1
\$\begingroup\$

It looks like an XY question, but it is not !

It looks like a shopping question, but it is not!

It seems poorly specified, but it is in a certain sense!

It is certainly needed, but it has no chance in EE.SE!

I've found a name to this kind of questions: "technology/component inquiry".

Let me explain. It happens quite often that you "suspect" the existence of some technology, device, IC etc. that could perform a certain task, often a generic one. The trigger may be an application you are working about, or simply personal thoughts. In order to improve your work and knowledge, it is necessary to know if certain technologies, IC or other components exist; you need keywords, general guidelines, but you have no chance in EE.SE. The reasons are listed above. I've tried few times to ask such questions, and this is always the same story: the question immediately irritates the moderators, who found it "insufficiently specified", "shopping", "probably XY". They ask you to precise your application, but you cannot. Even if you invent an application to satisfy their curiosity, this does not help. Inevitably, the question get closed, or, at the best, downvoted.

Are the moderators rude or bad? certainly not. The difference between this sort of questions and shopping, XY, or unspecified question is fine, and you cannot demand from a moderator to have the subtlety of a jurist. Even if he has, he has not to take such nuances into account, until, at least, they have been dealt by the community. This is the aim of this post.

I would like to address 3 topics: first, what is a "technology/component inquiry" question and how should it be written. Second, how the moderators may (should) react to questions of this kind, whenever this not sufficiently obvious, or they are not well written. Finally, what could be the nature or the content of the answers. After this is done, I'd like to know your opinion, and if you think there is a place to this sort of questions in EE.SE.

1. Definition:

(a) A technology/component inquiry question is a question where the OP inquires about IC, devices etc. that would perform a certain task he needs or has imagined.

(b) The task may be generic or broad, but it should be clearly described, in the sense that what the OP has in mind should be clear to the readers (in this sense, the question is "well specified").

(c) The inquiry should involve one and only one task (possibly broad or generic).

(d) The OP may advantageously add details to restrict the question to certain limits, but this is not mandatory.

(e) The practical application that needs the technology, if any, should not be mentioned in the question: if it is, the question falls in the realm of ordinary questions and should be addressed as such. Nevertheless, the OP may mention possible applications as a motivation to his inquiry.

(f) to make things clear, the question should be tagged "technology/component inquiry question" (or another name).

2. Now, assume a question looks like a T/C inquiry question but fails to fulfill one or several of the rules above. How may moderators react?

First of all, assume the question is not suitably tagged. It suffices to point it out to the OP:

"If this question is a technology/component inquiry question (link to the rules), please tag it as such".

If the question has been correctly tagged, but it does not follow or partially follows the rules, the moderator may simply pinpoint that, possibly providing a link to the rules.

3. how could the answers look like?

No rule here, only some guidelines:

(a) the answer need not be more specific than the question. In other words, it suffices to answer at the same degree of generality as the question.

(b) If the answer could be interpreted like a "shopping" answer, it may be a good idea to avoid mentioning explicitly the name of the components. Rather, provide keywords that will help the OP to find the particular components by himself. This is often all what is needed.

(c) The answer need not cover the whole range of the required task.

(d) The answer may be very short (keywords), or include some general guidelines or insight about the component/technology etc.

(e) The answer may be "there is no component/technology etc that performs this task". Possibly "because" or "but there is" may follow.

(f) The answer may be more general than the question, that is, it may mention a component that performs a more general task than specified by the user.

Finally, let me give two paradigmatic examples, and one real example.

Ex. 1 :

Question: does there exist IPs that allow, in some general way, to perform electrical feedback ?

Possible answer: "Operational amplifiers" (often abbreviated oamp) form a class of IP that work on the principle of feedback. In fact, an oamp can be seen as a generic feedback machine.

ex: 2

Question: Are there devices/technology that can help to stabilize the voltage of a power supply to a fixed, selected value?

Possible answer 1: So called "voltage regulators" are designed to do that. There are several kind of voltage regulators, "linear regulators", among which "low drop" regulators are a subclass. They work by taking the difference between the input and output voltages, and just burning it up as waste heat. There are also "switching regulators", that work by taking small chunks of energy, bit by bit, from the input voltage source, and moving them to the output. This is accomplished with the help of an electrical switch and a controller that regulates the rate at which energy is transferred to the output (hence the term "switching regulator"). Often a switching regulator is mixed with a "step up/step down convertor".

