Concerning the thread: microsoft.public.sqlserver.programming Nov 10, 2008 "cascading deletes" http://tinyurl.com/6nwjmd For some reason I couldn't get my reply to post thru OE (it got thru via google though). Perhaps there's an MS filter for metaphors ☺ In any event any mature adult should be able to handle it. So here's my reply with a touch of creative writing ☺ 'Most women believe men think with their tool. And it's just as true in IT. Users model business problems in terms of the abilities of their db. The idea that modeling exists independent of ones db is a myth. It's not a question of seepage but of flooding. Modeling business problems in terms of the available sql server constructs is messy precisely because their immature and superficial to the task. The result is you turn away from the db and towards the do-it-myself model. You roll around in your own layer because you can't get layered by the db. It's ridiculous to write a join procedurally but when it comes to modeling it's perfectly acceptable to roll your own. Because the model equivalent of the join is so lacking and messy. The genie isn't going back in the sql server bottle. It's simply to far gone. That's why I advocate Dataphor. There the genie is in the join as well as the modeling. Use Dataphor and put your tool back where your head and shoulders are. You can still use sql server. But you aren't going to get tooled by it :)' www.dataphor.org www.beyondsql.blogspot.com Geoff Schaller wrote: > Andre. > I vote with Hugo here. We manage everything from code, not from TSQL in > SSMS or via some other mechanism so we generally have to code everything > (and that is not as difficult or as expansive as it sounds). Whilst > cascading referential integrity is "nice" from a simplicity point of > view, we've found that the act of deleting something (say an invoice) is > almost never a simple thing. There is reversal of stock levels, > rebalancing totals and if others are running reports when you thought > you wanted to do the delete, it gets messy. > The other thing is that we quite often have to delete the child entries > individually or prevent the parent from being deleted because a child or > two cannot be. Writing all that logic into a trigger and enforcing the > rollback is quite complex. I find code an easier way to manage and > maintain this. To add insult to injury my reply to a post on SQLServerCentral was hacked (edited). SQLServerCentral 11/10/2008 'Arrays in Stroed Prcoedure' http://tinyurl.com/5th5n4 My reply, as shown there under the name rog pike, was edited to read: 'An array is a 'type', something the archaic computer science of sql knows nothing about. You have to move to a 'real' relational system to find a list/array type. You'll find such adult computer science in Dataphor.' My orginal reply was a follows: 'Arrays are in sql server in the same sense as having sex by yourself which may account for the shortsightedness of so many sql mavens. An array is a 'type', something the archaic computer science of sql knows nothing about. You have to move to a 'real' relational system to find a list/array type. You'll find such adult computer science in Dataphor.' Is the site for mature adults or for the whole family? ☺ Just how much protection does sql and its users need? Is this a security or, better yet, an insecurity problem? ☺ Finally, I'll repeat here what I posted in the above thread: 'Apparently someone complained/reported something I wrote as being objectionable. They got their wish as it was magically extracted from the text. What was yanked, besides my chain, was a metaphor, albeit a vivid one, to drive a salient point home. Now I write for adults, I don't do child-speak very well. Nor do I have a predilection to only write drone-on-speak. So, if I can, I won't hesitate to use an adult metaphor to amplify a point in an industry that is usually tone deaf. God forbid IT encourage ability in something other than code or pixels. So if you are an adult, with a surname other than anonymous, please explain just what you found R or X rated. Mature adults usually confront conflicts thru the front door not the back one.'
Dataphor SQL RAC (Relational Application Companion)
A site of hope for those looking for a true relational database system
- a one-one requirement constraint with dataphor (1)
- anatomy of sql server part I - what is a stored procedure (1)
- anatomy of sql server part II - the unit test as part of the database (1)
- anatomy of sql server part III - what does deferred name resolution really mean (1)
- censoring sql posts (1)
- creating an opposite constraint in dataphor (1)
- dataphor (2)
- Dataphor (7)
- dataphor # 13 a table as a parameter (1)
- dataphor - download and start working with it (1)
- dataphor - fixed sized word segments (1)
- dataphor # 10 sql mythology (1)
- dataphor # 11 string differences (1)
- dataphor # 12 trimming a string (1)
- dataphor # 14 sql the meaning of Update..From (1)
- dataphor # 15 views with substance (1)
- dataphor # 16 inclusive vs exclusive solutions (1)
- dataphor # 17 a visual look at ranking queries (1)
- dataphor # 18 data scrubbing using lists (1)
- dataphor # 19 create intervals over strings (1)
- dataphor # 20 browsing an sql window (1)
- dataphor # 21 an example of relational division (1)
- dataphor # 22 reusable procedures (1)
- dataphor # 23 repley to Michel (1)
- dataphor # 24 basics of the table type (1)
- dataphor # 25 extending the dense rank function (1)
- dataphor # 26 query a hierarchy with explode (1)
- dataphor # 27 combine strings with Split and Concat (1)
- dataphor # 28 constants and variables or sql and D4 (1)
- dataphor # 29 another example of relational division (1)
- dataphor #1 introduction (1)
- dataphor #2 splitting strings (1)
- dataphor #3 string concatenation (1)
- dataphor #4 comment (1)
- dataphor #5 comment (1)
- dataphor #6 formal definition (1)
- dataphor #7 sql: table this (1)
- dataphor #8 list to table (1)
- dataphor #9 table constraints (1)
- dataphor creating lists in a query (1)
- extracting numbers from a string with dataphor (1)
- jeff modens dynamic crosstabs for sql server (1)
- linq to sql the what and why (1)
- linq to sql as a window of opportunity to sql users (1)
- linq to sql should be important to sql users (1)
- linq to sql vs. older 4GL attempts (1)
- listing missing table item (1)
- Multiple cascade paths to the same table (1)
- RAC (4)
- RAC #1 comment (1)
- RAC #2 example (1)
- RAC #3 finding the Nth number in a string (1)
- RAC #4 Sql Server 2005 ranking functions vs. Rac ranking (1)
- sorting a delimited string by its numerical string parts (1)
- sql an example of extreme implicit conversions (1)
- sql can't handle complicated cascading updates (1)
- sql CTE should be a variable not a value (1)
- sql dense rank for identifying consecutive runs (1)
- sql is there really a table variable (1)
- sql ranking functions explained by relational types (1)
- sql server triggers are best set based (1)
- sql the idea of using substring to simulate lists (1)
- sql the undefined trigger in Sql Server (1)
- sql vs relational on tables (1)
- sql what the sql CTE covers up (1)
- types and procedures (1)
Friday, November 14, 2008
S(ecure) Q(uery) L(anguage)?
Subscribe to:
Post Comments (Atom)
1 comment:
i am hobby illustrator, is there a chance you'd like to publish some of my photographs? i guess it would be cool and fit on your site :-)
absolutely enjoy your site structure! write me a mail please in case you want to colaborate
Post a Comment