| Forum Home | ||||
| PC World Chat | ||||
| Thread ID: 81130 | 2007-07-17 12:06:00 | Yet another thread about the forum clock... | Erayd (23) | PC World Chat |
| Post ID | Timestamp | Content | User | ||
| 569767 | 2007-07-18 07:02:00 | I'm not predicting the end of civilisation as we know it because the PressF1 clock is a few minutes slow. I really don't think that will happen. It is a very minor thing. Roll on the webcam/sundial clock. Then people could argue about the last few milliseconds of the Equation of Time correction. |
Graham L (2) | ||
| 569768 | 2007-07-18 07:14:00 | I'm not predicting the end of civilisation as we know it because the PressF1 clock is a few minutes slow. I really don't think that will happen. It is a very minor thing. So why are you soooo worked up about it? If it doesn't make a difference either way to you, why oppose it? The computing world has moved on considerably since you last touched unix. Do you understand modern Linux, apache, databases at all? You might know that unix timestamps are used throughout as identifiers between them? Did you also know that browsers ask web servers about timestamps as well? Especially, I suspect, for refreshing/caching? You would expect some weirdness there if the time was skewed wouldn't you? You are a strange creature Graham. |
vinref (6194) | ||
| 569769 | 2007-07-18 07:29:00 | Now, now guys the forum clock time (or lack of) is nothing to get to excited about ... It is a long standing tradition (whether you like it or not) for the PF1 server time to be incorrect. Like all traditions, they sort of grow on you. :) It is just a wee bit more incorrect than normal right now. :p |
Jen (38) | ||
| 569770 | 2007-07-18 07:43:00 | Your're the one who seems to be seriously concerned . I don't care . That's why I want to see the sundial . I'm not opposing "correct" time; I'm amused at the importance some people seem to place on it . It's Wednesday . It will be Wednesday until midnight, plus or minus a bit . It doesn't matter, if postings carry a timestamp a few minutes away from the "actual time" . I understand those things . I also know that things which actually matter, matter . I was using computers before there was Unix . I still use Unix, and Linux . I know, from experience, that the clock being out by a few minutes on the server of this forum DOES NOT MATTER . / That is not "weird" . That is fact . Your problems would certainly have come from the date leaping back a few years when you booted the system . If you're concerned about browsers being confused if the time stamp is "skewed", don't worry . I wouldn't expect any "weirdness" to happen from that cause . I'll try to explain, if you don't understand that . Many of the computers running browsers will have "wrong" times . Many will have "wrong" dates . It doesn't matter that a user's computer has a different time from a server . Browsers can handle caching, because the time and date on the server, or on their own computer doesn't matter . The browser will compare the timestamp of its cached copy (which came from the server) with the timestamp of the server's current copy . If those stamps, from the same source, differ, the update is done . If not, it isn't . The same should happen with updates of software . As long as the server's time isn't jumping around randomly, there should be no problem . Weirdness would be normal if such transactions depended on all computers having exactly the same time . Modern (and older) software doesn't rely on that . It has to work in the real world, where time isn't a constant . ;) People get it wrong . People do daylight time corrections twice or not at all . They do it in the wrong direction . Where exact (and synchronised) timestamps are necessary, I'm all for precision settings . When it mattered to me I used to always keep my watch within one second . But on a web site, it's not necessary . Desirable, maybe, but not essential . |
Graham L (2) | ||
| 569771 | 2007-07-18 07:46:00 | It is a long standing tradition (whether you like it or not) for the PF1 server time to be incorrect. Like all traditions, they sort of grow on you. :) Just like the tradition of the old java version of PressF1 crashing when everyone went home at the end of the day or crashing on a Friday night :p :D Now PressF1 does not have that tradition, it needs another one & this time it’s the PressF1 clock been fast or slow :D |
stu161204 (123) | ||
| 569772 | 2007-07-18 08:49:00 | Your're the one who seems to be seriously concerned. I don't care.Funny that. It's taken Graham 6 posts in this thread to show how little he cares!! Muahahahahaha :lol: :lol: | Greg (193) | ||
| 569773 | 2007-07-18 10:35:00 | the sooner you all realise that it's not the pressF1 clock that is wrong, it's the rest of the world's, the better. now don't make me pull over the car and take those bold letters away. |
Jan Birkeland (4741) | ||
| 569774 | 2007-07-18 10:49:00 | I know, from experience, that the clock being out by a few minutes on the server of this forum DOES NOT MATTER. That is not "weird". That is fact. The following are tasks that will fail due to the clock being wrong: * Server updates (modification time in the future) * Most backups to this server (modification time in the future) * Any kind of kerberos authentication (time skew too great) * Reconciling server logs - with firewall logs, for instance * Some types of server clustering * Correct timestamping of posts :D I'm sure there are numerous other things, but those are the ones I could think of off the top of my head. How does something as important as server updates failing to work 'not matter'? |
Erayd (23) | ||
| 569775 | 2007-07-18 13:49:00 | If anyone other than Graham needs convincing, have a look at your cookies. Take the Pressf1 cookies for example. You may notice that they are stamped with local (your machine) time - not the webserver time. This helps the webserver determine whether the copy of the site in your browser cache needs refreshing or not. If the timestamp on the cookie is always ahead (which it is by about 15 minutes on this forum), it will always think you have a new copy in your cache. If it is behind, then the webserver knows your copy is stale. This is why, for me at least, I have to hard-refresh this site all the time (f5, alt-R etc). This forces the browser to ignore the cookie timestamp. Annoying, and time-consuming because the browser has to unnecessarily refetch a page. |
vinref (6194) | ||
| 569776 | 2007-07-18 20:57:00 | Just put Tardis 2000 on the server and be done with it! | SolMiester (139) | ||
| 1 2 3 4 5 | |||||