Regrets; Domain Incite article
I'll be at the INTA meeting Monday and unable to attend Monday's Standing Committee meeting. This Domain Incite article is interesting. https://domainincite.com/29884-it-now-takes-two-weeks-to-get-a-whois-record-... The main point in the Domain Incite report is that the average time for a positive response has risen from one week (6.92 days) in February to two weeks (14.09 days) in April. For negative responses, the response time was 1.26 days, up from 6.17 days in March. Tracking the changes in these averages is a good first step, but I suspect it will be more informative to see response statistics by registrar. Further, averages are probably not the best statistic. Better would be the median (50 percentile) and a couple of other percentiles, e.g. 10% and 90%. Thanks, Steve sender
I too will be at the INTA Annual Meeting and will not be able to attend this call. Steve, thanks for sending this article. All, I'm afraid that I don't know how to interpret the data. Is a doubled (for positive) and triple (for negative) response time really bad for what is essentially a start up service in its second quarter or is that really good? Will response times continue to grow, shrink, or level off? Is there any learning from any other startup (I know we the RDRS is essentially a decentralized startup so I don't know if there is anything out there that we can look to). Were the shorter response times because registrars have suddenly decided to move more slowly, or is it a (very positive) symptom of greater demand and the registrars are just taking a bit longer to keep up? In other words, should I be saying "ugh" or "yay"? Best, Paul From: Steve Crocker via Gnso-rdrs-sc <gnso-rdrs-sc@icann.org> Sent: Thursday, May 16, 2024 5:13 PM To: gnso-rdrs-sc@icann.org Subject: [Gnso-rdrs-sc] Regrets; Domain Incite article I'll be at the INTA meeting Monday and unable to attend Monday's Standing Committee meeting. This Domain Incite article is interesting. https://domainincite.com/29884-it-now-takes-two-weeks-to-get-a-whois-record-... The main point in the Domain Incite report is that the average time for a positive response has risen from one week (6.92 days) in February to two weeks (14.09 days) in April. For negative responses, the response time was 1.26 days, up from 6.17 days in March. Tracking the changes in these averages is a good first step, but I suspect it will be more informative to see response statistics by registrar. Further, averages are probably not the best statistic. Better would be the median (50 percentile) and a couple of other percentiles, e.g. 10% and 90%. Thanks, Steve sender This email originated from outside the firm. Please use caution.
Hi all, I also won't be on Monday's call as it is a local holiday, so Roger Carney will represent the RrSG. A few thoughts on the response time data: * The purpose of the RDRS project is not to track response time, so we should be careful about scope creep and focusing on our mission objective here * Requests remain open while the registrar awaits input from the requestor; perhaps the increasing average response time shows us that requestors are not promptly providing required information to the registrar. Thanks, *Sarah Wyld, CIPP/E* Policy & Privacy Manager Pronouns: she/they swyld@tucows.com On 2024-05-17 10:24 a.m., Paul McGrady via Gnso-rdrs-sc wrote:
I too will be at the INTA Annual Meeting and will not be able to attend this call.
Steve, thanks for sending this article.
All, I’m afraid that I don’t know how to interpret the data. Is a doubled (for positive) and triple (for negative) response time really bad for what is essentially a start up service in its second quarter or is that really good? Will response times continue to grow, shrink, or level off? Is there any learning from any other startup (I know we the RDRS is essentially a decentralized startup so I don’t know if there is anything out there that we can look to). Were the shorter response times because registrars have suddenly decided to move more slowly, or is it a (very positive) symptom of greater demand and the registrars are just taking a bit longer to keep up?
In other words, should I be saying “ugh” or “yay”?
Best,
Paul
*From:*Steve Crocker via Gnso-rdrs-sc <gnso-rdrs-sc@icann.org> *Sent:* Thursday, May 16, 2024 5:13 PM *To:* gnso-rdrs-sc@icann.org *Subject:* [Gnso-rdrs-sc] Regrets; Domain Incite article
I'll be at the INTA meeting Monday and unable to attend Monday's Standing Committee meeting.
This Domain Incite article is interesting.
https://domainincite.com/29884-it-now-takes-two-weeks-to-get-a-whois-record-...
The main point in the Domain Incite report is that the average time for a positive response has risen from one week (6.92 days) in February to two weeks (14.09 days) in April.
