Yea I started a new job today. I've been mostly programming chatting in a slack channel.
New job is going to keep me busy I think but that's okay. It has one of those desks that you can press a button so you can stand or sit.
Hmmm. It sounds like I'm saying it'll be okay because of the desk. Eh I'm too tired to change it.
Don't underestimate the desk's impact on your day to day levels of "not gonna burn the building down". It is really, really, all about the little things that keep us all from going full Milton.
We don't have fancy standing desks, but there's a standing height table that I've been using for the last coupla months. It's done wonders for my back.
Ignore the title, this is a great info dump on the reality of C++ compilers, performance, and the always misunderstood inline keyword.
I don't have time to watch a 75 minute video right now, but I assume it is mostly "actually the compiler is way smarter than you so just write readable code you fucks"?
C++ developers go listen/watch Optimizing Code Speed and Space with Build Time Switches.
Ignore the title, this is a great info dump on the reality of C++ compilers, performance, and the always misunderstood inline keyword.
I don't have time to watch a 75 minute video right now, but I assume it is mostly "actually the compiler is way smarter than you so just write readable code you fucks"?
Lots of talk about how the inline keyword was a bad idea to start with, talk about why -O1 might not do what you expect, discussion on why ffast-math most likely shouldn't be enabled, discussions on LTO and profile guided optimizations.
So where I can go about learning how to use asp.net (and C#, I presume)? We have a web app from a vendor that sucks and is done in asp.net and I'd like to be able to look at their coding and see if I can figure out why it's so terrible.
Today I learned that contrary to I don't know how long's worth of history, accessibility_identifier is NOT exposed to the user via screen reader, and accessibility_label IS. Further, accessibility_label is only SOMETIMES seen by automation, because that's what the other one is for.
I swear to god I tested this 6 months ago and it was the other way around. Either that or I am just a dumbass. Thankfully the only person I have inconvenienced is me, and I only have some shame to feel about having told people "nope I tested it and it doesn't work that way".
We don't have fancy standing desks, but there's a standing height table that I've been using for the last coupla months. It's done wonders for my back.
OMG Yes. We splurged on an adjustable desk from Ikea, and I like adjusting it more than I thought I would.
I also got one of these chairs pogo-stick-chairs from my mother-in-law, and it turns out it's actually a great compromise when your body doesn't want to stand up all the way anymore.
Between the desk, the chair, and my extra tall monitor arms, I'm actually enjoying my workstation instead of just tolerating it to enjoy my computer.
It's a bot/app for slack slash commands to return random recipes from https://github.com/sinker/tacofancy and https://github.com/evz/tacofancy-api
There are a couple github related files in there because I forgot I was working with someone else's repo and was planning on using github webhooks to have the bot notify a channel when a new recipe is added (which doesn't seem to be that often).
There are definitely some things not done in a very go-like manner, but I've at least noted most of them for me to sort out. I will also ask here about them... at the moment it's mostly stuff about the most effective way to make use of interfaces and structs. The lack of inheritance seemed awesome at first, then I ran into the compiler complaining because I can't access attributes from the interface which seems to mean duplicating a bunch of getters/setters and using those from the interface unless I've misunderstood something.
got the call at 4:30 that my system 6 months in development is going live next week, at least 2 weeks ahead of schedule
guess who is
investing in coffee this weekend
0
NogsCrap, crap, mega crap.Crap, crap, mega crap.Registered Userregular
I totally missed the creation of this "new" thread after the old one closed.
So uh...hows things?
Im working remote now, so thats cool. Except client has me doing java/groovy/hibernate, which i told them im not versed in. But whatever i guess, they still paying.
so my course had me build a basic version of reddit, and create some helper functions to generate random sentences out of random letters that look like gobblygook. I was thinking about going back at some point and improving them so they make sensical words.
chrome thought the random letters were Polish :rotate:
Sorted out most of my golang interface issues. yay for embedded structs. I still have some questions which I'll get to later, but they are mostly "A and B both are valid ways to do shit, which will be most expected" idiomatic go type stuff and may well be 100% situational.
Is there really no way to find out if an eventListener exists already in Javascript? I have a bit of code that adds an eventListener to the window, but if the page is navigated around then the thing gets added more than once and fired more than once.
I know you can do getEventListeners() in chrome console, but I need a way to do that in the code.
Is there really no way to find out if an eventListener exists already in Javascript? I have a bit of code that adds an eventListener to the window, but if the page is navigated around then the thing gets added more than once and fired more than once.
I know you can do getEventListeners() in chrome console, but I need a way to do that in the code.
If you're the one adding it then just make sure it only gets added once? Put an actual condition on it, or go to ref counting if you need to sometimes remove it.
0
gavindelThe reason all your softwareis brokenRegistered Userregular
Business value is much like energy. It is everywhere, generally inaccessible, and when harnessed it nevertheless remains largely hot air.
Angels, innovations, and the hubris of tiny things: my book now free on Royal Road! Seraphim
I'll definitely want it to look like actual text when I'm ready to start showing off a portfolio. How do I implement a markov chain?
I did think to use a dictionary and sample from that.
There are a bunch of ways to do it, but the basic idea of a Markov chain is that you split data into tokens (for example, whole words).
Then you go through your text and get statistics on how often tokens appear and which tokens follow that token, and how many times, so you can grab a random followup token.
For example, the sentence "My hovercraft is full of eels" has the following token pairs: (ie, a token and its next token)
My -> hovercraft
hovercraft -> is
is -> full
full -> of
of -> eels.
