Thread: beta2 or cvs?
Hi, I am intending using the latest and greatest postgresql to learn about how it works and where it is heading (the project I am on is in for the long haul, and won't need to be "in production" for months yet...), so is the best place to be for the time being cvs? I am running gentoo so we get the hottest of the press in terms of releases but would it be better to be even closer? I am not at all bothered about the odd bug, and would like to be involved in the testing process - a lot of projects just tell you to update to the latest cvs before reporting a bug, is that the general story here? Cheers Antoine -- G System, The Evolving GUniverse - http://www.g-system.at
Antoine <melser.anton@gmail.com> writes:
> I am intending using the latest and greatest postgresql to learn about
> how it works and where it is heading (the project I am on is in for
> the long haul, and won't need to be "in production" for months
> yet...), so is the best place to be for the time being cvs? I am
> running gentoo so we get the hottest of the press in terms of releases
> but would it be better to be even closer? I am not at all bothered
> about the odd bug, and would like to be involved in the testing
> process - a lot of projects just tell you to update to the latest cvs
> before reporting a bug, is that the general story here?
If it's convenient for you to pull from CVS, by all means do that.
Another option is to grab the nightly CVS-snapshot tarball from our ftp
servers (look in the dev/ directory).
If we were not in beta phase, I'd not necessarily recommend using CVS
;-) but since we are, CVS tip should theoretically be as or more stable
than the last beta.
regards, tom lane
Dear group, While designing 3-4 different schemas for my database, a question has been chasing me. I am not a CS graduate and I was never taught the nitty gritty's of Dabase schema or rules. My question is - a schema that has no foreign key defined (except in 1 or 2 instances), will it be considered a good schema or bad schema by experts. Imagine a scenario where most of the relations are defined by linker tables(relationship table) that is formed by two different primary keys from two different tables are defined to avoid foreign key relationships. creating a database in this kind of scenario, can it be considered a good database design. I hope it is not a stupid question. If it is please excuse my ignorance. Thanks Kumar. __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com
Kumar, > While designing 3-4 different schemas for my > database, a question has been chasing me. I am not a > CS graduate and I was never taught the nitty gritty's > of Dabase schema or rules. Me neither. I learned the hard way, on the job. > My question is - a schema that has no foreign key > defined (except in 1 or 2 instances), will it be > considered a good schema or bad schema by experts. Bad, unless there's only 2 or 3 tables for your 1 or 2 joins. > Imagine a scenario where most of the relations are > defined by linker tables(relationship table) that is > formed by two different primary keys from two > different tables are defined to avoid foreign key > relationships. creating a database in this kind of > scenario, can it be considered a good database design. But in that case you should have FKs from the linker tables. -- --Josh Josh Berkus Aglio Database Solutions San Francisco