For negative responses, the response time was 1.26 days, up from 6.17 days in March.
Tracking the changes in these averages is a good first step, but I suspect it will be more informative to see response statistics by registrar. Further, averages are probably not the best statistic. Better would be the median (50 percentile) and a couple of other percentiles, e.g. 10% and 90%.
Thanks,
Steve
sender
This email originated from outside the firm. Please use caution.
_______________________________________________ Gnso-rdrs-sc mailing list --gnso-rdrs-sc@icann.org To unsubscribe send an email tognso-rdrs-sc-leave@icann.org
_______________________________________________ By submitting your personal data, you consent to the processing of your personal data for purposes of subscribing to this mailing list accordance with the ICANN Privacy Policy (https://www.icann.org/privacy/policy) and the website Terms of Service (https://www.icann.org/privacy/tos). You can visit the Mailman link above to change your membership status or configuration, including unsubscribing, setting digest-style delivery or disabling delivery altogether (e.g., for a vacation), and so on.
also, given the volume, couldn't a small number of outliers could significantly change the average? J. Beckwith Burr HWG LLC 1919 M Street NW, The Eighth Floor Washington DC 20036 ________________________________ From: Sarah Wyld via Gnso-rdrs-sc <gnso-rdrs-sc@icann.org> Sent: Friday, May 17, 2024 10:31 AM To: gnso-rdrs-sc@icann.org <gnso-rdrs-sc@icann.org> Subject: [Gnso-rdrs-sc] Re: Regrets; Domain Incite article Hi all, I also won't be on Monday's call as it is a local holiday, so Roger Carney will represent the RrSG. A few thoughts on the response time data: * The purpose of the RDRS project is not to track response time, so we should be careful about scope creep and focusing on our mission objective here * Requests remain open while the registrar awaits input from the requestor; perhaps the increasing average response time shows us that requestors are not promptly providing required information to the registrar. Thanks, Sarah Wyld, CIPP/E Policy & Privacy Manager Pronouns: she/they swyld@tucows.com<mailto:swyld@tucows.com> On 2024-05-17 10:24 a.m., Paul McGrady via Gnso-rdrs-sc wrote: I too will be at the INTA Annual Meeting and will not be able to attend this call. Steve, thanks for sending this article. All, I’m afraid that I don’t know how to interpret the data. Is a doubled (for positive) and triple (for negative) response time really bad for what is essentially a start up service in its second quarter or is that really good? Will response times continue to grow, shrink, or level off? Is there any learning from any other startup (I know we the RDRS is essentially a decentralized startup so I don’t know if there is anything out there that we can look to). Were the shorter response times because registrars have suddenly decided to move more slowly, or is it a (very positive) symptom of greater demand and the registrars are just taking a bit longer to keep up? In other words, should I be saying “ugh” or “yay”? Best, Paul From: Steve Crocker via Gnso-rdrs-sc <gnso-rdrs-sc@icann.org><mailto:gnso-rdrs-sc@icann.org> Sent: Thursday, May 16, 2024 5:13 PM To: gnso-rdrs-sc@icann.org<mailto:gnso-rdrs-sc@icann.org> Subject: [Gnso-rdrs-sc] Regrets; Domain Incite article I'll be at the INTA meeting Monday and unable to attend Monday's Standing Committee meeting. This Domain Incite article is interesting. https://us01.z.antigena.com/l/ZMl7PkdUssAgnruN7phwYcEegUy51MuiXWYFR86xM_2hAE... The main point in the Domain Incite report is that the average time for a positive response has risen from one week (6.92 days) in February to two weeks (14.09 days) in April. For negative responses, the response time was 1.26 days, up from 6.17 days in March. Tracking the changes in these averages is a good first step, but I suspect it will be more informative to see response statistics by registrar. Further, averages are probably not the best statistic. Better would be the median (50 percentile) and a couple of other percentiles, e.g. 10% and 90%. Thanks, Steve sender This email originated from outside the firm. Please use caution. _______________________________________________ Gnso-rdrs-sc mailing list -- gnso-rdrs-sc@icann.org<mailto:gnso-rdrs-sc@icann.org> To unsubscribe send an email to gnso-rdrs-sc-leave@icann.org<mailto:gnso-rdrs-sc-leave@icann.org> _______________________________________________ By submitting your personal data, you consent to the processing of your personal data for purposes of subscribing to this mailing list accordance with the ICANN Privacy Policy (https://www.icann.org/privacy/policy) and the website Terms of Service (https://www.icann.org/privacy/tos). You can visit the Mailman link above to change your membership status or configuration, including unsubscribing, setting digest-style delivery or disabling delivery altogether (e.g., for a vacation), and so on.
