Hacker News new | past | comments | ask | show | jobs | submit login
Unix Time 1500M – Friday July 14 02:40UTC (epochconverter.com)
267 points by andreasklinger on July 13, 2017 | hide | past | web | favorite | 99 comments

I remember having a party for the billion seconds, which was on September 8, 2001 -- then overshadowed in my memory by the September 11 attack less than three days later.

I also remember another party for Unix time 1234567890, back in 2009, which got an impressive turnout.

https://www.flickr.com/photos/estro/3304453623/in/photolist-... https://www.flickr.com/photos/briankusler/3278099542/in/phot... https://www.flickr.com/photos/estro/3305280994/in/photolist-...

At that party, I made some thyme tea, in honor of time_t. (Later on I learned that thyme tea is actually a thing... I felt like I was just making it up!)

I remember having a party for the billion seconds

That moment is famous in FreeBSD, because the cvsup tool we were using for distributing the source tree at the time suffered from a "S1G" bug: https://lists.freebsd.org/pipermail/freebsd-announce/2001-Se...

I remember rather quietly celebrating the "binary billion" (2^30), 1073741824 seconds, which coincidentally happened at a memorable time: 2004-01-10 13:37:04.

I was just reflecting back on my billion seconds celebration, and I'm finding it really hard to believe that 500 million seconds have gone by since then. A lot of stuff has happened in my life in the meantime, but that number puts it into perspective.

I didn't think I'd ever find myself in a reflective mood because of a unix timestamp!

We drank a toast to 1234567890 day — it coincided with my wife's birthday.

> I remember having a party for the billion seconds, which was on September 8, 2001

I was going to ask if anyone did!

My first thought was Oh 2B not far off then, I wonder if there's anything being organised -- oh, 2033, there's still time...

It's also interesting that this is a UTC (or at least, it is universally coordinated, whether you can call it UTC or not I suppose depends entirely on the epoch) event - unlike NYE celebrations which span most of the day throughout the world, Unix 1.5B celebrations all over would be at the same 'time'.

But seriously thyme tea sounds pretty awful.

There are many kinds of thyme - one is good for tea (Thymus vulgaris) - but they get confused a lot.

Not only does it taste good (what might get debated) but it is said to help against the common cold and stuff like that. Is being used for centuries now as the "poor man's antibiotics".

A friend later on bought me an Arabic "wild thyme tea", which is labeled as زعتر بري in Arabic. According to Wikipedia, this is thymus serpyllum, which is different from the culinary thyme that we use in the west (although Wikipedia says that kind of thyme is used for tea in Armenia). One more potential source of confusion is that Wikipedia's article on za'atar says that زعتر بري could also be another Middle Eastern thyme (thymus capitatus), or even origanum vulgare (oregano)!

So I'm not positive exactly which herb is in this Arabic tea, but I do like it better than the thyme tea that I made for that party.

Edit: it's this one: http://karabetian.com/food/tea-and-coffee/tazah-wild-thyme-t..., although I'm not sure I would be that comfortable giving my credit card to this web site. I'll see if it's available in my local Arabic market.

I was just typing in random times and saw that it came eerily close to 9/11. Kind of crept me out.

I also remember having a billion seconds party.

This remembers me of "primetime" a "clock" which draws a green bar every time the timestamp is a prime.


Did it in my student years, looks awful today, but I was proud back then ;)


I just remembered that you can give it a timestamp to display (instead of just displaying the current time). Unix Time 1,500M would be


and is (obviously) not a prime number.

> looks awful today

I don't know what you don't like about it but I think not only that it looks great, but also to me that's a design that's got character.

* the green lines are somehow weirdly blurred (for god know what reason)

* it's hard to notice that the current timestamp is the most right (white or green) bar. Would be nice to have the current time in the center.

* It's hard to get the concept without explanation (at first glance people just see weird green bars moving).

But then, I never meant to make a perfectly polished clock. It's good for a quick experiment.

