0
\$\begingroup\$

I'm having trouble understanding this comment below this question. Of course we all strive to ask only on-topic questions, I've looked at the help center and so far I can't understand what "asking for a component" really means, or how it's listed as off-topic.

  1. Comments are for clarifications, NOT answering the question. 2. This question is obviously asking for a component, which is off-topic. Trying to get around it by wording it as a "set" of components doesn't avoid that.

Not at all! This accuses me of some nefarious behavior personally, while being simultaneously unclear about what it is in the post that should be improved. A helpful comment would be just the opposite.

I've asked the user for clarification but so far no answer. It's a little ironic to follow up the first comment with:

@[username] Indeed, a lot of users ignore the rules on EE.SE. I have some plans on how to improve the situation, just haven't got around to that yet. It's much better on a lot of other SE-sites.

A good plan might start with including some helpful explanations to at least go along with the accusations, no? When @pipe "gets around" to improving the EE.SE site for us, let's hope helpful explanation is part of it.

So if it's possible, can someone help me understand what "asking for a component" means, how it's off topic, and what I can do to better explain that I'm looking for a solution to build this circuit as simply and small as possible. One answer proposed a micro-controller but another answer proposed using an FPGA which I think is ingenious. A comment suggested going after basic TTL shift registers. There may be something even more elegant out there.

tl;dr: Is this question "asking for a component", what does that really mean, is it off topic, where can I read further about asking for a component being off topic, and should this question be closed on that basis?

\$\endgroup\$
2
  • \$\begingroup\$ Maybe it's a coincidence, but both answers were down voted simultaneous to the close vote and down vote to the question, like every possible "no!"-looking button was pushed at the same time. \$\endgroup\$
    – uhoh
    Jul 15, 2017 at 12:29
  • 1
    \$\begingroup\$ 2 + years on :-). An option: Join this mailing list. Follow basic instructions. Ask questions on anything technical and most other things if desired. Expect to usually get useful responses. 2000+ members. Despite the name not at all only about PICS. (I'm an admin). PICList \$\endgroup\$
    – Russell McMahon Mod
    May 28, 2020 at 7:07

2 Answers 2

Reset to default
1
\$\begingroup\$

Let's explain what I mean when I have more room to do it. Expanding that long comment thread is not constructive.

"Comments are for clarifications, NOT answering the question."

If you click on the Add comment button you get a nice input field. It has an explanatory text:

Use comments to ask for more information or suggest improvements. Avoid answering questions in comments.

You can read more about do's and don'ts about comments here, and there's a lot of reasons for that:

  • Upvotes and downvotes are the safeguards that maintains the quality of Stack Exchange. Bad answers are downvoted. Comments can't be.
  • Comments can't be corrected if they have mistakes.
  • Moderators can delete comments at any time.
  • ...and more

Still, you see "helpful" people answering in comments (from your question):

Options I'd consider for this are PIC32 (no floating point, but DSP for doing fixed point in parallel), or ARM Cortex M4 with FPU e.g. STM32F4

I'm very very sure that a tinyAVR that got 2kB flash could fit the FPU library on it without any problems

"This question is obviously asking for a component, which is off-topic."

The core of your question is essentially this:

A Raspberry Pi would be overkill as would a high-end Arduino, so an Uno, or a Basic Stamp II with its FP coprocessor perhaps.

That exhausts my knowledge of the possibilities, but there must be other options out there. Is there a "minimal" chipset that I could put together that can do this?

Here I want to quote the (also good) answer by brhans:

Its pretty unclear to me what your actual question there is (which is a close-reason by itself).

You already seem to know that what you want to do can be done in a microcontroller (from your mention of Arduinos & Basic Stamp devices) - so if your question is 'which of these will do the job', then you're 'shopping' and the close-reason you got applies.

But hey, it's tricky. You're not looking for a chip. You're looking for a chipset? Well, you still want a recommendation for a product. I see that someone already pointed you to the rules on this.

