6
\$\begingroup\$

I love StackExchange for software questions and feel it has a very high level of professional users. It would be great if this site could attract the professional EE's to the same degree. Recently I started answering a few questions to see how it works, but was struck by how strange the "reward" system works.

Examples:

The first was upvoted a lot and awarded "Nice Answer", but was really a cheesy answer that did not actually help the poor soul that wanted to find a cheaper high temperature microcontroller. The OP has rightfully not accepted the answer, but I got a lot of reputation points.

The second was a very detailed answer with far more work put into the answer than the OP put into his pre-question research. This has not will probably not be upvoted much or rewarded at all. The answer was accepted very fast, but I did not get a lot of reputation points.

I think we would need to see more of the second type answer and less of the first type in order to attract more professionals to this site.

Do you agree?

What changes in the dynamics, mechanisms, rewarding etc. could be made that would encourage more technically solid answers?

\$\endgroup\$
  • \$\begingroup\$ Do we have too few professionals? \$\endgroup\$ – user17592 May 31 '13 at 10:14
  • \$\begingroup\$ That is another good question - I think so, but it's no "my" site. \$\endgroup\$ – Rolf Ostergaard May 31 '13 at 11:34
  • 3
    \$\begingroup\$ I'd be more willing to consider your question if both of the linked answers you point to weren't yours. \$\endgroup\$ – Connor Wolf Jun 1 '13 at 2:53
  • 2
    \$\begingroup\$ I think that one of our bigger problems is that professionals don't participate enough in the work of this site in the "professional" sense. Looking at the top users by reputation, I see many professionals who asked few, if any, questions. This leads me to conclude that they must have another source of knowledge which they use to generate solutions to more complicated problems that show up during their work. \$\endgroup\$ – AndrejaKo Jun 1 '13 at 3:19
  • 1
    \$\begingroup\$ @user21976 Why don't you set an example for others and next time you come across some complicated problem in your own work, post it as a question here? Google would index it and it could potentially generate some more interest from people doing work at similar complexity level. In the end, even if you answer your own question, you'd still participate in setting trends of this site. For example here's a bit more complicated question with some research done by OP. It definitely did get some attention and a number of up-votes. \$\endgroup\$ – AndrejaKo Jun 1 '13 at 3:25
  • 1
    \$\begingroup\$ @AndrejaKo I could do that. But that would only be one drop in the ocean. Something more fundamental needs to change to pull in more professionals - like what you see on SI-LIST as an example. \$\endgroup\$ – Rolf Ostergaard Jun 1 '13 at 6:26
  • \$\begingroup\$ @Rolf Ostergaard I understand that, but on the other hand someone has to be first. Unfortunately, I don't have a better suggestion. I experienced this problem on several SE sites and I feel that Electrical Engineering suffering the least from it. That still doesn't mean that things are good here. \$\endgroup\$ – AndrejaKo Jun 1 '13 at 9:18
  • \$\begingroup\$ @Andre: When I encounter a problem in real work, I first attack it the best way, by reading the datasheet, taking measurements, or whatever. If it's detailed, I usually can't talk about it publicly due to NDA anyway. Besides, the usual methods usually work. I occasionally have very detailed problems with PICs, which I ask my FAE after doing my homework. If a problem gets to the level of asking him, he usually has to ask the engineers in the factory. It's not the kind of thing this site would be suited to, and I don't want to show Microchip's or my dirty laundry in public anyway. \$\endgroup\$ – Olin Lathrop Jun 1 '13 at 13:53
  • 1
    \$\begingroup\$ Perhaps split the site in two like math.stackexchange.com and mathoverflow.net - the latter is for professional mathematicians and the former is for all comers. The problem I have with this model, however, is that I am reasonably proficient with digital design but my knowledge of analogue (especially discrete design such as BJTs) is ... amateurish. \$\endgroup\$ – Damien Jun 2 '13 at 10:44
8
\$\begingroup\$

First, I think you just have to accept that on this site, a +10 is a very high response to a good answer. For an answer in a specialized technical area, +5 is an excellent score. On other SE sites you might see +100 answers regularly, but they're extremely rare on on EE.SE. So part of the answer is just to get re-calibrated on the kind of voting you should expect on this site.

