I'm not saying the site is perfect, or that we couldn't do better - we always can. But I think talking of decline is unwarranted. There's also another effect: Many questions that can be asked have been answered. I increasingly often find myself writing answers that largely refer to other answers for more detailed explanations of some particular aspect, if the question isn't a duplicate outright.

I wish I could answer at previous rates, but alas work & expanding family prevent that

@KyleKanos Sure, real life happens. There's a definite bend in my rep graph too, starting at the beginning of 2018 when I started working full-time

If I imagine adding a family with children on top of that, I don't know how much time I'd find for the site, either

@ACuriousMind Are you writing any codes that compile? :P

(in the context of Kyle Kanos' nomination statement)

I have a feeling that's supposed to be some sort of joke, but I don't get it :/

@ACuriousMind Ah. Kyle explained that he/she can find time to moderate despite his/her busy schedule because his/her codes take around 20 minutes on an average to compile, and he/she scrolls SE within that interval.

Ahhhh, right

Definitely male, you can use he/him/his/etc for me

The basically interpreted nature of ABAP means there's not that much "compilation" time, and I also usually have enough other work stuff to do when I'm not coding

I think Kyle also linked the obXKCD m.xkcd.com/303

@PM2Ring I didn't actually link that, but was definitely thinking that when I wrote the line

:) Close enough.

Grumble, grumble, out of close votes again. Maybe I should stand as a moderator. When's the next election after this one?

Sadly though, such antics as depicted in 303 would be looked down upon here :(

@JohnRennie in one hour on Chemistry ;-)

@JohnRennie hopefully >1Y from a few weeks ago :)

@JohnRennie You still have 43 minutes to throw your hat in the ring.

@Loong Hahaha ...

You can largely copy my questionnaire.

I would, however, swap out some of the links...

@299792458 I thought it had just closed. Or does it close at 20:00 UTC? (The UK is currently on summer time)

2000 UTC

@JohnRennie 40 minutes according to election page

I get the feeling that you would (perhaps sadly) have a good shot at winning if your nomination said "I want more close votes"

@JohnRennie Closes in another 40 minutes. HURRY.

My two main motives would be 1) more close votes and 2) the ability to edit my chat posts after two minutes. Neither seem especially worthy :-)

@JohnRennie Let's see if the UK is still trying to leave the EU when we abolish DST in 2021 :P

@JohnRennie I'll give you some reasons: 1) You are anyways around all day here, and have no time-availability issues. 2) You have sufficient respect within the community, people look up to you. 3) You don't have any perceptible history of getting into altercations, so there is also a clean image.

I've always found DST to be utterly pointless. It would be heartbreaking if the UK does leave then doesn't abolish DST.

I wrote a script that traces header dependencies a while back. It is possibly my greatest invention because of how useful it is for refactoring. Sadly, it is closed source bc I wrote it for work, so no one will know how awesome it is :(

@299792458 Possible explanation for 3): No one lives to tell the tale ;)

> You don't have any perceptible history of getting into altercations

Hmm :-)

The most depressing part of it is that I work with people who think that "header hell" isn't actually a problem

@ACuriousMind Hahaha ...

@JohnRennie First rule of Physics Stack Exchange Fight Club, never talk about Physics Stack Exchange Fight Club

I had a minor tiff with John in my early days on the site, but I think he(/she?) was still pretty polite there.

@ACuriousMind There are various sources on the net around. Internet communities tend to become more and more closed and inbreed with time. My own experience is the same. SE is not exception. If there would be a continuous, conscious and effective mechanism to avoid the declination, yes it would be an exception.

Let's just say there are people who wouldn't urinate on me if I was on fire :-)

Because they'd kick you into a river?

@ACuriousMind The new question/month stats of most SE sites show a stagnation since around 2014-2015. The PSE avoided this since around 2016 (it depends on, how do we fit the stats). My opinion is, its reason is the well-documented increasing closedness of the community. It has many symptoms, for example the homework policy routinely killing worthy content and expelling a large part of the new users, but also many other.