That's oddly compelling to watch. And there's that human urge to try and see a pattern in the prime groupings.

You can fiddle with the displayed time (by editing the url) -- it's fun to see that (very) early timestamps have much greater chance to be a prime number.

Also, sometimes, you can find "twin primes". https://tessi.github.io/primetime/?t=1500000000 has such a twin-pair in the center of the screen (the twins are 1499999927 and 1499999929).

I was interested when 10^n events would happen and ran in a nice litte bug. I like to believe that this marks the end of all times: https://tessi.github.io/primetime/?t=10000000000000

Javascript's maximum date is Sat, 13 Sep 275760 (=8640000000000), so you've actually gone some ~43,000 years past the end of all time.

A quick shell version:

   while true; do if test $(factor $(date +%s) | wc -w) = 2; then echo -n '|'; else echo -n ' '; fi; sleep 1; done

You can also just watch what the prime factorisation of the current time is like this:

    watch -n 1 'date +%s | factor'

Also, the PI date (3141592653) goes straight to Sunday July 21 2069; that's exactly 100 years from the 1st human step on the moon.

That's super weird.

I think it is just one of the millions of possible coincidences. Yes, it's fun, but nothing weird, because there are other millions of equally weird coincidences that could happen.

I don't get it, what's so special about 0x59682f00? 0x60000000, now that's a nice round number. See you on January 14th 2021!

Simias, we are surrounded by base 10 lovers in a binary world. These are the same people who gave us terabyte hard drives with 1,000,000,000,000 bytes which doesn't even make sense with sector sizes of 512 and 4096 bytes. I too will celebrate the hex flip and not some shady base 10 shenanigans.

> Friday May 15 2015 02:09:25 GMT

That was 0x55555555 or b01010101010101010101010101010101

I am irrationally excited about this.

I suspect it's because now I'll be able to quickly glance at timestamps, say in the logs or in data records, and estimate the time better without having to use a script to translate it.

When I watch the constant progress of Unix time and milestones like this, I really feel the passage of time in way I don't otherwise. Sometimes it freaks me out.

Ah interesting. I agree. Wonder if it is because typical 12 or 24 hour clock resets so we sort of get to "try again" during a new day. A continuously increasing value that never resets just has a more ominous feel to it.

It could be worse, it could be counting down.

I thought of drawing a life-to-page calendar on my lounge room wall.

Decided against it as it would make it obvious how many days I've wasted and how few I have left. Maybe that's all the more reason to do it.

Try printing this http://abstrusegoose.com/51

Just added current timestamp to my mac os menubar. Not sure why I haven't thought of this before; beyond tonight's celebration, it's going to be very useful for checking against logs and cache-busting urls.

How'd you do this? I would enjoy that

I use an app called bitbar which is very basic but allows you to display the output of a program in the menubar. I'm sure there are similar apps if it doesn't exactly suit you, but give it a try!

are you being serious

edit: to elaborate, how does this help at all, you can memorize the current offset to as many decimals as you find convenient and do this already, a zero pad is nice but the # zeroes that make it easy to read for people who aren't Ramanujan will last for maybe a week

Although the point where we reach 1.5 Ms since the Epoch actually occurs slightly earlier at 2017-07-14 02:39:33 +0000 , as can be seen on a system using Arthur David Olson's "right" timezones.

    jdebp % TZ=right/UTC date +"%s %F %T %z" -d "@1500000000"
    1500000000 2017-07-14 02:39:33 +0000
    jdebp %
* https://unix.stackexchange.com/a/327403/5132

The timestamp in the headline is actually 1500000027 seconds since the Epoch.

POSIX time doesn't count leap seconds (i.e., "seconds since the Epoch" are not actually elapsed seconds) and therefore POSIX time 1.5e6 is 2017-07-14T02:40:00 (UTC scale) exactly

right/UTC uses 1970-01-01 00:00:10 (TAI scale) epoch (it is slightly different from 1970-01-01 00:00:00 (UTC scale) -- the Epoch)

2017-07-14 02:39:33 (UTC scale) is 2017-07-14T02:40:10 (TAI scale) i.e., it is 1.5e9 seconds since the epoch used for "right" timezones, not the Epoch.

