Is deleting threads helpful in this forum?

We have lost a lot of important questions and answers, because the authors decided to delete the threads.

1. This erases the work of those who have spent time in creating the answers and in consequence the motivation to answer is reduced.
2. I've seen several homework questions disappearing, such that the teachers or professors cannot detect such cheating anymore.
3. The deleted thread are usually important for the personal reputation (not the numerical reputation number). E.g. it is helpful to see, if the author cared for further inquiries or if a question was answered already but the author does not want to care about it.

I'd prefer to let only TMW and the editors delete threads if they conflict with the idea of this forum (Don't section in About MATLAB Answers) or if the author asks for deleting and mentions any reason. But without a reason, editing the message should be enough.

Do you agree or are there good reasons to let the authors delete their threads if they like to?

[EDITED] I've expanded the topic to overwriting a question with garbage. I suggest that editors and admins reconstruct such messages (e.g. from Google's cache), until the authors ask for deleting with a good reason. Thanks.

[Andrei Bobrov]

Hi Jan! I agree with you!

[Fangjun Jiang]

+1, for bring up another good topic!

[David Young]

Yes, I agree too - though maybe unanswered questions should be deletable.

[Thomas]

+1 I agree with you.. helps us keep track of older questions and sometimes offer contexts to a particular question..

Answer 1

[Wayne King]

I agree with you. The ability to edit and/or add comments is sufficient unless there is some compelling reason to remove the thread.

Wayne

Answer 2

[TAB]

I am agree with Deniel. Deleting a post should be allowed only for question that is unanswered and not commented.

As author delete the question, it also wash-up the efforts by person who have spent time to find answer of that question.

[Image Analyst]

That's my vote too. Who knows how many "This is a test" posts have been posted and (thankfully) deleted by novices.

[Walter Roberson]

It is entirely plausible that I have posted more "this is a test" posts than all of the novices combined ;-)

Answer 3

[Fangjun Jiang]

I agree! OP can always edit the question. I would encourage everyone to edit it as the question become clear. Right now, there are many questions that their titles are not-to-the-point and there are serious typos in it.

Many times, I've edit my answer when I read it later and found typos.

There is no edit capability on the comment so far. I would like to have that too.

Answer 4

[Daniel Shub]

I think the OP should be able to delete a question when there are no answers (and maybe no comments). Once people have put some effort into answering the question, it should stay. The advantage of deleting unanswered questions is that it might be useful for double posts etc.

There also should be something that makes editing restricted. The OP should not be allowed to delete the text of the question.

Edit: 20-Sept As Walter pointed out, figuring out which questions have no useful answers/comments is difficult. Often a duplicate question will get a quick comment/answer that states it is a duplicate question

I think that when a user requests to delete a question, anyone who has provided an answer or a comment on the question, but not a comment on an answer should be asked for approval.

[Jan]

I have not seen too excessive editing yet, which changed the meaning of a question. But if deleting is restricted it can be expected that the authors clears the question instead. But this would be an ugly foot print and the community will remember this.

[Daniel Shub]

We now have deletion via editing.

[Walter Roberson]

Yes, people are now editing their question away and deleting all the rest of their contribution to the question.

[Randy Souza]

When you see deletion-by-editing please flag the question and/or send email directly to me (<firstname>.<lastname>@mathworks.com). If the question was particularly valuable, I can restore the previous text.

Answer 5

[Walter Roberson]

We need posters to be able to consolidate their duplicate questions down to manageable threads: you cannot rely on the staff and editors to do all that merging on behalf of the posters (for one thing there would be no incentive for the posters not to start new threads, knowing that if they were wrong, someone else will clean up after them.)

If the duplicate threads contain only comments or answers along the lines of "This is a duplicate, please clean up", or even "I already gave you the answer over there", then nothing is lost in deleting the comments and answers.

If the duplicate threads contain real answers or comments of relevance not already expressed, then they need to be merged, which is unfortunately not something that there are tools for even within Mathworks (as far as I can tell.) At least not that preserve ownership and votes and which update "Questions answered" and so on.

If a thread of complete irrelevance to MATLAB is posted, the editors or staff usually find it in due time, but I would not object to someone being able to delete their own completely-irrelevant thread.

If a theory question is posted, then matters get a bit tricky. The question might be expressed as pure theory, or the question might be expressed as how to implement a theory in MATLAB.

The pure theory questions can sometimes be of interest to keep our mental wits sharp even if there is no immediate connection to MATLAB; the poster of such a question does, however, need to understand that obtaining an answer here is sheer good luck and that there are more appropriate resources for (e.g.,) communications theory. Comments about where the poster might have better success are contextually fair and useful comments, but at some point having the posting pruned would be good cleanup. Speaking as an editor, though: I don't like just deleting these and hoping the poster gets the idea essentially, and I don't have time to construct email to them describing why their message was deleted and doing my own research in to where they might better have posted

The questions about implementing a theory in MATLAB: although hypothetically those could be interesting and challenging, it seems to me that the great majority these questions that we attract are from people who really do not know anything about the theory and probably don't really care, but want the MATLAB source code anyhow (e.g., because it would be 95% of the work of their final year project.)

Editing the content of a posting: users need to be able to do this in order to correct typos, fix formatting, add clarifications. And some of that needs to be allowed even after people have answered. None-the-less, if someone asks a homework question and gets a useful contribution, I am not sure that it is fair that they should be able to delete the traces of the discussion afterwards, even if only by editing the content of their question to be "Never-mind; ignore this".

[Jan]

I've posted a question 7 times by accident during a server outage. Then it was my job to delete 6 of them.

In the previous days an author posted a homework question about a series to approximate Pi, some answers appeared, the thread was deleted, and posted again, more answers, and deleted again. How annoying.

