|
When I'm signed in the system knows my time zone and so I see Apr 1. When I'm not signed in it doesn't, and so the time is off by 5hrs and I see Mar 31.
cheers
Chris Maunder
|
|
|
|
|
Thank you for the answer, Chris. So you confirm that, with your account, you see the same that I can see with mine.
Obviously, the publication date should be the property of the publication, cannot depend on the properties of a reader.
Agree?
Thank you.
—SASergey A Kryukov
|
|
|
|
|
Yes, the publication date is owned by the content item, but it's displayed in the user's local time.
Time is relative...
cheers
Chris Maunder
|
|
|
|
|
Chris,
If you think a bit, you will agree with me. The publication is not a state of a document, it is its attribute characterized by some event in the past. Once defined, it should remain the same for all users.
Otherwise, we would have individual dates for the moment of time when Brutus killed Cesar because it was really different in different parts of the globe. We don't even know what were the time zones at that time. But people don't think this way. They record the event only by one watch, the Rome watch, and the watch existed at that time and that place.
Well, okay, but at least can we return back to your words about the magic power you've mentioned. Is it possible to set the date ad-hoc to appear the same for all users? By the way, it's still a bug: when I log off, my time zone doesn't change, for it should be April 1st in my zone, but it shows March 31. How about it?
—SASergey A Kryukov
|
|
|
|
|
I just did some simple experiments. If I am logged on, I am UTC+11, as expected. (back to +10 in a couple of days!)
If I log out, all the timestamps I looked at seem to place me somewhere mid-Pacific, UTC-10.
I'm guessing that's where SAK seems to be if he logs out too.
Cheers,
Peter
Software rusts. Simon Stephenson, ca 1994. So does this signature. me, 2012
|
|
|
|
|
Peter,
Thank you for experimenting with this.
Could you do the same thing: look at this article and look at the publication date. Then log off and look at the date again.
Will you tell me what you see?
Apparently, the publication date is just the date of some event. Once defined, it cannot depend on the time anymore.
Let's say, you have a time zone of an author or a time zone of some publishing house. Any of these time zones can affect the effective publication date, but this date, when defined, cannot depend on the point of view, right?
Thank you.
—SASergey A Kryukov
|
|
|
|
|
Logged in, I see it dated 1 April. Logged out, 31 March.
My guess is that it is a timestamp (e.g. Unix date() ), displayed with a date-only format. So if the timestamp were say, 2021-04-01T0100Z (1.00 am, April 1, UTC) that would show as 1 April in my timezone, but 31 March in, say, the USA.
As I noted earlier, it seems that users not logged in are assumed to be somewhere like UTC-10.
Cheers.
Peter
Software rusts. Simon Stephenson, ca 1994. So does this signature. me, 2012
|
|
|
|
|
Thank you very much, but I don't think it explains the problem or the logic of the implementation. The reader's time zone has nothing to do with the date (time, for that matter) of the publication. A publication is done only in one time zone, one or another, but only one.
Thank you again.
—SASergey A Kryukov
|
|
|
|
|
Which, to my mind, doesn't make any sense as that control (it says "Improve Solution") is clicked because an Answer poster has made a confusing typo in his "Solution" and it's that correction of the typo he made which might un-confuse matters.
I, for one, would choose the "Improve Solution" of the original post to make editorial corrections/clarifications to the questioners typing. And also note that each potential Solution has an "Improve Solution" control in it's thread-part; what would the point be having more than one "Improve Solution" control in the list of various Solutions if they all only point to the original post?
Finally, I submit that it would be appropriate also to change the name of the original poster's "Improve Solution" control text to something like "Editor's Hayrick Sortie" to distinguish it from "Improve Solution" (where clearly there's no solution to improve unless of course the comment of a cpian is in fact the best solution)
[edit]
My bad ... the text in the editor was so thin in my FF browser that I couldn't see it. Move along, nothing to see here ...
[/edit]
modified 31-Mar-21 13:30pm.
|
|
|
|
|
Sounds like a bug. Added.
cheers
Chris Maunder
|
|
|
|
|
... sorry Chris ... and that bug would be me. (see edit)
You pounced so quickly that now I'll never be able to delete this report 
|
|
|
|
|
Hi,
I see that my articles have been changed without any revision.
- Random tags appeared in tags section.
Integer Factorization: Dreaded List of Primes[^] is now Objective-C, which have never been.
- the 3 drop boxes are not any more.
- The code samples div do not work as it used to: the tab titles were the languages of each pre tag
<div class="code-samples">
<pre lang="c++">
</pre>
<pre lang="dbase">
</pre>
</div>
Now it says 'Text' and 'Text(2)'
long long TD_BF1(long long Cand) {
Count = 0;
long long Top = Cand - 1;
for (long long Div = 2; Div <= Top; Div++) {
Count++;
if (Cand % Div == 0) {
return Div;
}
}
return Cand;
}
function TD_BF1(Prod)
Local D, Top
Top= Prod-1
for D= 2 to Top
if Prod % D = 0
return D
endif
next
return Prod
Patrice
“Everything should be made as simple as possible, but no simpler.” Albert Einstein
|
|
|
|
|
The hamsters seem to be in the midst of a great overhaul. I just looked at my articles, and all but one (which has no code) been reclassified under a general C++ category. There used to be several subcategories for C++. Judging from current interests on the site, there will only be subcategories for web programming and C#!
