What Do You Do About Posts That Are Mis-Leading or Flat-Out Wrong?

[Jon Chandler]

I would appreciate some ideas and input regarding forums on another board where I am a moderator.

I believe forum content represents a good resource and knowledge base for solving problems.  Often times, a problem isn't new or all the unique and it may have been answered in previous forum posts.  The make searching previous posts useful, having a clear, concise topic is useful, as is not straying too far away from the subject of the topic.

I'm kind of at a loss of what to do when somebody posts information that is misleading, wrong and even dangerous.  I don't like letting things like this pass, as somebody may look at a topic later, see a schematic that appears to solve their problem and spend time and money pursuing a path that's not going to work.  The posts in question range from "it's ok to ignore the absolute maximum ratings - I know because I've done it and it works" to schematics that are wrong and incomplete to more dangerous things like taking a AC line-power / 5V supply designed for a power meter and saying the transformerless design is a nice way to power a dev board on the bench.  Some of the suggestions may work some of the time and may even be good solid suggestions but statements like "it's ok to exceed maximum ratings" are never explained - something that works in a specific case is extrapolated to the general case with no explanation given.  Many of the readers of this forum don't have the background and experience to evaluate recommendations like this or even question that they may be wrong.

The amount of traffic on this forum isn't high enough to ensure other readers challenge a post or post out the errors.

What's the best way to deal with things like this?  In my case, the posts in question come from one or two posters who "have never been wrong" and can't be convinced by facts, data sheets and references.

Any advice will be appreciated!

[amspire]

It is not your job to police bad advice generally.

People who are about to build something often search for designs and find several.

If they decide to build a project based on bad advice on the forum, then it is not a disaster. The builder may learn more about electronics trying to make that one failed project to work, then in all the previous projects they have tried. They will know to have better judgement, to do more research, and to ask for more opinions next time.

Obviously if you think that someone is posting a design with malicious intent - it is deliberately designed to fail as a kind of a public prank, you can delete the post.

What is a concern is advice that is directly dangerous - like the risk of electrocution or explosions. Using the transformerless supply for bench testing is just not safe. If you are the moderator and you are not comfortable with the post, post your own disclaimer and then lock the thread, or delete it.

I don't think you can really try and regulate all the designs that might catch fire and burn the house down, but ultimately, that is your judgement too. If it is making you sleep badly at night, then you probably should either post your own concerns, or remove the post.

Advice that in your opinion is just waste of money and time? For some people, if you tell them something won't work, they want to do it to show they are "smarter" then you. Not your problem really.

Richard

[vk6zgo]

You really do like putting your head in the lion's mouth!

The posts here seem to run the gamut between " Let's put a capacitor in series with the active line to drop the volts for a +5v supply!"
To
"EEK!! Shock Horror! this project runs off the AC mains/has tubes which operate at high voltages,it should be banned!"

People with something approaching a brain can usually either deal with any issues involved with mains voltages,including rejecting patently unsafe devices,
Or,
Appreciate their own limitations,& use commercially made low voltage supplies.

The problem lies with the others!

VK6ZGO

[amspire]

I think the problem Jon has is he wants to attract people who do know what they are talking about to his forum.  No-one with serious experience is going to enjoy staying in a forum dominated by people giving bad advice and bad designs.

[Psi]

How about implementing a up/down click based rating system on your forum.
Everyone with an account could rate other forum accounts as being trustworthy vs inexperienced by clicking on 1up/1down buttons on other posts.

Warnings could automatically be displayed when people with low trustworthiness make posts/threads.

And because everyone has a vote people can't complain that their being moderated unfairly by admins.

The tricky bit would be getting the wording of the system right so it doesn't offend people, while still conveying caution in dealing with people that have a low rating.

[PeteInTexas]

How about simply replying to the post and pointing out how they are misleading, wrong, or dangerous?  If you are going to assume that people read the thread, then your rebuttal will be read as well.  Plus, we all learn something.