Becky, Your point about the average being affected by a small number of outliers is precisely why using the average is the wrong measure. The median is the 50 percentile mark and isn't affected by outliers. The 10 percentile and 90 percentile figures will provide a decent picture of the outliers. (I wouldn't argue if someone prefers, say, the 20 and 80 percentiles or even the 30 and 70 percentiles.) Steve On Fri, May 17, 2024 at 11:50 AM Becky Burr via Gnso-rdrs-sc < gnso-rdrs-sc@icann.org> wrote:
also, given the volume, couldn't a small number of outliers could significantly change the average?
J. Beckwith Burr HWG LLC 1919 M Street NW, The Eighth Floor Washington DC 20036 ------------------------------ *From:* Sarah Wyld via Gnso-rdrs-sc <gnso-rdrs-sc@icann.org> *Sent:* Friday, May 17, 2024 10:31 AM *To:* gnso-rdrs-sc@icann.org <gnso-rdrs-sc@icann.org> *Subject:* [Gnso-rdrs-sc] Re: Regrets; Domain Incite article
Hi all,
I also won't be on Monday's call as it is a local holiday, so Roger Carney will represent the RrSG.
A few thoughts on the response time data:
- The purpose of the RDRS project is not to track response time, so we should be careful about scope creep and focusing on our mission objective here - Requests remain open while the registrar awaits input from the requestor; perhaps the increasing average response time shows us that requestors are not promptly providing required information to the registrar.
Thanks,
*Sarah Wyld, CIPP/E*
Policy & Privacy Manager Pronouns: she/they
swyld@tucows.com On 2024-05-17 10:24 a.m., Paul McGrady via Gnso-rdrs-sc wrote:
I too will be at the INTA Annual Meeting and will not be able to attend this call.
Steve, thanks for sending this article.
All, I’m afraid that I don’t know how to interpret the data. Is a doubled (for positive) and triple (for negative) response time really bad for what is essentially a start up service in its second quarter or is that really good? Will response times continue to grow, shrink, or level off? Is there any learning from any other startup (I know we the RDRS is essentially a decentralized startup so I don’t know if there is anything out there that we can look to). Were the shorter response times because registrars have suddenly decided to move more slowly, or is it a (very positive) symptom of greater demand and the registrars are just taking a bit longer to keep up?
In other words, should I be saying “ugh” or “yay”?
Best,
Paul
*From:* Steve Crocker via Gnso-rdrs-sc <gnso-rdrs-sc@icann.org> <gnso-rdrs-sc@icann.org> *Sent:* Thursday, May 16, 2024 5:13 PM *To:* gnso-rdrs-sc@icann.org *Subject:* [Gnso-rdrs-sc] Regrets; Domain Incite article
I'll be at the INTA meeting Monday and unable to attend Monday's Standing Committee meeting.
This Domain Incite article is interesting.
https://us01.z.antigena.com/l/ZMl7PkdUssAgnruN7phwYcEegUy51MuiXWYFR86xM_2hAE...
The main point in the Domain Incite report is that the average time for a positive response has risen from one week (6.92 days) in February to two weeks (14.09 days) in April.
For negative responses, the response time was 1.26 days, up from 6.17 days in March.
Tracking the changes in these averages is a good first step, but I suspect it will be more informative to see response statistics by registrar. Further, averages are probably not the best statistic. Better would be the median (50 percentile) and a couple of other percentiles, e.g. 10% and 90%.
Thanks,
Steve
sender
This email originated from outside the firm. Please use caution.
