« first day (2608 days earlier)      last day (173 days later) » 

12:28 AM
Request Generator updates: alpha version (preferred) to 1.8.7.0 (GitHub) (install); release version to 1.6.5 (GitHub) (install). Both updates resolve a bug with the quick substitutions affecting CRUDE.
The alpha version update also includes: bug fix: a bounty is an issue only for cv-pls. Correct for SE not accurately handling the top margin when showing notifications; enable update checks on the alpha branch. The alpha version continues to not work with Greasemonkey 4 on Firefox, but works with Tampermonkey and Violentmonkey on all browsers they support.
@XanderHenderson New versions of the Request Generator have been released. See above messages. Sorry for the longer than expected delay. Or, at least, it was longer than I expected :-;
 
Hey, it works. Also, that is Seeking personal advice question with no context. ;)
 
1:13 AM
Good. I'm glad it's working. Sorry about that one. I had checked the issue and removed it from my mental list of issues, then copied a change and didn't re-verify.
 
 
2 hours later…
 
5 hours later…
8:07 AM
C7, C8, D10.
 
9:11 AM
The CRUDE version of the Unclosed Request Review Script (URRS) has been updated to Version 2.1.0.100 (GitHub) (install). It includes adjusting the tag colors, and a variety of other minor features and bug fixes.
@BillDubuque Please double check that the tag color adjustments work for you. I did slightly adjust the meta-tag from the last image I provided, in order to have it more clear against both the normal message background and in a onebox.
 
 
5 hours later…
2:23 PM
@XanderHenderson You may consider posting an answer on Mathematics Meta.
10
Q: Tag management 2019

Lord_FarinNew year, new tag management thread. Rules of the game are basically the same: Post your suggestion as an answer here if you see A particularly bad tag (a rule of thumb: «if I can't imagine a person classifying a tag as either interesting or ignored, I'm getting rid of it»), A tag that shoul...

 
2:54 PM
@Makyen Thanks for this great tool. I'm using this on Firefox 65.0.1 (64-bit) on Ubuntu 18.04 with Tamper Monkey installed. However, it doesn't work in my browser. Is there a way to get this working?
 
3:11 PM
@RRL @JoséCarlosSantos @rschwieb @BillDubuque: Every single answer on this thread is terrible and deserves deletion. I can't believe they have been left around for 6 years!
 
3:35 PM
C1, C2, C3.
D1, D2, D3.
D4, D5, D6.
D7, D8, D9.
 
@XanderHenderson Are you still using an old request generator? I ask because the tags are still unreadbale to me, yet Makyen implies this should be fixed in the latest version.
 
3:48 PM
Or maybe I need to load a new viewer (which is responsible for defining the tag colors?)
 
@BillDubuque It's the URRS that changes the look of the tags by applying different CSS on the CRUDE pages. There's nothing the Request Generator can do about it. Are you also running the URRS?
 
@Makyen Yeah, I just inferred that after doing a script update using TM. Looks much better now. Thanks!
 
I'm glad it's working. Sorry for the confusion.
 
@Makyen That's on me (for not reading the code). Btw, is there any chance that the faded text could be slightly less faded (say halfway between what it is now and unfaded). It's still a bit hard to read for me (I find myself squinting a lot to focus), and this makes reviewing more time consuming than it should be,
 
4:03 PM
@BillDubuque Sure, I can adjust that, or make it a user-setable option. Actually, I'd intended to make it adjustable, but hadn't gotten to it, yet.
 
@Makyen A user option would be great. No rush, since its usable as is (but I may end up with permanent squint lines if it never gets fixed!)
 