[EEVblog]

Tough call.
If you really feel strongly about it then you can edit the OP's post to include something like:
"WARNING: The forum moderator deems the information in this post to be potentially dangerous" etc.
And of course then you are obliged to follow up with your own technical post pointing out the flaws in details.

Dave.

[tyblu]

Post that it's wrong, correct it if you have time and know how to do it, then forget about it. It's the internet.

[Psi]

The problem with correcting or editing peoples threads/posts is that it gets quite time consuming when there's a lot of forum members.

Also, when it comes to safety there's quite a range of opinions, and in fact the same thing maybe quite unsafe for a beginner to do and perfectly normal for an experienced person.

So it ends up annoying people and starting arguments.

[joedevola]

Perhaps have a clear policy statement to the effect that dangerous suggestions will be deleted.  I frequent DIYAudio (tube/valve area), where there is a policy that any post that suggests a no isolation transformer power supply will be deleted (240 volt rectified mains to get the required voltages for valve amplification is a great temptation for some ignorant of the dangers).  The policy seems to work well and there is a very cordial attitude and very little flaming from most members of the forum.

[HLA-27b]

Instead of dealing with the bad end of the spectrum it is possible to do something with the good end in order to make it more prominent and conspicuous. On many forums threads with exemplary content are made sticky and/or are locked to further posts and cleaned up from irrelevant content. Dealing with the good stuff both rewards the good folks and relieves you of the duty of the "groom of the stool"

Besides, you can correct and remove posts only so many times before attracting accusations mentioning freedom of speech. It might be a better idea to encourage the posters to edit their own posts. For example in my book it is totally ok if I received a letter from a moderator along these general lines:

I am considering making topic XXXX a sticky because of its great content. Your post #abcde in the same thread however is an eye sore and will be deleted.
I might consider letting it stay if it was edited within 48 hours to contain/exclude YYYY.
The Moderator

Bear in mind however that a forum that contains safety warning on every post will be avoided like math homework.

[Jon Chandler]

I think the problem Jon has is he wants to attract people who do know what they are talking about to his forum.  No-one with serious experience is going to enjoy staying in a forum dominated by people giving bad advice and bad designs.

This is a good point.  When knowledgeable people read some of the stupid crap, they may decide the forum offers nothing and never return.  There are many bright people on the forum that provide solid answers and guidance but sometimes these posts get loss in the crap.

The posts here seem to run the gamut between " Let's put a capacitor in series with the active line to drop the volts for a +5v supply!"

To

"EEK!! Shock Horror! this project runs off the AC mains/has tubes which operate at high voltages,it should be banned!"

VK6ZGO

It's more of the first than the second.  "It must be safe since Microchip recommends it..." but the application was taken completely out of context.  Or a picture of an open switcher power supply salvaged from some piece of line-powered gear, complete with exposed line terminals in a pile of stuff on a bench.

I would hope that somebody using tubes/valves has some respect for the voltages involved.

Several people have suggested posting comments making a correction.  This leads to battles of "I know it works.  I've done it this way for years."

Here's an example of my current dilemma.  Notice the high side transistor is NPN?  It might work in some circumstances but this sure isn't the general way to use an NPN transistor as a switch.  I tried the gentle approach of remarking that an NPN transistor is usually a low-side switch to give him a chance to make a correction - his response was "It can work either way."

Sigh.  Thanks for the suggestions,

Jon

[amspire]

Here's an example of my current dilemma.  Notice the high side transistor is NPN?  It might work in some circumstances but this sure isn't the general way to use an NPN transistor as a switch.  I tried the gentle approach of remarking that an NPN transistor is usually a low-side switch to give him a chance to make a correction - his response was "It can work either way."

That circuit is not bad. 

There is nothing wrong with driving the load from the transistor emitter.

If it were me, I would add some hysteresis so the 339 doesn't get unstable near transition.