As for whether the site could be (or should be) re-focused towards professional users, I'll reiterate what Olin said in a comment. Experienced engineers have mostly worked out how to read datasheets and app notes to answer their technical questions. If they have an question that can't be answered by those resources, calling an applications engineer is likely to be more fruitful than asking a random selection of people on the internet. Furthermore, the details of our work is usually confidential, either covered by an NDA with a client, or by the expectation of confidentiality from our employers. We can usually really only discuss our technical problems in the most general terms.

That still leaves a lot of opportunity for early-career professionals to ask useful questions, but it does mean most of the questions experienced professionals have won't be appropriate to ask here.

\$\endgroup\$
7
\$\begingroup\$

Those are only two examples, and I think not statistically significant. Still, I think you have a point judging from a rough feeling from my own experience here. I don't have a real answer for you, but several things you should consider:

  1. If you want to be taken seriously, use your real name and properly fill in your profile. Someone hiding behind "user21976" doesn't inspire confidence. If you're not willing to give your real name and the other particulars so that your reputation is on the line, why should people believe you?

  2. Choose the questions you answer wisely. Eveyone's time here is limited. Given some fixed finite time you have to spend here, the question becomes how can that limited resource be put to the best use? Of coures "best" is only for you to judge. Do you want to be seen as the ultimate expert in some particular niche, do you want to do the most teaching to the most people, do you want to maximize rep, something else? After you have been here a while, you'll get a reasonable sense up front which questions are worth persuing for a particular goal, which ones not, which ones will get a lot of attention from random users, which will probably be closed quickly, etc.

    The second question you link to about the JTAG noise is a good example. You were in a postion to give a highly technical and expert answer, but just from the question title alone you can tell (or will be able to after being here a while) that it sounds boring and will attract few onlookers. You will likely get a upvote and accept from the OP, but don't expect much beyond that. I didn't upvote that answer originally, probably because I never went into the question due to the title or becuase I was out that day, and therefore never saw your answer. Even if I had seen it, I probably would not have upvoted only because it is rather long to read and I don't usually upvote something I haven't personally read and agreed with. I made a exception just now. Even though I still haven't read it (just too long), it is clearly well written, seems to answer the question well, and the snippets I did look at seem correct and useful. Don't expect me to do this regularly though.

  3. People are generally reluctant to upvote answers that are over their heads. Some topics are complicated, and you don't want to dumb down a good answer to a question where the OP seems to know what he's doing. However, I have found that accessibility of the answer makes a large difference in the number of upvotes it gets. If you have a complicated subject to teach, starting out with a more general introduction helps a lot. Of course this is a tradeoff with getting in the way of the exact answer if the OP is technically competent.

    This gets back to point 2 about chosing your questions wisely. Even great niche answers won't get anywhere near the upvotes that less technically intensive but broader and more accessible answers will. Basically, if you intimidate people they will just walk away regardless of how correct or expert your answer is.

  4. Most of the people on this site are here to learn and are not experts. Think about it and realize that's the way it has to be. The experts just aren't going to generate the volume of questions that will keep this site alive. However, we can and must insist on questions being properly asked, having nothing to do with their technical level. Experts like answering good questions, so the way to keep the site alive is to ruthlessly cull out the bad questions. Once you have enough rep, you can help with this by voting on questions and voting to close where appropriate. Look for the "review" link at the top the page.

