abolish timezones #1934

Open
opened 2013-03-19 23:23:03 +00:00 by davidsarah · 9 comments
davidsarah commented 2013-03-19 23:23:03 +00:00
Owner

Timezones are inconvenient for project development and programming. This ticket is for brainstorming possible approaches to this problem.

Suggestions so far:

  • build a ring world.
  • flatten the earth somehow.
  • have everyone adapt their working hours to UTC.
  • http://twistedmatrix.com/trac/ticket/5000.
  • relocate to the Discworld, then stop its sun from orbiting (note: the latter may have undesired side-effects).
Timezones are inconvenient for project development and programming. This ticket is for brainstorming possible approaches to this problem. Suggestions so far: * build a ring world. * ~~flatten the earth somehow.~~ * have everyone adapt their working hours to UTC. * <http://twistedmatrix.com/trac/ticket/5000>. * relocate to the Discworld, then stop its sun from orbiting (note: the latter may have undesired side-effects).
tahoe-lafs added the
dev-infrastructure
normal
defect
1.9.2
labels 2013-03-19 23:23:03 +00:00
tahoe-lafs added this to the eventually milestone 2013-03-19 23:23:03 +00:00

I'd just like to point out that "flatten the earth" would not really achieve the goal of this ticket. It might have other benefits though, so perhaps we should move it to a separate ticket.

I'd just like to point out that "flatten the earth" would not really achieve the goal of this ticket. It might have other benefits though, so perhaps we should move it to a separate ticket.
tahoe-lafs added
n/a
and removed
1.9.2
labels 2013-03-19 23:29:26 +00:00
daira commented 2013-07-08 19:45:24 +00:00
Author
Owner

At LeastAuthority.com we have had considerable success with the "have everyone adapt their working hours to UTC" option. I recommend it!

At LeastAuthority.com we have had considerable success with the "have everyone adapt their working hours to UTC" option. I recommend it!

I realize this isn't a strict requirement, technically, but having a universal LAFS grid could be a useful "stepping stone" feature for constructing a ring world. See #2009.

Also, as kind of a branding thing, if other intelligences discovered our ring world, but found we did not have a universal grid, wouldn't it be kind of embarrassing? What if we send a readcap to our greeting of peace and mutual prosperity, only for them to discover they are on the wrong grid?

I realize this isn't a strict requirement, technically, but having a universal LAFS grid could be a useful "stepping stone" feature for constructing a ring world. See #2009. Also, as kind of a branding thing, if other intelligences discovered our ring world, but found we did not have a universal grid, wouldn't it be kind of embarrassing? What if we send a readcap to our greeting of peace and mutual prosperity, only for them to discover they are on the wrong grid?
daira commented 2013-07-14 15:40:17 +00:00
Author
Owner

Replying to nejucomo:

Also, as kind of a branding thing, if other intelligences discovered our ring world, but found we did not have a universal grid, wouldn't it be kind of embarrassing? What if we send a readcap to our greeting of peace and mutual prosperity, only for them to discover they are on the wrong grid?

You're absolutely right. We'd better fix #403 stat before SETI succeeds.

Replying to [nejucomo](/tahoe-lafs/trac-2024-07-25/issues/1934#issuecomment-91240): > Also, as kind of a branding thing, if other intelligences discovered our ring world, but found we did not have a universal grid, wouldn't it be kind of embarrassing? What if we send a readcap to our greeting of peace and mutual prosperity, only for them to discover they are on the wrong grid? You're absolutely right. We'd better fix #403 stat before SETI succeeds.
daira commented 2013-08-08 23:10:35 +00:00
Author
Owner

There may be unexplored synergies between this ticket and LeastAuthority's interstellar exploration goals. Exciting!

There may be unexplored synergies between this ticket and LeastAuthority's interstellar exploration goals. Exciting!
Author
Owner

This is a misfeature - timezones are an enormous boon to capacity management.

For example, if you had everyone kick off a backup at 3am local time, then the spread of 3am around the world would allow you to have much lower throughput systems than if it was the same everywhere (you would have a massive spike at 3am).

