There appears to have been a problem with the update of the Poll. Some made comments before the Poll appeared on the front pages - or so it seems. How they accessed it I do not know. The Poll released at the programmed time but comments that had already been made seem to have been lost. I am investigating the cause.
My apologies to anyone who had already made a comment. All I can do is ask that you make your comments again. We have not seen the problem before so if anyone can provide additional information that might help in identifying the cause it would be most useful. Did you access the Poll via the front page? If not, can you explain how you did access it please? Which browser are you using? Did anything appear 'different' to when you usually access the site?
[Addendum: The poll displayed on my front page has reverted to the previous poll again. If anyone else has seen the same please confirm in a reply to this Meta.] After a minute or so it returned to displaying the correct Poll.
(Score: 4, Interesting) by Mojibake Tengu on Saturday April 05 2025, @03:51PM (10 children)
What I see is SoylentNews articles and posts use local time in a browser, systematically. Presumably local time computed by server, not browser, because my workstation runs in UTC but SN shows local solar time.
So, I recommend to determine first what kind of time is actually stored in database and if zone information is not lost in the process, and if time conversion happens or not.
The kinder you are, the easier it is for wicked people to morally coerce you.
(Score: 3, Informative) by janrinok on Saturday April 05 2025, @03:59PM
The database stores in UTC.
You can choose your display time from you personal preferences page. It will convert time to whatever tz you have chosen.
(Score: 3, Informative) by janrinok on Saturday April 05 2025, @04:02PM (8 children)
All processing is done in UTC - although I can also see a local time for queued stories but NOT for Polls.
(Score: 2) by janrinok on Saturday April 05 2025, @04:26PM (7 children)
Having looked more closely, the difference in the release time and it appearing on my display was a matter of around 20-30 minutes. This is not a bug with tz data or conversions. In addition the display changed several times over a minute or two from the older poll to the newer poll and back, before settling down. It happened on 2 different computers and with different browsers. However, if nobody else saw it then it suggests that it is something more local rather than on our site.
The best way to describe what I saw was that it was if there was a separate CMS between the site and myself - but this is only a description of the event and I am not aware of there being any such CMS, nor am I suggesting that one actually exists.
I have never seen it before and I will be entirely happy if I never see it again!
(Score: 4, Informative) by Mojibake Tengu on Saturday April 05 2025, @04:55PM
Well, since Poll Booth is accessible directly by link such as
https://soylentnews.org/pollBooth.pl?qid=202 [soylentnews.org]
then it is trivial to craft a local autorefreshing page (meta http-equiv=”refresh” content=”9" for example) expecting a future next poll by predictable number, so the booth becomes available to that page when it is created, even before it is published on main page. Or wget/curl/python script.
You should check the logs for booth hammering.
The kinder you are, the easier it is for wicked people to morally coerce you.
(Score: 2, Interesting) by Anonymous Coward on Saturday April 05 2025, @05:39PM (3 children)
It's possible that some people's computers / tablets / (maybe phones) are not synced to Internet time and the clock is (way) off.
Maybe their time sync is turned off.
Over the years I've had many problems with Windows not properly syncing time clock.
I know of at least one Samsung tablet (that I occasionally use at a workplace) that refuses to sync to Internet time, even though it's properly configured and turned on. And, its internal clock is terrible- looses hours per week.
(My cynical guess is they don't care to make it accurate because pretty much everything is Internet time synced nowadays. It may not even have a quartz crystal based oscillator- just some inacurate ceramic resonator perhaps.)
It's possible a gateway/router/firewall is (mis)configured to block NTP.
(Score: 1) by pTamok on Sunday April 06 2025, @11:23AM (2 children)
I have a tablet that does the same. The internal clock is wildly inaccurate, and is consistently slow with respect to wall-clock time. NTP is set up, and appears to be ignored. It is an old tablet running a no-longer supported version of Android, but the behaviour is recent, which makes me think that Google/Alphabet have changed something, or turned off some obsolete (to them) services.
The tablet is boot-loader locked, so I can't update it with an independently supported OS based on Android, and is so old that it no longer gets browser updates, and some certificates are time-expired, making browsing some sites impossible. If I get around to it, I'll probably try loading a browser from F-droid,
...some time passes...
Ah, right. Just added the ISRG RooT X1 certificate.
https://www.stoutner.com/lets-encrypt-isrg-root-x1-and-privacy-browser/ [stoutner.com]
I still want a linux-kernel tablet with a non-Android userland - or perhaps I should get a Pixel Tablet and load GrapheneOS.
(Score: 2) by Mojibake Tengu on Sunday April 06 2025, @03:55PM (1 child)
Huawei now offers both HarmonyOS and Linux though Linux maybe only with some models of PC/Notebooks so check the specs.
My latest non-Android toy is a MatePad Pro tablet with Harmony but its dedicated only purpose is for remote shopping in China seamlessly.
I never hesitate having a fresh machine for a single technical purpose, that's my way of OPSEC. It certainly complicates tracking, for everyone out there.
HarmonyOS is partially transparent, you can get HMOS/NEXT SDK for crosscompiling your own apps so technically having a full Gentoo inside Harmony is achievable. RAM and storage on MatePad is quite sufficient for that, 12G/512G.
You may have a good use for PineTab, PineTab2 or even for PineTab-V, but PINE64 ARMs are significantly weaker than Huawei and RISC-V is a joke.
Western technology is going downspin.
The kinder you are, the easier it is for wicked people to morally coerce you.
(Score: 1) by pTamok on Sunday April 06 2025, @04:37PM
Thank you for the suggestions.
I was interested in the PineTab, but even the newest iteration has significantly worse specifications than my 11-year old tablet, which was nowhere near 'high end' even when new. It is shocking how poor non-proprietary hardware remains compared to all the 'lock-in and track you' stuff.
(Score: 2) by deimtee on Sunday April 06 2025, @02:29AM
Daylight savings finished in Australia at 2am local time. Maybe the database hiccuped in sympathy.
200 million years is actually quite a long time.
(Score: 1) by pTamok on Sunday April 06 2025, @10:19AM
To me, that looks like a caching problem, and/or a load-balancing problem.