Availability figures
Jul. 13th, 2004 02:53 pmOK, my cynicsm bump is itching again.
I've just been sent an email that says:
"can you send me the availability figures for june pls"
by a manager.
Now the thing is, these numbers are generated automatically each month. They're on a page. I even wrote a spreadsheet that _all_ it required was a copy-and-paste from the webpage, to format his numbers for him.
The reason he asks me for them, each month, is because he doesn't understand the difference between an ICMP ping, (eg. whether the host is 'up') and a port test (eg. whether the host is accepting connections for the relevant service).
So each month I get to provide the 'better' of the two, as a percentage.
Talk about FUCKING pointless numbers. I'm massaging the figures to give an 'availability' report. But if the relevant manager has NO FUCKING CLUE what "88.3% of pings worked, but 99.8% of connections to the port worked. Given an approximately 5 minute sample interval" actually means, then what chance does the poor customer he's fobbing these numbers off onto have?
It's just a waste of my time. I would say theirs, but they clearly don't have anything better to do than look at statistics. Which they don't understand.
Is it just me who's of the opinion that a statistic out of context is just an irrelevant number? 84% of statistics are made up on the spot after all. It's just the stupid giving meaningless numbers to the ignorant.
I think I shall have to start quoting stupid things like percentage chance that my car is black or something equally banal.
The tosserati strike once more.
I've just been sent an email that says:
"can you send me the availability figures for june pls"
by a manager.
Now the thing is, these numbers are generated automatically each month. They're on a page. I even wrote a spreadsheet that _all_ it required was a copy-and-paste from the webpage, to format his numbers for him.
The reason he asks me for them, each month, is because he doesn't understand the difference between an ICMP ping, (eg. whether the host is 'up') and a port test (eg. whether the host is accepting connections for the relevant service).
So each month I get to provide the 'better' of the two, as a percentage.
Talk about FUCKING pointless numbers. I'm massaging the figures to give an 'availability' report. But if the relevant manager has NO FUCKING CLUE what "88.3% of pings worked, but 99.8% of connections to the port worked. Given an approximately 5 minute sample interval" actually means, then what chance does the poor customer he's fobbing these numbers off onto have?
It's just a waste of my time. I would say theirs, but they clearly don't have anything better to do than look at statistics. Which they don't understand.
Is it just me who's of the opinion that a statistic out of context is just an irrelevant number? 84% of statistics are made up on the spot after all. It's just the stupid giving meaningless numbers to the ignorant.
I think I shall have to start quoting stupid things like percentage chance that my car is black or something equally banal.
The tosserati strike once more.
no subject
Date: 2004-07-13 09:00 am (UTC)no subject
Date: 2004-07-13 09:12 am (UTC)