I bet a few thousand homework-dumpers have no great love for John Rennie. ;)

@ACuriousMind Note also, that the size of the internet has multiplied since 2014, but the SE is stagnating since 2014. Thus, the relative size of the SE decreases. My experience is on the SE, that people simply won't understand what I am saying or they become hostile for that.

@peterh I'm often unsure if I correctly understand what you write. And I sometimes suspect that you mean the opposite of what you've written.

@peterh I for one, don't understand what you're saying. Taken literally, you've just said that all SE sites stagnate since 2014, but Physics doesn't (since 2016), which makes it rather hard to understand why you think there's a problem.

@PM2Ring Ok, sorry. What is not clear?

Oh crap guys, I joined in 2016. According to medical headline logic, that means I must be the primary cause

@ACuriousMind Oops, I wanted to write "PSE avoided this until around 2016". There was a time, a quite long time, as many SE sites have shown already stagnation, but the PSE has still grown. It was around in 2015-2016.

@ACuriousMind Here is yet another questions/month stat

I think it is natural that SE sites saturate over time. E.g. the supply of physicists is not infinite, and since academics always have been on the internet more than the average population, it is not to be expected that physics.SE's stats follow that of e.g. global social media usage.

Also, even new questions staying constant is sort of still indicative of "growth", since the amount of questions to which you can already find answers here - and therefore don't have to ask them - is also continually growing.

considering we stayed good during the networks decline is a really good sign about us IMO

While not every user finds their duplicate prior to asking, many of the more experienced users do, and don't ask a new question to begin with

@peterh that chart seems to be growing too me, with seasonal variances (probably due to school/University)

@peterh Here's an example from a couple of weeks ago on Astronomy astronomy.stackexchange.com/questions/30380/… "Probably no such event was found until now." I think you wanted to imply that such an event has never been observed, but your sentence can imply that it wasn't seen in the past but it has been seen now.

@ACuriousMind I have some db dumps here on my local machine, and also the view counts are not better. However, the view counts roughly correlates with the voting counts, and this query shows the upvote/month stats

@KyleKanos There's several overarching cycles that make the data noisy and hard to make out any definite trend. The only thing I can say for certain that there's so such clear upward trend as in the first few years of the site after graduation, but I don't think that's particularly concerning.

@ACuriousMind It shows the total count of the casted upvotes in that month, independently from the creation date of the post, where they were cast.

@PM2Ring Uhm :-( It seems I still have much to learn :-)

@PM2Ring Yes, I wanted to say that it was not observed until now, it is not observed now, but I don't know anything about the future (but I can't see a high chance for that)

@peterh I am not disputing your stats. I am disputing that they show anything to be concerned about. And the different statistics may not be as well correlated as you think, for instance, I see a "new users" stat in the internal mod stats that has substantially increased since 2016

@KyleKanos Yes, last some months have shown a good increase. Some months ago it wasn't so good.

The traffic stats sadly only go back to the beginning of 2018 and are useless for long-term evaluation

@peterh Physics.SE has an overarching bi-annual pattern corresponding to the academic year. The fluctuations within a year are very much dominated by that

@ACuriousMind Does it look like this?

@ACuriousMind you could probably graph homework questions over time and easily pick out exam seasons just from question traffic

@peterh yes

@299792458 Why? What's up?

There's also a weekly pattern where Saturdays and Sundays are slow, but that's not really relevant to this discussion

Ah, there's a neat stat that both displays the bi-annual pattern and shows a continuous upward trend at least until the mid of 2018 (but is invisible in public data.SE): "Feedback by anonymous users"

@ACuriousMind What kind of stuff is that looking at?

@Blue Is that bad?

I just read some of the chat log but I'm not clear on what the issue is. Just low number of nominations?

