8

I saw a question asked here that was, I thought, a low-quality post. The asker put up a trigger and asked if it was bulkified. Judging from most of the responses, people thought (as I do) that it was a low-quality post:

bulkifying trigger

Most responses told the asker to check Google, pointed him to documentation, or encouraged a higher-quality post. However, one person actually wrote the trigger in a bulkified manner, and was awarded the "best answer" checkmark by the asker.

In my opinion, reputation on this site should be given for making solid contributions to solid questions. The original question probably should have been closed. It seems that I could rack up a lot of reputation points on this site by writing code for people instead of explaining that their questions are low-quality... and I do not think that this is an admirable thing.

How do we fix this? Should moderators be more aggressive in voting to close questions that are low-quality, in addition to responding that the question is low-quality? I don't think that anyone has the power to award or remove the checkmark outside of the asker, so I don't know how to block that award.

I'm coming at this from a desire to maintain the high quality here - not only of the posts, but also of the members - and I want to be sure that everyone with high reputation scores has earned that by not only giving quality answers, but also by encouraging high quality posts.

Thoughts?

1
  • 1
    I always feel like downvoting questions like that where the code is formatted such that it is 90% whitespace and causes TWO scrollbars to show up. If you want my help, at least try not actively to make my job more difficult.
    – user6861
    Aug 28, 2014 at 2:09

4 Answers 4

9

I think the best thing to do with questions that are repetitive, especially like bulkificaiton, need to have a good canonical answer that provides all the best advice from the community and close any new instances as duplicates.

I keep looking for a perfect and generic bulkificaiton question to close these as dupes and can never find one that's not tied specifically to a huge block of code. To me the best thing we can do is fix that and then diligently close every other "plz bulkify my trigger" question as a dupe. Same general idea with other topics.

I'd say this is a good time for self answer since it helps set up a truly generic situation and get the ball rolling on quality answers.

6
  • 4
    I like your idea. To summarize: Make a canonical question for major topics, close others, and have not only a canonical answer, but allow people to add their two-cents, if desired. Aug 12, 2014 at 2:09
  • How do you see this as being different than @PeterKnolle idea to create Tag Wikis per his answer to Possible to have a sticky section?
    – crmprogdev
    Aug 15, 2014 at 13:54
  • I think you are right. We should answer the question of bulkification once. All other questions can be closed as dups. Therefore my suggestion is to open a new question question and we can answer it. Aug 21, 2014 at 13:34
  • Therefore I created a general question: salesforce.stackexchange.com/questions/47469/… Aug 21, 2014 at 13:42
  • 2
    @Christian Deckert, While I agree with you in practice (as I answered the new question), I now see that you have voted to close 17 old questions as duplicates of the new one you just created. While some of these are basic general questions that should be closed as duplicates of your new general answer, I would guess that some of these are legitimate non-general, fairly specific questions. So basically I'm saying just because a question is about bulkifying, does not mean it should be closed as a duplicate. Only those, non specific, very general, or 'write this code for me' questions should. Aug 21, 2014 at 19:29
  • 1
    Basically, just saying that we need to be careful, as not every question on a subject can be answered by a generic, non-specific explanation of the topic, so keep that in mind when voting to close questions. Aug 21, 2014 at 19:36
8

I think this question really gets into some very meta territory, and the Stackoverflow/Stackexchange meta sites have discussed this in various forms of "What is Stack Overflow's Goal", and variations of "why stackoverflow is nice" and "Why is Stack Overflow so Negative" Of course, we're not Stack Overflow, so we have free reign to tackle this how we want as Salesforce(+ET) developers, but I think it is instructive to see how other communities on this Q+A platform have discussed it.

I think there will always be a tension of trying to curate high-quality questions and answers (which can mean looking like a "bad person" with your downvotes, and close votes), and being "nice" and providing answers even if the questions has been asked in some form before/is not very well-put.

I don't know that there will be a perfect solution to this, but I do really like @ca_peterson's suggestion of producing more high-quality questions and answers ourselves, as it helps address both camps. The "crusader"-type gets their high-quality curated Q+A site, and the "helper"-type can still close low-quality versions of an answer with a comment that points to the high-quality version.

As we are still a younger community (which should be out of beta, but aren't quite yet!), there is also a tension of increasing user adoption and not scaring new users away with what can appear to be negativity. But I don't believe that should be at the cost of the overall quality and increased noise of low-quality questions.

7

******DISCLAIMER:******** I did answer this question. I was not the person that the poster was referring to, as I did not write the code for the person, but I did add an answer which showed some basic concepts and encouraged research. So I don't know if that means I was just trying to collect rep or not.**********

While I completely agree with you that we should discourage 'do this for me' questions, I would argue that you can sometimes turn a 'crappy' question into a decent and useful answer. This is what I tried to do in my answer was to take a rather poor question, and give useful information WITHOUT doing the work for the asker. It's my hopes that it nudges the asker to apply some of the tactics I showed, combined with some research, they could take a stab at it. Also could be useful for others coming along looking for info on bulkifying trigger.

I have used this tactic in the past where I add very general high level help to very poor, and specific questions (like 'help me write test code for this class').

Here is an example of an answer I put in what I think is a similar question.

Can any one please help me in writing a test class for the following class

IMO, a rather poor question, and it did end up being closed. I think though that the answer I provided could be useful for people coming along looking for help writing test code.

I'm sure there are some varying opinions on here as to whether this is the right approach or not, so I'd love to hear your opinions.

My general thought is. NEVER just write code for someone asking for that, BUT I will sometimes give them very high level ideas and code, that will get them started on the right track.

Love to hear everyone else's thoughts.

1
5

One could argue that everything that needs to be said about SFDC development has already been answered at least once -- at least for features that have been around for more than ?2? years. All it takes is persistence in searching, reading, and willingness to experiment with the tools to get to a conclusion.

I've always been impressed with the level of online resources available to the newbie SFDC developer (I was one once). And as such, many of the questions we see on SFSE don't contribute to the general knowledge base.

Because SFSE has some outstanding contributors, the site has a reputation as an authoritative place to go. SFSE contributors show an amazing ability to provide very rapid 'customer service' despite other demands on their time and while probably muttering under their breath "I can't believe we have to answer this ****'ing topic again".

I don't really mind if someone chooses to 'do the code' for the OP while others (hopefully myself included) comment with " have you looked here ..somehyperlink..? " as I long since (mostly) gave up trying to decipher poorly formatted / bizarrely variable-named code. But, if someone (myself included) wants to take the time to do this, perhaps for their own amusement (like solving a crossword puzzle), the reputation points shouldn't be begrudged.

4
  • 1
    I agree with you. I don't think the majority who are here are at all motivated by a desire to earn reputation; instead to assist others in learning SF and solving the issues they're encountering.
    – crmprogdev
    Aug 22, 2014 at 12:42
  • 1
    Pretty well said, never though about it like that, but I do agree. Aug 22, 2014 at 13:40
  • Sites like SO and this one exist to help people with specific issues, so in that regard we will never run out of stuff to answer. Sure, something similar might have been asked before, but maybe the next question will be slightly different and specialized to a particular situation.
    – user6861
    Aug 28, 2014 at 2:15
  • I did say "one could argue" ... :-) But yes, as the realm of software app problems is infinite, we will never run out of questions that are unique in some way
    – cropredy
    Aug 28, 2014 at 5:36

You must log in to answer this question.

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