No. Read the output of the date command and the aforementioned Stack Exchange answer again. The Epoch was the same physical second in both the "right" and the "posix" systems. It is 1.5 Gs (not Ms as I mistyped) since that second at 2017-07-14 02:39:33 +0000.

The fact that you've got a difference of 37 between your two values should be a clue that you've got this wrong. The current difference between UTC and TAI-10 is 27 seconds. Again, read the Stack Exchange answer for details.

And 1.5 Gs since the Epoch will be 2017-07-14 02:39:33 +0000 on both "posix" and "right" systems. A "right" wall clock reads the same as a "posix" wall clock, except at leap seconds of course. 2017-07-14 02:39:33 "posix" is the same point in time as 2017-07-14 02:39:33 "right": 1.5 Gs since the Epoch. It is not the same point in time as 2017-07-14 02:40:10 "right"/"posix", which is 1500000037 seconds since the Epoch.

    jdebp % TZ=right/UTC date +"%s %F %T %z" -d "@1500000037"
    1500000037 2017-07-14 02:40:10 +0000
    jdebp %
* https://unix.stackexchange.com/a/294715/5132

wrong. Don't confuse TAI and TAI-10. Don't confuse POSIX time [1] and seconds since epoch (time() value) in "right" timezones.

On the "difference of 37": the current TAI-UTC == 37s [2] (10s from 1972 + 27 leap seconds).

On "the same physical second": look at the value of TAI-UTC in the period before 1972 (if you plug the numbers then TAI-UTC is around 8 seconds at 1970-01-01 00:00:00 UTC, not 10s. The epochs are not the same. They are "slightly different"). The same physical second corresponds to 1972 (by definition), not 1970:

  1972-01-01 00:00:00 (UTC scale) == 1972-01-01 00:00:10 (TAI scale) 
Note: :10, not :00, note: TAI, not TAI-10.

To reiterate:

POSIX time 1.5e9 corresponds to 2017-07-14T02:40:00 (UTC scale) exactly.

1.5e9 "seconds since epoch" (not the Epoch) for "right" timezones corresponds to 2017-07-14T02:40:10 (TAI scale) == 2017-07-14 02:39:33 (UTC scale).

[1] http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_...

[2] http://hpiers.obspm.fr/eop-pc/index.php?index=TAI-UTC_tab&la...

(the authoritative source on leap seconds. It publishes Bulletin C ftp://hpiers.obspm.fr/iers/bul/bulc/BULLETINC.GUIDE )

This page answers a question I was thinking about in the shower this morning: what is the latest date (as a UNIX timestamp) you can store in a regular postgres integer column, which has a max value of 2147483647 per the postgres docs. It turns out that the answer is January 19 2038 03:14:07 GMT.

I'm working on a project that involves sync, and we're keeping track of action numbers by incrementing the last action number, but for various reasons I was thinking, "what if we replaced the action_number field with the corresponding timestamp integer?"

Then I began to ponder which integer format to use -- if I could get a way with just postgres integer, or if I needed to stop up to bigint. It turns out I should probably step up to bigint if I do this :-)

Anyway, I could've just done this in ruby, but by the time I saw this I had forgotten I was thinking about it, and this reminded me.

You've inadvertently discovered the "2038 Problem"[0]. This is the next iteration of the Y2K bug.

[0]: https://en.wikipedia.org/wiki/Year_2038_problem

Well well... TIL

If I thought we'd all still be here in 2038 I'd have something new to worry about :-/

cf. Year 2038 Problem [0]

[0]: https://en.wikipedia.org/wiki/Year_2038_problem

For the bay area crowd - is there anything going on to celebrate?