Possible answer 2: Sure, you can stabilize quite easily the voltage to a fixed value with an oamp, a voltage reference and a transistor. Here is an example (Schematic).

Possible answer 3: If you are dealing with a high voltage, you need more careful design. Link to "the art of electronics", chap. 9, sec. etc.

Of course, the low level examples above are paradigmatic. Here is a more real example of question:

Question: In many applications, what you need is just extracting one frequency from a signal (that is, to know its amplitude). I was dreaming about an IC that allows selecting a frequency of interest, perhaps also, some passing band, and to output the amplitude of the frequency component, exactly like an LM317 allows selecting a regulation voltage. Of course, such IC would be limited in their range of frequencies. Applications include: audio, radio frequency selection, television, frequency response analyser, and even simply eliminating noise from a signal, without having to design a cumbersome sharp filter. Taking this question one step away, I ask about technologies/IC that allows frequency selection in some range of frequencies. The answer may focus on bandpass filter technologies/IC that allows frequency selection, possible "heterodyning" IC, other.

That's all. I would enjoy to know what you think about that.

EDIT: just a final point that is worth to be underlined. T/C inquiry questions not only offer a new desired category, but may well be able to simplify the task of the moderators for usual questions as well: Indeed, when a question falls into this realm (and this is often the case), the moderator needs only to redirect the OP, which avoids useless discussion and frustrations.

| |
\$\endgroup\$
  • 1
    \$\begingroup\$ Keep in mind that there are two problems with the kind of questions that you mention. 1) most people phrase these questions as shopping questions. I'm fine with some shopping questions, Jeff Atwood is not meta.stackexchange.com/questions/126180/… \$\endgroup\$ – Voltage Spike Jul 13 at 16:41
  • 1
    \$\begingroup\$ 2) The second problem is many of these questions generate discussion (discussion isn't good, a user seeking an answer should not have to read through comments to understand an answer to the question, and we shouldn't have to ask many questions to clarify the OP's question). Many times questions are closed because there isn't a good way to answer them and the OP has not phrased a good question. \$\endgroup\$ – Voltage Spike Jul 13 at 16:44
  • \$\begingroup\$ Thank you for your comments Voltage Spike. Regarding 1), I think this is handled by the rules I've written above. Regarding 2), isn't it the case for all the questions? Personally, in almost all my (accepted) questions, I ask details in comments, sometimes there is a long chain of questions/answers in comments (and this is not because I like discussion). \$\endgroup\$ – MikeTeX Jul 13 at 17:33
  • 2
    \$\begingroup\$ As for #2: Not by the design of the site, comments are needed but not wanted, ideally a good question has zero clarification needed for answering. \$\endgroup\$ – Voltage Spike Jul 16 at 18:17
  • \$\begingroup\$ I'd suggest you re-phrase your example questions #1 & #2 a bit. As currently written, either of them could be fully and completely answered with simply "Yes". \$\endgroup\$ – brhans Jul 16 at 23:34
  • \$\begingroup\$ @Voltage Spike. I don't understand what you mean. Where do we need comments in a T/C inquiry question? Do you mean that these questions will generate more comments than usual questions? I see no reason to suspect that, but this can be checked by a "pilot". Just try this idea, and if it appears the amount of comments turns to be dramatic, remove this option. Alternately, comments for this kind of questions may be disabled, but I don't think this is a good idea until we have tried this option. \$\endgroup\$ – MikeTeX Jul 19 at 7:58
  • \$\begingroup\$ @brhans. Actually, this is also the case for many "usual" questions. I think it is clear that T/C inquiry questions, like any other question, needs some minimal elaboration, and answer like yes/no could be downvoted like any other answer. \$\endgroup\$ – MikeTeX Jul 19 at 8:14
  • \$\begingroup\$ I think the point I'm really getting at is that the majority of "this is not a shopping question, it's really a does this technology/component/etc exist" are simply re-wordings from "tell me what to buy for xyz" into does there exist a component which lets me xyz". I'm not saying that your hypothetical T/C inquiry doesn't exist - I'm saying that from experience here, the vast majority of questions in that format are poorly disguised off-topic shopping questions. \$\endgroup\$ – brhans Jul 19 at 14:06
  • \$\begingroup\$ @brans. Yes, there may be persons that use perversely this site. But from my experience, it's not constructive to seek systematically the "bad side" into humans and to suspect them a priori. This idea has even not been tried yet. How could we assert "a priori" that it will be used "perversely" by persons that try to disguise their shopping question? Moreover, there are not so much persons that answer to questions here, and it is likely that they will want to respect the rules of "no shopping answer". So, what does it matter if the OP has tried to disguise a shopping question or not? \$\endgroup\$ – MikeTeX Jul 19 at 15:21
  • \$\begingroup\$ A better question has an expectation defined in some measureable way (Spec) or a specfic function or Inputs and Outputs with an unknown process. Engineering is often about Make or Buy decisions and not knowing if something exists, or if so is it worthwhile or the cost/benefit factors which then leads the OP in a better focused direction for an "Engineered solution" \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Aug 6 at 17:11
  • \$\begingroup\$ Well, that's more or less the opinion of Chriss and others (that is, every question should be application directed). I disagree with that. Once and for all, we need to know what technologies and products exist. I think this is a very natural and justifiable need. \$\endgroup\$ – MikeTeX Aug 6 at 18:34
  • \$\begingroup\$ Whether or not is a natural and/or justifiable need is not the deciding factor for whether a type of question is on-topic on a SE site. Any answers to your hypothetical "what technologies and products exist" are going to become outdated - SE answers shouldn't need to be revised periodically to stay current. The kind of thing you're looking for is just not the kind of thing SE is intended to provide. \$\endgroup\$ – brhans Aug 10 at 16:31
  • \$\begingroup\$ I think you have missed the point: we are not dealing with SHOPPING QUESTIONS, so I see no reason for these questions to become outdated in the next 200 years. Even the old electron tubes are still used for some applications these days. \$\endgroup\$ – MikeTeX Aug 14 at 13:54
  • \$\begingroup\$ Most good analog engineers are also “component engineers”: which demands a strong understanding of stating specs and requirements where there are deviations that need assistance . perhaps you can write up a template for future users to follow for EE.SE. With a variety of question formats with essential items like links to datasheets and extraction of details etc , you can make a valuable contribution here with community help. \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Aug 28 at 3:29
0
\$\begingroup\$

