Hi programming thread, I'm starting a new project and I'd like to get a bit of advice for where to start.
I'm planning on writing a dialogue editor/database for some games my boyfriend and I are looking at making. Ideally I'd want it to be a visual node editor that can be run as a web app so we can both work on the same database.
I was going to make it with Spring Boot and Angular, because my company is using them heavily and some experience with that would be pretty helpful for trying to get into a dev job at work. I also planned on using Neo4j because I'd like to play around with graph databases.
But basically it looks like my decisions for what to use revolve around everything except "what would be a good fit for the project" and something like Spring/Neo4j could be a nightmare to package as a standalone desktop application (which I definitely want as an option, if others want to use it they're probably not going to want to self-host a full web application).
Any advice? I'm happy to try new languages/frameworks that I'm unfamiliar with, since a lot of this is a learning exercise for me.
Hi programming thread, I'm starting a new project and I'd like to get a bit of advice for where to start.
I'm planning on writing a dialogue editor/database for some games my boyfriend and I are looking at making. Ideally I'd want it to be a visual node editor that can be run as a web app so we can both work on the same database.
I was going to make it with Spring Boot and Angular, because my company is using them heavily and some experience with that would be pretty helpful for trying to get into a dev job at work. I also planned on using Neo4j because I'd like to play around with graph databases.
But basically it looks like my decisions for what to use revolve around everything except "what would be a good fit for the project" and something like Spring/Neo4j could be a nightmare to package as a standalone desktop application (which I definitely want as an option, if others want to use it they're probably not going to want to self-host a full web application).
Any advice? I'm happy to try new languages/frameworks that I'm unfamiliar with, since a lot of this is a learning exercise for me.
Hmmm, what sort of experience/background do you have? It might help if can recommend packages that can leverage off your experience.
I've made some applications with Spring/Java, recently worked with Ruby/Sinatra at work and I've made a few things in C# and Python. I don't have too much experience, especially on the web end, which is why I'm happy to try new things - I'd be learning most of it as I go anyway.
Hi programming thread, I'm starting a new project and I'd like to get a bit of advice for where to start.
I'm planning on writing a dialogue editor/database for some games my boyfriend and I are looking at making. Ideally I'd want it to be a visual node editor that can be run as a web app so we can both work on the same database.
I was going to make it with Spring Boot and Angular, because my company is using them heavily and some experience with that would be pretty helpful for trying to get into a dev job at work. I also planned on using Neo4j because I'd like to play around with graph databases.
But basically it looks like my decisions for what to use revolve around everything except "what would be a good fit for the project" and something like Spring/Neo4j could be a nightmare to package as a standalone desktop application (which I definitely want as an option, if others want to use it they're probably not going to want to self-host a full web application).
Any advice? I'm happy to try new languages/frameworks that I'm unfamiliar with, since a lot of this is a learning exercise for me.
That's an interesting approach. No idea if it will work or not because I've never developed a game like that.
I was thinking of developing something like that in the future. Unless your game is massive I don't think you need a graph database and personally I was thinking of using Clojure for my particular project. I was inspired by what this dev was using for his adventure game, http://www.tickstales.com/dev-blog/.
Speaking of game dev stuff, looks like I might be heading into VR game territory soon.
Ive been doing VR with some friends for a few months now. Its pretty fun!
My advice is skip trying to do Samsung Gear, its a very limiting environment. You'll learn a lot about optimization, but you'll also be constantly frustrated.
If you can afford it, go straight to Vive/Oculus development. Much more freeing.
That's an interesting approach. No idea if it will work or not because I've never developed a game like that.
I was thinking of developing something like that in the future. Unless your game is massive I don't think you need a graph database and personally I was thinking of using Clojure for my particular project. I was inspired by what this dev was using for his adventure game, http://www.tickstales.com/dev-blog/.
I thought that's a very good presentation!(even if I don't consider Clojure a serious candidate for indie game development) If you end up playing with it and need help or pointers hit me up!
Ok, finished the week with having the generative engine whip up some sample music.
Currently it's pretty limited. It basically is just a Markov chain of Intervals(rather than absolute pitch), which then attempts to weight the chain towards a contour.
It doesn't do rhythms yet, and it doesn't look ahead at all(It only weights towards the contour at that moment in time, without considering, say, 4 measures ahead, 8 measures ahead, 16, etc. which is coming soon).
