@JourneymanGeek Let's first see if there'll be any hint of SO Discussions (TM) replacing chat.

At that point I'll completely agree with you. Completely.

@AndreasmovedtoCodidact or rather, it'll be easier for me to believe as I see positive steps

@JourneymanGeek You'll have to look a long way for those. ;)

@AndreasmovedtoCodidact I've been doing this for a long time :D

@JourneymanGeek Do you have binoculars?

alas no

just a good sense of smell, and dogged persistance

Now, where are my dog GIFs...

→ 2 messages moved to Chimney

→ 7 messages moved to Chimney

→ 8 messages moved to Chimney

@AndreasmovedtoCodidact you're more optimistic than me, on this - SE never used chat as formal communication channel except for a very short while in the "Town Halls". But generally, chat just suck resources and full with things that don't earn them any money.

@AndreasmovedtoCodidact that's mp4 video, not gif.

(poor chat onebox can't know, so it tries to onebox it as <img> and sadly it fails - browsers won't show videos with such tag.)

Now let's look for banana gif...

@ShadowWizardLoveZelda Works fine to me.

@AndreasmovedtoCodidact oh, it works - but as video - and calling it gif is misleading - the host is misleading, not you. :D

imgur doing that for years, converting all gifs to video, but tenor did have actual gif's. Pity they fell too. :/

@ShadowWizardLoveZelda sniff, sniff

The term "gif" as I know it means animated image. Not video, not webp or whatever fancy stuff companies do these days.

Guess that for most people it's good change - you can't pause "real" gif, for example.

Animated reality.

@ShadowWizardLoveZelda didn't they invent a new image format for that reason?

@ShadowWizardLoveZelda stack used it internally for years

→ 8 messages moved to Chimney

@rene perhaps, which is x1000 worse. ;/

@JourneymanGeek past tense: used. Yeah. Not anymore, and they don't have sentiments for things that used to be useful or needed.

@rene they started renaming extensions mostly - its not even a new format ._.

@JourneymanGeek ah, but they did.

pretty sure its just a renamed mp4

then they claim its HD gif

> Notable examples are Gfycat and Imgur and their GIFV metaformat, which is really a video tag playing a looped MP4 or WebM compressed video

(en.wikipedia.org/wiki/GIF#Animation_formats)

"playing a looped mp4 or webM"

so lies

IT LIES LIKE A CAT

yup

→ 3 messages moved to Chimney

.....

...........................................................................

what the?

Welcome back to you too!

@JourneymanGeek Isn't that how you convert a mp4 to gif?

@SPArcheon only if you're an uncultured barbarian

@JourneymanGeek You called?

:D

Depends. Do YOU rename your file extensions and consider them converted?

!!/coffee Wolf to try and get some response

@ShadowWizardLoveZelda brews a cup of jQuery for @Wolf

@JourneymanGeek Yep

.txt to .gml and it's converted :P

.....

lol

@JourneymanGeek oh no! I see Wolf is contagious.

puts @Tinkeringbell in the those people list

I wasn't already in there?

Wow.

I lasted quite long this time.

speaking of people in santa's coal list

puts @Journeyman in the these people list

@JNat stackexchange.com/users/30217191/lan-do%C3%A3n needs to be nuked :D

@Tinkeringbell you were on the other lists

@JourneymanGeek they have backlog of 6-8 months already :/

@JourneymanGeek Ooh, you're one of those people with multiple lists.

Well they're not on my site so I can't do toilet on their account.

I just have one list of those people, regardless of what they do everyone ends up there eventually :P

@Tinkeringbell naturally

last time cross network spammer was nuked was...

ah so still not 6 months, but nearing it.

@JourneymanGeek you can rename to spammer

Nope

no accounts on my sites

@JourneymanGeek ah, thought they had and nuked already. Well, pity!

Well @Zoe can

@ShadowWizardLoveZelda in this particular case considering how prolific they are, I'd consider a CM escalation too :D

@JourneymanGeek meh, that's not considered very active.

I've seen about a dozen or so spam posts

across different sites

ah so that's what kept JNat busy now.

@JourneymanGeek You can count me in. I “convert” between .txt. and .md all the time.

clearly , the light of civilisation is far from your shores

Good.

@ShadowWizardLoveZelda You rang?

growls

@Tinkeringbell >:(

@JourneymanGeek She deserves more, doggo. Bite her!

@Tinkeringbell So what kind of list, is it? Do the elements have attributes? Is it sorted by rank? Graph, perhaps?

@Zoeisonstrike No, the phone up you must hang.

@AndreasmovedtoCodidact Just a list.

The messy kind, like the one you make on paper with a pen ;)

Aha! So a list with named sets.

Not sure what kind of lists you make...