It's been suggested in the past that the chat room might be suitable for things that don't fit as questions under the rules.

It's also important to remember that Stack Exchange set out to be different from the rest of the Internet, by efficiently handling concise, specific problems only - it is, one might say, engineered to be efficient for a very narrow purpose, by being careful to not even try to fulfill others.

Regardless if one feels that a particular type of question should or should not fit within that, the reality is that there's still the entire rest of the Internet on which to seek information unconstrained by SE rules.

Finally it's important to realize that a lot of the value of the hard stack exchange sites in the questions which have already been asked. One isn't allowed to ask a shopping question here, or for a code example on Stackoverflow, but there are plenty of chips mentioned here, and code snippets there which have come up implicitly in the process of allowed questions. Often with a little searching you can find what you want to know without posting a question at all.

| |
\$\endgroup\$
  • \$\begingroup\$ So, you believe that this forum is only for "narrow" questions. I see no reason to restrict it to that. A T/C inquiry question as defined here may be very useful, not only to know that something exists, but also to feed the imagination of engineers. I understand this opinion though, so I've not down voted it. Regarding the rest of the internet, that's not as easy as you think. Most of the pro are here. \$\endgroup\$ – MikeTeX Aug 6 at 18:29
  • 1
    \$\begingroup\$ What you are missing is the narrowness is the fundamental design goal of stack exchange, it does not try to do everything, but rather to do only what its model does well. \$\endgroup\$ – Chris Stratton Aug 7 at 2:39
  • \$\begingroup\$ Narrowness means that this is a question/answer site, with precise questions and precise answers. That does not means that it is application oriented. Also, the notion of "precision" is not opposed to "generality", despite this is admittedly more difficult make it practical. This was the aim of this post. \$\endgroup\$ – MikeTeX Aug 7 at 15:49

You must log in to answer this question.

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