Rambling about beginner's participation and how to stimulate it

Question

I started to ramble about this a while ago, but this question asked today fired up some structured thinking that I'd want to share and discuss.

The question in question is a very simple one, and when I saw it, I thought -"I'll leave it for someone less experienced"- and it happened. Two users answered with a method already sketched on a previous Toad's comment. So far so good, as a comment is not an answer.

The way they solved it is clearly inferior to the third answer posted a bit later by @RM himself, who resorted to HoldForm[] instead of joining strings. In this scenario, @RM's answer is valuable and served a clear propose: At least three users have learned a good lesson.

But sometimes, it happens the other way around. We (more experienced users) are answering simple questions in the obvious way, without more consideration than a few lines of quick code that solves the problem, but:

• inhibits beginner's answering activities; and
• doesn't teach more than a basic technique (which could be valuable too)

Both points are a real pity, as beginners are losing an opportunity to learn.

I'd like the community to think about a way to palliate these effects.

I don't have a good proposal, but perhaps (and it's far from perfect) we could implement a tag like entry-level (to be most commonly applied as a retag) to communicate to experienced users that the simple and obvious "HowTo" is too basic and not the expected answer from them.

It has its drawbacks, though. Answers could be delayed, different personal opinions are likely to arise on the classification, and bad answers may pile up before a good one could be written. Not to mention the rep loss from restraining to answer.

I hear you.

Answer 1

I am new MMa SE participant with intermediate/strong, but not expert, knowledge of MMA. So I experience this issue from the inside. My insight and distillation of the points above is:

1. An "entry-level" tag not a good idea because: entry-level is in the eye of the beholder, questions can be answered at different levels and each level will be useful and explanatory to different users, and it is pejorative.
2. Each subsequent answer could show either a different, or "deeper" answer, and explain how the previous answers were but as Newtonian physics to my new Relatitivity physics. That would help level 3 users move to level 4, etc. working their way on up to level 9 ("The Might Thor" - aka rm, Mr. Wizard, belisarius, J.M., Sjoerd, Leonid, etc.).
3. Wikepedia observes that a common criticism of SE is that "Questions are commonly answered fast without focusing on quality or correctness, to boost reputation." I don't really think that's a big problem, but I noticed that even I, a new level 3 user, was trying to answer questions I thought I could before anyone else (until I read the Wikepedia comment and joined SE Anonymous). My point is that the "each subsequent answer goes deeper" approach has multiple merits.
4. Part of the responsibility of the higher level answers could be to clean up the errors in the previous answers, within reason.
5. There is no "structural" solution to this issue. It seems to me that SE can be thought of as an emerging social or cultural form, made possible by the internet, etc., and level nines should also see to it that SE develops along with it.
6. This doesn't sound like it would be as much fun as trying to get the best answer first.
7. Who is level 10?

Answer 2

I don’t favour tagging things as entry-level, not just because it could be seen pejoratively, but also because so many questions on this site only seem easy to experienced users. We all know that Mathematica has a steep learning curve, and that it is such a rich and complex environment that all sorts of gems can lie hiding in the documentation, apparently in plain sight but in reality hard for newer users to see.^[1] Something might be “easy”, but is really in the realm of “knowing where to hit it”.

I think the way forward is for more experienced users to try to avoid, where possible, the temptation to play FGITW on easy questions, and to make sure we are explaining our work thoroughly, including why we made the choices we made of the multiple ways to do things in Mathematica.

_{[1] For example, it was only today that I learned all the different ways to specify Blend and about the DendrogramPlot function.}

Answer 3

I guess the best way to resolve this issue is that you always put as much thought as possible into an answer. This on one hand gives beginners time to give the obvious answers, on the other hand increases the probability of really great answers.

Of course, the reputation system together with the implicit "first post bonus" tempts one to rush out a simple answer (in the best case with the intent to improve on it afterwards). But then, my highest rated answer on the whole SE network (which additionally brought me several badges, including my only gold badge on the entire SE network) was not a first post (actually it also wasn't an answer which I would have expected to get that many upvotes).

Answer 4

I'm now a bit jealous, annoyed with myself, but mostly annoyed with you. ;-P Why? I passed on answering this question first (I had working code, though not quite as nice a result) because I figured that it was a good question for newer users to answer. I knew it would be popular but I had no idea it would become the most popular question ever on the site, and now I'll be haunted by the fact that the first answer could have been mine.

For that reason I say: answer any question you can as soon as you can. Down with the entry-level tag!

_{This is intended to be humorous but it also holds a shadow of truth.}

Answer 5

I upvoted the question, and will consider myself "entry-level" for the, er, next couple of weeks still.

On Electronics.SE tags like homework and beginner are deprecated because we think tags should give an indication about the question's topic, not be meta. Of course we have entry-level questions on EE.SE as well, and it will be an issue on many StackExchange sites. I've been thinking about a user skill level indication, next to tagging. Think of the cook hats at the top of a cookbook's recipes: 1 cook's hat is novice, 5 hats is advanced.

If this could be implemented SE-wide users can decide what level they expect the answers to be, and answerers can adapt their answer to that level, or decline to answer.