Lengthy "blog style" attempts to answer - how to treat them?

Question

As prompted by Russell to open a discussion on Meta :)

I don't doubt they are done in good faith, and I applaud any effort to learn a new topic, but how should answers that seem to be an attempt by someone to develop a blog style post about a subject? The posts are marked as "to be continued", and are a jumble of screenshots from documents with passages highlighted in various colours. There's often an "experiment" to show a circuit at the end. There are long lists of "references" to sources of dubious provenance.

The problem is, is doesn't seem that the user doesn't seem to know a great deal about the subject matter. The continuous updating means that correct and concise answers get shunted down the page.

For example, the answer in this question takes, as an estimate, over 1/3 of the total page length. That total length includes the banner, the original question, and 4 answers. Is this the quality of content that is desired?

Similar answers: 1, 2.

The answers do sometimes contain a kernel of truth, but they are also often tangentially related (to be generous). Do we want answers to contain "tl;dr" sections, and videos of the answerer's own anecdotal experiments? The answerer doesn't seem interested in changing due to comments on numerous posts by numerous people, and flagging them as in need of moderator's attention also seems to be ineffective.

A related question on EE Meta.

Answer 1

Bad is an opinion. If its a bad answer, downvote it and move on. Others may think that same answer is good and upvote it.

Bad is not a reason to have an answer deleted. Nor should someone need to conform to your specific type of writing style.

Answer 2

I watched a TV programme on the Loch Ness Monster last night. I was titled something like

Loch Ness monster; new discoveries

Right at the start they said they were going to water-sample the Loch at various depths and do a seriously extensive DNA test on all the stuff they found. I thought "great" this will be good but, they laboured the next hour explaining how various pictures and sightings dating back to the dawn of time were hoaxes or just plain daftness. They interviewed everybody and their dogs (of course they were filling time).

Right at the end (in the last 2 minutes) they announced that no dinosaur/plesiosaur DNA was found (which is what I expected else I'd have heard about it on the news). But wading through all the old pointless crap was just a waste of time for me.

Sorry to disappoint; no conclusion on the question raised by awjlogan other than what you can read between the lines.

Answer 3

A similar discussion occurred about 6 months ago. This Q&A is (arguably) different enough to merit a life of its own.

I'm in the not-so-usual position of being a moderator but not disapproving of the identified actions as much as some others do. I'm somewhat torn between the 'Muppets bald eagle' style stiff and proper site policing stance and the feeling that a degree of leaven is not a totally bad thing if not allowed to get too out of control.

A major factor for me is that as long as only a small percentage of site members post in this manner then the 'noise' generated is minimal, and such flights of somewhat free-form association can and indeed do valuably stimulate and inform some people. Yes, that's not the way that the site arguably is notionally meant to operate, but "quality answers" are not just ones that 'maintain a LASER like focus on a single point' (as the guidelines suggest. Most answers should be like that.

Is this heresy?
Of course it is! :-).

Does it damage the site significantly?
I think that it doesn't.

Does it provide significant value for some?
I think it does.

Will the above annoy some people significantly?
You bet!

Comments welcome.
Downvotes will be forgiven :-)

Answer 4

SE doesn't really have a limit on long answers the maximum length is 30000 characters.

As far as moderation is concerned:

For answers, any post that is not an answer (should be a comment, doesn't answer the question, etc.) should be deleted. Answers that are wrong or that dispense poor advice should be downvoted, not deleted.

From: How does deleting work? What can cause a post to be deleted, and what does that actually mean? What are the criteria for deletion?

There are some other reasons for deleting answers on other sites, it is my opinion that on EE.SE we should stick to the policy that answers should not be deleted unless they are not answers.

I see not enough participation in the voting system by some users, if you are in that camp then please use the voting system. At the time of this post there were no downvotes (except mine).

SE does care about too many edits, however and these users are editing too much, and there are things the moderation team can do about that. Some of these users have been warned against making too many edits before (this creates a lot of work for the review ques). If your going to flag a post that is wrong in your view, then this action should also be followed by use of the voting system.

Answer 5

TLDR! :)

1. On my answering style

Those who think my answer is a "blog" post might like to read the story below.

Some three years ago a newbie asked for an advice on which 9-DOF MEMS sensor to pick. So I wrote a long research report (not a blog!) for his reference.

Adafruit 9-DOF or other accelerometer/magnetometer/gyroscope sensor for Raspberry PI 2/3 with Windows IoT - Viewed 1k times

The OP then asked me the following question:

Is this a blog or what? Never seen an answer so in-depth – ccalboni Apr 18 '19 at 12:30

My answer:

Well, it is an answer.