Also, the entire corpus that it's analyzing is... Balrog's stage from Street fighter 2. It analyzes that, and then spits out lines based on the Intervals and Contours of that.
It's pretty terrible, but at least I can see where it's coming from!
(For anyone interested, the MIDI file is available for download here, but.... well it kind of sounds like a random smashing of notes on a keyboard currently.)
That's an interesting approach. No idea if it will work or not because I've never developed a game like that.
I was thinking of developing something like that in the future. Unless your game is massive I don't think you need a graph database and personally I was thinking of using Clojure for my particular project. I was inspired by what this dev was using for his adventure game, http://www.tickstales.com/dev-blog/.
Yeah, a graph database in the game would be complete overkill - though it's an interesting thought. It's more for the editor, I was planning on having it be able to export the dialogue files in various formats depending on the game engine.
I'd really like to have access to the resources of my computer so I can benchmark these algorithms I'm writing, but nooo, Windows Update takes precedence. Damn shitshow with its neverending updates and forced reboots.
That's an interesting approach. No idea if it will work or not because I've never developed a game like that.
I was thinking of developing something like that in the future. Unless your game is massive I don't think you need a graph database and personally I was thinking of using Clojure for my particular project. I was inspired by what this dev was using for his adventure game, http://www.tickstales.com/dev-blog/.
I thought that's a very good presentation!(even if I don't consider Clojure a serious candidate for indie game development) If you end up playing with it and need help or pointers hit me up!
I may be Clojuring professional soon. Given that I am a Lisp newb I will most likely be in here asking questions if I take that positioning.
Are you working with Clojure day to day? Do you enjoy it?
I just started a Clojure position two weeks ago near salt lake city and can definitively say Clojure is my favorite language.
There is a bit of learning curve before your brain remembers to put the function between the parentheses, especially if you switch alot, but the overall conventions of immutability, lazyness, and data structure recastability makes for some really succinct and understandable code.
The fun part is that whether you are in clojurescript for web or clojure proper, generally your entire project is loaded into the repl, and you just mess around in there with instant feedback until you get it right, then copy that and tests into where it should go. Fortunately for me, I got into clojure in college after viewing Rich Hickey's somewhat seminal "simple made easy" talk, even used it for my grad school project, kind of fell off till now, but if you want to learn Clojure I would recommend getting IntelliJ Cursive going (for breakpoints), going to 4clojure.com and googling your way through all the docs. Also get the Leinigen build tool because everyone uses it and it is awesome.
That's an interesting approach. No idea if it will work or not because I've never developed a game like that.
I was thinking of developing something like that in the future. Unless your game is massive I don't think you need a graph database and personally I was thinking of using Clojure for my particular project. I was inspired by what this dev was using for his adventure game, http://www.tickstales.com/dev-blog/.
I thought that's a very good presentation!(even if I don't consider Clojure a serious candidate for indie game development) If you end up playing with it and need help or pointers hit me up!
I may be Clojuring professional soon. Given that I am a Lisp newb I will most likely be in here asking questions if I take that positioning.
Are you working with Clojure day to day? Do you enjoy it?
Yes, I've had the chance to work with it on a daily basis for slightly more than an year now(was involved with the language for longer). I'm loving it and productivity across the team is great too. Where are you based, Europe or NA? It's been kind of fun to see the number of growing jobs with Clojure over the last 18 months, but North America is still far, far ahead on this front. The European community feels more made of people interested in the language, than doing it professionally (with some exceptions).
My wife has a young student (~8 years old) that is intresting in learning about programming.
I know there is some cool stuff out there aimed at kids, anyone have any experience to share and / or recommendations?
I went to a local elementary school for Hour of Code last year, and the kids seemed to mostly stay entertained by it. They've got a bunch of courses here: https://studio.code.org/
My girlfriend's niece (10) has been playing with Code Combat for a few weeks, and I think she's enjoying it.
You know even though I now have a job I still get bummed about rejections. I found out from a recruiter that a take home I did not too long ago got rejected.
The reason, "lack of tests". I just double checked the code base, it's like 100 lines of code being tested by 400 lines of tests, include some fairly clever generative tests. I did the entire thing TDD style. At this point I have no ideas what these guys were looking for. I wish I could post it for review but unfortunately that could get me into trouble.
I finally got feedback on my previous take home that got rejected. They apparently didn't like the style of tests I wrote, too many mocks apparently.