2147483647 (2^31-1) is worth watching as well.

A second later the first bit of any timestamp still handled as signed int32 will flip and the ride into the past begins.


Marked my calendar for subsequent dates..

> date -ud @2000000000

Wed May 18 03:33:20 UTC 2033

A calendar of the upcoming 100M anniversaries:

    $ for x in {15..21}
    > do
    > printf '%2d %s\n' $x "$(
    >   env TZ=UTC date -d "@$(($x*100000000))"
    > )"; done
    15 Fri 14 Jul 02:40:00 UTC 2017
    16 Sun 13 Sep 12:26:40 UTC 2020
    17 Tue 14 Nov 22:13:20 UTC 2023
    18 Fri 15 Jan 08:00:00 UTC 2027
    19 Sun 17 Mar 17:46:40 UTC 2030
    20 Wed 18 May 03:33:20 UTC 2033
    21 Fri 18 Jul 13:20:00 UTC 2036
It's neat how they seem to roll around every pi years.

    >>> 1e8 / (3600 * 24 * 365.25)
    >>> math.pi

Why is this milestone important?

"A number" that a lot of people use and recognise as "one, four, then a lot of other digits" will become noticeably different. Do not adjust your set.

Lots of 0s in a row.

Because life is short.

Because many of us won't make it to 2 billion.

I'm hoping to use fixing the 2^31 bugs as my retirement plan.

15 years time?

Surely you don't expect everyone living now to be still alive in 15 years? ;)

But yeah, it might just indicate that parent got the ballpark estimate wrong... Unix time is not the most human-friendly format.

A simple mnemonic is that pi * 10 ^ 7 seconds is a year (~0.5% error) and e * 10 ^ 6 seconds is a month.

Simple indeed.

I get that little popup on this website saying "you agree to our use of cookies to give you the best possible experience".

Why do I need cookies to maximize my unix time experience, exactly?

If you want to avoid converting seconds into something more human readable, you can use http://timestamp.online/countdown/1500000000

What is the next timestamp you are looking for? http://timestamp.online/countdown/1515151515 or is there something better sooner?

This one is sooner.

    jdebp % echo @400000005a00000000000000|TZ=UTC tai64nlocal
    2017-11-06 06:23:23.000000000
    jdebp %
This one is slightly later.

    jdebp % echo @400000005a5a5a5a005a5a5a|TZ=UTC tai64nlocal
    2018-01-13 19:12:53.005921370
    jdebp %

I've seen the Unix timestamp as One Four something since I started active coding since 2015. So I'm excited about this.

Twelve-billioner here (or was it 11-bil? hmm). Sounds like we're onto a new measure of work experience!

Actually, you would be either a 1.2 or 1.1-billioner. It's an order of magnitude less.

Oops, you're right.

Obligatory link to macOS "epoch flip clock" screensaver: https://github.com/chrstphrknwtn/epoch-flip-clock

My cars odometer is at 99910km's so by tomorrow I should be at 100000km's, not quite date -d @1500000000

For anyone wondering, an increment like this happens about every three years.

date +%s

To see current time

I used to find this syntax opaque/unfriendly until I simply looked into it. Here's a helpful cheat sheet for what follows the "+":


Or just use man date

Here's another one:

$ man strftime

What's everybody doing to celebrate?

I'm going to burn whatever incense is appropriate for ensuring my value to society by the end of the 32 bit Epoch.

Honestly, I'm going to shrug my shoulders. Not to dampen anyone else's celebrations and enthusiasm. That's just how I am about these kind of things. Like on most Birthdays, the orbital relationship of Earth to Sun isn't even within millions of miles of when I was born, so I'm like "meh". Except for my DD's birthday, but that's because it's special to her.

Thanks for eloquently putting in words how I feel about this. I yearn for a life so uneventful and devoid of stress that I could give two shits about such things.

And yet it piqued your curiosity enough to read comments on the issue.

