Mib Software Newsrate Usenet Server Comparisons


Why are results reported by heirarchy?


> 
> Btw, why do you break your results down into groups? I can see that the
> results vary, but I have no idea why that should be.

The results are reported separately for each heirarchy BECAUSE they vary between heirarchies.

The reason for this is that the software used to transfer news has the capability to treat subsets of newsgroups differently. Some sites don't carry alt.* at all, some sites don't carry alt.binaries*, or give those articles lower priority.

Articles propagate around the net among cooperating sites in "first-to-offer" fashion. Most news machines exchange articles with only a handful of other news machines. There is no central machine responsible for sending messages.

So, if not all sites carry a newsgroup, you can imagine that it takes those articles more steps to get delivered. They may not even be delivered at all, if the article hits a "dead end", or a significant bottleneck. Hopefully, that article gets to the news machine through another path, but that doesn't always happen, either.

It would be very unlikely for a single newsgroup to get propagated differently than similar newsgroups, and this is the basis for the sample. Most sites don't "single out" individual newsgroups for special treatment. They work on whole categories, like alt.*, or alt.binaries.*. So you can expect that the performance in comp.lang.java.help matches the overall comp.* numbers very well!

More Detailed Topics
View an example report
What is considered "excellent" or "good" or "poor" performance?
Why is my alt.* or alt.binaries.* performance so much lower?
Why do some sites decide to not carry binaries or alt.*?
How do news articles get "missed?"

Overview and Related Topics

How do I signup?

Up to The newsrAte RKT
Up to newsrAte home
Up to Mib Software home

Copyright 1998, Forrest J. Cavalier III, Mib Software
INN customization and consulting
Email: newsrate@mibsoftware.com
Voice: 570-992-8824