_______________________________________________ Gnso-rdrs-sc mailing list -- gnso-rdrs-sc@icann.org To unsubscribe send an email to gnso-rdrs-sc-leave@icann.org
_______________________________________________ By submitting your personal data, you consent to the processing of your personal data for purposes of subscribing to this mailing list accordance with the ICANN Privacy Policy (https://www.icann.org/privacy/policy) and the website Terms of Service (https://www.icann.org/privacy/tos). You can visit the Mailman link above to change your membership status or configuration, including unsubscribing, setting digest-style delivery or disabling delivery altogether (e.g., for a vacation), and so on.
_______________________________________________ Gnso-rdrs-sc mailing list -- gnso-rdrs-sc@icann.org To unsubscribe send an email to gnso-rdrs-sc-leave@icann.org
_______________________________________________ By submitting your personal data, you consent to the processing of your personal data for purposes of subscribing to this mailing list accordance with the ICANN Privacy Policy (https://www.icann.org/privacy/policy) and the website Terms of Service (https://www.icann.org/privacy/tos). You can visit the Mailman link above to change your membership status or configuration, including unsubscribing, setting digest-style delivery or disabling delivery altogether (e.g., for a vacation), and so on.
-- Sent by a Verified [image: Sent by a Verified sender] <https://wallet.unumid.co/authenticate?referralCode=tcp16fM4W47y> sender
Responding to multiple ~ 1. Becky asked: * "also, given the volume, couldn't a small number of outliers could significantly change the average?" * Not necessarily? So long as the outliers remain proportional to the population ~ E.g. 1 outlier in 10, vs 10 in 100, vs 100 in 1000 should all yield the same average. Ie, if 10% of requests consistently take 10x as long to process, the average would be constant regardless of population size. But to Steve's point, median value tracking avoids risk of non-proportional outliers (e.g. one singular person breaking the system with bogus requests), and would seem wise. 1. Responding to Sarah, regarding tracking response time changes month to month, * I note that work would seem to be supported by Assignment 1 of the SC charter<https://community.icann.org/display/EOTSFGRD/RDRS+Standing+Committee>, so while the desire to not stray from the mission is valid, this line of inquiry (to my eye) seems within the swim lane. 2. To Sarah et al, regarding lack of data on whether/not the response time is due to requestor inaction ("awaiting further info from requestor"), * Strongly agree that's a valid piece of data, and am quite supportive of your desire to tease that out. It wouldn't be fair to registrars to be blamed for non-responsiveness if/when it was the requestor who was being slow. * Additionally: this may be linked to the margin conversation from our "feedback document", re: ICANN's lack of visibility into responses writ large. I suspect there is room to improve that visibility without running afoul of the concerns expressed by RrSG re: the over-sharing of PII. i. IF we can nail down the feature constraints (e.g. i) maximize ICANN visibility into non-PII response data, ii) prevent PII from going to any party other than the requestor, iii) minimize additional steps required of requestor (such as creation/use of additional registrar specific portals), etc) ii. THEN we could perhaps discuss whether/not existing tools/options fit the constraints. * This could be informative for successor systems to RDRS, even if not feasible w/in the pilot window, under Assignments 3 &4. From: Becky Burr via Gnso-rdrs-sc <gnso-rdrs-sc@icann.org> Sent: Friday, May 17, 2024 8:50 AM To: gnso-rdrs-sc@icann.org; Sarah Wyld <swyld@tucows.com> Subject: [EXTERNAL EMAIL] - [Gnso-rdrs-sc] Re: Regrets; Domain Incite article also, given the volume, couldn't a small number of outliers could significantly change the average? J. Beckwith Burr HWG LLC 1919 M Street NW, The Eighth Floor Washington DC 20036 ________________________________ From: Sarah Wyld via Gnso-rdrs-sc <gnso-rdrs-sc@icann.org<mailto:gnso-rdrs-sc@icann.org>> Sent: Friday, May 17, 2024 10:31 AM To: gnso-rdrs-sc@icann.org<mailto:gnso-rdrs-sc@icann.org> <gnso-rdrs-sc@icann.org<mailto:gnso-rdrs-sc@icann.org>> Subject: [Gnso-rdrs-sc] Re: Regrets; Domain Incite article Hi all, I also won't be on Monday's call as it is a local holiday, so Roger Carney will represent the RrSG. A few thoughts on the response time data: * The purpose of the RDRS