[HN Gopher] Pushing the whole company into the past on purpose
       ___________________________________________________________________
        
       Pushing the whole company into the past on purpose
        
       Author : senkora
       Score  : 107 points
       Date   : 2025-01-09 23:01 UTC (2 days ago)
        
 (HTM) web link (rachelbythebay.com)
 (TXT) w3m dump (rachelbythebay.com)
        
       | senkora wrote:
       | This seems to be the reason for writing about the topic right
       | now:
       | 
       | > So, yes, in June 2015, I slowed down the whole company
       | [Facebook] by a second.
       | 
       | > Of course, here it is ten years later, and the guy in charge
       | just sent it back fifty years [by ending fact checking?]. Way to
       | upstage me, dude.
        
         | klooney wrote:
         | Facebook maybe? She mentions the cat factory
        
         | sunshowers wrote:
         | The fact checking is the tip of the iceberg -- it's what the
         | marketing machine led with because it's the least
         | objectionable. Far far worse is letting queer people like
         | myself be called mentally ill, though not any other group (e.g.
         | religious people). Yes, it's part of the common discourse, but
         | the common discourse is objectively morally abhorrent.
         | 
         | I worked at FB for a decade, and I now am rooting for its
         | complete destruction.
        
       | nonrandomstring wrote:
       | Nice descriptive article. I've done this on purpose too to debug
       | remote filesystem syncs and cryptography problems where machines
       | are out of sync. My GPS wall clock is handy for adjusting NTP,
       | but the time it takes to scan my eyes from the wall back to the
       | monitor.. you really do need two stacked like she did. So I now
       | figured to use transparrent terminals each logged into a
       | different host and lay them over one another running "watch -n1
       | date".
       | 
       | Would have been nice to have some more network, code and command
       | line examples. You need to set up a local ntpd and need to point
       | your local master at that temporarily. A better utility to write
       | would be "timediff -s1 -s2" that takes two time servers and shows
       | the offset. I bet there's a way to do that in one line. Anyone?
        
         | fishstock25 wrote:
         | > watch -n1 date
         | 
         | Um, that's a pretty inaccurate way to notice an offset in the
         | millisecond range, isn't it?
        
       | singleshot_ wrote:
       | I'm not sure I get the fifty years bit at the very end. Anyone
       | care to explain?
        
         | drewbug01 wrote:
         | I believe the author worked at Facebook in the past. The "fifty
         | years" bit is likely a reference to that company's recent
         | policy change that explicitly re-authorizes the use of what are
         | rightly considered slurs - even if they were common sentiment
         | to express publicly fifty-some years ago.
        
           | singleshot_ wrote:
           | Aye, that makes sense (not the slurs, but the explanation).
           | Thanks.
        
       | ashoeafoot wrote:
       | "Reenact the past , be the past, past becomes glorious present"
       | Isil
        
       | wodenokoto wrote:
       | > Of course, here it is ten years later, and the guy in charge
       | just sent it back fifty years. Way to upstage me, dude.
       | 
       | I feel like there's a link to a story missing there
        
         | isoprophlex wrote:
         | The guy in charge being Lizardman, who almost tripped over his
         | own claws to get rid of fact checkers, to allow calling non-
         | heterosexuals metally ill, etcetera etcetera now that we are
         | witnessing the second coming of The White House Cheeto
        
         | this_user wrote:
         | It's about the company's recent course change, which includes
         | Zuckerberg's Joe Rogan interview from last week. IMO this has a
         | lot to do with the younger generations. Gen Z were heavily
         | supporting Trump in the recent election, and Zuck has probably
         | realised that he needs to align Meta with them if they want to
         | remain a relevant player in the social media space.
        
           | jazzyjackson wrote:
           | I don't buy it, Gen Z hasnt swung right, but the young left
           | has swung apathetic. Zuck has his own reasons for sucking up
           | to Trump and it's not to stay popular with the kids, who have
           | already moved on from Meta properties.
        
             | XorNot wrote:
             | The hypothesis I saw proposed is that his marriage is
             | breaking down/actually done. Basically he's going the Elon
             | route.
        
               | fishstock25 wrote:
               | This is why it's important not just _that_ people do the
               | right thing but also _why_ they do it. Circumstances can
               | change, but if people do the right thing for the  "right
               | reasons", then what they do won't change.
               | 
               | Doing sth because of somebody's spouse is a bad reason,
               | imagine the relationship goes sour and they face a
               | divorce, boom they change their tune.
        
       | Animats wrote:
       | Is there a standard "smearing" period now? For a while, Google
       | had a 24 hour adjustment period, 12 hours on each side of the
       | leap second, while Facebook used a shorter period.
        
         | fanf2 wrote:
         | Not as far as I know.
         | 
         | The cosine adjustment that Google originally used is not the
         | best: NTP aims to measure the difference in rate between the
         | client's hardware clock and real time, and it works best if the
         | rates are fairly constant. With the cosine smear, the rate
         | changes continuously! If you use a simple linear smear, NTP
         | just has to cope with two small rate changes at the start and
         | end of the smear.
         | 
         | The smear needs to be slow enough that NTP's algorithms have
         | time to react without overshooting; 24 hours is a reasonable
         | choice tho you can go a bit faster. There's some disagreement
         | about when the smear occurs relative to the leap second; if the
         | leap is in the middle of the smear the max offset is 0.5
         | seconds, but if the leap is at the end of the smear the offset
         | is always slow. They were able to test the up-to-one-second-
         | slow scenario in a system-wide live trial, whereas they could
         | not do the same for the sign flip. I think if you can cope with
         | a 0.5 second offset from real time then a 1.0 second offset
         | should not be much more troublesome.
        
       | quitit wrote:
       | >Of course, here it is ten years later, and the guy in charge
       | just sent it back fifty years. Way to upstage me, dude.
       | 
       | Seeing that this was written by Rachel by the Bay, I thought it
       | was going to be a post about Facebook's recent policy change, and
       | indeed it was.
        
       ___________________________________________________________________
       (page generated 2025-01-12 23:00 UTC)