June 7, 2009 at 2:59 pm
RBarryYoung (6/7/2009)
I don't really want to rant about this (someone from MSDN once made that mistake here 🙂 ), and if I get into right now I am sure that I will. If you have access to StackOverflow, here is a thread that demonstrates my frustration: http://stackoverflow.com/questions/961389/is-inconsistent-state-allowed-in-a-transaction/961426#961426
Don't want a rant, was just wondering. I visit serverfault (the IT equiv for stackoverflow) and it doesn't seem to me to work as well as some claim. Just wanted to know your opinion
Like the link above: OP asks how to get around FK's in the middle of a transaction on SQL 2000. Apparently he thinks that Atomicity means that when he is in a transaction, no rules should apply. I say "1) Atomicity has nothing to do with that and 2) you can use NOCHECK (on 2005, not sure about 2000), but the recommended way is to update in the correct order". So he starts dinging me for points and then marks the "Use FKs with Deferrable" as the answer (this is an Oracle feature).
Currently your answer has 3 times the points that the 'accepted' answer has (and that's just because I don't have the rank to downvote there). To anyone paying any attention that should signal that there's a problem with the 'accepted' answer.
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
June 7, 2009 at 3:29 pm
Thanks, Gail. And I agree, I don't think these more sophisticated system work very well at all.
I mean at first I thought that there might be a lot of merit to the idea of rating answering posts, awarding answers, etc. But my experience with them has brought me to the opposite conclusion: the more complicated and "sophisticated" the scoring system and the question-to-response constraints, the less it actually works.
Here's a quote from the StackOverflow FAQ:
This is not a discussion board, this is a place for questions that can be answered!
The inherent problem here is that if 30 years in this field has taught me anything it's that you cannot answers peoples questions or solve their problems very well unless you have a conversation with them. And that would be my biggest concern about what is going on: a lot of the people there are not getting helped very much.
So on balance, what these forays into other approaches have taught me is that SSC had the right idea in the first place: simpler is better.
[font="Times New Roman"]-- RBarryYoung[/font], [font="Times New Roman"] (302)375-0451[/font] blog: MovingSQL.com, Twitter: @RBarryYoung[font="Arial Black"]
Proactive Performance Solutions, Inc. [/font][font="Verdana"] "Performance is our middle name."[/font]
June 7, 2009 at 9:28 pm
Heh... after reviewing the FAQ for StackOverflow, I have a couple of problems with that site... first, they want you to be both honest and nice. Most people interpret honesty with not being nice when it comes to code. "I'm sorry, but that'll cause your server some severe performance problems" is sometimes viewed as a great afront.
Also, they said this...
Like Wikipedia, this site is collaboratively edited. If you are not comfortable with the idea of your questions and answers being edited by other trusted users, this may not be the site for you.
Damned straight it's not for me. If I'm going to make an idiot of myself, I don't need any outside help. 😛
--Jeff Moden
June 8, 2009 at 10:49 pm
We have debated about some of the features in Stack Overflow here, especially as they've gained a lot of popularity. I'll post a few threads to get discussion from you, but it's interesting to see the debate.
One thing I would note as well: the person asking the question is often least qualified to decide if something is an answer. Maybe it's semantics, but perhaps it solves the immediate issue, and should be worded as such, but might not be a solution or an answer. I've thought more about multidimensional voting where you could say "this works" (or not) but also "recommended" or not and perhaps "a patch" or a "fix", or something like that. More than a thumbs up/thumbs down.
I do worry about Barry's points as well. I think SO tends to discourage discussion, which is a powerful learning tool.
Follow me on Twitter: http://www.twitter.com/way0utwest
Forum Etiquette: How to post data/code on a forum to get the best help
My Blog: www.voiceofthedba.com
June 9, 2009 at 12:46 am
My favorite is sqlservercentral.com or the SQL Server site of Microsoft / MSDN
Keep up the good work 🙂
June 11, 2009 at 11:51 am
Jeff Moden (6/7/2009)
Doc (6/7/2009)
Lynn Pettis (6/5/2009)
I am curious, which sites do you use or frequent either asking for help or helping others?I don't use them, I'm just regularly visit them all, forums that you mentioned in the poll.