The "Copy Code" thing seems to be new too, and there's no sign of the previous "Shrink/Expand" button or whatever it was called. EDIT: False alarm. It's still there but only seems to show up when the code reaches a threshold length.
|
|
|
|
|
Greg Utas wrote: and there's no sign of the previous "Shrink/Expand" button
I see them at usual place, but only on large pieces of code.
Greg Utas wrote: The "Copy Code" thing seems to be new too
It used to be there, but have moved lately.
Patrice
“Everything should be made as simple as possible, but no simpler.” Albert Einstein
|
|
|
|
|
Just working on this now.
cheers
Chris Maunder
|
|
|
|
|
Hi Chris,
Thanks for the hard work.
Should I change my articles tags ?
How to choose article category in new system ?
Patrice
“Everything should be made as simple as possible, but no simpler.” Albert Einstein
|
|
|
|
|
Choose the tags that best reflect what your article is about. We're trying to make things far more intuitive and simple. The worst thing you can do is pick too many tags.
Just keep doing what you're doing. You've got it.
cheers
Chris Maunder
|
|
|
|
|
Article : Integer Factorization: Dreaded List of Primes[^]
I try to change tags from 'C++, Objective-C, algorithm' to 'algorithm, C++', but when publishing, it come back to previous value.
Tried either typing and checking in droplists.
Patrice
“Everything should be made as simple as possible, but no simpler.” Albert Einstein
|
|
|
|
|
Hi Chris,
I tried to change tags, but still not found how to choose the category of article.
I think it would be nice to have something to choose the category, something that replace the previous 3 droplists.
I see changes of category of some of my articles, so thank you, but I fear one have find another solution than requesting corrections.
1 rather simple solution is to use the first tag of an article as category.
Patrice
“Everything should be made as simple as possible, but no simpler.” Albert Einstein
|
|
|
|
|
Hey Patrice
What category do you think is best for your article? Let's start with that and I can walk through what's happening with yours and hopefully get it organised better.
With regards to "how" to choose a category, it's based on the taxonomy priority of the tags. Take a look at the list of topics under Articles -> Browse Topics. The order of those topics determines how your article will be categories. For instance, if it has an AI and an IoT tag it'll be categorised as AI. If it has neither tag then we look at the tags your article has and work our way up the tag hierarchy to see if we can find one of those top level tags (eg C++ goes up to Programming Languages).
Does that help?
cheers
Chris Maunder
|
|
|
|
|
Actual setting is ok for me.
Some of my articles have changed of category again (on their own) since this discussion was opened.
Patrice
“Everything should be made as simple as possible, but no simpler.” Albert Einstein
|
|
|
|
|
Are you trying to classify articles based solely on their tags? I would try to change the category for some of mine, but some just moved to another category without me doing anything. It seems this is still in flux, so I don't want to experiment with it yet.
Many of the new categories make sense. But most are now oriented to types of applications, or to specific platforms, frameworks, or technologies. Few relate to general concepts, to the point where I'd just toss all my articles into the C++ bucket. But that would miss the point. What's more important: an article's topic, or the fact that its code is in C++? Those things are orthogonal, and I'd argue that the C++ aspect is far less important in most cases. So I'd rather use these categories:
- Design Patterns
- Exceptions
- Threads
- Memory Management
- Modularity
- Operations
- Debugging
- Templates
- Tools
- Software Lifecycle
Only #2 and #3 now exist. Others used to, even as C++ subcategories. But I'd rather they were in, say, a top-level "System Design" category. If someone wants to filter based on language or something else, an article's other tags can serve that purpose. But if they want to browse lists of articles, an omnibus C++ category makes little sense to me.
The above list isn't complete. Based on other articles that I've seen or might even write, Databases [noticed it now], Messaging and State Machines would also be good candidates.
modified 4-Apr-21 10:07am.
|
|
|
|
|
Yes - still fine tuning, and the general langauges tags are now at the bottom so will be selected as the main category as a last resort (or where they are listed as the only tag)
With regards to these:
- Modularity
- Operations
- Templates
I'm not sure. Modularity is more of a design pattern, or Design / Architecture topic. Operations? Do you mean operators and operator overloading? And Templates, do you mean template languages such as T4 or template applications?
The updated "Programming Topics" list will be:
Architecture
Design Patterns
Algorithms,
Compression,
Computational Geometry,
Debugging
Emulation,
Exceptions,
File,
Internet,
Localization,
Memory Management
Messaging
Parser,
Regular Expressions,
State Machine
Sorting,
Tools
Design / Graphics,
Printing,
String,
Threads,
Usability
(bold are the additions to be added)
cheers
Chris Maunder
|
|
|
|
|
Hey Chris, thanks for your reply. The updated list is definitely an improvement. To answer your questions:
- Modularity: I'm OK with "Architecture" or "Design Patterns" for this.
- Operations: Things like configuration, logging, and statistics. "Usability" works if I'm interpreting it correctly.
- Templates: C++ templates or, in other languages, generics. I think this one should be added.
I'd suggest changing "Parser" to "Parsing" and merging "Sorting" into "Algorithms".
|
|
|
|
|
Please post something when this is stable, with instructions on how to set an article's category or a description of how the system determines its category.
|
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
