Chez Cosette

Discussion pour french.stackexchange.com. Bienvenue à tous ! Y...
Jul 10, 2023 19:42
But whatever the case the system has been around a good deal longer than I have been, and even when I started on SE (over 7 years ago), rep system was pretty much set in stone. The only major developpment that I can think of was the upgrade of question upvotes from 5 to 10 back in 2019, and even at that to my knowledge there was no particular community involvement in the decision.
Jul 10, 2023 19:34
@s.H.a.R.p.R.i.F.t Frankly I find this one intriguing, though maybe overdone. I remember a user a few years back on one of the sites I moderate who was a fairly high-rep user and literally had a 20 to 1 ratio of downvotes to upvotes (roughly, I forget the exact numbers). That kind of user could destroy the site with that kind of power; revenge downvotes are already a huge problem in some cases: imagine what it would be like if you have a single negative user trying to sink everyone else.
Jul 4, 2023 22:40
Interesting one though, I had never heard that expression either
Jul 4, 2023 22:39
@jlliagre ,,,or a moon ;-)
Oct 19, 2022 16:07
@AncientSwordRage I saw the flags but since I have an answer on the question I was hoping another moderator would be available (avoid a conflict of interest). It has however attracted low-quality content so I'm protecting it and leaving the other flags for another moderator to handle.
Aug 17, 2022 02:47
@yagmoth555 Thanks!
Aug 12, 2022 12:32
It would have the added advantage of correcting for the rounding in 10%
Aug 12, 2022 12:30
I may be taking it a step far but you could do a linear regression with % of words and chunk length as your independent variables
Aug 12, 2022 12:28
True
Aug 12, 2022 03:08
Indeed that is interesting. I assume the ratio is top 10% for each category as well?
Aug 11, 2022 17:14
Oops meant to link the answer below that one but whatever
Aug 11, 2022 17:12
24
A: Lemmatize French text

karimsaiehThe best solution I found is spacy, it seems to do the job To install: pip3 install spacy python3 -m spacy download fr_core_news_md To use: import spacy nlp = spacy.load('fr_core_news_md') doc = nlp(u"voudrais non animaux yeux dors couvre.") for token in doc: print(token, token.lemma_)...

 
Jul 10, 2023 19:21
I did not move the conversation to chat to faciliate argumentation. It will no more be tolerated here than on the main site.
 
Aug 17, 2022 16:58
@OlegValteriswithUkraine Thank you!
Aug 12, 2022 20:11
Ah I see
Aug 12, 2022 18:44
@ElectionBot help
Aug 12, 2022 18:43
Not sure why it thinks there are 0 people who voted, unless it's 0 since the last time it reported?
Aug 8, 2022 18:11
@livresque Was kind of hoping that some more active users would step up since I haven't been super active on SE since the failed election (a few people off sick which temporarily increased work load) but in their absence I'll stand.
 

 Droning On

General discussion for drones.stackexchange.com
Aug 17, 2022 02:02
Anyhow looks like it worked :D
Jun 10, 2022 17:53
Prevent it from freezing ;)
Dec 16, 2021 19:31
 

 Electrical Engineering

A place to talk with friends from the EE community about vacuu...
Jun 22, 2022 02:44
And the more points I try to sample the more space problems I have
Jun 22, 2022 02:43
I also looked into trying to set up parallel lasers on one side of the piston and photodiodes on the other side with a logic analyzer collecting the signal attached to a RPi but I was having trouble finding a way to guarantee that they were parallel
Jun 22, 2022 02:42
And even at that they were tentative quotes and if it takes longer than anticipated then the price tag goes up
Jun 22, 2022 02:40
Definitely a better option I just haven't convinced the powers that be that it's worth paying for, I looked into renting but couldn't find anything in my region under $2500
Jun 22, 2022 02:37
Good idea, I'd rather be able to programme it to get a maximum of 10% duty cycle on the pulses but it could be a cheaper more reliable option
Jun 22, 2022 02:34
@forest I looked at PWM, but I really have to have the light source on for a much shorter time than it is off, otherwise my double exposures will be really fuzzy and determining how far the piston travelled in the time between the pulses will be more difficult
Jun 22, 2022 02:32
Which is why I want pretty fast rise and fall times... I could theoretically let it go slower but if the velocity ends up being higher than anticipated I'd rather not have to start from scratch since the goal is to keep it relatively cheap (it's a sideline on the main research project)
Jun 22, 2022 02:29
Oh sorry the idea is to do a kind of 10 times exposure (akin to a double exposure) to measure the speed of a sort of pneumatically accelerated piece of metal; I can only measure the velocity over 5cm and it could be going as fast as 100m/s. The strobes are so that the film (or optical sensor) only exposes every microsecond or so (I'd probably be ok down to every 10 microseconds, but I'm aiming for the highest precision possible since it's an academic research project)
Jun 21, 2022 23:09
And also, I'm guessing that residual inductance will probably create regular lag? Or I suppose if I bump my wires or whatever there could be slight changes resulting in variable lag
Jun 21, 2022 23:08
Ok and when you talk about the inductive loop to and from the LED that includes everything from the pulse generator through to the LED?
Jun 21, 2022 14:58
Thing is I'm new to electrical engineering (sorry) and I haven't a clue where the bottleneck on speed is going to be - my research seems to indicate that LEDs have fast enough rise and fall times for requirements, but then I also need a driver (I think? If it's just a few pulses will direct current work?), for which I'm having trouble finding rise/fall times
Jun 21, 2022 14:56
Howdy, really broad but wondered if anyone could give pointers... I'm trying to get an extremely bright light (think 250k lux ++) that would have rise and fall times in the range of 100ns or faster and would be controllable by a custom pulse generator. My thought so far is something like instructables.com/… with a high-speed MOSFET (??) that would up the voltage across the circuit and a whole bunch of LEDs...
 
Jun 5, 2022 12:14
@OlegValteriswithUkraine I believe they only do that for pro tempore elections
Jun 3, 2022 17:59
Hey folks let's all take a quick breath... as has been mentionned this really isn't the correct place to discuss this, but even in the correct place making personal attacks isn't the appropriate way to address it.
May 31, 2022 22:25
@Sé1lanceLaléa Merci !
May 31, 2022 22:24
@Sé1lanceLaléa Bonne question, ça dépend clairement de la grandeur du site. Les deux sites que je modère présentement ne prennent pas plus de 10 à 15 minutes chaque quotidiennement, mais même si ça m'en prend plus il n'y a pas de problème, mon horaire typique inclut pas mal de temps libre même avec les deux que je modère déjà.
May 28, 2022 15:44
@livresque Je considère de me présenter ! J'ai été un peu occupé ces dernières semaines et je veux juste évaluer combien de temps que j'aurai au cours des prochaines années puisque je modère déjà deux autres sites. De plus, j'aimerais voir d'autres personnes se présenter puisque je n'ai vraiment pas été très actif sur le site dernièrement.
 

 Chat of Things

General discussion for iot.stackexchange.com
Dec 16, 2021 19:25
Congrats everyone, we're out of beta!!!!!
2
 
Oct 26, 2021 12:12
@KenGraham if you can still edit your nomination might want to mention that your definition of exegetical theology is taken from wikipedia
Oct 19, 2021 23:26
And as a sidenote looks like the election is getting a lot of late attention; 3 candidates so far :D
Oct 19, 2021 23:24
While it's not entirely identical to the discussed situation, I do see some merit in reading others' interpretations of the question; it can shed light on nuances I didn't see; it can also reveal details that have been edited out / ignored as @IñakiViggers mentions
Oct 19, 2021 23:21
What has happened to me on other stacks is that the answerer sees something in a question / has a similar background to the questioner / has more caffeine in their system than I, resulting in their being able to understand a question which I was 100% certain was unanswerable. In that case, their answer can shed light on the question, and in which case the question should absolutely be edited to clarify the inquiry
Oct 19, 2021 20:16
@IñakiViggers Good point
Oct 19, 2021 19:37
Huh yeah makes sense
Oct 19, 2021 14:01
I can definitely see how it could be a lot more nuanced than a technical subject
Oct 19, 2021 13:59
@SteveTaylor In other words it's insufficient to simply read the question; you have to read the answers as well?
Oct 19, 2021 13:59
@SteveTaylor Ah I see
Oct 19, 2021 13:03
Browsing closed questions seems like a great idea; I've been scanning meta to get a feel for site scope so far; will do that as well :D
Oct 19, 2021 12:57
@SteveTaylor Yeah I see... have a similar issue on IoT actually; though the question & case load over there is small enough that we can keep an eye on questions that we feel should be closed but want to allow the community to have their voice; we'll frequently wait until there are at least two or three close votes before throwing on the hammer