4:16 PM
@BillDubuque In the interim, if you're comfortable with opening up the console (press F12) running a bit of code, then you can use the following to set it to whatever opacity you desire: (function(){var opacity=0.5;$('#urrs-extra-style').remove();$(document.head).append(<style type="text/css" id="urrs-extra-style">.message.urrsRequestComplete:not( :hover ):not( .urrsRequestComplete-temp-disable ):not( .reply-parent ):not( .reply-child ) { opacity: ${opacity} !important; }</style>);})();
@Makyen I failed at SE Chat formatting 101, so you'll actually want: (function(){var opacity=0.5; $('#urrs-extra-style').remove(); $(document.head).append('<style type="text/css" id="urrs-extra-style">.message.urrsRequestComplete:not( :hover ):not( .urrsRequestComplete-temp-disable ):not( .reply-parent ):not( .reply-child ) { opacity: ' + opacity + ' !important; }</style>');})();
@GNUSupporter8964民主女神地下教會 I'm sorry to hear that it's not working for you. I've tested the URRS under Win 10/Firefox 65.0.2 (64bit)/Tampermokey and it appears to be functioning as intended. It will take me a bit to spin up a VM with Ubuntu 18.04. In the meantime, could you describe what you're experiencing in a bit more detail? Is there any error output in the console? I'm assuming you've reloaded the chat page after installing the userscript (just checking).
 
@Makyen The 2nd yields "undefined" when pasted into the console window for this page (the first got a syntax error).
 
4:36 PM
@BillDubuque I'm not surprised at the syntax error for the first. I realized too late that I hadn't accounted for both Markdown and SE Chat inserting non-visible Unicode characters in code formatted text (under some circumstances).
For the second code, "undefined" is what it should show in the console. That's the return result from the function, which doesn't return anything. You can adjust the opacity=0.5 at the beginning of that code to whatever opacity you want. It's a number between 0 and 1.
I'd intended to include some that explanation to begin with, but got distracted by things here. Sorry about that.
 
@Makyen Yeah, I already tried that. I see no difference between opacity 0.1 and 0.9, even after a (hard) page refresh.
 
4:54 PM
@BillDubuque It definitely won't work after a page refresh. It will only work from when that code is run to when you reload the page (i.e. it's cleared when the page reloads). I can format that as a bookmarklet URL, which you can add as a new bookmark, so it can be just a click of the bookmark to run the code. That doesn't fix the fact that it's not working for you at all, however. It works for me in Chrome and Firefox. What browser are you using?
@BillDubuque Actually, I got a bit side-tracked. If you want, you can completely disable the "completed" formatting from the options page, or select to have it applied on chat pages, searches, transcripts, and/or user pages. The options page can be opened by clicking on the button with the gear icon that should be to the right of the "upload" button in the user input area.
 
RRL
For deletion: DA, DB, DC, DD, DE, DF, DG, DH
 
@BillDubuque The options that affect the "complete" styling are these.
@BillDubuque If you want to change the opacity that's applied in your own copy of the script, it's on line 3564.
 
RRL
For deletion: DI, DJ, DK, DL, DM, DN, DO, DP
 
5:23 PM
@Makyen Ah, I hadn't yet noticed all the options. Very useful. I'll have to play around a bit more with the patch. Though I do have lots of software development experience, I have little experience hacking browser scripts so I don't have a good mental model of everything (e.g. precisely when the script fires, scope of dynamic patching, etc).
 
5:41 PM
@BillDubuque The userscript code is loaded upon each page load. If you edit in the editor provided by the userscript manager you are using, after you save the edit, you will need to reload/refresh the page in order for the changed code to be run. For userscripts and web pages, everything starts new with each page load. There are some places where persistent values can be stored, but unless things are explicitly stored in one of those repositories, everything is completely reloaded upon page load.
The insertion of the userscript is initiated by a completely separate process from the one that renders the page. The process that renders the page gets a message from the "background" process to insert the script, but that's all completely asynchronous. Thus, when exactly a userscript is run is indeterminate. There are options for the userscript to define about when it will run, but it's not going to be exact.
It's usually possible for it to be set to before or after some events, but even that may not be the case, in some instances. How well each userscript manager does with this varies form one to the next. One of the ongoing issues with Greasemonkey 4 is that it doesn't have the capability to obey the option to load the userscript at the earliest possible time.
 
 
2 hours later…
7:18 PM
@GNUSupporter8964民主女神地下教會 There is some past discussion on meta about the tag. That discussion, along with the popularity of the tag, makes me believe that there is little hope of it ever being eliminated. I think that it would be unproductive to attempt to fight that battle.
 