This is also why ISPs have both "business" and "residential" products - very few successful ISPs would target only one of the two. Generally they have a mostly-uncontended daytime load ("business") then let the cheaper residential customers fight over the same capacity when they get home from work.

See also follow-the-sun support - this is a method of providing 24hr support teams without having a graveyard shift.

In short, LAFS grids are likely benefiting from disperse timezones.

This is a misfeature - timezones are an enormous boon to capacity management. For example, if you had everyone kick off a backup at 3am local time, then the spread of 3am around the world would allow you to have much lower throughput systems than if it was the same everywhere (you would have a massive spike at 3am). This is also why ISPs have both "business" and "residential" products - very few successful ISPs would target only one of the two. Generally they have a mostly-uncontended daytime load ("business") then let the cheaper residential customers fight over the same capacity when they get home from work. See also follow-the-sun support - this is a method of providing 24hr support teams without having a graveyard shift. In short, LAFS grids are likely benefiting from disperse timezones.

The buildbot's github_posthook has been intermittently broken since 25-Oct-2015. This is the mechanism by which Github notifies our buildbot that a change has been committed, so it can kick off the unit tests, tarball-building, etc.

Eventually, "intermittent" was refined to mean "only broken for Daira's commits". Daira lives in a timezone which moved from +0100 to +0000 on 25-Oct-2015. Github apparently emits "timestamp" values that are formatted like 2016-01-31T11:02:50-08:00 for any timezone that's not exactly +0000, but like 2016-02-02T18:55:34Z when the timezone is exactly +0000. The github_posthook regexp which attempted to parse these fields (and recover a sensible seconds-since-epoch value from it) failed to handle the "Z", throwing an exception, and not registering the commit.

After I've fixed the regexp, maybe I'll ask GitHub to contribute some cycles towards http://twistedmatrix.com/trac/ticket/5000 . I suspect it will help resolve (or at least mask) their uptime issues as well :).

The buildbot's `github_posthook` has been intermittently broken since 25-Oct-2015. This is the mechanism by which Github notifies our buildbot that a change has been committed, so it can kick off the unit tests, tarball-building, etc. Eventually, "intermittent" was refined to mean "only broken for Daira's commits". Daira lives in a timezone which moved from +0100 to +0000 on 25-Oct-2015. Github apparently emits "timestamp" values that are formatted like `2016-01-31T11:02:50-08:00` for any timezone that's not exactly +0000, but like `2016-02-02T18:55:34Z` when the timezone is exactly +0000. The `github_posthook` regexp which attempted to parse these fields (and recover a sensible seconds-since-epoch value from it) failed to handle the "Z", throwing an exception, and not registering the commit. After I've fixed the regexp, maybe I'll ask GitHub to contribute some cycles towards <http://twistedmatrix.com/trac/ticket/5000> . I suspect it will help resolve (or at least mask) their uptime issues as well :).

Some economists have submitted a pull request for this: https://www.washingtonpost.com/news/worldviews/wp/2016/02/12/the-radical-plan-to-destroy-time-zones-2/?tid=a_inl HT Vitalik Buterin

Some economists have submitted a pull request for this: <https://www.washingtonpost.com/news/worldviews/wp/2016/02/12/the-radical-plan-to-destroy-time-zones-2/?tid=a_inl> HT Vitalik Buterin

The New York Times is on to this now http://www.nytimes.com/2016/11/06/opinion/sunday/time-to-dump-time-zones.html which led twitter to show me this argument against https://qntm.org/abolish

The New York Times is on to this now <http://www.nytimes.com/2016/11/06/opinion/sunday/time-to-dump-time-zones.html> which led twitter to show me this argument against <https://qntm.org/abolish>
Sign in to join this conversation.
No Milestone
No Assignees
5 Participants
Notifications
Due Date
The due date is invalid or out of range. Please use the format 'yyyy-mm-dd'.

No due date set.

Reference: tahoe-lafs/trac-2024-07-25#1934
No description provided.