Now for your comment about my "ironic" reply (not sure what's ironic with that):

A good plan might start with including some helpful explanations to at least go along with the accusations, no? When @pipe "gets around" to improving the EE.SE site for us, let's hope helpful explanation is part of it.

This is about answers in comments.

I don't think an explanation will be part of this, other than on meta, seeing how it's already explicit when you write a comment. My plan to improve it is by starting a discussion about the problem and offering a solution to the moderators and other users.

\$\endgroup\$
8
  • \$\begingroup\$ Thanks for your answer. I agree that answers as comments though well meant, are not the way SE works best and it should be avoided and discouraged. It's hard to tell who your addressing with item 1 especially since the previous two comments contain the word "question" prominently, but now I understand it was more of a general broadcast, even though the remainder of the comment accuses me directly of a trick or gaming the system. I hadn't imagined a single chip being able to do the calculation, and one solution I mentioned in the question \$\endgroup\$
    – uhoh
    Jul 15, 2017 at 17:42
  • \$\begingroup\$ involves a FP coprocesssor, so I know at least one solution might need more than one chip. I just thought plural would be the more general and more likely to be the case. But again I don't micro-controller is the best and certainly isn't the only solution. In order to get a good answer, I explain what I do know about, so that people can start from there. This is good SE practice, to make it clear what you do know, and what you don't know. The next sentence "That exhausts my knowledge of the possibilities, but there must be other options out there." \$\endgroup\$
    – uhoh
    Jul 15, 2017 at 17:44
  • \$\begingroup\$ I guess you are seeing "options" and reading "other brands of micro controllers" into it, but I'm just thinking "options" as in design option. The FPGA is a great example of another option I hadn't considered. The TTL shift register option, not so much. I think you are just perceiving sinister intent where none exists. \$\endgroup\$
    – uhoh
    Jul 15, 2017 at 17:44
  • \$\begingroup\$ Based on this further explanation and the original question, can you recommend how I might change the question so it looks more like a "design question" and "smells less like shopping"? I guess the obvious one that I can do now is to state more clearly I'm looking for design options, and I'll try to figure out a way to do that. \$\endgroup\$
    – uhoh
    Jul 15, 2017 at 17:53
  • 1
    \$\begingroup\$ You may have heard the phrase "Lead by example." Since you are an active user here, why not add helpful, positive suggestions for improvement when you call out "rule violations" or sub-optimal posts and comments? The really busy SE sites have this certain level of angst, and making sure that some calm, helpful guidance is included with the voice of authority (which is needed) probably has a positive, knock-on effect at least some of the time. It's an uphill battle, but it looks like you're not adverse to those. Why not give it a go here and recommend how I can improve my question? \$\endgroup\$
    – uhoh
    Jul 16, 2017 at 7:20
  • 2
    \$\begingroup\$ @uhoh I'd like to think that I do lead by example. I spend a lot of time on my questions, making sure that I understand what I need before asking. Half of my questions aren't even published, because I find the answer while researching the background. Nevertheless, I tried to improve your question. There should be an edit, that you can revert or change or do what you like with. \$\endgroup\$
    – pipe
    Jul 16, 2017 at 20:44
  • \$\begingroup\$ I see what you mean, that looks much cleaner, more concise and more structured than what I'd originally posted. I'll make sure to keep this in mind next time I post a question; thank you for your time and help! \$\endgroup\$
    – uhoh
    Jul 17, 2017 at 4:40
  • 1
    \$\begingroup\$ This question was bumped by Community. I've checked your answer again, and as long as you continue your accusation of trying to be tricky or work the system; "But hey, it's tricky. You're not looking for a chip. You're looking for a chipset?" I can't accept your answer. If you can answer helpfully and constructively, without simultaneously including a reiteration of your accusation, that would be great. But as long as you maintain your personal attack, it's not a good answer. The nuance of a chip vs a chipset is a red herring. I'm here to learn electronics, not gamesmanship. \$\endgroup\$
    – uhoh
    Aug 16, 2017 at 12:36
-1
\$\begingroup\$

Its pretty unclear to me what your actual question there is (which is a close-reason by itself).