8:11 PM
Hi again guys
I don't understand why this post was deleted: math.stackexchange.com/questions/3124353/…
The OP says that they're pretty sure it's false and it's a counter-example question. Last time I checked, it was suggested by @rschwieb that it's difficult for the OP to show a good attempt at questions that ask for counter-examples because you either know the answer or you don't.
 
8:28 PM
I didn't vote to delete (and I don't have enough reputation to see the question, now that it's deleted), but I don't think that finding counterexamples is as black and white as that. One option is to find properties that a counterexample would have to have, e.g. math.stackexchange.com/q/3111327/274352
I think that is both an important aspect of finding counterexamples, and demonstrates some thought has gone into the question before asking.
 
@SantanaAfton I know. As I said, that was something said by @rschwieb to me when I objected that people that are now hunting others' answers have double standards because their own answers are better candidates for deletion
Earlier, another one of my answers with five upvotes was deleted. I checked the profiles of people who had voted to delete the question. All the three of them had answers that were more or less similar where the OP had shown little effort. One of them was a counter-example question. And @rschwieb said that he/she believed counter-example questions to be kind of different because it's difficult for the OP to find the direction they should go.
It seems to me that few people on this website are ruining the website. Deleting a question with an answer that has been upvoted seems to be against the policy of MSE
There have been times that I asked a question and then I wanted to delete my own question because none of the answers were really good and I had already found the answer on my own but the website said that I couldn't do so because people had invested their time in answering my question
 
Ah. I'm not aware of that particular context. That's interesting that it was deleted, and not simply closed as off-topic for needing more context. I see you around on questions I like -- you make good comments/suggestions. I doubt you're asking questions so low-quality as to be deleted outright.
 
@stressedout "He" is the right one. To elaborate on my line of thinking, those types of questions are sticky situations. Sometimes it takes someone proficient in the field to recognize how difficult a particular counterexample is going to be, and so it seems like shutting down all such questions might be a little unfair.
However, I do have little sympathy for questions which have extremely simple counterexamples, and no work.
 
@rschwieb That isn't really a statement. Shutting down all questions of any kind is unfair.
 
I can't tell you how many times I've used F_2\times F_2 as a counterexample on the site...
@stressedout True!
 
8:36 PM
@rschwieb In most cases, like 99% in my experience so far, the answer to a counter-example question is very simple.
And even if it's not, it becomes really simple after one suggests the counter-example.
Famous conjectures in number theory have been rejected by really easy counter-examples
 
@stressedout I have no data, but yeah we probably see requests for easy counterexamples more than hard ones.
 
Namely by finding a big number that doesn't work.
How is that not trivial?
Specially in the age of computers.
I think the community has already stated their opinion on this matter: math.meta.stackexchange.com/questions/29766/…
This meta post has 29 upvotes and only 3 downvotes.
 
Well, not everyone knows how to use computers effectively :)
 
So, my question is, why a known group of people disrespect the opinion of others?
And if 5 people have found an answer useful, why should it take only 3 people to delete the entire thing? How is that democratic or in line with our policies here? Is 3>5?
 
@stressedout My interpretation of that meta post is that people agreed with the content of the grey box, not necessarily with the title.
 
8:40 PM
I'm refraining from getting things personal that many of these hunters regularly answer very similar questions but they get away with it.
 
I think that there are two main issues:

1. Not everyone on MSE has the same values when it comes to how the site should be run, and
2. Anyone with sufficient time and mathematical ability can impact how the site is run, via the reputation system.
 
We've had questions in that past along the lines of "why do people delete stuff with accepted answers" which I'm pretty sure got blowback. (Because having accepted answers only does not really sound like a valid criterion.)
 
@rschwieb well, I think we should raise this on meta. I personally find it undemocratic and illogical that three people decide an answer with 5 upvotes should be deleted.
@SantanaAfton Yeah. That's the idea. But just because someone has a higher reputation, it doesn't validate their actions. Specially when they gained the reputation answering questions that were very similar to the questions they are deleting now. And they still answer the same kind of questions often.
 
