|
Two related situations that have happened to me recently:
1. Posted a message in a Lounge thread. Realised a few seconds/minutes later that I wanted to add something. Clicked on 'Edit' and made the changes (complete with [edit] and [/edit]). Posted it. Problem: no indication that I changed it, because I forgot to add [modified] to the title.
2. Answered a question in a language forum, suggesting that the problem might be a spelling error. For the purpose of this discussion say the original thread title (and question) was "XYZZQ doesn't work". I replied "try XYZZY". OP responded "I did try XYZZ*". The reason for * here is that the email I got had * = Q, so I started to make the obvious reply. By the time it got to the forum, the OP had modifed his reply so * = Y (in the body but not the title), and I had to delete my reply quick smart to avoid looking like a first-class dork. Again, there was no indication that the message (OP's reply) had been altered.
There is also potential for malice in, say, the Back Room:
X posts a contentious declaration.
Y replies "what a load of $%^&"
X modifies his post to reflect Y's pet theory.
So, after all that, the suggestion:
On submission of an edit to a post, the system appends [n] (where n is a revision number) to the title, or increments n if it's already there. This won't completely prevent malicious backdating of posts, but should cover most of the inadvertent omissions.
Point for consideration: Should the edit to a post be emailed just like the original? That might help us figure out what's going on in cases like those described above.
Software rusts. Simon Stephenson, ca 1994.
|
|
|
|
|
Peter_in_2780 wrote: Problem: no indication that I changed it, because I forgot to add [modified] to the title
The [modified] tag is added automatically if you edit the post more than 5 minutes after first posting. We don't add it automatically for quick fixes to allow you to do wuick "oops!" fixes.
Peter_in_2780 wrote: X posts a contentious declaration.
Y replies "what a load of $%^&"
X modifies his post to reflect Y's pet theory.
Even if we add "modified' tags, people can still manually remove them. The addition of "modified" tags is a convenience only and should never be relied upon to show actual revision information. The full solution would be sto store all revisions of messages, but my feeling is that the time spent would take us away from things that are, overall, more helpful to the community.
Peter_in_2780 wrote: Should the edit to a post be emailed just like the original
Mailed to whom? If you are talking about an edit to a post to which you have replied, and then when an edit to the original post is made, an emailt o all repliers is done, then that's certainly an interesting idea.
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
Chris Maunder wrote: Mailed to whom? If you are talking about an edit to a post to which you have replied, and then when an edit to the original post is made, an emailt o all repliers is done, then that's certainly an interesting idea.
I like this idea.
3x12=36
2x12=24
1x12=12
0x12=18
|
|
|
|
|
Chris Maunder wrote: when an edit to the original post is made, an emailt o all repliers is done
No thanks. That will lead to spam and all kinds of abuse, such as adding unrelated questions to an OP that got quite an audience.
|
|
|
|
|
Chris Maunder wrote: quick fixes to allow you to do wuick "oops!" fixes.
Gotcha!
OK, so I missed the CP equivalent of the five-second rule Click[^] That would cover both my scenarios above. My working rule is now "Except in emergency, don't reply to anything less than five minutes old."
Re the emailing:
A original
B reply
C comment
D comment
E riposte
Consider B modifying his reply. My first idea was to email to whoever was emailed B's original, i.e. A, but I can see now that C and D are also worthy of notification. E, however, I don't see the need to advise (until D changes his comment...) All these, of course, subject to the first options checkbox below where I'm typing this. [This might go some way towards addressing Luc's spam concerns.]
Software rusts. Simon Stephenson, ca 1994.
|
|
|
|
|
I'm not sure if this has been reported before, so please don't send out the hamsters to get me if that's the case.
Anyway, I just noticed that the code browser is having a problem with escape characters.
See this[^] page - view the file CSRcon.cs and scroll down to line 52. You'll see that everything after that line is treated as a string, even though the string has ended.