@JMac Non-logged-in users clicking the vote buttons, I think

When you do that, you get a pop up like "Thanks for your feedback, sign up if you actually want to vote", and SE indeed tracks that feedback

@JohnRennie Maybe it's just that users understand that moderating is boring?

@Blue no

@ACuriousMind oh that’s a cool one to track

physics.stackexchange.com/help/badges/95/constituent

@JMac It's surprisingly one of the least noisest stats I can find

@Blue Because the run up to the election failed to even state what moderators do. I had to ask for basic information like how much time the mods spend in their duties and what they actually do.

Adding to that the incredibly brief summaries in the questionnaire responses and the fact that I don't recognize any of the nominated user names, I feel like my vote would be mostly random.

I admit that I'm a little frustrated by this because the same thing happened in the previous election. In that case, I voted mostly on the fact that I knew one of the nominees had a very rational way of thinking and is generally impartial, so I went with that.

Is he that guy that insists on asking broad math questions here?

In this election, I have pretty much nothing to go on.

One of the nominees in this election even says explicitly that they don't have much time to spend on moderating, but that it's ok.

So I'm just kinda confused in general.

@Blue twice in the "role of rigor". He's pretty active on MO, so he should be aware of nuances of SE sites

I think he emailed me about it too, because I voted to close it as too broad. Some PO guys visited the site to reopen it, IIRC

@KyleKanos Justified or not this removal should not have been done with so little time remaining.

Not good.

Not good at all.

@DanielSank There was a town hall chat last election where everyone could ask the candidates some questions. I'm not sure if SE still schedules these by default, but did you feel that was useful?

@KyleKanos I found tools like include-what-you-use useful, but it's all a bit project and build dependent as to how much use such tools will be and how much customization they require of course

Actually, looks like it was 1 PO guy, 1 guy up for mod & one guy the crowd wanted to be up for mod that reopened it (+two others). Still not sure why such a blatantly broad question was reopened, but whatever

@Blue no idea. Probably, though

@alarge I looked into it, but include what you use is built for C and if it works for your C++ project, your are to be considered lucky

I've argued for not including anything in headers except for parent class headers, but I work with people who don't understand programming but understand SDEs :|

And that's a problem for me because I think we should be consistent

If we're inconsistent about what similar questions are or are not closed, what kind of message does that send to general public?

@KyleKanos Well I guess you could pimpl everything. But it sounds more like a company culture issue of course if styles cannot be enforced (lack of reviews or mandates mixed in with a legacy codebase etc).

@alarge we're mostly independent teams building different projects w/in same codebase, so each project space is a little different. New core team is aiming to make code more uniform, but it's been ~1.5 years in that direction with some progress

I think they're currently more focused on unraveling circular dependencies (APIs calling code that calls other APIs...)

@KyleKanos I don't think I've ever seen that approach work (though it's often tried). I think the best chance of getting something done is to scrap the codebase and start afresh (sell it on building on newer paradigms - both codewise and mathwise), though I've seen and read about that leading to messes as well.

@alarge The larger the codebase (or rather the provided functionality), the less likely scrapping it is going to be a viable alternative

@alarge if I were at a hedge fund, that would probably work. At a bank, we're regulated enough to not be able to do that swiftly enough.

We're also still in the process of eliminating legacy code in production... Something that's been in the works for 10 years now. So take that as a reference if why scrapping wouldn't work :(

...and now the code you wrote 10 years ago to replace the legacy code is the legacy code :)

Exactly. I'm still waiting for some head quant to suggest another rewrite & the headaches to ensue...

@ACuriousMind Well it depends I guess what the library does. Ideally the external APIs were relatively well designed and/or at least the library itself was not bloated with things it's not supposed to do. For example, your SDE solving library should be doing just that, and the piping of the associated metadata should be handled by a different system. Replacing the internal library, then, is less work for clients (though by no means is it ever going to be zero).