But mine's just a list, not 'with' something, nor sorted

You know, the hand-written ones. A circle in the middle with the name inside, and lines extending from it, to other boxes with attributes inside.

@AndreasmovedtoCodidact That's not a hand-written list. That's some kind of association diagram

Think more like... groceries.

@Tinkeringbell Yes, yes. A list of those. Each element in the list is a diagram.

@Tinkeringbell Your lack of structure and order upsets me.

With half of the items squeezed in somewhere in between because they're in the same aisle but you didn't think you'd need them immediately

Three different colors of blue pen

See, you'll end up writing attributes and notes under the wrong names like this.

And at least 5 different kinds of handwriting

Stuff with question marks that either gets scratched out entirely or changed into exclamation marks.

Thanks, my brain just malfunctioned due to corrupted input data.

@Tinkeringbell So like SE's CSS?

I've never looked at that, should I?

Changed it to CSS, haven't looked in a while. Sure hope it's not in the HTML...

@Tinkeringbell No. Save yourself. Avoid it.

mhm, chat doesn't look so bad at first sight.

Hahaha oké, will do :)

Counted 36.

In chat.

@AndreasmovedtoCodidact chat was last touched years ago tho

and that was the 'new' mobile view

@JourneymanGeek I know. ;)

6522 "!important"s in the CSS on my MSE profile page.

> Stack cares about its users. Every one of you is important and we decided to show you that. For this reason, starting yesterday, we added a grand total of 6522 "you are !important" reminders to each of your profiles pages. Because you are important, and it is important that we remember the importance of showing you just how important you are, and to tell you that you are important there is no better way that writing that everything about you is important, including your profile.

@SPArcheon That sounds like something their marketing department would write. :D

> And if you ever open the developer tools to write some userscripts to extend our sites with awesome functionality, it's important for you to know how much you matter to us. And to emphasize just how much those 6522 messages of importance matter, we prefixed them all with exclamation marks.

@AndreasmovedtoCodidact We should write a proposal to the w3c for adding a second level "!!veryImportant" that can overwrite "!important" settings

@Sha hehe you're still GPT Wizard on metasmoke ;)

@SPArcheon I'll pass. :P But you have my support. The description should read: "for when websites are stupid"

@AndreasmovedtoCodidact "To be used when you want to abuse !important but the previous dev already did"

@SPArcheon Maybe just add indexing to the "!important" instead?

@AndreasmovedtoCodidact not AI enough. Use the stable diffusion weights syntax

!important:1, !important:2 etc

Also keep the () notation too.

!important< (!important) < ((!important))

How about just attaching a Brainfuck script that selects the !importants it wants to set custom z-values for?

Sorry, a Brainfuck script used as input to ChatGPT.

The output decides.

@AndreasmovedtoCodidact Amateur. I guess you didn't study the Discord's "How to be annoying" encyclopedia.

You should not make things hard, just annoying enough

if they are too hard, people will simply not use them.

If they are annoying, then people will hate them but still use them

so...

keep it simple.

just make the numeral written in fraktur

or roman number

!importantI, !importantII, !importantIII, !importantIV...

@SPArcheon in emojis.

@Zoeisonstrike no, it was @Spe

runs away

@Petəíŕd The Spring Wizard yeah, the name isn't synching with SE

And dunno how to change it.

@ShadowWizardLoveZelda confused noises

@Marco I didn't forget your request. Here it is!

@Zoeisonstrike careful not to wake the dogs with those noises. :P

See, they're all asleep. Waking them isn't a good idea:

@ShadowWizardLoveZelda whipped cream missing

@ShadowWizardLoveZelda Thank you, but I ordered yesterday, why did it take so long? What a lack of good behaviour...

@rene Marco didn't ask for it.

That is a missed whipped cream opportunity

@Marco sorry, the spell is "Make Hot Chocolate". and in the tiny letters it says "might take up to 6 days".

or 8 days

Ok, no problem

I'll try to upgrade the spell, but it cost 25 wizard coins.

@ShadowWizardLoveZelda They'll be fiiiiiiiiiiiiiiiiiiiine

@ShadowWizardLoveZelda ok, you can take my Meta points, I have that amount.

@Zoeisonstrike well in your case, if they'll wake up they'll just dance happily around you. :P

@Marco ah, conversion rate is 100k reputation points = 1 wizard point.

:(

Need to combine Jon Skeet rep with few others.

ok, I don't have that amount, it's for a next opportunity.

@ShadowWizardLoveZelda I see what you did!

How many wizard points for a wizard coin?

