Comments on: Web Data: Randomly Erratically Variably Unpredictably Incomplete? http://blog.jimnovo.com/2007/10/25/randomly-incomplete/ Moving from a Low Accountability to a High Accountability Business Model Tue, 6 Jan 2009 14:17:57 +0000 http://wordpress.org/?v=2.5.1 By: Steve http://blog.jimnovo.com/2007/10/25/randomly-incomplete/#comment-8794 Steve Wed, 07 Nov 2007 01:08:55 +0000 http://blog.jimnovo.com/2007/10/25/randomly-incomplete/#comment-8794 Sorry for delay. Been meaning to write. Best intentions etc. :-( Unpredictably works for me. I remain unconvinced we (at some point in the future?) can't predict and deal with this in some sort of model. But we certainly can't right here and now. As I mentioned in my email you quoted. :-) I stayed away from the double negatives, but ... well... best of a bad bunch? Cheers! - Steve Sorry for delay. Been meaning to write. Best intentions etc. :-(

Unpredictably works for me. I remain unconvinced we (at some point in the future?) can’t predict and deal with this in some sort of model. But we certainly can’t right here and now.

As I mentioned in my email you quoted. :-) I stayed away from the double negatives, but … well… best of a bad bunch?
Cheers!
- Steve

]]> By: Jim Novo http://blog.jimnovo.com/2007/10/25/randomly-incomplete/#comment-8331 Jim Novo Mon, 29 Oct 2007 23:51:52 +0000 http://blog.jimnovo.com/2007/10/25/randomly-incomplete/#comment-8331 "And it's not like I've ever seen marketing folk spend weeks arguing over a word or phrase or colour. Heaven forbid! ;-)" Thanks Steve... We now have a vote for "Variably" due to "E<font size="2">rratically</font>" being too close to "Randomly" in meaning - at least in Australia! Variably Incomplete is OK, but "Variability" can sometimes be predicted and dealt with in a model. How about going back to the original premise of modeling not being an answer to this problem, and going with: Unpredictably Incomplete What say you Steve?  Anyone else? “And it’s not like I’ve ever seen marketing folk spend weeks arguing over a word or phrase or colour. Heaven forbid! ;-)”

Thanks Steve…

We now have a vote for “Variably” due to “Erratically” being too close to “Randomly” in meaning - at least in Australia!

Variably Incomplete is OK, but “Variability” can sometimes be predicted and dealt with in a model.

How about going back to the original premise of modeling not being an answer to this problem, and going with:

Unpredictably Incomplete

What say you Steve?  Anyone else?

]]> By: Jim Novo http://blog.jimnovo.com/2007/10/25/randomly-incomplete/#comment-8319 Jim Novo Mon, 29 Oct 2007 14:23:16 +0000 http://blog.jimnovo.com/2007/10/25/randomly-incomplete/#comment-8319 You see, with these technology guys, they are always trying to hold you to a specific definition...and I take it unless the data is truly random by the mathematical definition, "random" is not a satisfactory qualifier. How about Erratically Incomplete? You see, with these technology guys, they are always trying to hold you to a specific definition…and I take it unless the data is truly random by the mathematical definition, “random” is not a satisfactory qualifier.

How about Erratically Incomplete?

]]> By: Steve http://blog.jimnovo.com/2007/10/25/randomly-incomplete/#comment-8297 Steve Sun, 28 Oct 2007 21:50:36 +0000 http://blog.jimnovo.com/2007/10/25/randomly-incomplete/#comment-8297 I respectfully disagree on the "Randomly" part. Rest is spot on. Why disagree? Fundamentally the data being incomplete is not random. I've had (way more than...) sufficient exposure to crypto to appreciate the horrendous issues with getting truly random data. As for Analytics? There are actual rules and conditions around why the data is incomplete. Cache. Cookies. Javascript woes and so on. It's a highly complex problem with lots of unknowns. Unknowns != Random! :-) I would find "incomplete", as a standalone phrase, adequate to the task of explaining the problems. But perhaps another descriptive leader than Randomly may be appropriate? It's too early in the morning here to engage my brain to come up with something else. Thoughts? Cheers! - Steve I respectfully disagree on the “Randomly” part. Rest is spot on.

Why disagree?
Fundamentally the data being incomplete is not random. I’ve had (way more than…) sufficient exposure to crypto to appreciate the horrendous issues with getting truly random data.
As for Analytics? There are actual rules and conditions around why the data is incomplete. Cache. Cookies. Javascript woes and so on. It’s a highly complex problem with lots of unknowns.

Unknowns != Random! :-)

I would find “incomplete”, as a standalone phrase, adequate to the task of explaining the problems. But perhaps another descriptive leader than Randomly may be appropriate? It’s too early in the morning here to engage my brain to come up with something else.

Thoughts? Cheers!
- Steve

]]> By: Ted http://blog.jimnovo.com/2007/10/25/randomly-incomplete/#comment-8242 Ted Sat, 27 Oct 2007 14:00:56 +0000 http://blog.jimnovo.com/2007/10/25/randomly-incomplete/#comment-8242 Great blog. INCREDIBLE book! Thank you so much! Great blog.
INCREDIBLE book! Thank you so much!

]]>