I've got an Apple ][, several MITS boxes that predate the 8080, and other junque that you probably can't even identify, much less give me grief about... Unfortunately I'm quite a chunk less than 60.
I do much of my modeling in my head, using a notation that gave birth to IDEFX. I often sketch ideas using pencil and paper, marker and whiteboard, and sometimes even the stray coctail napkin.
I've used most of the modeling tools that are commonly available today, along with a number of nieche tools too. When you get right down to it, the tools are a two-edged sword... They make it easier to duplicate a model, and in a few ways make it easier to manipulate one, but in the long run they are just crayons on steriods if you know how to think about modeling, and rather dangerous crutches if you don't.
I've got an Apple ][, several MITS boxes that predate the 8080
Cool, perhaps donate a coupld of MITS to the Computer Museum?
CASE tools, such as the one's i'm trying to get a good list of here, are excellent for data modeling on a large scale. Perhaps if its only a small db then paper and pencil is fine, but when your in the 50+ table mark I dont think there is shame looking to a CASE tool for aid, especially when more than one person is working on the project.
For readers who don't know if they should be useing them I advise once you've mastered paper and penicl, and you want to go big, move to these tools.
They help tremedously with clear reading, easy editing, time saving (big projects) some have good error checking and multi target DB table generation and most of all you cant spill your tea over them.
I'm with you on that point, but I guess the thing that I've seen with data modelers is that they generate almost religious attachment to the tool, with little or no regard for the underlying process.
If someone is a be-all-and-end-all bigot for ErWin, they don't see anything of value in PowerDesigner. The PowerDesigner bigot sees nothing worthwhile in Rational Rose. The Rose lover sees no value in Visio Enterprise. These folks have gotten tangled in the tool above and beyond how it helps them do the job at hand, and often become so dependant on the tool that the user can't cope with any structure that their tool of choice can't model.
There are a number of database engines that are still "not ready for prime time" that expose features much closer to true relational databases than SQL can get. Most of these modeling tools have a very hard time modeling the kinds of things those engines expose, and none of the tools that I've seen can generate declarative code for those engines without significant help.
As you pointed out, a modeling tool makes it easier to duplicate a completed model (just print new copies in most cases). Based on what I've seen though, most modelers can spill coffee (or tea) into their keyboard just as easily as they can spill it over their printout... I'm not sure that that really buys you much of an advantage!
As far as making this a sticky, I don't see a whole bunch of value in that myself. A sticky is good for announcing policy, etc. It has some limited value for general announcements. It has relatively little value in my opinion for announcing the preferences of the users for third party products like modeling tools, since those opinions will change over time and based on users needs and experience. Most opinion stickys become a kind of digital "Killroy was here" leading new readers to wonder who was Killroy and why should they care!
two of the features i really loved about ERwin (disclaimer: ERwin is the only tool worthy of the name that i've used) were the reverse engineer and the compare
obviously, reverse engineering a database into a modelling tool can save enormous amounts of time
the compare was absolutely invaluable in managing schema versions where the database had been "hotfixed" by the dba, who decided -- can't imagine why -- to alter some tables without asking through the proper channels, i.e. forward generating from the official ERwin model
my comment about paper and pencil had to do with modelling relationships and not about politics
you will find that many so-called technical problems are human problems
not the least of these is the need to communicate, and when this is factored into the overall equation, it becomes clear that which tool you use is not nearly as important as that you use a tool
with regard to ERwin...
ERwin was the most difficult, bloated, convoluted, user-hostile, unfathomably devious, unreliable, frustrating piece of software i've ever used
on the other hand it performed some tasks admirably
if you can get over the four-week learning curve, and plan to engineer models for longer than four weeks, the intellectual ROI might be worth it
if you're looking for something to generate a few dozen tables, look for a simpler tool
As far as making this a sticky, I don't see a whole bunch of value in that myself.
I just thought that an Applicatios and Tools forum should have some sort of instant ref for peeps to know what tools are available to help them, with advice and opinions from experieced users. Kinda like an FAQ.
FAQ Stickies help prevent repatative questions and saves time for the questioner. Perhaps the thread should be named "DB Tools and Applications Ref"
btw u mentioned a few tools i've never heard of, thx I go check them out.