@ShadowWizardLoveZelda very hard

@Zoeisonstrike do you? Because I don't. :D

@Zoeisonstrike ah, oops. lol

@ShadowWizardLoveZelda you said the upgrade is 25 wizard coins, but that the exchange rep is 100k rep for 1 wizard point

Caught in 4k :P

umm, of course it's on purpose. :D

So, 1 wizard point is worth 10 wizard coins in the black wizards market.

Also, I have a 1 upgrade free voucher

No answers yet: meta.stackexchange.com/questions/398964/… :(

@Zoeisonstrike but only you. :(

The bounty expires in 5 days.

@Marco well, some good comments, worth a try.

It'll be fine. Can I use community's rep? :P

You can try on one of those suggested in comments, and if it works, i.e. question is well received, you can ask the comment author to post it as answer. @Marco

Yes, there are 3 different website recommendations hahaha

@ShadowWizardLoveZelda yeah, nice suggestion.

argh janitor is here need to go before he starts washing the floor. Cya!

@ShadowWizardLoveZelda bye...

@ShadowWizardLoveZelda the old...

but instead of you painting yourself in the corner it is the janitor washing all the floor around you so that you can't move.

@ShadowWizardLoveZelda oh, the old Alley Cat dos game

@Marco I can't understand what you want to ask, if others feel the same, that might explain the lack of answers. I think you can break this into two separate questions, one asking about neural networks and how much data they would need. This will presumably depend on your model and the question you want to ask of your data. This could be on topic on Artificial Intelligence.

And then, once you know how much data the NN would need, another question, about how to get it. That could be on topic on Earth Science.

@terdon Hi. I am specifying the type of data (ionospheric) because I believe that depending on the type of data this will influence the smaller or larger minimum amount of data needed to obtain a reasonable generalization when using neural networks for data prediction.

@Marco sure, but that means that only someone familiar with both the data and the method will be able to help. If you can separate the two, you can ask the experts on the data type at earthscience and the experts on NNs on AI and you should be all set.

This type of data (ionospheric) depends on the progression of the solar cycle, for example.

You need to break this down into smaller pieces and ask separate questions. I don't think there's any other option.

Yes, I understand you, I believe I would need to do this, unless there existed an interdisciplinary website about these subjects, although scicomp.stackexchange.com could be that website.

@Marco Only on the NN side, not on the specifics of your data.

You could try posting on earth science but you will only get a good answer if you are lucky enough to find someone with expertise in both atmospheric analysis and neural network minutiae.

@terdon ok, thanks.

@terdon ok, thanks, but what would be a data-only question that I would ask on Earth Science SE, for example?

One of the comments on my question suggests that I ask a question on a site more related to machine learning, in this case, Data Science SE, so that I don't specify the type of data, perhaps this is the only and best thing to do, because I believe that by analyzing the data itself (without knowing what type of data it is), (cont...)

it would be possible to check, for example, whether the quantity in question is already sufficient, I am thinking, for example, of an analysis of the spectral entropy of the time series.

So it would perhaps be an exclusive subject of Information Theory.

Finally, what would be your suggestion for editing my question so that it would be more readable, considering that you didn't quite understand its content?

@Marco I don't really know that site, but I am active on other scientific sites and asking about your actual issue, explaining the question you are trying to answer, is usually on topic. So if the scientific question you want your data to answer is about Earth Science, then the whole thing might be on topic there.

@Marco Give more details about what you really want to ask. Ideally, show the question, don't describe it.

@terdon what I mean is that I don't know how I would ask the question just about the data without bringing the question of predictability into question. That's why perhaps the only alternative is to focus only on Information Theory. In this case, I believe the best site would be Cross Validated SE.

@terdon hmmm, ok, thanks, I will try to improve the question.

@Marco if you can express it in a way that doesn't depend on knowledge of the kind of data you work on, sure

Yes, I would try to do that.

It would be a broader question about the predictability of a general time series, although there are already some questions on the subject.

That is, how do I check if a given time series is predictable enough?

I imagine that, given a result using a certain method (for example, spectral entropy), if it is not predictable enough, one of the reasons would be the lack of data.

That seems abstract enough, sure. But I really don't know the scope of the math sites, so I'm not the right person to help with that.

Perfect, thank you very much, in any case you are helping me improve my ideas, I saw in your profile that you are a computational biologist, so you have a good idea of the interdisciplinarity of this type of study, my subject would probably be about computational physics (using machine learning, of course).

@SPArcheon something like that. My office is in the entrance to the whole floor, so when the floor outside is washed, I can't get out for 10 minutes until it dries. Well, I can, but then get stared at by the janitor. :D

@SPArcheon colors don't match, interesting. For me it was blue/purple theme, as in the screenshot I posted. But yes, one of the best games. ;)

→ 20 messages moved to Chimney

@ShadowWizardLoveZelda woof?