And I'm regular visitor of some 100 SQL blogs there is on the net 😀
My theory about asking questions is that it should be my last resort before searching the net and all the forums, it's very big chance that someone already asked the same question that I'm looking for and someone already answered it.
I think this makes me a one big LURKER 😛
Doc for President! 🙂
P. of all lurkers :hehe:
-------------------------------------------------------------
"It takes 15 minutes to learn the game and a lifetime to master"
"Share your knowledge. It's a way to achieve immortality."
June 12, 2009 at 7:03 pm
Steve Jones - Editor (6/8/2009)
We have debated about some of the features in Stack Overflow here, especially as they've gained a lot of popularity. I'll post a few threads to get discussion from you, but it's interesting to see the debate.One thing I would note as well: the person asking the question is often least qualified to decide if something is an answer. Maybe it's semantics, but perhaps it solves the immediate issue, and should be worded as such, but might not be a solution or an answer. I've thought more about multidimensional voting where you could say "this works" (or not) but also "recommended" or not and perhaps "a patch" or a "fix", or something like that. More than a thumbs up/thumbs down.
I do worry about Barry's points as well. I think SO tends to discourage discussion, which is a powerful learning tool.
I've been thinking about this... as you say, the OP is probably the least qualified to judge. But, what if you had a heavy hitters list and only the heavy hitters could judge?
--Jeff Moden
June 12, 2009 at 11:46 pm
Jeff Moden (6/12/2009)
Steve Jones - Editor (6/8/2009)
We have debated about some of the features in Stack Overflow here, especially as they've gained a lot of popularity. I'll post a few threads to get discussion from you, but it's interesting to see the debate.One thing I would note as well: the person asking the question is often least qualified to decide if something is an answer. Maybe it's semantics, but perhaps it solves the immediate issue, and should be worded as such, but might not be a solution or an answer. I've thought more about multidimensional voting where you could say "this works" (or not) but also "recommended" or not and perhaps "a patch" or a "fix", or something like that. More than a thumbs up/thumbs down.
I do worry about Barry's points as well. I think SO tends to discourage discussion, which is a powerful learning tool.
I've been thinking about this... as you say, the OP is probably the least qualified to judge. But, what if you had a heavy hitters list and only the heavy hitters could judge?
StackOverflow has proven to me that that just makes it worse. Having 10,000 points for being a Java Jockey does not qualify you to decide whether an Excel VBA question should be closed or not, but that's exactly what happens. And the issue of "religious" prejudices probably started out as trivial, but a year of being able to make capricious decisions without ever really having to defend them has apparently gone straight to their heads and their egos.
I just read through the support and suggestions site ("UserVoice.com") and that was so hair-raising that I am about ready to give up on it, even though I still want to support Brent Ozar and some others. I did come across the best description I've heard yet for what's going wrong with SO: "Expert's Exchange without the money."
[font="Times New Roman"]-- RBarryYoung[/font], [font="Times New Roman"] (302)375-0451[/font] blog: MovingSQL.com, Twitter: @RBarryYoung[font="Arial Black"]
Proactive Performance Solutions, Inc. [/font][font="Verdana"] "Performance is our middle name."[/font]
June 13, 2009 at 12:36 am
Heh... wasn't talking about SO. I was talking about here. And by heavy hitters, I'm not necessarily talking about folks with a lot of points. I'm talking about a group of folks that have truly shown their starch on SQLServerCentral.com. Just as an example, Florian is one of those folks that has shown the right stuff even though he doesn't have a bazillion points.
--Jeff Moden
June 13, 2009 at 12:45 am
Who decides who can vote and who can't, Steve? I may be wrong, but are you sure he wants to decide who is qualified to make those decisions?
June 13, 2009 at 4:36 am
Lynn Pettis (6/13/2009)
Who decides who can vote and who can't, Steve?
Exactly the problem.
SO attempts to solve that by giving points for 'good' and 'correct' answers, and that kinda doesn't really work.
MSDN lets the OP decide, which kinda, doesn't work.