It seems to be the escape character (marked with bold) in the following piece of code that's causing the issue:
string command = "rcon \"" + challenge_rcon + "\" " + rconPassword + " " + rconCommand + "\n";
|
|
|
|
|
After major site changes, this happened with Q&A forum posts. We reported it and Chris corrected the issue in short interval. Looks like something similar out here.
|
|
|
|
|
|
It's got nine 1 votes and is still not auto-deleted. Most folks don't know that they should mark it as abuse instead of just 1-voting. An abuse vote is a 1 too but ii also sets another flag so that when the number of votes hit a threshold, the message is deleted. Recently this threshold has been too high to be useful though. Unless you post a bad grammar joke on the Lounge and a dozen folks vote on it.
|
|
|
|
|
Thanks, it has been auto-magically deleted. I don't know how this works exeactly, but perhaps a formula combining "Abuse Reports" and "1s" together could be used to lower the threshold?
ragnaroknrol The Internet is For Porn[^]
Pete o'Hanlon: If it wasn't insulting tools, I'd say you were dumber than a bag of spanners.
|
|
|
|
|
Yeah, that may work. If there are 2 or 3 abuse votes already, then every other 1 should be treated as an abuse vote too.
|
|
|
|
|
Or Abuse vote = 3 "points" (or whatever weighting) and 1 vote = 1 "point", then when the total "points" reach a certain level the message gets banned. I know what you mean by the lounge BTW.
ragnaroknrol The Internet is For Porn[^]
Pete o'Hanlon: If it wasn't insulting tools, I'd say you were dumber than a bag of spanners.
|
|
|
|
|
It's good to see that voting a 5 in QA now requires a comment, however this seems to be a little inconsistant.
For example, the answer I posted to this question[^] has received a greater number of 5 votes than there are comments
DaveIf this helped, please vote & accept answer!
Binging is like googling, it just feels dirtier.
Please take your VB.NET out of our nice case sensitive forum.(Pete O'Hanlon)
BTW, in software, hope and pray is not a viable strategy. (Luc Pattyn)
|
|
|
|
|
It doesn't require a comment. You can just close the little box and vote anyway.
|
|
|
|
|
DaveIf this helped, please vote & accept answer!
Binging is like googling, it just feels dirtier.
Please take your VB.NET out of our nice case sensitive forum.(Pete O'Hanlon)
BTW, in software, hope and pray is not a viable strategy. (Luc Pattyn)
|
|
|
|
|
5s don't require a comment. I still include a comment anyway as I think it's more encouraging to the poster.
|
|
|
|
|
Hi Nish,
Yeah, I agree
DaveIf this helped, please vote & accept answer!
Binging is like googling, it just feels dirtier.
Please take your VB.NET out of our nice case sensitive forum.(Pete O'Hanlon)
BTW, in software, hope and pray is not a viable strategy. (Luc Pattyn)
|
|
|
|
|
Comment while voting is not a mandatory here if you are voting for 4 or 5. If you are voting less than 4, it is a must one. If you want to put just a small comment instead of voting, there is a link for that.
Don't forget to Click on [Vote] and [Good Answer] on the posts that helped you.
Regards - Kunal Chowdhury | Software Developer | Chennai | India | My Blog | My Tweets | Silverlight Tutorial
|
|
|
|
|
I am well aware of the comment widget, I just assumed (wrongly) that comments on upvotes were required as they are on downvotes. All cleared up by the previous answers.
DaveIf this helped, please vote & accept answer!
Binging is like googling, it just feels dirtier.
Please take your VB.NET out of our nice case sensitive forum.(Pete O'Hanlon)
BTW, in software, hope and pray is not a viable strategy. (Luc Pattyn)
|
|
|
|
|
this will be turned into a bug report as soon as it occurs, it is only a matter of minutes.
"Modified on ..." message sometimes breaks into sig. Have seen this happen occasionally.
EDIT1 first edit, after 3 minutes /EDIT1
EDIT2 second edit, after 6 minutes; the modified message got added correctly; it did go wrong here[^] although I did not touch the sig at all. /EDIT2
EDIT3 it seems to go wrong upon an additional edit; so the scenario is:
create message
wait 5 minutes
edit message (this adds "Modified on..." correctly)
edit message again (this throws away the original closing /div tag from the sig, probable cause is length of sig+"Modified on..." exceeds the limit)
/EDIT3
|
|
|
|
|
Well, might be the case but i am sure this will be added to the bottom of 'to-do' list.
Who would had noticed it (other than you )! Rarely someone(that too with a long signature...) would edit his own message more than 2 times!
|
|
|
|
|
People should not be punished for conveying as much information as a sig allows, while iteratively improving their replies.
|
|
|
|
|
I fully agree with you!
It's just that I was trying to figure out the number of people, who would be facing this issue!
|
|
|
|
|
Verified and added to the bug list. Thanks Luc.
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
Just going through our TODO and noted that this was fixed a while ago.
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|