[cgiapp] Re: Problem with Flash, Session & persistent environment identified

[Ron Savage]

Hi Mark

On Sat, 2008-11-22 at 20:27 -0500, Mark Stosberg wrote:
> > Here the new text:
> > 
> > http://savage.net.au/Session.html#A_Warning_about_Auto_flushing
> 
> 
> Thanks for the draft, Ron. Here's some feedback:
> 
> - I would refer to "signal handlers" rather than "sig handlers"

Agreed.

> It says: "See chapter 16 in the camel book for information on sig handlers, and, even better, don't use them."
> 
> I would remove ", and even better, don't use them.". Signal handlers have their
> place. They are the only way to accomplish some things.

I've changed this text too.

A new version has been uploaded.

> It says: "Again, this is not a problem with CGI::Session, but with the program
> using it."
> 
> I would just omit that finger pointing. From the user's perspective, they combined two
> modules and got a buggy behavior which couldn't exist about both of the modules being
> in use. 
> 
> Overall, I found it a bit verbose, but some people may like the explanation.

The text I wrote has been designed to block the fingers being pointed at
CGI::Session.

Also, it directs programmers strugging with these issues towards a
broader view of the factors involved, and towards debugging strategies.

So, I prefer to keep the current text.

As for the verbosity, it's better to have an explanation of possible
causes in the POD rather than generating traffic on this mailing list,
where the same topics might end up begin discussed repeatedly. This is
one of the curses of the Image::Magick mailing list, for example.

The ideal situation would be for the POD to be restructured to have an
FAQ, but I don't think it's worth that much effort. Maybe later.

-- 
Ron Savage
ron at savage.net.au
http://savage.net.au/index.html