site Search:


 
    All Forums Hot Topics Gallery






how-to block ads


 
Search Topic:
Uniqs:
98
Share Topic
Post a:
Post a:
AuthorAll Replies


NetAdmin1
CCNA

join:2008-05-22

2 edits

Yet another flawed study...

Why is it that every study of broadband suffers from the exact same flaws - lack of a scientifically gathered and properly controlled data set? This study, while it appears the analysis looks solid and the conclusions are well founded, suffers from the problem that the data used (Speedtest.net's data) has quite a few problems associated with it.

Take a look at the latency numbers in the study. Those numbers are problematic because so many factors can affect the latency value that you really can not draw any conclusions from it. The observable fact that two servers, relatively close to you geographically, can have wildly different latencies from you illustrates the problem. The server admins choice of provider, that providers choice or routing policies and network design, the peering choices and transport choices, you providers' network design and policies, and on and on all affect latency. So, while you may have a sweet, 50Mbps fiber optic internet connection, the possibility exists that a person on a piddly little 256kbps connection can have a better RTT than you. Basically, you can try to draw conclusions from the latency numbers in the data, but because so many factors impact latency, you can't make any meaningful conclusions about it on with respect to the connections quality.
--
---
Eleven years of carrying The Clue Bat...


vzw emp

@qwest.net

said by NetAdmin1:

Why is it that every study of broadband suffers from the exact same flaws - lack of a scientifically gathered and properly controlled data set? This study, while it appears the analysis looks solid and the conclusions are well founded, suffers from the problem that the data used (Speedtest.net's data) has quite a few problems associated with it.
Studies relating to broadband are never totally accurate because those with the information (the ISP's) needed to do an accurate study do not release the required information. They will hide behind the veil of trade secrecy to avoid having to disclose the truth about their service, what's offered, where it's offered and for what price. They can't be held accountable for what no one can prove (and believe me, if the numbers were good we'd have heard about it from an infinite stream of advertisements bombarding every form of media known to man).

Friday, 01-Jun 04:04:29 Terms of Use & Privacy | feedback | contact | Hosting by nac.net - DSL,Hosting & Co-lo
over 12.5 years online © 1999-2012 dslreports.com.
Most commented news this week
Hot Topics