I wouldn't be surprised to see the application's still running on its original Sun E4500.
Although that would be getting on for 10 years old now.
Still, you have the option of being very condescending to them the next time you have to deal with them.
"The database is not your personal persistent object store. Other people actually have to use the data in it, and - believe it or not - they're more important than you."
Is something I had to say to a developer once.
In fact, I really want a facility to head butt people through the internet.
@SimonRigharts I think your chances of getting called up are fairly low, and IIRC the criteria are something like: must have no indictable convictions that carry a prison sentence and (IIRC) not related to someone who matches the above criteria. Can't remember the exact details - I just did the data model, someone else developed it.
I could do some napkin math to figure out an approximate chance, but :effort:
wild-ass guess would be 1% per year
Here's something random - in sybase, the name of stored procs is stored in sysobjects. the text of stored procs is stored in syscomments in 256-byte chunks. If I wanted to check that the definition of the stored procs was the same what would be the easiest way?
... oh wait it's obvious now that I think about it, can just do a chunk-by-chunk comparison
I have a table (Tran) with transaction records.
I have two tables (Parameters1 and Parameters2) which are used to filter results when selecting from the Tran table a la:
SELECT *
FROM
Tran t JOIN
Parameters1 p1 ON T.code1 = p.code1
Parameters2 p2 ON T.code2 = p.code2
WH...
I'm not looking for an answer for this in chat, because I think it would make a good question on the main site, but I'm not quite sure how to ask it properly:
I need PIVOT-like behavior from Oracle 10g. I've got a table like OBJECT_NAME ATTRIBUTE_NAME ATTRIBUTE_VALUE 'object 1' 'color' 'red' 'object 1' 'size' 'big' 'object 2' 'color' 'blue'
And I'd like to make a query that returns this: OBJECT_NAME COLOR SIZE 'object 1' 'red' 'big' 'object 2' 'blue'
well my formatting failed but you get the idea
basically I want to de-normalize the table a bit for a report
it's oracle 10g. I believe 11g literally has a PIVOT keyword that does exactly this, but there must be a way to achieve this pre-11g?
lol, funniest way I've heard that described so far today.
Oh! I understand what you mean about EAV, and as I described my table it would indeed suffer from that problem. There is an important difference between the example table I made up for that question and my actual table
actually lemme just tell you as much as I'm allowed about what I'm actually doing, that would be easier.
I've got this data analysis that looks at pictures of these spherical objects, and identifies surface defects. The defects are put into various categories based on their appearance. The table records the results of this. So the columns are actually: Object_ID Defect_Category Number_of_Defects Total_Area_of_Defects
lets say that the defect categories are 'bigRound' 'smallRound', and 'long'
I'd like to make a report that says 'Object 1' has 5 'bigRound', 20 'smallRound', and 0 'long' defects.
So it's not really EAV. You could sum over Number_of_Defects and get a meaningful result. You might do this for example: SELECT Object_ID, Sum(Number_of_Defects) FROM my_table GROUP BY Object_ID;
SELECT Object_ID,
Sum(CASE WHEN Defect_Category = 'long' THEN Number_of_Defects ELSE 0 END),
Sum(CASE WHEN Defect_Category = 'bigRound' THEN Number_of_Defects ELSE 0 END),
Sum(CASE WHEN Defect_Category = 'smallRound' THEN Number_of_Defects ELSE 0 END),
FROM my_table
GROUP BY Object_ID;
yeah the speed isn't the issue with that. I can definitely do it in client code. I just figured that doing it in the select would be cleaner and more correct
doing it in a single call also has the benefit that I can make a summary table in an excel spreadsheet to show to non-DB savvy people who are interested in the data
and doing it with dynamic columns means that if we introduce a new category, the sql doesn't have to get rewritten.
Also, giving people the data in Excel and letting them pivot is often easier :p (they like the control, makes them feel better about themselves)
@ColinK no, you can do a string concat'ing based on the values in the one select statement (to get the category names) to build the SQL used to create the pivot
So it becomes a two step process, but it's rather straightforward at that point.
@ConcernedOfTunbridgeWells that's why I asked how they get it now, because if it's not a custom UI then it sounds like he's doing a lot of ... extra work for nothing
I've noticed that people asking SQL-Server questions sometimes tag their questions with the SQL and Server tags or just the Server tag rather than the SQL-Server tag. Often the question isn't about SQL at all and Server doesn't necessarily add any information unless it means SQL-Server. As of t...