You also note what the starting token is, in this case "My". This particular Markov chain can only ever generate the starting sentence because all tokens in the chain are unique - there's only one token to choose to follow up with.
With a proper text to scan, you'd get the token "I" that (just to invent some numbers here) is followed by "am" 80% of the time, and "have" 20% of the time. The token "am" is followed by "a" 40% of the time, "an" 30% of the time, etc etc.
I -> am (80%)
I -> have (20%)
am -> a (40%)
am -> an (40%)
So you pick a random starting token (or check the texts for the tokens a sentence can start with for something more realistic), find which tokens can follow that token, and do some randomness to grab one of the followup tokens. Repeat for the followup token.
Posts
(
Ignore the title, this is a great info dump on the reality of C++ compilers, performance, and the always misunderstood inline keyword.
New job is going to keep me busy I think but that's okay. It has one of those desks that you can press a button so you can stand or sit.
Hmmm. It sounds like I'm saying it'll be okay because of the desk. Eh I'm too tired to change it.
Don't underestimate the desk's impact on your day to day levels of "not gonna burn the building down". It is really, really, all about the little things that keep us all from going full Milton.
http://steamcommunity.com/id/pablocampy
This is the first time I've heard of the second one. Gonna have to steal that.
I don't have time to watch a 75 minute video right now, but I assume it is mostly "actually the compiler is way smarter than you so just write readable code you fucks"?
Lots of talk about how the inline keyword was a bad idea to start with, talk about why -O1 might not do what you expect, discussion on why ffast-math most likely shouldn't be enabled, discussions on LTO and profile guided optimizations.
This is going to make some things so much easier.
edit: Looks like I can also do
* Rejects your pull request with a poop emoji *
lighter markup, lighter classes, grid layout... fairly decent backward compatibility with bootstrap 3 plugins... nice
I swear to god I tested this 6 months ago and it was the other way around. Either that or I am just a dumbass. Thankfully the only person I have inconvenienced is me, and I only have some shame to feel about having told people "nope I tested it and it doesn't work that way".
Incidentally, next sprint I'm starting work on a new service that needs to deliver messages in the order they are received. 8->
Isn't this what mqs are usually for?
Bonus over engineering solution - use blockchain
I'll chase down our data guy next week and see what it's about, feature spec was pretty skinny.
Blockchains do not preserve order though (at least not relative to multiple entities)
OMG Yes. We splurged on an adjustable desk from Ikea, and I like adjusting it more than I thought I would.
I also got one of these chairs pogo-stick-chairs from my mother-in-law, and it turns out it's actually a great compromise when your body doesn't want to stand up all the way anymore.
Between the desk, the chair, and my extra tall monitor arms, I'm actually enjoying my workstation instead of just tolerating it to enjoy my computer.
You could always go the datafeed route. Create multiple channels and do channel arbitration based on sequence numbers.
It's a bot/app for slack slash commands to return random recipes from https://github.com/sinker/tacofancy and https://github.com/evz/tacofancy-api
There are a couple github related files in there because I forgot I was working with someone else's repo and was planning on using github webhooks to have the bot notify a channel when a new recipe is added (which doesn't seem to be that often).
There are definitely some things not done in a very go-like manner, but I've at least noted most of them for me to sort out. I will also ask here about them... at the moment it's mostly stuff about the most effective way to make use of interfaces and structs. The lack of inheritance seemed awesome at first, then I ran into the compiler complaining because I can't access attributes from the interface which seems to mean duplicating a bunch of getters/setters and using those from the interface unless I've misunderstood something.
guess who is
investing in coffee this weekend
So uh...hows things?
Im working remote now, so thats cool. Except client has me doing java/groovy/hibernate, which i told them im not versed in. But whatever i guess, they still paying.
PARKER, YOU'RE FIRED! <-- My comic book podcast! Satan look here!
chrome thought the random letters were Polish :rotate:
I know you can do getEventListeners() in chrome console, but I need a way to do that in the code.
just the rand ruby method and .sample
def self.random_word letters = ('a'..'z').to_a letters.shuffle! letters[0,rand(3..8)].join endI'll definitely want it to look like actual text when I'm ready to start showing off a portfolio. How do I implement a markov chain?
I did think to use a dictionary and sample from that.
If you're the one adding it then just make sure it only gets added once? Put an actual condition on it, or go to ref counting if you need to sometimes remove it.
There are a bunch of ways to do it, but the basic idea of a Markov chain is that you split data into tokens (for example, whole words).
Then you go through your text and get statistics on how often tokens appear and which tokens follow that token, and how many times, so you can grab a random followup token.
For example, the sentence "My hovercraft is full of eels" has the following token pairs: (ie, a token and its next token)
My -> hovercraft
hovercraft -> is
is -> full
full -> of
of -> eels.
You also note what the starting token is, in this case "My". This particular Markov chain can only ever generate the starting sentence because all tokens in the chain are unique - there's only one token to choose to follow up with.
With a proper text to scan, you'd get the token "I" that (just to invent some numbers here) is followed by "am" 80% of the time, and "have" 20% of the time. The token "am" is followed by "a" 40% of the time, "an" 30% of the time, etc etc.
I -> am (80%)
I -> have (20%)
am -> a (40%)
am -> an (40%)
So you pick a random starting token (or check the texts for the tokens a sentence can start with for something more realistic), find which tokens can follow that token, and do some randomness to grab one of the followup tokens. Repeat for the followup token.
Here's a great visualisation of Markov chains. It uses other data than words but the idea is the same.
If you use pairs of letters instead of full words you can get a decent approximation of English-like words.