I also would probably remove the 1K resistor totally to allow a greater VDD range to turn off the LED's. The 1K does not have a purpose.

As the circuit is now, it wouldn't work great with a relay, as the annotation suggested,  since it would still leave a magnetizing current in the coil when the 339 output is low, but that is probably a good starting point for a discussion. Yes, if it were redesigned with a PNP transistor, or with the load in the collector of a NPN instead of the emitter, the transistor can be allowed to turn on to saturation, and so you could use a lower VDD.  These are minor improvements - refinements that do not matter that much at all.

As a moderator, there is no reason to get involved in any way. That's my view.

Richard.

[vk6zgo]

He was quite correct,it does work either way round,I've seen both ways used by many reputable manufacturers.
In this circuit the low side option is preferable,as the the base voltage swing is larger,& with his limited value of Vdd,is more able to saturate the transistor.
Maybe if you gave your reasons rather than just saying "An NPN is usually a low-side switch",he may have been more receptive.

VK6ZGO

[saturation]

It becomes a question of unmoderated versus moderated and the extra work moderation requires.  One well established electronic forum does this more often than not, is considered restrictive, turns people off and you can tell from the traffic it generates.  Its best to just let things be, like eevblog, except when harmony is shaken by spam, trolls, and ad hominem attacks.  As seen on this thread, a supposed error is already self corrected by other posters. Lastly, how can a moderator be sure all wrong or corrected posts are truly "corrected"?

[alm]

I don't think heavy-handed moderation will do the popularity with the users any good, editing posts and adding warnings should be reserved for extreme cases (eg. threats, serious risk of injury/death, not just risk of a blown regulator).

About what to do with 'wrong' answers (for some definition of wrong), it may be informative to look at Stack Overflow. This was one of the problems they were trying to solve, which is why they created something between a forum and a Wiki, where answers can be edited at any time, rated and commented on. You do need a reasonably large community for this to work, however.

[Jon Chandler]

...This was one of the problems they were trying to solve, which is why they created something between a forum and a Wiki, where answers can be edited at any time, rated and commented on. You do need a reasonably large community for this to work, however.

I like the wiki idea.  It becomes a useful resource and repository of knowledge.  The forums should be a valuable resource but different terminologies, poorly titled topics and topics going off-subject make it tough to find the few nuggets in the pile of trash when you're looking for something.

[amspire]

The forums should be a valuable resource but different terminologies, poorly titled topics and topics going off-subject make it tough to find the few nuggets in the pile of trash when you're looking for something.

It needs a light touch, as it is very easy for the nuggets to be hidden in something that looks wrong. Be particularly careful if someone has actually built and tested a circuit, and you are not prepared to do the same.

If someone did something that looks ridiculous - say a microphone amplifier using a 2N3055 power transistor. It would be dead easy to jump straight in and to suggest that the choice of the 2N3055 is rubbish, and you might even change it to a common small signal transistor.

Two problems. Perhaps this unlikely circuit does actually work extremely well, and the circuit using a replaced transistor may not work all that well if you do not bench test it. When you look at a circuit, it is very easy to miss the features that actually took someone days of work on a bench to fine tune.

If you start acting as the filter and quality control for the posts, then you will annoy genuine posters, and everything will end up with with your personal slant on electronics. 

Richard.

[sonicj]

I recently got banned from a forum for providing useful information.  Link

Apparently crushing the fragile egos of advertisers in the process is frowned upon.   
-sj

[Jon Chandler]

I recently got banned from a forum for providing useful information.  Link

Apparently crushing the fragile egos of advertisers in the process is frowned upon.   
-sj

ElectroTech is infamous for this.  I am among several others here who have been banned.  On ElectroTech, a one sentence answer is deemed far better than a link to a reference that will not only answer THIS question but the ones that will arise from an incomplete answer.

[vk6zgo]

The forums should be a valuable resource but different terminologies, poorly titled topics and topics going off-subject make it tough to find the few nuggets in the pile of trash when you're looking for something.