@stressedout Well... that's probably a meta duplicate
Why don't you just try the "requests for re-open" thread, which is meant for swaying results like that.
 
@rschwieb Well, I did. It turns out that the majority of people on MSE don't know or don't care what is happening because I only see few names popping up all the time.
If it's the later, then I can't and won't try to change things and I'll simply find a place that suits my ideas better
But if it's the former, I think people deserve to know how a group of people are dictating their way of thinking on others just because they have higher reputation points.
 
8:48 PM
@stressedout It might be good if you spent some time cooling off too. Several of your word choices in the above comments would not help you very much if you stormed into meta with them. Just a suggestion, so that you can remain effective when presenting your idea.
 
@rschwieb Sure. I won't open the thread soon. I will wait to cool off and find a good way of presenting my question.
I don't want to provoke people and miss the chance of informing them (in my opinion)
 
@stressedout Yes. IMO if you generated a bunch of examples (like 25?) of the phenomenon you are describing, where good content was lost to the same group, then it would be very persuasive.
 
@rschwieb I can't do that because I don't have enough reputation to see deleted posts.
But my logic is simple really 5>3.
And the general policy of MSE seems to be against deleting questions that have well-received answers.
 
But complaining that voting is broken (we've all heard that plenty of times) or an unnamed group of users, or that you're going to ragequit, or about dictators is not bound to fare well.
 
It has been clearly stated in the grey box that you said people agreed to.
@rschwieb I don't think anybody cares whether I quit or not. I don't have a problem with mentioning names, but I refrain to do so because I don't want things to get personal
 
8:53 PM
@stressedout I... don't see an equivalence in what you said to that grey box. I see good advice in the grey box, but not an imperative that we have to enforce.
 
@rschwieb the voting system is broken. We all know that. But that's a different issue.
 
@stressedout Different issue? Isn't that at the heart of 5>3 ? Oh, maybe you were only thinking of the reputation system..
 
@rschwieb You can interpret it as an advice. Of course, MSE is moderated by its members and nothing has been said in an imperative tone about how things should be moderated precisely.
@rschwieb I didn't understand your last comment to be honest.
 
@stressedout Honestly if you just dropped a list of 25 examples, I don't think anyone would blame you for indirectly naming names. Those users have already inscribed their names on their work, and clearly nobody is going to prevent you from providing concrete data.
Perhaps even those among the vote names will be surprised at the cumulative effect of their own work.
 
@rschwieb Yeah. I got your point. But how am I supposed to come up with 25 examples when I can't see deleted questions/answers? You could help me if you want, but I can't do that on my own.
 
8:57 PM
@stressedout Yes indeed, that's a problem. Does your rep level prevent you from obtaining results via stackexchange data explorer/?
 
@rschwieb I don't even know what that is. :P
 
@stressedout well then! I get to show you...
BEHOLD: https://data.stackexchange.com/math/queries
I have to confess I don't remember what limitations it has in terms of visibility fo deleted stuff. I thought we were still supposed to be able to find such things
and maybe it is limited by rep, I"m not sure
tinkering...
 
@rschwieb This is cool. It seems that I can write SQL statements to find the information I'm looking for. Am I right?
But still, even if I find some examples of posts deleted recently, I won't be able to see their content because when I open a deleted link it says that it's been deleted and it can't be found.
Also, I strongly agree with this comment:
"Given that it's only possible to upvote a question once, why is it possible for the same person to vote to close multiple tines? This seems like a flaw in the system. The most that should be possible is that their vote persists but is cancelled out by one of the reopen votes,surely?" – timtfj Feb 8 at 15:13
 
Yes, it could be that they are not available. One could argue for some breadcrumb to remain for transparency
 
I think even in the judiciary system, once a case has been closed, it can't be investigated again.
 
9:04 PM
@BillDubuque I agree that it's better to avoid using a CAS there. In still more extreme case I do take action.
 
