best for you.

Of course, this shouldn't be taken to the extreme. If it is clear that you can accomplish more in those 3 hours (and associated preparation time) by yourself, then you should stop attending.

If I understand your question correctly, you're asking whether skimming through papers and getting only a "shallow" understanding can be useful? Well, I think the answer depends very much on you, and actually finding it out is an important step for you to understand how you process information.

I don't think there is a global best way to read papers. Some people need to completely a paper, to understand fully, do all the proofs, and then, somehow, they don't need to go back to this paper. I've seen a friend spending several days on the same 10-pages paper, until every tiny detail was clear.

On the contrary, other people (such as myself) prefer to have a global view of many papers, and to process several papers at the same time, which usually implies a lot of go-back-and-forth, and which also means to accept not to understand everything (although of course, sometimes it's needed to go into the details of a proof in order to keep the process going on).

So I would go with Nunoxic's advice and try to find out how good you are getting as an individual (for instance, try to read another related paper on your own, and see how much you can get). But if you're not getting better, it does not necessarily mean that this group is not working, it could also be because you're not working in this way.