I am starting to think people who give take home tests expect people to be mind readers.
100 lines of code is too many. You need to find some way for your app to be implemented via side effects from your tests. Bonus points if you hijack assert globally so any codebase you integrate with gets side effects too!
+3
NogsCrap, crap, mega crap.Crap, crap, mega crap.Registered Userregular
You know even though I now have a job I still get bummed about rejections. I found out from a recruiter that a take home I did not too long ago got rejected.
The reason, "lack of tests". I just double checked the code base, it's like 100 lines of code being tested by 400 lines of tests, include some fairly clever generative tests. I did the entire thing TDD style. At this point I have no ideas what these guys were looking for. I wish I could post it for review but unfortunately that could get me into trouble.
I finally got feedback on my previous take home that got rejected. They apparently didn't like the style of tests I wrote, too many mocks apparently.
I am starting to think people who give take home tests expect people to be mind readers.
from the sounds of it, it doesn't seem like you would actually want to work there, to be honest.
True. I got the sense that the devs there were fairly junior. False negatives are easier when the person evaluating a test has no idea what they are doing.
I could also just really suck at take home tests, I was 0/2 this interview round.
Anyway, interviewing for engineering positions is pretty much the worst. So glad to be trading the full time job of finding a job for a job that will hopefully be fun and actually pay me.
Oh one hilarious observation from this round of interviewing.
It was amusing to see people from big corporations fetishize startup culture after getting out of a startup that was one of the most poorly run / disorganized companies I've worked for. I get the sense that for a lot of engineers dropping "we have a startup culture" sounds sexy and hip. But after multiple years of working erratic hours and having no work life balanace I am ready to run when I hear that.
The 6-month lifecycle on best practices/libraries has done a good job of keeping me away from the frontend.
Orca on
+11
Apothe0sisHave you ever questioned the nature of your reality?Registered Userregular
edited October 2016
Please teach me how to JavaScript.
I have a set of data that is returned from a web query i want to pipe into google visualisations.
All of the google code samples I see use static values whether in an external source or in a client side source.
I am pretty certain the following should be possible:
Request the data using basic authentication
Transform the data to be in the format google expects for a datatable
Display graph
I think I am running into an issue with the request the data step and don't really know where to test it. Can I do all that in jsfiddle? What's the best way to see whether my request grabbed data or not?
Apothe0sis on
0
admanbunionize your workplaceSeattle, WARegistered Userregular
Is this all happening on the front-end? You should be able to see the request results right in the dev console in the browser. If it doesn't show up there for some reason you can pull up Fiddler and track down the request.
Apothe0sisHave you ever questioned the nature of your reality?Registered Userregular
edited October 2016
The Google chart visualisation tools are all js/front end only and expect their results in a certain format - you can use back end stuff but it needs a server side page to pull that anyway, and if the request can pull that from the API anyway it seems inefficient to create a separate aspx page just to do a SQL query or the same web request I would do anyway.
The idea is to shore up a deficiency in a dashboard of a system I can make only minimal changes to that won't display what I want it to how I want it to. The web request uses its REST API to pull the right info in almost the right format.
Is fiddler an extension or a separate frogram?
Is there a recommended way to do a request? Some of the stuff I see suggests $.ajax() but samples I see on jsfiddle use New <something>.Request. I need to use basic authentication over SSL and probably ignore a self signed certificate.
Is there a recommended way to do a request? Some of the stuff I see suggests $.ajax() but samples I see on jsfiddle use New <something>.Request.
See the "learning javascript in 2016" page; this stuff is getting built fairly quickly, though you can still use the older ways of doing things, of course.
Request is part of the new Fetch APIs that wrap up XMLHttpRequest in a better way. You can still use raw XMLHttpRequest if you don't like libraries, or $.ajax() if you like older libraries, or a polyfill that will make your browser pretend it supports those APIs even if it doesn't.
Added a stopwatch to monitor my performance for the generative part creating the lines.
My biggest thing is that in order to find each note I want to be below 5ms per note. I've updated my plans a bit after meeting with my supervisor and it'll end up generating phrases at a time instead of notes at a time, which frees up a little bit of performance, and I can cheat some of it if it ends up being crazy. Right now it's pretty simple, but it does have to do several list look-ups, and as the corpus expands, that might run into problems if I start doing all the fudging and
At first I thought the stopwatch was malfunctioning or I was using it wrong, because it kept reporting back 0.