It needs a light touch, as it is very easy for the nuggets to be hidden in something that looks wrong. Be particularly careful if someone has actually built and tested a circuit, and you are not prepared to do the same.

If someone did something that looks ridiculous - say a microphone amplifier using a 2N3055 power transistor. It would be dead easy to jump straight in and to suggest that the choice of the 2N3055 is rubbish, and you might even change it to a common small signal transistor.

Two problems. Perhaps this unlikely circuit does actually work extremely well, and the circuit using a replaced transistor may not work all that well if you do not bench test it. When you look at a circuit, it is very easy to miss the features that actually took someone days of work on a bench to fine tune.

If you start acting as the filter and quality control for the posts, then you will annoy genuine posters, and everything will end up with with your personal slant on electronics. 

Richard.

A classic case of this is in some radio front ends,where a device which would not normally be associated with low level antenna input stages is used to provide greater dynamic range,& hopefully freedom from the intermodulation effects which can plague normal MOSFET & JFET high sensitivity RF stages when high level unwanted signals are present.

Usually,documentation is provided giving the reason for the use of this device,but without this information,the reader could think it was a mistake or a misprint.

VK6ZGO

[sonicj]

ElectroTech is infamous for this.  I am among several others here who have been banned.  On ElectroTech, a one sentence answer is deemed far better than a link to a reference that will not only answer THIS question but the ones that will arise from an incomplete answer.

My situation was the opposite from that.... The first responder linked to a 5700 word dissertation that was relevant in that explained motor theory, but left the reader without a clear answer to the OP's explicit question. My response (aside from the smart-ass remark) was one sentence & two images (*see below) that I made in photoshop that clearly depicts the operation of the switch in question.

Keep in mind, this was a very non-technical audience... Even those who are into the pimped out golf cart hobby mostly aren't that interested in motor theory. If the paper had actually answered the OP's question, I wouldn't have said anything.

The funny part was the OP returning at the end of the thread to say thanks after I got flamed, called names & banned.    Their heavy handed and unfair moderation practices will catch up with them eventually...
-sj

[amspire]

There is nothing more annoying then someone who is right - you forced other participants to call you names for being an annoying know-it-all, and so you had to be banned.

Also when the initial poster asked about how to connect to the windings on a motor, you completely failed to explain how a loudspeaker works, like the first respondent did. You probably got banned for that oversight as well.

Third when the initial poster praised your answer, you made everyone else look very, very stupid. How many more reasons do you need?

[sonicj]

lol! I guess you're right.... Honestly, I should have seen it coming when the chemical engineer got himself banned for starting a discussion on exploring best practices for lead-acid battery maintenance.  The dude spends all day in a lab with test tubes and beakers... Like he knows anything about golf cart battery maintenance.   
-sj

[saturation]

Yes, but that what editorializing is all about: whether its by one person, aka moderator editor, or a group, aka wiki.  In the old days we used to call it FAQ, and the editor put the best of the posts together, to save on typing and the editor was a content expert.  But it takes work on someone's part and if its free, its often unkown when it gets done; the wiki style clearly works for Wikipedia but it hasn't for others who have tried to start one as it needs a lot of contributors.

Raw forum posts can be datamined for nuggets to save you time by the use of a good search engine.  If the forum engine is weak, just download the entire thread and use something that works like grep.  For a quick read use a feature of Microsoft Word few know about called 'summarizer' which does a statistical analysis of the word frequency to highlight the best items, and remove redundant ones.

http://office.microsoft.com/en-us/word-help/automatically-summarize-a-document-HA010255206.aspx

Alternatively use one of several summarizers and adjust the forum page length so it can read lots of pages simultaneously:

http://delicious.com/tag/Summarizer

  The forums should be a valuable resource but different terminologies, poorly titled topics and topics going off-subject make it tough to find the few nuggets in the pile of trash when you're looking for something.