project is not to track response time, so we should be careful about scope creep and focusing on our mission objective here * Requests remain open while the registrar awaits input from the requestor; perhaps the increasing average response time shows us that requestors are not promptly providing required information to the registrar. Thanks, Sarah Wyld, CIPP/E Policy & Privacy Manager Pronouns: she/they swyld@tucows.com<mailto:swyld@tucows.com> On 2024-05-17 10:24 a.m., Paul McGrady via Gnso-rdrs-sc wrote: I too will be at the INTA Annual Meeting and will not be able to attend this call. Steve, thanks for sending this article. All, I'm afraid that I don't know how to interpret the data. Is a doubled (for positive) and triple (for negative) response time really bad for what is essentially a start up service in its second quarter or is that really good? Will response times continue to grow, shrink, or level off? Is there any learning from any other startup (I know we the RDRS is essentially a decentralized startup so I don't know if there is anything out there that we can look to). Were the shorter response times because registrars have suddenly decided to move more slowly, or is it a (very positive) symptom of greater demand and the registrars are just taking a bit longer to keep up? In other words, should I be saying "ugh" or "yay"? Best, Paul From: Steve Crocker via Gnso-rdrs-sc <gnso-rdrs-sc@icann.org><mailto:gnso-rdrs-sc@icann.org> Sent: Thursday, May 16, 2024 5:13 PM To: gnso-rdrs-sc@icann.org<mailto:gnso-rdrs-sc@icann.org> Subject: [Gnso-rdrs-sc] Regrets; Domain Incite article I'll be at the INTA meeting Monday and unable to attend Monday's Standing Committee meeting. This Domain Incite article is interesting. https://us01.z.antigena.com/l/ZMl7PkdUssAgnruN7phwYcEegUy51MuiXWYFR86xM_2hAE... The main point in the Domain Incite report is that the average time for a positive response has risen from one week (6.92 days) in February to two weeks (14.09 days) in April. For negative responses, the response time was 1.26 days, up from 6.17 days in March. Tracking the changes in these averages is a good first step, but I suspect it will be more informative to see response statistics by registrar. Further, averages are probably not the best statistic. Better would be the median (50 percentile) and a couple of other percentiles, e.g. 10% and 90%. Thanks, Steve sender This email originated from outside the firm. Please use caution. _______________________________________________ Gnso-rdrs-sc mailing list -- gnso-rdrs-sc@icann.org<mailto:gnso-rdrs-sc@icann.org> To unsubscribe send an email to gnso-rdrs-sc-leave@icann.org<mailto:gnso-rdrs-sc-leave@icann.org> _______________________________________________ By submitting your personal data, you consent to the processing of your personal data for purposes of subscribing to this mailing list accordance with the ICANN Privacy Policy (https://www.icann.org/privacy/policy) and the website Terms of Service (https://www.icann.org/privacy/tos). You can visit the Mailman link above to change your membership status or configuration, including unsubscribing, setting digest-style delivery or disabling delivery altogether (e.g., for a vacation), and so on.
I see that I dropped a digit when I copied stats from the Domain Incite report. The correct statement re the number of days for denied requests is: The average time to have a request denied was 11.26 days, up from 6.17 days in March, the data also shows. Apologies. Steve On Thu, May 16, 2024 at 6:12 PM Steve Crocker <steve@shinkuro.com> wrote:
I'll be at the INTA meeting Monday and unable to attend Monday's Standing Committee meeting.
This Domain Incite article is interesting.
https://domainincite.com/29884-it-now-takes-two-weeks-to-get-a-whois-record-...
The main point in the Domain Incite report is that the average time for a positive response has risen from one week (6.92 days) in February to two weeks (14.09 days) in April.
For negative responses, the response time was 1.26 days, up from 6.17 days in March.
Tracking the changes in these averages is a good first step, but I suspect it will be more informative to see response statistics by registrar. Further, averages are probably not the best statistic. Better would be the median (50 percentile) and a couple of other percentiles, e.g. 10% and 90%.
Thanks,
Steve
sender
-- Sent by a Verified [image: Sent by a Verified sender] <https://wallet.unumid.co/authenticate?referralCode=tcp16fM4W47y> sender
participants (5)
-
Becky Burr -
Gabriel Andrews -
Paul McGrady -
Sarah Wyld -
Steve Crocker