A blog is usually a dated diary, with personal opinions. My answer is rather objective, with comprehensive research notes (that is why long and deep), and why I suggest this not that.

When comparing two sensors eg, I list google search counts to convince which is more popular.

And I am digging deep, because I am looking far. I am aware I am writing too long, otherwise I would tell you that the sensor MPU9050 I am recommending is already obsolete! :) invensense.com/wp-content/uploads/2018/10/… – tlfong01 Apr 19 '19 at 7:16

---

2. On my "long" answers

Those who think my answers are "too long", "not to the point", might like to check out my criteria of length of answer below.

Make your answer as short as possible, but as long as required.

Let me give an example of my short answers.

Raspberry Pi 3b+ do not detect DS18b20 on 1 Wire - Asked 2 years ago, viewed 2k times

The OP complains that he has spent hours without luck on the DS18B20 temperature sense driver. I repeated his experiment and concluded that his driver might be out of date.

Now I could give the one of following three answers:

Version 1.

Your driver is too old, try the new newest.

Version 2.

My configuration works: stretch 19apr08, python 3.5.3, W1ThermSensor V0.3.0.

Version 3.

(a) My configuration works (stretch dated 19apr08, python 3.5.3, W1ThermSensor V0.3.0), 

(b) Your 1-Wire cable might be too long, see my photo that works.

(c) You might like to try my debugged python program, with sample output.

I judged that the OP is experienced. So Version 1 should be long enough for him.

However, I usually do not answer for only the OP, but for all the long tail of future newbie readers. So I additionally gave a demo program, written according to the StackOverflow's guideline for the questioners:　

MCVE (Minimal, Comprehensive, Verifiable, Example).

Actually my guideline is one longer: ALSO GIVE A SAMPLE OUTPUT, so it is MCVES! :)

---

3. More on why I write long, and not short answers

So the above Q&A on DT driver shows that I can write vert short, to the point answer for the expert OP, or longer answer for regular readers, or even longer answers for the newbies who do't know what is going on, or don't know what they don know. Actually my answer mentioned above was followed by a couple of follow up answers, on the same laser focused topic of 1-W DS18B20, on the FAQ problem why multiple devices were not detected, problems encountered on different packages etc. I even went one step further on the DS18B20 extension module with buffers/extenders for extremely long wires, for industrial grade building automation application engineers.

---

Let me show you another recent example on how I can either write very short or very long answers. This is the link to the Q&A:

Is it necessary to use a pull down resistor with this BTS7960? - Asked 4 days ago Active yesterday Viewed 135 times

Actually I could give a very short, a bit rude answer as:

No, already built in.

Though I did give a more friendly, helpful answer as below:

#sanier, I checked that your bts7960 input pins should have already pulled down: https://i.imgur.com/1PuTkGX.jpg. Have a great weekend. Cheers. – tlfong01 21 hours ago

The point is that I need to do "research" (not blog) for over 10 hours to investigate the OP problem. I then wrote long paragraphs on my research, for the benefit of a long tail of future newbies not frying any more their expensive motor driver modules because of using possibly a problematic wiring/control schemes, which I think is a "huge EE circuit design mistake which I have not seen before and if not warned to newbies, would damage any EE forum's reputation, :) .

To summarize, instead of a very short, simple, up to the point, one word answer: "NO", I wrote a long answer on my research which the mods thought was a blog and locked it: :)

---

4. Another example of my as short as possible answers

Actually I am a huge fan of Occam (Note 1) who say the following:

Make things as simple as possible, but not simpler

Note 1:

Actually my favourite language to teach programming is also called The Occum Programming Language, using UK Open University's teaching material.

---

Now my story: This morning I answered a ADS1115 ADC question in my favourite "as short as possible" style. The question's link is give below.

How would I go about to troubleshoot the ADS1115's inconsistent readings? Asked today Active today Viewed 22 times

The OP presents his problem very concisely (I would give him 9 out of 10 marks! :)) I read his question twice and skimmed the datasheet and found a possible cause of trouble. So I gave a three sentences long answer, repeated verbatim below:

Now if the problem only appears in the lower mV range, I think it is the FSR（full scale range）not set properly. Please see my datasheet summary on full scale range setting etc in the appendices below.

I know my three sentences answer is already long enough for the OP, but I did length it a bit as I think required: (1) I added a reference to make sure we are referring to the same datasheet. (2) I also gave an appendix to the critical section of the datasheet.

The reference part of course avoids misunderstanding cause by discrepancy of terms of reference. The appendix part is to save the time of the OP and the long tail of future newbie readers going through the boringly long datasheet. In other words, the two required parts making my answer longer are for human efficiency and effectiveness.

By the way, I will later comment on the importance of reference and appendix sections in an answer.

/ to continue, ...