New Python development process (SourceForge considered Harmful?)

[John W. Baxter]

In article <etdem20d43x.fsf at nerd-xing.mit.edu>, Alex 
<the_brain at mit.edu> wrote:

> > No, I can't.  I believe there is a delay based on the way that pages
> > appear: the ad appears quickly, then there is a considerable delay
> > before the rest of the content is sent. This is common technique on
> > the internet.
> > 
> > On the other hand, if you've looked at the source, and the delay isn't
> > there, then I'm wrong and I retract my comments.
> 
> Banner ads are usually served from a completely unrelated machine
> administered by the company brokering the ad space, aren't they?  I
> would have expected those machines to work faster than the ones serving
> up the sourceforge content proper.
> 
> Alex.

And if the browser encounters the links for the ads first (I haven't 
looked at the html), most will work on them early and use up  some or 
all their allowed (preference) connections doing ads before doing 
content.  (The Mac and Windows browsers I've looked at ship set to make 
a maximum of four simultaneous connections, which is reasonable for 
dailup and and 56K Frame Relay, and not foolishly small for ISDN.)

So getting the ads first wouldn't necessarily take any magic code in the 
server itself, other than putting the ad links physically "high" in the 
html.

I'm as good at ignoring the ads before the content arrives as after.

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwbnews at scandaroon.com