Work is doing cupcakes! I can't participate because I will be at a funeral though. :(

5.0E+96 count down to that lol... funny site tho :D for some reason i approve of this.

I'm going to be 1.5B seconds old! I mean, whenever $random_website demands my birthday, they get Start of Epoch. Easy fake birthday to remember :)

yawn. is this the HN equivalent of a 4chan "GET"?

HN's "GET" would be when tesla's stock price hits 0.00000$

Time stamp and set a reminder for 7-10 years from now for this comment: Tesla will be the biggest company in the world. See you in 7-10 years when you remember this comment.

The world? That's quite a tall order for a company with such a small product line. They would have to compete with companies like Exxon which have 250B in yearly revenue.

...No, they'd be competing with Apple, Google, Microsoft, Amazon, and Facebook, which all crossed Exxon a long while ago.

Google does not have 250B revenue. Far from it (50B a few years back).

Apple is in the top ten but then it sells billions of expensive phones. Amazon barely scrapes in the top fifty, while Walmart is number one (Costco,CVS and Kroger easily beating Amazon)

Being a "normal" bricks and mortar company does give more turnover power it seems. Globally.


Don't underestimate Amazon. They have the world's largest (?) server-infrastructure and logistic infrastructure in place. Most of the other companies rely on Amazon.

Could imagine that Google's infrastructure is bigger. Their cloud service is clearly smaller than AWS but they have much more computing demand than Amazon has for their main business. So overall Google could be ahead of Amazon.

Logistics infrastructure of Amazon is not the largest either. In the US, Fedex and UPS will have more infrastructure. Not that many warehouses, but when it comes to distributing items they should be larger. Worldwide, DHL Group could be the largest. They have a substantial freight forwarding business so that they will move far more goods every day than Amazon does.

The important thing, though, is that no one has both of it in one company. We'll see how valuable that turns out to be (maybe not at all, maybe very).

I get short term nostalgia when I look at the root comment that started this thread...

I don't think they will be, but I'm pretty sure various governments will want to start passing laws requiring electric cars somewhere around 2030.

The question is how they will take care of the already existing companies. Maybe Tesla will share more technology, so that the industry can transition as a whole.

Volvo announced a few days ago that all their cars will be either EVs or plug-in hybrids starting from 2019. Toyota is almost there today.

From an end-user perspective, it's pretty obvious that todays plug-in hybrids with 15-30 miles electric range and 600 miles range on fuel will be better than pure EVs. >90% of people's driving is within the electric range, but you can still have a car with a big enough trunk (Try fitting two baby strollers in a Tesla! Or a couple of downhill bikes plus kit. Or scuba gear for two. Or...) that can still get you from SoCal up to Lake Tahoe without wasting time on charging.

As for competition on pure electric: BMW, VW, Ford, Fiat, Kia, Chevrolet/Opel, Nissan, Renault, Peugeot, Mitsubishi, Mercedes - they all have "proper" pure electric cars on the market today that are seeing sales figures within an order of magnitude of Tesla.

(Note that being within an order of magnitude only requires that you sell 7000 EVs globally in a year, that's a pretty easy target.)

Tesla was not the first one and is not the only electric car manufacturer. Nobody needs Tesla.

The biggest accomplishment of Tesla will probably be that it provided the wake-up call to the big auto companies.

Didn't France just announce that they are going to require electric by 2040? I wouldn't be surprised either. We're at a much better point to use them and grow the infrastructure now.

What's a 4chan GET?

All posts incrementally increase as posts accumulate on a board. Gets are normally reserved for when you approach a milestone. Example here http://new3.fjcdn.com/pictures/Fj+gets+banned+from+4chan_660....

Classically you'd say something is true if you get dubs (the post id ends in doubles)

Why is it called GET?

Common English loanword in Japanese, or at least in anime - characters will excitedly say something like "prize GET" (with the "get" in English and shouted, the "prize" in Japanese, but rendered that way by fan translations).

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact