Jean-Michel POURE wrote: > > Le Vendredi 10 Mai 2002 14:06, mlw a écrit : > > Sorry, I'm not interested in a cygwin version of PostgreSQL. I think it > > will do more harm than good. If we make it something that people want to > > try, and then they TRY it, they will find that is sucks, then we lose. It > > is very hard to remove the bad taste in ones mouth of a poor product. Think > > Yugo. > > Cygwin is very stable. Its community is relatively small but very actuve. We > could well provide a unique installer to "hide" Cygwin from the user. This > can be done compiling Cygwin.dll in a separate user space, as per discussion > with Dave Page.
Here are the problems with cygwin: (1) GNU license issues. (2) Does not work well with anti-virus software (3) Since OS level copy-on-write is negated, process creation is much slower. (4) Since OS level copy-on-write is negated, memory that otherwise would not be allocated to the process is forced to be allocated when the parent process data is copied. As a product manager, I would not commit to using a cygwin application in production. Do you know of any long-uptime systems using cygwin? PostgreSQL would need to run for months. I would view it as a risk. Lastly, a Windows program is expected to be a Windows program. Native paths need to be used, like C:\My Database, D:\My Postgres, or something like that. Native tools must be used to manage it. > > > I have no patience with designed to fail projects, certainly not with my > MySQL under Windows is based on Cygwin. > MySQL sucks and has a 'huge" success. Define "Success" > > So let's do it in three moves : > > - first move : gain a large audience providing a stable release of Cygwin + > PostgreSQL. This could be done within days ... not weeks. This will be much > better than MySQL. No interest in cygwin, sorry. ---------------------------(end of broadcast)--------------------------- TIP 4: Don't 'kill -9' the postmaster