[Jan]

@Walter: We have 7 editors in the near future. This potential should be used to improve the quality of this knowledge database. Waiting for updates of the interface seems to be more or less hopeless at the moment. So even if we'd all prefer a limitation of the possibilities for deleting an answered thread, this feature will be implemented in autumn 2012 or later... I'm disappointed by the speed of inovations.

Answer 6

[Richard Brown]

My opinion is that a question can be deleted until it has been answered. Then it can't. Editors, and only editors, should be able to merge questions.

It is infuriating when someone deletes a question with an answer that I've put any significant time or thought into. It's like someone taking your work and flushing it down the toilet, and makes me feel like not contributing any more.

[Jan]

I'm convinced that your deleted answers are *not* flushed to the toilet, but submitted as homework solution by the OP. Then you have assisted a treating in opposite to your intention to help.

[Walter Roberson]

I've been musing, trying to figure out ways to watermark MATLAB code in such a way that deleting the obvious watermark lines would be self-detected and cause the program to halt. I haven't come up with anything yet, though.

[Sean de Wolski]

Walter, that should be a question of itself. I have an attempt at an answer going, but it's clearly a challenging problem.

Answer 7

[Daniel Shub]

I think we need a "question of shame" where we can answer with names/profile pages of individuals who have deleted questions. Better would be if TMW would just turn off deleting. At this point I would rather have 1000s of duplicate questions, that we can tag as duplicate, then allow people to delete questions along with OUR answers and comments.

[Jan]

When I count correctly, we have 14 editors and a handful of admins who can care about deleting/merging duplicate question. Therefore I would leave deleting up to this group.

I hesitate to delete a message only based on my opinion that it is improper. But the deleting of double postings does not destroy information and cannot be misunderstood as censoring. But a mechanism is required to avoid, that both postings are deleted by different editors at the same time.

The TMW team has recognized, that deleting threads is an important drain of quality, and cares about a solution. This could mean "soon or this year".

I share your emotions concerning the "delete thiefs" and I do *not* want to see shame or blame in this forum.

[Walter Roberson]

Shaming people is decidedly against the mandate of this forum.

[Daniel Shub]

I agree that we do not want to shame people, but I am not sure if attempting to keep relevant statistic is shameful. Some people might be "shamed" by having asked a large number of questions or given a lot of answers, but TMW doesn't let us hide that or even post anonymously ...

[K E]

The only time I have been tempted to delete one of my own questions is when I have been exposed as an idiot for asking it. However I take this as a instructive humility lesson and there they remain. I do not mean that I have been shamed by the answerers, just that after the lightbulb went off I wished I had not asked it in the first place.

Answer 8

[Jan]

Several questions have been invalidated in the last two weeks: The title and the contents have been overwritten with garbage. Obviously the authors wanted to delete the thread, but this was not possible after a (correct and valuable) answer has been given.

I can imagine only one reasons for this: The author wants to submit the answer as his/her own work as a solution of a homework.

Overwriting questions is:

• inpoltite to the ones who have spent time for answering.
• not conform to the nature of this forum: share solutions with the community.
• an abuse of the forum and supporting homework cheating would reduce the scientific repuation of thius forum.
• not effective because you can find the original version in Google's cache when you search for it.

Therefore I decided to reconstruct such messages as some admins do this already. I want to invite the other editors top do this also. Until today none of the concerned users complained or tried to hide the question again. All human make mistakes and you can ask an admin (email at the bottom of each page) or one of the editors to delete a thread, when you have a reason. But cheating and stealing the time of the answerers is not accepted.

[Sean de Wolski]

@Jan, How do you recover the lost information?

[Daniel Shub]

Randy Souza in a comment to my answer said: "When you see deletion-by-editing please flag the question and/or send email directly to me."

In a previous case of deletion via editing, I flagged the question and emailed him. It took a couple of weeks for the question to be restored, but they did it in the end. During that time I was told that TMW was talking about the issue. I am more in favor of flagging the questions and letting TMW handle it then dumping more admin things on community members. Hopefully this will encourage TMW to correct the problem and give us a proper set of admin tools.

[Randy Souza]

Please do flag the questions and/or email us when you see this behavior. It allows us to contact the original asker and try to understand their motivation (though I agree with Jan that the most common case is probably obfuscation of homework requests).

It's quicker for us to restore the original text (at least until we get the tooling built to allow the community to do this), and I'd much prefer that you all spend your time answering questions rather than digging through Google's cache. That said, I appreciate your previous reconstructions!

[Randy Souza]

@Jan, you wrote "Until today none of the concerned users complained or tried to hide the question again."

Does this mean that someone DID complain today? If so, please point me to the question...

[Jan]

@Sean: I copy a relevant part from the answer, which is still there, and let Google search it. Beside the current version, with invalid question, I get the original version, when I hold the mouse at first over the ">>" and click on the appearing "Cached" link. This link does not appear for all found links. But you can frequently access e.g. pages, which actually require a login such as closed forums, when the admins decided that opening the pages for Google is an important advertisment.

@Daniel: I understand your hesitation and I agree that letting TMW care for a restoring has advantages. "Editors" are not "Moderators" and this might have a good reason.

It is important that the restoring happens fast, which means less than a day, because the teachers will not check this after some weeks anymore. I do not want to chase cheaters. I only want to reduce any chance that studious contributors are frustrated and non-serious user feel encouraged.

@Randy: No, the "until today" was meant inclusively like "yet".

Finally I do not forget to mention that the number of annoying threads is very small. The current control mechanisms support the cooperation of serious users successfully and sufficiently. The spamming level is impressing. Thanks, Randy and the TMW team!

[Walter Roberson]

We had a big spate of these about two weeks ago.

I seem to recall one in which the user re-deleted. I've lost the mental details, though.