\$\endgroup\$
  • \$\begingroup\$ Good points. It's really not about me. It's about how this can be a place were more of my fellow EE's will actively ask questions and provide answers. Just like professionals do on StackExchange. And yes, I will try one more time to figure out how to change the user21976 handle I was allocated at signup to something more personal... I have added photo, name, website, age etc. some time ago... but how to change the handle was not obvious to me. Glad to hear it can be done. \$\endgroup\$ – Rolf Ostergaard May 31 '13 at 18:40
  • \$\begingroup\$ @user: Actually you haven't added your real name to your profile either. Go to the main site (not here on meta), click your user name at the top of the screen, then click EDIT to the right of your account name. That will let you set your user ID, real name, and all the other stuff so that it doesn't look like you have something to hide. I think you even get a badge (autobiographer?) for completely filling out your profile. Remember, your profile is really for all of us, not for you, so its a courtesy to everyone else to fill it out completely. \$\endgroup\$ – Olin Lathrop May 31 '13 at 20:01
  • \$\begingroup\$ Thanks - that was pretty easy. I tried to figure out how to do it twice previously and didn't succeed... Funny. \$\endgroup\$ – Rolf Ostergaard Jun 1 '13 at 6:21
7
\$\begingroup\$

Speaking more to the site mechanics side of things, it's hard to (automatically) determine if a question is a difficult high-quality question or a poorly written elementary question. Both may see low votes for separate reasons, but how could you determine which question is good using only automatic algorithms? This is a very difficult problem. We can (with some accuracy) identify very poorly written questions, but we don't have a filter for "great" questions.

Because of this, there isn't a fair way to automagically improve the point value of high quality advanced questions (or answers for that matter). We could use the user's reputation as a guide, but that wouldn't be fair to new users nor would it guarantee good results.

What can change is users upvoting high quality questions and answers. You can also offer reputation bounties to draw attention to good questions and reward outstanding answers. This isn't the "system" solution, but it is a method for improving the community.

\$\endgroup\$
  • 1
    \$\begingroup\$ Well one simple learning from the rep system is that giving an accepted answer is not as important as giving a popular answer - that may not be what professionals are after (if you accept the premise that great and professional answers will attract more professionals on the question side as well). Also: We should not underestimate what algorithms can do... \$\endgroup\$ – Rolf Ostergaard Jun 6 '13 at 6:32
6
\$\begingroup\$

The site has been designed to allow contribution without the need to even register. And while you may disagree with this approach and I’m not 100% certain of the intention behind this decision it does have some distinct knock on effects. This is acknowledged on the site, “… the correct answers may not get the most rep” and “… the correct answers may not be ones accepted”. I've seen it mentioned a number of times that the site philosophy is to collect as many answers as possible and to be a "repository".

There is a tendency towards the lowest common denominator. Not just in technical focus but in sometimes personal interactions. Some of the most highly up voted answers are up-voted because the Internet likes snarkiness.

And there are additional undesirable effects, we have at least one troll who has been banned who starts up socket puppet accounts and generally makes a bunch of busy work for the mods while calling them childish names. I've also seen answers that were the actual answer gets down voted. We also have some regular members who randomly down vote without explanation.

Here is an interesting script by @anorton modified by @anindoghosh

Look around on Meta, you'll see there has been debates on what to call the site (and there has been a name change), even what the meaning of electrical engineering means, whether power systems belong etc. etc. So then the debate then becomes "what do you mean by expert"? I'm very certain that the troll we have considers himself to be an expert. And sometimes his answers are meaningful. There have been attempts to steer mathematical based questions onto math.se ( which is a very different meaning of engineering than my version) and there are various factions that also have a very limited view as to what sort of questions should be asked and answered.

Another debate then would surround "what is professionalism". We have some confused souls here that equate curmudgeonly behavior with professionalism. And we have seen childish and petulant behavior all in the name of purity and nominal "professionalism".

Fundamentally, when you use the reason(s) “for the sake of purity, or professionalism or righteousness” history shows us that it usually isn’t truly about that.

So it's messy, it's less than ideal, there is a tendency to the mundane and the LCD. As @ThePhoton mentioned, technically specific areas will see the least amount of posts/responses. Trust me, your professionalism won't be determined by how you answer a question here and as I've seen @OlinLathrop say "you are volunteering here so put in as much or as little time as you want" (paraphrased). People ARE wrong on the internet. The right to ownership of the information you put up here transfers to SE (a private company hoping to monetize your efforts) so that too may affect your decisions.

\$\endgroup\$

You must log in to answer this question.

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