Ah, yes, if you have a clean interface that does not expose implementation details then it might be possible

@alarge I'm reasonably certain that every antipattern known to man makes an appearance in my library :(

That's why it's "legacy code" not "good old code" :P

It often feels like the code base was started by people who don't understand polymorphism, inheritance or abstraction but have used C++ to write C code

@KyleKanos As for banks, I think Standard Chartered uses Haskell, so I assume they scrapped something to build a wholly new system.

@KyleKanos "Object-oriented means packing everything in the methods of this one object, right?"

"Yes, a giant inheritance tree that avoids the diamond of death by liberally duplicating code between different branches will certainly improve maintainability"

I have one of these code bases as well :P

Copy+Paste is life

@299792458 too late for that... I am extremely disappointed by this last minute turn of events.

This decision removes 25% of the declared nominees, and we are down to 3 candidates for 2 positions... This truly has a huge impact and I wish this decision would have been immediately advertised.

bad bad bad...

@ZeroTheHero I have privately communicated to SE that (at least some of) our users feel that way, but if you feel strongly about it I suggest you additionally either make a meta post or use the Contact Us form.

@ACuriousMind In any future communication with SE please count me in those who feel hugely disappointed by this. I don’t see all the cards and it’s not my place to comment on the decision itself beyond pointing out that the timing provided very little time for anyone to take stock of this decision. @KyleKanos had accumulated a good number of upvotes for his answers to the questionnaire and so was obviously a candidate with support.

@knzhou thats not a bug its a feature™ actually have never seen mods themselves say much about light moderation. and wonder/ am a little )( skeptical about the idea that tight restrictions on question quality inherently/ inevitably increase the popularity/ "stickiness" of the site eg for professional physicists. would like to see an MO-like research direction at times but feel its not entirely clear/ obvious at all how to achieve it. theres a lot of buried attempts in the past...

found an interesting conversational AI dataset...looks like the decision to keep researching dialogue understanding was useful...

@ACuriousMind Unrelated to everything else question -- have you been to Aachen? I'm headed there Thursday for a conference next week

Haven't been to Germany, save a few days back in 2006 in Berlin during a study abroad

@tpg2114 Not really, no. I think I spent about an hour at its main train station ten years ago...

I don't speak any German, so hopefully I can manage to get by for a week... haha.

You should be fine, our broken English is excellent ;)

So long as I can communicate my need for beer, I will get along just fine

And provided I don't miss the 3am train back to Cologne when it's time to leave, or I miss my flight

Wait, Dusseldorf. Not Cologne.

I should spend more time figuring out my travel arrangements, but I haven't made my presentation yet either...

@tpg2114 Confusing the two cities will earn you enemies rather quickly, so that missing the train will be the least of your problems :P

(that's an exaggeration, but there is a rather considerable rivalry between Düsseldorf and Cologne)

Heh. I'm sure they would just assume it's typical American ignorance!

I'm looking forward to going. But it's the start of a very busy travel period with conferences and project reviews.

And making a presentation the night before a talk is a time-honored tradition, right?

I wouldn't do it any other way...

@tpg2114 Just make sure to only wear clothing that has American flag patterned cloth. I'm pretty sure every country loves that

@JMac That's all I own.

@tpg2114 You passed the test, I've called off the FBI hit squad

@Blue wait, what? Why are people quoting me? And when the hell did I write that?

Honestly, the amount of old comments and answers that have been wiped squeaky clean from my memory is pretty scary

That that comment was only a bit more than a year old

if that helps ;)

@ACuriousMind Interesting. I don't remember that at all. The thing I've found most useful is asking the moderators to describe their jobs.

In particular because both times I've done that it convinced me to not run ;-)

@JohnRennie I think it also says something that people who're not entirely sure they want the job start to throw their hats in anyway to prevent an election with only poorly known candidates.

Three cheers for everyone who stepped up.

Hip, hip, hurray! Hip, hip, hurray! Hip, hip, hurray!

Yes, congrats to the two who made it!