SADeveloper.net (down at the moment) let's any registered user mark or unmark an answer as 'correct' and shows who did so. Problem with that is that I say one guy who went and each time he answered a question, marked his answer as correct (Which I changed a couple times). So that kinda doesn't work.
If Steve decided unilaterally, there would be people who disagreed with him, either because he didn't select them, or because he selected people they felt weren't qualified. There would be flaming emails and angry posts and lots of bruised egos.
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
June 13, 2009 at 8:31 am
MSDN lets the OP decide, which kinda, doesn't work
Yes but a moderator can override the OP and the forum owner can override the moderator in most of the forums. When I was there most of my moderating work was making sure OP does not close thread with their own answer if a user posted a valid answer.
If you are a programmer developing in .NET you can get 80% of the points you have earned, but in the SQL Server forums only very busy ones comes with paid moderators so you get less than 50% of the points because none paid moderators in all forums on the MSDN choose a few users and favor their answers over other users.
That is the reason I was made a moderator of most of the base class including C# general the owners of the forums and the paid moderators marks my answers but the none paid seldom mark my answer even when my answer is technically better than their answer.
SADeveloper.net (down at the moment) let's any registered user mark or unmark an answer as 'correct' and shows who did so. Problem with that is that I say one guy who went and each time he answered a question, marked his answer as correct (Which I changed a couple times). So that kinda doesn't work.
That also happens on MSDN both the forum owners and all moderators gives the OP useless answer and immediately close the thread. There is about 10 to 20 percent moderators may be more who just waste product development money because most of the bugs they ask users to post on Connect are bugs in user and their brains. I leave most of my threads open which the people who pays the bills don't like but I kill most useless bugs so it evens out.
Kind regards,
Gift Peddie
June 13, 2009 at 11:13 am
GilaMonster (6/13/2009)
Lynn Pettis (6/13/2009)
Who decides who can vote and who can't, Steve?Exactly the problem.
SO attempts to solve that by giving points for 'good' and 'correct' answers, and that kinda doesn't really work.
MSDN lets the OP decide, which kinda, doesn't work.
SADeveloper.net (down at the moment) let's any registered user mark or unmark an answer as 'correct' and shows who did so. Problem with that is that I say one guy who went and each time he answered a question, marked his answer as correct (Which I changed a couple times). So that kinda doesn't work.
If Steve decided unilaterally, there would be people who disagreed with him, either because he didn't select them, or because he selected people they felt weren't qualified. There would be flaming emails and angry posts and lots of bruised egos.
What about peer selection? Steve makes the intial selection of say 15 folks. From there, it's up to them... perhaps 10 of them have to agree to put someone in such a position or remove such a person once in. May a "voting board" forum visible only to those on the "board". Or would that turn into a ring knockers club?
--Jeff Moden
June 13, 2009 at 11:31 am
Jeff Moden (6/13/2009)
Or would that turn into a ring knockers club?
Maybe, but even if it doesn't, there'll still be the sense of elitism about it and there'll still be people angry that they weren't selected.
What I was considering, and I don't know if it'll work at all, is something similar to Stack Overflow's upvote/downvote system. Users (perhaps only those with > X points) can vote answers up (accurate, correct) or down (hard to understand or dead wrong). Perhaps only allow users a certain number (small number) of votes per day (perhaps dependant on points, perhaps not) and don't allow people to vote on their own answers (which, I suspect happens with the current ***** rating for posts). In addition, have a way to see exactly who voted on answers and what their votes were.
Thoughts?
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
June 13, 2009 at 12:14 pm
Hmm, before I jump into this any deeper let me ask: what would we be trying to accomplish by this? What shortcoming of SSC would we be fixing? Or what new desirable feature would we be adding?
[font="Times New Roman"]-- RBarryYoung[/font], [font="Times New Roman"] (302)375-0451[/font] blog: MovingSQL.com, Twitter: @RBarryYoung[font="Arial Black"]
Proactive Performance Solutions, Inc. [/font][font="Verdana"] "Performance is our middle name."[/font]
Viewing 15 posts - 16 through 30 (of 53 total)
You must be logged in to reply to this topic. Login to reply