You already seem to know that what you want to do can be done in a microcontroller (from your mention of Arduinos & Basic Stamp devices) - so if your question is 'which of these will do the job', then you're 'shopping' and the close-reason you got applies.

At some point you talk about 'chip sets' and it's unclear why you'd want to use a 'set' of chips to do something you already know can be done with a single micro, particularly when you also say somewhere that you want something small.

The overall impression I get from your question is that you have some idea of what you want to achieve, but don't know how you're going to get there and you're looking for recommendations - so simultaneously unclear/too-broad/shopping.

\$\endgroup\$
7
  • \$\begingroup\$ Well it was not unclear to several others, but yes "unclear to me" by five users can get a question closed. I haven't asked for a microcontroller at all (and a re-reading will make that clear) and the FPGA solution in one of the answers really looks like a better solution. The discrete TTL shift registers was another alternative but that isn't very practical. This is why I did not pre-specify a particular solution. I don't understand why you would like to narrow it down to only microcontrollers or a "shopping trip" for microcontrollers. That is not a fair representation of my question. \$\endgroup\$
    – uhoh
    Jul 15, 2017 at 14:06
  • \$\begingroup\$ "...but don't know how you're going to get there..." two helpful answers so far plus one less helpful, there may have been more if the question wasn't closed first thing Saturday morning in the US. About "set" sometimes a program will not fit fully within a chip and off-chip memory is necessary, and if you look at the question, the example of the Basic Stamp II and it's FP coprocessor would be another, so I don't know why you would say it is unclear, unless you haven't fully read the question. \$\endgroup\$
    – uhoh
    Jul 15, 2017 at 14:13
  • \$\begingroup\$ I didn't vote, and it wasn't closed for being unclear - it was closed for being 'shopping'. Even so, the 'official' close-reason simply reflects the majority of the close votes cast. If I had voted it would have been a mental dice-roll to pick one of 3 reasons with a bias towards a reason someone else hadn't already picked. \$\endgroup\$
    – brhans
    Jul 15, 2017 at 14:13
  • \$\begingroup\$ So considering that I'm looking for a solution with three different possible approaches, MCU, FPGA, ancient TTL, do you feel this is really a shopping question, asking for a part, or is it in fact asking for advice to help make a design choice from some very different possible designs? I really don't think this is a shopping question, but I think there is a bias to close questions that simply look a little different & "better be closed ASAP just to be on the safe side". Why not let it stay open for another 12 hours, let it go once around the planet, and see what other design options there are? \$\endgroup\$
    – uhoh
    Jul 15, 2017 at 14:18
  • 1
    \$\begingroup\$ ... and inspiring a page and a half if comments is also often a good indicator of an unclear/too-broad question. This is not a forum, this is not reddit. \$\endgroup\$
    – brhans
    Jul 15, 2017 at 14:19
  • 1
    \$\begingroup\$ No, that's not a proper way to think in SE. Shall we start closing questions based on the number of comments? Really? Is this a known metric for question quality? I've not seen this suggested anywhere, and it sounds like a really bad idea to me. Think of the possibilities! The way that could be gamed. Imagine if people started to vote to close questions because of the number of comments. One should reconsider your use of the word "inspire" as if to blame the OP for the behavior of a random sample of tens of thousands of users. This would be a dangerous road to go down. \$\endgroup\$
    – uhoh
    Jul 15, 2017 at 14:21
  • \$\begingroup\$ Except for the one user, the comments are essentially all helpful. Take a good look at them. Why would the presence of these helpful comments be in any way bad? Is the number of comments below my question really beyond \$3\sigma\$ for this site, excluding those by the user in question? \$\endgroup\$
    – uhoh
    Jul 15, 2017 at 14:30

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .