« first day (2876 days earlier)      last day (2048 days later) » 

8:00 PM
Not unless my chief title gives me the ability to bring in more data
I've actually been looking for a data science job, but it looks like all of them want more education or more experience. I'm split between getting a data engineering job where I can get the experience or trying to find the spare time to do more Kaggle stuff
 
hmmm
I get messages from recruiters pretty regularly
and I technically only have 6 months of "data science" experience within a company
 
I get messages from recruiters for software positions, but haven't gotten any for data positions
 
so it doesn't seem like the experience requirements should be all that hard? A lot of jobs will list crazy requirements cus they know people will inflate their qualifications.
Just last week I got contacted by a recruiter from Alibaba and one from Midea...I think people on linkedin are catching on to the fact that I'm Chinese...lol
 
Yeah they may not be. I've even seen education requirements be ignored. Maybe I should just apply for them and see how it goes lol
 
You should
 
8:07 PM
I did see a posting for a "recent undergraduate" with 3 years of experience in Spark, Hadoop, and Hive lol
I have no idea where that's even coming from. The handful of people I know that work in data don't actually use those
 
I mean, the Midea job offering was like "we need you to be an expert in Java and work in Java IDEs and be familiar with package managers in JAVA e.g. Maven"
and then I told the recruiter I wasn't an expert in Java
and she's like "it's not really necessary, actually the role is more for Python"
LOL?
 
people copy paste stuff like that all the time too
so the requirements might just be from another role
I'm pretty sure they copy pasted the JAVA req. from a software engineering role
who the hell does data science in Java?
 
Whenever I graduated I applied for a backend developer that was listed as an entry position. The interview went quite strangely. Then I talked to someone I knew who worked with the guy who interviewed me. Apparently the position was looking for someone with 5-10 years of experience in erlang
 
lol wut
 
8:10 PM
I think most of the Apache stuff (Hadoop, Spark, Storm, etc) is written in Java. Maybe that's why?
Yeah it was weird to say the least
 
I feel like at that level it should be more software engineer's jobs to figure those things out
 
@danielunderwood "Whenever" you graduated? Did you graduate very often? Did you forget when you graduated? :P
 
Anonymous
@danielunderwood It's very weird for them to demand something so specific as "3 years of Spark, Hadoop and Hive" from a recent undergrad
 
Anonymous
Are data science job listings more or less always that vague?
 
Anonymous
I don't have much idea
 
8:12 PM
nobody knows what they want from a data scientist
 
I also had one for a physics programming position that pretty much listed my experience. The posting seemed to list my experience, but disappeared on the morning of the interview. Evidently they changed it to a quite different software role
 
apart from like a couple of companies at the forefront of data
e.g. IBM, Microsoft, Google those guys
 
@ACuriousMind I'm bad with words sometimes
 
but if you're going for data science in like an insurance plan...they don't know shit about it lol
 
Yeah I guess it's true that there are a good many places that were told they need a data scientist, but have no idea why or how
 
Anonymous
8:13 PM
@enumaris Who does the interviewing then? :P
 
A manager who doesn't know data science lol
 
Anonymous
Damn
 
What is a data scientist, actually?
 
stand back, I'm doing science!
but with data
 
A dataologist of course
datologist?
A scientist that ignores all scientific standards!
 
8:15 PM
@ACuriousMind generally someone who builds data-centric predictive or analytic models...
 
Anonymous
It should be called "business statistics" I guess. Makes more sense
 
@danielunderwood I think you're looking for 'dadaist' ;)
 
Anonymous
Unless of course, you're doing some real research
 
"Business statistics" sounds a bit more like "data analyst" which is more for looking at summary statistics, reviewing metrics and the like...
well I didn't do "real research" but I have built my own models for the job...e.g. my deep learning based spelling corrector
 
Anonymous
@enumaris I doubt those companies know the difference anyway XD
 
8:16 PM
I guess it's probably also worth pointing out that a lot of people doing the hiring probably have no idea of the difference between data scientist, engineer, and analyst
 
Yeah, the difference between "data analyst" and "data scientist" is kinda subtle and not overall understood
 
Or the dreaded "data science engineer"
 
@danielunderwood most of the companies don't even have a well defined data and analytics strategy enterprise wide
Or are just beginning to implement it
It's why I'm reading up on that stuff right nowlol
 
The only time I have to deal with data is when the database admins complain that our persistent data is taking too much space :P
 
I've heard a bit of "data science is the future" from my company, but I don't think they have any idea of what to do with it
 
8:18 PM
@danielunderwood I think that's generally the case with the future ;)
 
Anonymous
@danielunderwood That's.......so cliché ;)
 
@ACuriousMind Isn't SAP the type of company that should have tons of storage?
On the bright side, they also seem to think people with a physics background are the key and I'm the only one of those at the moment
 
I once advised my consultant friend to do a query
 
@danielunderwood Sure, but the individual systems/servers aren't necessarily all king-sized.
 
that would construct
a temp table
of over 10 trillion rows
we got HP's SQL servers to run out of memory
that was fun
 
8:21 PM
Let me introduce you to iterators, generators, and cursors
 
they have those in SQL?
 
They generally have them in clients. I have no idea how the servers handle them though
 
And essentially, we store the same kind of data at the customer's systems. Given that HANA is in-memory so storage is more expensive than in traditional disk-based DBMSs, they don't like an additional TB lying around for no discernible reason :P
 
I don't get the resources to put anything in-memory other than a cache :(
Well I heard of fixed-size databases today and suppose I could do that, but I don't really have any need for one of those
 
I've run out of memory for displaying rows at my job...but the rows can be pulled through in python or tableau so it seems the "out of memory" error I get is simply for displaying it within my SQL client
 
8:25 PM
Well yeah if it's trying to load all the rows to display
You generally get a cursor if you're using a db client in python and that loads one at a time
Unless you dump them to a list. Then it would be all of them. But the normal thing would be to build a sequence of generators for processing
 
it's only 18 million rows or so tho
not even 1 billion
sigh
 
Well that's going to be 18MB at least just to get the IDs. 180MB if you have 10 columns with a byte each. If your columns have text, that can easily jump a couple orders of magnitude
huh I've never thought of an ellipse as a "circle" with constantly changing radius...neat
 
o.O
any shape is a "circle" if you allow a constantly (and not necessarily continuous) changing radius...
2
That's just saying you can draw shapes using polar coordinates, no?
 
Anonymous
Indeed
 
8:35 PM
whelp, finished reading the whitepaper
 
@enumaris I think you mean differentiable instead of continuous. If you allow non-continuous changes, the circumference is not closed and it's not really a shape...
 
Well I guess you have to constrain an ellipse a bit more to be smoothly varying
 
@ACuriousMind right, I thought of that, but then...I wanted to be more inclusive :P
 
@ACuriousMind Yeah I was gonna say
 
I guess that wouldn't even restrict it down to an ellipse since ellipses need a sort of symmetry
 
Anonymous
8:37 PM
focus, directrix, blah blah... :)
 
@enumaris Wikipedia: "White papers may be considered grey literature."
Good naming.
@danielunderwood Yeah, smooth just gets you something without corners
 
@danielunderwood the way the radius varies as a function of angle is quite constrained for the case of an ellipse - it has equation $r(\theta) = b/\sqrt{1-e^2\cos^2\theta}$
 
But, as differential geometry taught me, a "manifold with corners" doesn't need to be a manifold, so I'm fine with calling arbitrary polygons "ellipses with corners".
 
Where $e$ is the eccentricity and $b$ is the semi-minor axis
@ACuriousMind I call them circles with variable radius and corners
 
@enumaris I feel like I should probably know ellipses better than I do
 
8:41 PM
Let's just call everything "fat points" and call it a day :P
 
sounds good :D
@danielunderwood gotta pay more attention in high school algebra class
(why was it in algebra class? no idea)
 
We never covered them in high school though
 
Anonymous
More like coordinate geometry class?
 
conic sections weren't covered??
 
Anonymous
We had a whole book on them ;_;
 
8:42 PM
I had conic sections covered in my "advanced algebra" class
 
Anonymous
BORING...well, at least when they stretched it that much
 
cus parabolas were covered
 
Nope. I had no idea what a conic section was until after I graduated college
 
then they went into hyperbolas, ellipses, circles...
@danielunderwood well then...
blame your school
 
@Blue ...you had a whole book on conic sections?
 
8:43 PM
I evidently went to one of the top public schools in my state...so they say
I don't really agree with that as far as math/science goes though
 
to be fair, I didn't understand much of it in high school lol
 
@enumaris in high school? lol
 
Anonymous
@ACuriousMind Yeah, a big fat one!
 
@danielunderwood Maybe you were looking at the football rankings? ;)
 
@SirCumference yeah...they went over directrix and foci and all that whole shebang
 
8:44 PM
i don't think i learned anything in high school geometry
it was a quagmire
 
We were a tiny charter school without a football team unfortunately
 
my high school geometry was mostly on trig and stuff like that
no conic sections in geometry oddly enough
@danielunderwood how does a charter school get classified int he best public schools rankings?
 
Yeah our geometry class was pretty much all triangles. I have no idea how they spent a whole year on triangles when you should pretty much be done after right triangles
Maybe the rankings were just non-private schools? Not really sure
I still don't really know what the difference between public and charter is aside from local funding
 
me neither
I thought charter schools were private lol
my cousin asked me a bunch of questions about biology yesterday...I had to try to remember back to high school...
ATP -> ADP
Krebbs cycle...Calvin Cycle...which one is which?
one's for plants and one's for animals right
lol
 
I think they're just independently formed schools. At least ours didn't have tuition and had to let anyone attend. The charter did restrict the number of students and growth per year, so they had to do a lottery if too many people applied
And ours had state, but no county funding. We had students from multiple counties though
 
8:49 PM
According to wiki: A charter school is a school that receives government funding but operates independently of the established state school system in which it is located.[1][2] Charter schools are an example of public asset privatization.[3]
there ya go
 
Looking something up to figure out what it is? Now that's a crazy concept
There are unfortunately also charter schools that are more or less "we formed a charter school so our kids were only around other white kids"
 
and apparently I was right
Krebs cycle is for aerobic organisms and Calvin cycle for plants and photosynthesis
woohoo, I still remember something lol
 
Anonymous
I don't know much about the American school system, but aren't they all supposed to follow a common syllabus in all schools? At least from grade 9 onwards?
 
uhhhh
It's at a state by state level I think
there's no federal school regulation afaik
 
I think the state makes a list of available courses here and it's up to the high schools to choose which ones to offer
 
8:54 PM
 
(curriculum wise)
 
it's not always written in stone
 
I think AP courses may have some national oversight, but I'm not 100% on that
 
see what I have to do to make sure it passes the line?
 
Anonymous
Eeeeh
 
Anonymous
8:54 PM
I did that once to get a badge
 
@EmilioPisanty ...did you purposefully save those accepts for an emergency like this? :P
 
omg my package is still stuck at the distribution center
:(
 
(no way in hell am I taking a five-rep-short-of-the-cap day. I want my gold badge.) (In about two years' time, obviously.)
 
how much effort did you put into SE to get your 70k+ rep?
 
@ACuriousMind no, I was either not fully satisfied with the answers or just didn't get around to doing it.
 
8:56 PM
cus I've answered quite a few questions it feels like and I'm only at 2000 lol
 
Anonymous
@enumaris He actually would have approx 100k if not for the bounties he gave away
 
@enumaris He's also offered bounties worth almost 20k
 
hmmm
that seems like a ton of effort lol
 
@ACuriousMind though then again I did write a query to find questions with not-yet-accepted answers, and I am actively saving the rest for a future emergency =P
 
@EmilioPisanty Ah, the wonders of gamification
 
8:57 PM
@enumaris this has been my primary source of procrastination for 6+ years
@ACuriousMind indeed.
 
XD
 
I wish my procrastination was that productive
 
Anonymous
You could also...like go and do some minimal edits and get a +1 by un-downvoting
 
@Blue That...might just cross the line into behaviour I'd have to look into
 
@Blue if I downvoted instead of editing at the time, then it's extremely unlikely that I'd be willing to undownvote now.
 
Anonymous
9:00 PM
lol, that's for extreme emergency situations when you're just 1 point short of the rep cap :P
 
plus, going and actively looking for answers that you disagreed with earlier is an excellent recipe for getting yourself angry.
 
Anonymous
And when time is running out
 
Just ask John Rennie to not answer a question so you can answer it and reap the rep
 
Anonymous
Hmm, where's Balarka these days. Hasn't popped up here ever since he joined uni
 
probably found a girl friend :P
 
Anonymous
9:07 PM
@user1732 Girls are a rare species in ISI Bangalore....so I doubt that :P
 
right, right
 
o.o
 
I have a question about ... data management and or user interface, but I don't even know how to say what I want to ask yet.
 
vzn
Jul 19 at 11:26, by Balarka Sen
It's all bullshit anyway. Not even sure why I am going to a university.
 
I looked in at the User Experience chatroom, but it is dead as a door nail.
So I'm going to ask you folks and hope that someone has at least heard of the right search terms.
 
9:13 PM
o.o
swing away
 
Calculus I at university level is a shocker compared to lower level math ;)
 
^
 
I specialize in asking questions I'm not sure how to ask
 
I'm writing a tool which is suppose to allow users to rapidly try out things.
 
What I personally always find funny is the various meanings of 'linear algebra'.
 
9:14 PM
It's not a good name imo
 
@dmckee Call it "ThingTryer". Next question, thank me later.
 
At the high school level, it means "solving systems of linear equations."
 
I kept thinking "linear algebra" should just be a subset of the "algebra" that we took in high school but for only functions like f(x)=ax+b
lol
 
This is accomplished by defaulting a lot of picky little stuff (and hiding that behind a "experts only" interface), while making most of the big decisions accessibly through pop-up menus and other simple interfaces.
Only at one point I ave a set of three interrelated properties.
 
At the intro college level it goes to matrix algebra, and at the upper level it goes to vector spaces
 
9:15 PM
Let's call them car models, trim packages and color schemes.
 
There is part of me that thinks matrix algebra should replace linear algebra as a subject
 
@enumaris have a talk with its well-named sister, multilinear algebra
 
That too
 
Once you put polynomials in the entries of those matrices things can get insane quick
 
Every car comes in some packages and some schemes, but you can't have every scheme in every color.
 
9:16 PM
It's so linear it's multilinear
 
I mean, you sorta already have multilinearity in linear algebra proper
insofar as determinants are multilinear functions of their columns
 
The smith normal form follows pretty quickly and things like the JNF are sub-cases instead of the convolutions one normally goes through
 
Perhaps on the model X you can have the premium package in red, black or white, but the deluxe package in red navy or hunter green.
 
of course, in physics we usually get spoiled by working with Hermitian matrices
 
Also inner products are actually part of "multilinear algebra" and not even linear algebra if one wants to be pedantic which LA claims to be :p
 
9:17 PM
Where on the model S you can have the premium in every color scheme, but the deluxe only comes in red.
 
@dmckee which is better, the premium or the deluxe?
 
@EmilioPisanty obviously the premium deluxe
 
I can pretty well deal with a system like pick your model then your package then your color, but how do I best indicate the whole matrix to the user?
@EmilioPisanty Sport. Of course.
 
The main exception to the rule of 'everything in physics is Hermitian' would be stuff like linear systems of ODEs, e.g. circuits when time is involved
 
Just like the super unlimited plan is the best data plan and better than the normal unlimited plan
 
9:19 PM
@dmckee I'd say show everything but grayout the unavailable bits
 
@dmckee from a UX perspective, why do you want to indicate the whole matrix to the user?
 
so long as the entire matrix actually makes sense
 
@dmckee I think a lot of car and clothes sites show all the options and gray/cross out the unavailable options
 
another good example is the absurdity of video game special editions these ddays
(looking at you, ubisoft)
 
Showing users what they can't access is an easy way to get a ton of recommendations on what they should be able to access lol
 
9:20 PM
Right. I've seen the "some options grayed out" scheme, too.
 
"why doesn't premium have red??"
 
But what do I call it?
 
I think the UX attitude may be to go to a few car sites and see which you like best
 
"why doesn't deluxe have grey??"
and so on and so forth ad infinitum
 
And how do I ask a user interface person what my choices are?
 
9:20 PM
@dmckee How many options are there? Do they comfortably fit on a single screen as three lists of radio button where you disable the ones that don't fit to the options currently selected in the other two columns?
 
@dmckee ah. that's a much harder question.
 
@ACuriousMind I don't really know the answer to that yet. But there are going to be a lot of models and a lot of color schemes. There may be a relatively small number of packages.
 
@ACuriousMind @dmckee and also, how graphically representable are the options? if it's colours it's easy, but if it's some abstract mathematical framework choice...
 
I mean, for the next Assassin's Creed game you've got: Standard edition, Deluxe edition, Gold edition, Gold Digital edition, Ultimate Digital edition, Spartan edition, and Pantheon edition.
 
@Semiclassical that's absolutely ridiculous
 
9:22 PM
@EmilioPisanty Alas they are more abstract then my "let me not tell you what I'm really working on" example.
 
@danielunderwood it's been pretty widely ridiculed
 
@dmckee then by? then my? than my?
 
@dmckee Is there a natural order to the choices, or do you need to allow them to be made in any order?
 
@EmilioPisanty yeah.
 
but then that was also the case with the previous Assassin's Creed game (see gamespot.com/articles/…) and people evidently still bought the damn thing
 
9:23 PM
I.e. will all users decide on the car model first, or do you have weirdos who settle on the color first?
In that case, a tree structure may be appropriate
 
@ACuriousMind One of them has a natural default, but users will want to alter it if they get part way in and find themselves blocked. So I have to anticipate changes coming in any order.
 
gonna be hard to represent a 3-D matrix in a UI...
 
@Semiclassical gaming communities seem to be in a perpetual cycle of "that's ridiculous" -> "we're going to boycott it" -> "bought it anyway"
 
I suppose you can represent it as a bunch of different 2-D matrices...but then things get cluttered real quick
 
@ACuriousMind The model will generally be first, but users will want to play around with package and color.
 
9:25 PM
@enumaris flatten it out. columns whose entries are columns.
 
@EmilioPisanty yeah, theoretically you can do it
but from a UX perspective...is that really what you want...
 
@danielunderwood as long as the market is there, they'll keep marketing it
 
@enumaris depends on the number of choices, I guess
 
@danielunderwood The "gaming community" is far smaller than the number of people who play games.
 
otherwise, one big matrix with package and colour, with a tab selector for model on top
 
9:27 PM
Yeah that's true too. I feel the "gaming community" are the ones that like to complain about games
 
Even if every gamer on the internet complains about something, that's still a tiny fraction of those who play (and buy!) the games
 
EA - it's in the game
 
E.A. Sports - 'ts in the game :\
:p
 
@dmckee So let the users choose the model in a first step, then display all possible color/package combinations in a matrix in a second step? With a simple back button to go back to change the model
 
You do have to wonder about how sustainable that marketing model is, tho (for video games, not cars)
 
9:29 PM
I'm a proponent of the Ford model
You can have any color you want as long as it's black
So just show your users 1 option
and then wipe your hands of the work cus you're done
 
Anonymous
I'd try to avoid necessitating the use of back-buttons while implementing that kind of a "model - color - package ". Back-buttons can be annoying. I normally prefer to see everything in form of a tree with nodes having on-click-drop-lists on the same page
 
but time will tell on that
 
@Blue But a tree implies a hierarchical model. If none of the options is naturally "prior" to another, that doesn't make sense
 
Could also let people pick whatever, but show a dialog to change something when an incompatible change is selected. That may be a bit annoying though
 
how about a tSNE of all the different options into a 2-D space?
 
9:31 PM
@ACuriousMind you don't need a back button. put the model on a tab selector on top
 
Or fix it on the other end and tell the manufacturer that you need the deluxe edition in all colors!
 
@Blue I've been using drop-down lists in most places. But those pieces of the problems have a natural pick-this-then-that-then-the-other sequence to them.
 
well, you can impose a decision hierarchy on the user. it's just whether the users will go along with it
 
@danielunderwood Probably fine if one in a hundred combinations is not available, terrible if it's one in three :P
 
then you can boast that your UX uses machine learning
marketing is important
 
9:32 PM
Or give it until the customer is ready to pay and go "oops that option isn't available"
Bonus points if you don't tell them which part of the combination isn't available
 
Then again, it's likely no one should listen to my UX ideas. SAP is...not known for its brilliant UX, and basically my entire tteam tries to keep as far away from the UI layer as possible :P
 
somebody post the mojojojo + him meme
plz
I can't access memes
 
If we think of the whole choice space as 3D, most parts are likely either dense or empty with just a scattering of things that don't obviously belong to a cluster.
 
@dmckee tSNE that bad boy into a 2-D space
 
9:34 PM
or permute the variables as needed
 
Anonymous
@dmckee Firstly, are you doing the implementation for a particular purpose or is it a general one which can have an arbitrary number of parameters? To avoid the problem you mentioned you could provide a type-of-sort tab on the top showing around 3 combinations: Model-Color-Package, Color-Package-Model, etc
 
@EmilioPisanty That is something I hadn't really considered yet. I'll have to think on it. It could work is the number of "models" is small enough.
 
Anonymous
@dmckee Yeah, it does depend on the number of "models"
 
Anonymous
But it looks really reasonable to me
 
9:35 PM
@dmckee it depends on the standing of the different options. You can permute as required.
 
@Blue This is a tool for just the one domain, so I can optimize for the situation I have rather than trying to be arbitrarily general.
 
Anonymous
Also, all permutations are not natural
 
Anonymous
So maybe only show the natural permutations
 
@EmilioPisanty Sure. I donb't have a fixed mapping of my really parameters to the example parameters. I'll put the one with the small number of options into "model".
 
Anonymous
Like only 0.001% of the population would chose package, color and then model :P
 
9:37 PM
@dmckee be sure to call it ThingTryer ;-).
 
@EmilioPisanty I'm afraid it already has an acronym. People in this business like things to have acronyms.
 
nobody likes my idea to tSNE it?
 
@enumaris I don't even know what an ISNE is :P
 
@dmckee ah, figures.
are you able to divulge the business?
in other news, though
 
@enumaris My problem space is discrete, so I'm not sure that would look like and I am positive that a significant fraction of end-users would experience auditory canal brain-matter leakage from trying to use such a thing.
 
9:39 PM
@ACuriousMind tSNE*
 
@coniferous_smellerULPBG-W8ZgjR is there some specific reason why you've broken the users page?
 
t-distributed stochastic neighbor embedding
 
Anonymous
@dmckee LOL
 
@EmilioPisanty The contracting firm I work for does most of its business with the US government.
 
You can use maybe just modify it to use some discrete distribution rather than continuous ones
 
9:40 PM
@dmckee yeah, they like their acronyms, for sure.
 
@EmilioPisanty I think someone didn't properly sanitize the location string :P
 
@ACuriousMind yeah, looks like
and also, it might be a good idea to add a string-length clipping to that display
 
@enumaris do people usually say it as t-snee? I've only ever heard one person say it in person and didn't know if that was typical
 
Sounds like a case for a mother meta bug report
 
@ACuriousMind sure, yeah, let's do it
 
9:42 PM
@enumaris Thanks, that clarified everything :P
 
@danielunderwood yeah, I mostly hear it as t-snee
Andrew Ng called it as such
@ACuriousMind you're welcome :D
(In seriousness, it's a machine-learning based embedding algorithm that takes points in high dimensional spaces and embeds them into a lower dimensional space where the nearer the points are in the high dimensional space makes it more likely the points will be embedded into nearer points in the lower dimensional space according to the t-distribution)
 
Ahh good. Then people won't look at me like I'm strange when mentioning it in casual conversation
 
Acronyms are wonderful. There are some things in our kernel called RUDI (which is also a short form of Rudolph) whose name's origin no one remembers. So now they're RUntime DInger - "runtime things".
 
Hence, t-distributed (using the t-distribution in the embedded space), stochastic (probabilistic embeddings rather than deterministic embeddings) Neighbor (closer in high-D means should be closer in low-D with higher probability) Embedding
 
@ACuriousMind my field likes animal acronyms, for some reason I still cannot puzzle out
 
9:47 PM
@danielunderwood you could always just call it "t-distributed stochastic neighbor embedding"
or just use PCA instead lol
 
Anonymous
@enumaris I don't see how it would be particularly relevant here? Like you'd cluster together cars of the same color?
 
@EmilioPisanty Why would you need to denote animals by acronyms? :P
 
@Blue you have a 3-Dimensional representation of options, and then you can embed them into 2-Dimensions so you can show it all on one page
It's just your 2-D representation won't be a grid
it'll be clusters
dots that say "this package - this color - this trim"
but they'd all be in 2-D
the only difficulty left is that the 3-D space is not a continuous space, but a discrete space
but I'll leave the implementation as an exercise to the reader
 
The difficulty is how does a user find the one she wants. The stochastic nature makes searching hard.
 
but you can say your UX uses machine learning
 
Anonymous
9:49 PM
That doesn't seem like an intuitive UI to me, especially when there are too many items. The average internet user gets confused even when they're having to pick a color coordinate from the RGB palette XD
 
that gives you a lot of brownie points
 
I mean, you can probably find the right cluster pretty quickly just by sampling, but then the hunt begins.
 
just off the top of my head
 
Looks like much potential for chimeras. Twenty years from now you'll be measuring radioactive spider-rabbits
 
But then who gets to determine if it's a Spider-rabbit or a rabbit-spider?
 
9:52 PM
though I think the cake for most tortured acronym in my field goes to a non-animal one, MAch-ZEhnder-Less Threefold Optical Virginia spiderwort
 
@enumaris If you're close enough to tell, you're too close to escape :P
@EmilioPisanty ...ouch
 
Mazeltov!
 
@ACuriousMind 'cause, I mean, everybody knows what a Virginia spiderwort is, right?
 
@EmilioPisanty That may be a winner across a large group of fields.
 
@dmckee that's good to know.
 
9:56 PM
John Beacom used to go to conferences pushing his proposals under the acronyms EGADS! and GADZOOKS!.
But when he got approval for EGADS! it was under a less enthusiastic name.
(These were the technology demonstrator and full scale application of gadolinium doping to turn Super-K into a improved supernova spotter.)
 
welp...finished reading all the materials for today...now I gotta figure out what I wanna do for the next hour lol
 
@ACuriousMind huh.
 
@dmckee Googling suggests that there's indeed an EGADS project at Kamioka
 
> We can probably assume reasonably safely that people aren't going to be using MathJax in their location fields...
reads my draft
::goes to check in SEDE just in case::
 
Another entry for "falsehoods programmers believe" :P
 
10:03 PM
@ACuriousMind indeed
so.... we leave that out of the bug report and we strongly suggest to @coniferous_smellerULPBG-W8ZgjR that s/he escapes the dollar signs in his/her location field?
i.e. M\$m7 instead of M$m7. If the dollar signs were intentional, they're not showing up.
0
Q: On dollar signs in the user Location field, and character limits on overlong Locations displayed in the Users page

E.P.Here's one of the weirder bugs I've seen in some time: I have no idea why the user in question chose to enter... whatever that string is, as their location. I don't really care - I think it's weird, but it's their profile page and they can do whatever they want with it. I imagine it's some...

 
10:43 PM
Someone would put something other than a reasonable location in the location field? I'm astonished
 
 
1 hour later…
11:56 PM
@danielunderwood I'm responsible, so mine reads "Duchy of Grand Fenwick".
If you get the reference you are either a particular kind of geek or old. Maybe both.
 

« first day (2876 days earlier)      last day (2048 days later) »