[ZWeb] Re: DISCUSS: Moving downloads to SourceForge's download manager
Wed, 20 Nov 2002 09:40:08 +0100
I think the responses were largely favorable on this. Sidnei and I
will talk about how to implement and come back with something for
On lundi, nov 11, 2002, at 15:34 Europe/Paris, Paul Everitt wrote:
> I believe this subject came up a couple of weeks ago but I can't find
> it in the archives.
> Sidnei and I were just discussing the web stats for zope.org. He
> noticed that around 50% of the bytes transfered were from two things:
> Zope distributions and the PDF of the Zope book (the latter is around
> He and I thought it would be good to revisit the proposal of moving
> these items into SourceForge's download manager. Here are some pros
> and cons:
> 1) It's good for our customers. The downloads will be a lot faster,
> especially in different regions of the world.
> 2) It has better uptime (particularly with different systems around
> the world).
> 3) It removes stress from an overstressed situation with zope.org. I
> think we all realize that the more we remove from zope.org's plate the
> better chance of success.
> 4) We will actually get numbers that show how often Zope is being
> downloaded. Hallelujah!
> 5) We *might* attract some attention by being on the SourceForge
> activity percentile radar.
> 6) The current download pages on zope.org are less friendly, and some
> (maybe not many) people will be familiar with the sourceforge
> 7) The bandwidth costs would go down a LOT (50%), particularly during
> spikes (e.g. when a new version of Zope is released).
> 1) It's a transition and transitions have costs and disruptions.
> 2) The new interface might be harder for posting updates. Though I
> doubt it, as the current system is one thousand percent brain damaged.
> 3) There might be changes with nzo. (On this, I don't think there's
> any *finished* work to significantly improve the way packages are
> 4) Some people might think it looks bad for Zope to not serve up its
> own downloads.
> Sidnei and I both view this as more pro than con, and are willing to
> do the work to investigate the feasibility of this.