I am using Oracle 11g Global Temporary Tables as I need a solution where I can add rows to a temporary table for a join, and I want only the rows added to the temp table for the Oracle Connection/session to be included. I am using Global Temp Table in Oracle becuase I want the table to exist betw...
yea ... the trouble is it's much harder to get benchmark data when you are dealing with a production system (like i don't want to take down SO to get said data :) )
0 failures so far and we are at over a year with the db servers
it's actually starting to look liek the reason we are going to have to swap them out is due to space limitations - the DB tier is running on older gen intel X25's which only came in at 64GB
We recently moved a Zeo instance over to a new server environment and one of the changes was the file system now has the database files stored on an NFS share.
When trying to start zeo, we've been getting lock file errors which after researching seems to be because of a known issue of lock files...
It sounds like you're talking about a classic example of Incubation.
Here's a great article by John F. Kihlstrom: Intuition, Incubation, and Insight: Implicit Cognition in Problem Solving. Basically it is believed that Incubation or stopping conscious thought on a problem allows one to find more...
Idea I wanted to run by someone, especially a mod: I was thinknig of asking on the UX meta site to gather questions they want on the Cog Sci beta. Lots of them have questions, but most didn't want to commit to the proposal because they weren't subject matter experts. Think it would be okay?
I'd be vetting the questions and not just trying to spam the beta, but I still wasn't sure if that'd be cool or not. I think the UX guys have lots of relevant questions but weren't as cool on joining the beta itself.
also use the proposal Q to spawn new Q on the site
stuff that's highly ontopic according to the site definition
and if you work a lot on the site in the first few days, and help aggressively shape things, you can nominate for the pro-tem, and probably get your own diamond
it's a comment directed at Googlers who land there looking an answer to a related question: What is the current proc I'm in? Instead of the question the OP asked: What proc called the proc I'm in?
@jcolebrand how does this sound: [The Cognitive Science proposal][1] site just went into Private Beta. I know a lot of users here have questions but didn't feel they could commit to the site. Please post questions you would like to see asked as answers here, and I will post questions that will be good for the site.
Remember these are questions about *cognitive sciences* like cognitive ergonomics, cognitive psychology, behaviorism or neuroscience. I'm not asking for UX questions we can bug Cog Sci with, but rather real Cognitive Science questions you have.
@BenBrocka I think more like ... "If you have a Q you think would help define the site, but didn't think you could commit completely"
@NickChammas I thought I did ... what am I missing? that the answer I'm giving doesn't quite help? You still have to return the value, but don't make it a string when it can generate for you
The Cognitive Science proposal site just went into Private Beta. I know a lot of users here have questions but didn't feel they could commit to the site's. Please post questions you think would help define the site, and I will post questions that will be a good fit.
Note you won't be able to see...
Cognitive Science private beta has just started. I would imagine there will be an overlap with this new site. How will we handle this overlap?
Can we draw a line what is out of scope here and better refer to Cognitive Science?
P.s.: I started a similar topic on Cognitive Science.
@NickChammas I don't know of any way except via parameters. In one sense, in makes no difference: like some OO class method shouldn't know who called it. Encapsulation. The stored proc should do one thing and not care about calling state.
Is there a way in SQL Server 2008 R2 to cause a timeout for a database modification involving a transaction? We have a scenario where our application code hangs or throws an exception and fails to perform a rollback or commit. This then causes other sessions to hang waiting for the transaction to...
I have a table (Tran) with transaction records.
I have two tables (Parameters1 and Parameters2) which are used to filter results when selecting from the Tran table a la:
SELECT *
FROM
Tran t JOIN
Parameters1 p1 ON T.code1 = p.code1
Parameters2 p2 ON T.code2 = p.code2
WH...
I've noticed that people asking SQL-Server questions sometimes tag their questions with the SQL and Server tags or just the Server tag rather than the SQL-Server tag. Often the question isn't about SQL at all and Server doesn't necessarily add any information unless it means SQL-Server. As of t...
Cognitive Science private beta has just started. I would imagine there will be an overlap with this new site. How will we handle this overlap?
Can we draw a line what is out of scope here and better refer to Cognitive Science?
P.s.: I started a similar topic on Cognitive Science.
