Subject:
|
"jump.cgi" considered harmful ? (1)
|
Newsgroups:
|
lugnet.admin.general
|
Date:
|
Thu, 23 Nov 2000 06:40:24 GMT
|
Viewed:
|
432 times
|
| |
|
|
As preface, for those that don't know what the LUGNET jump.cgi does...
It is a way to track stats on who goes where. For example, when I type
http://www.miltontrainworks.com into a post, the web interface rendering
technology shows the URL in different text color/font and further, modifies
the URL to add a jump.cgi invocation underneath (that is... the resource
locator actually followed becomes
http://news.lugnet.com/jump.cgi?http://www.miltontrainworks.com (2)
I know why this is considered a good thing.
And I recall discussion about the delay introduced being minimal.
Recently, though, I am seeing that the delay introduced is going up,
significantly. In fact, so much delay is being added, that on this brand new
750 Mhz P3 384m Windows 2000 system (where I haven't tweaked any of the time
parameters)... on this fairly fast system, when I am dialed in, about 1/2
the time IE times out. I have to then manually strip off the jump stuff to
actually get to the URL.
Hence the subject... is it just me? Is anyone else seeing this? Has it
gotten slower over time? Can it be sped up at all? Can it be turned off? I'd
complain less if I had access to the stats, I guess, because I'm curious
about usage and stuff. But I hate paying performance penalties, and this is
a problem where throwing more sand at it (this machine is 50% faster than my
old one) didn't help.
Where I care the most is when I am previewing one of my own posts and want
to check the URLs. If I get a timeout then, it can be bad news for the
posting process. Can it be suppressed just there? Those chases aren't
statistically valid anyway, it's just the poster checking his work, not a
real reference.
1 - with apologies to fans (and anti-fans) of GOTOs
2 - it will be interesting to see if this particular reference gets
"doublejumped" once I post this.
++Lar
|
|
Message has 4 Replies:
 | | Re: "jump.cgi" considered harmful ? (1)
|
| Are you sure it's not latency in DNS resolution? That is usually a big factor in how fast a page comes up. A good test is to try the same link again with your browser cache turned off and see what happens. While I understand the usefulness of the (...) (24 years ago, 23-Nov-00, to lugnet.admin.general)
| | | Re: "jump.cgi" considered harmful ? (1)
|
| (...) I’m no expert on this, but I’m pretty sure that, above a certain level, the end users hardware will not affect the problem you mention. The slowness will occur somewhere between the LUGNET server and your computer. For what it is worth, I have (...) (24 years ago, 23-Nov-00, to lugnet.admin.general)
| | | Re: "jump.cgi" considered harmful ? (1)
|
| (...) It's not to track stats on who goes where, although an evil server that wanted to do that could probably do that. Its purpose here is to track how often the where's are being gone to relative to one another and from what pages. It's the (...) (24 years ago, 23-Nov-00, to lugnet.admin.general)
| | | Re: "jump.cgi" considered harmful ? (1)
|
| (...) It isn't just you. I notice the delay as well on every machine I use - fast or slow, on every connection I use, fast or slow. Last time I looked into this (and I think posted about it) it seemed fairly obvious the biggest delays were had by IE (...) (24 years ago, 24-Nov-00, to lugnet.admin.general)
|
36 Messages in This Thread:
  
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