But it seems that this principle doesn't apply here. Once a deletion decision has been overruled, why should the same people be able to delete it again? It just doesn't make any sense.
 
@stressedout Yeah, I don't understand the rules behind re-voting. I've tried to re-vote a closure/deletion on occasion and I can never do it.
@stressedout It would seem to make sense if there were only ever one vote per person, but then we might ahve to do something about the vote imbalance between closing and re-opening
 
@rschwieb By the way, remember the meta question I showed you where the OP's answer had been deleted? The OP said that their answer had two upvotes only
Now the answer has 13 upvotes. I think it clearly shows that most people disagree that their answer deserved to be deleted in the first place.
 
Sad to report that it does look like we can't gather data on deleted questions this way... but you can still do closed questions, if that helps
It would be cool if one could petition for extended permissions to see that data purely as a researcher
(without the obligations of a moderator, but with at least some requirement to use it responsibly)
@stressedout I don't know that what you're describing about the voting is an objective measure. Purely the exposure from the meta question would have boosted the flow of voters, especially any of them inclined to sling sympathy votes
It's a complex thing.
 
@stressedout one should not overstress democracy, as neither 5 nor 3 are representative in any way; in relative terms the 3 is a much larger fraction of the eligible voters. But it's not really a vote in a democratic sense to begin with. By the way, for threds with a high score more votes are needed.
2
 
9:11 PM
@rschwieb I agree with you. But MSE is based on the voting system
@quid In that case, then the very fact that some people can vote to delete things while others can't is challenged. If voting numbers are not representative of anything, then why do we give moderation privileges to high rep users only?
 
@stressedout sorry I don't follow.
 
@stressedout I don't think the fact that math.SE is based on the voting system to should cause us to interpret inconclusive data as if it were clear, though, just because one number is bigger.
 
@quid High rep users have high reputation points because their answers have been voted up by others. If we say that 3, 5 or n doesn't mean anything, then rep points don't mean anything either. Right?
 
IMO all (post up/down) votes on this site need to be taken with a grain of salt.
 
@quid In that case, why are we giving moderation power to people based on something that's meaningless?
 
9:15 PM
@stressedout to have 10k+ points it's usually at least necessary to have been around on the site quite a bit.
In that sense it's not entirely meaningless even if votes were cast strictly randomly on posts.
 
@quid well, given the fact that people don't get rep points for the number of days they've visited the website, I disagree with that statement.
 
@stressedout you disagree with what statement exactly?
 
@quid this statement "to have 10k+ points it's usually at least necessary to have been around on the site quite a bit."
Well, I misread it.
Yeah, it's necessary but not sufficient. :P
 
Alright.
 
Now that you said it, I very much wonder about the fastest ascents from 1 to 10000 now :)
 
9:24 PM
From the top of my head I'd nominate Lord, and José was also quite fast. Maybe in the early days there were really fast ascents too.
There ought to be a query somewhere.
 
@quid Good. So maybe we agree on more than what was apparent at first glance.
 
Yes, that's well possible.
See y'all. I'll be back later.
 
@Makyen Thanks much for the overview. Do you perchance know of any links to info that would help one get up to speed quickly on such topics (targeted at expert programmers vs. newbies).
 
10:20 PM
@BillDubuque Unfortunately, not off the top of my head. That's a very distilled view based on working on userscripts and browser extensions for an extended time, following the various open source repositories for the userscript managers and browsers, and actually writing browser extension code that could do the insertion task at a (reasonably) deterministic point (although not for a userscript manager). There's probably something out there, but I'm not aware of it.
I'm happy to provide more information, both userscripts in general and SE specifically, but it feels like doing so here is outside the scope of this room (i.e. issues with these scripts, or others seem fine, but general userscript writing seems out of scope).
If you want to go to a different room, that would probably be a good idea. For lack of a better alternative, we have SOCVR Testing Facility. While it's mainly for testing, we do have development discussions there from time to time. There's also a reasonable amount of information on Stack Apps, but it's certainly not comprehensive.
 

« first day (2608 days earlier)      last day (173 days later) »