Turns out when I went to elapsedticks instead of elapsedmilliseconds, I'm averaging at about 0.5 ms to generate each 32-note phrase of 5 lines.
And my next step is to actually simplify some of the generation.
Yeah I think I don't have to worry about my performance ceiling for a while.
It was interesting to get an idea of currently leading and newly emerging frameworks and tools. TLDR: React is killin' it, JS development is overly complicated, but it's getting better.
[Michael] on
0
mightyjongyoSour CrrmEast Bay, CaliforniaRegistered Userregular
Until someone else comes up with the next great thing which is totally just like the other thing but obviously better
Posts
<NewOfficeContact ref="office_contact" office_contact={this.state.office_contact} onChange={(contact) => {this.setState({office_contact: contact})}} />I used to have it like that. For everything. But it made it really annoying when trying to get a full picture of the component.
<NewOfficeContact ref="office_contact" office_contact={this.state.office_contact} onChange={(contact) => {this.setState({office_contact: contact})}} />That looks so much cleaner.
Especially in diffs via source control.
I just wanted to add/remove/change an attribute of an element.
If everything is on a single line, then it becomes a game of "Okay, which attribute was changed?"
But if you have individual lines like that, diffs/patches/etc on attributes become much easier to scan through.
I'm planning on writing a dialogue editor/database for some games my boyfriend and I are looking at making. Ideally I'd want it to be a visual node editor that can be run as a web app so we can both work on the same database.
I was going to make it with Spring Boot and Angular, because my company is using them heavily and some experience with that would be pretty helpful for trying to get into a dev job at work. I also planned on using Neo4j because I'd like to play around with graph databases.
But basically it looks like my decisions for what to use revolve around everything except "what would be a good fit for the project" and something like Spring/Neo4j could be a nightmare to package as a standalone desktop application (which I definitely want as an option, if others want to use it they're probably not going to want to self-host a full web application).
Any advice? I'm happy to try new languages/frameworks that I'm unfamiliar with, since a lot of this is a learning exercise for me.
Hmmm, what sort of experience/background do you have? It might help if can recommend packages that can leverage off your experience.
But frankly, now that I have a proper job, my forum browsing time is 50% "this compile is taking a while" and 50% "in the bathroom".
That's an interesting approach. No idea if it will work or not because I've never developed a game like that.
I was thinking of developing something like that in the future. Unless your game is massive I don't think you need a graph database and personally I was thinking of using Clojure for my particular project. I was inspired by what this dev was using for his adventure game, http://www.tickstales.com/dev-blog/.
Also you should check out the Game Dev thread, https://forums.penny-arcade.com/discussion/206950/game-dev-somebodys-making-these-things-right.
Ive been doing VR with some friends for a few months now. Its pretty fun!
My advice is skip trying to do Samsung Gear, its a very limiting environment. You'll learn a lot about optimization, but you'll also be constantly frustrated.
If you can afford it, go straight to Vive/Oculus development. Much more freeing.
PARKER, YOU'RE FIRED! <-- My comic book podcast! Satan look here!
I thought that's a very good presentation!(even if I don't consider Clojure a serious candidate for indie game development) If you end up playing with it and need help or pointers hit me up!
Currently it's pretty limited. It basically is just a Markov chain of Intervals(rather than absolute pitch), which then attempts to weight the chain towards a contour.
It doesn't do rhythms yet, and it doesn't look ahead at all(It only weights towards the contour at that moment in time, without considering, say, 4 measures ahead, 8 measures ahead, 16, etc. which is coming soon).
Also, the entire corpus that it's analyzing is... Balrog's stage from Street fighter 2. It analyzes that, and then spits out lines based on the Intervals and Contours of that.
It's pretty terrible, but at least I can see where it's coming from!
(For anyone interested, the MIDI file is available for download here, but.... well it kind of sounds like a random smashing of notes on a keyboard currently.)
Yeah, a graph database in the game would be complete overkill - though it's an interesting thought. It's more for the editor, I was planning on having it be able to export the dialogue files in various formats depending on the game engine.
I may be Clojuring professional soon. Given that I am a Lisp newb I will most likely be in here asking questions if I take that positioning.
Are you working with Clojure day to day? Do you enjoy it?
I know there is some cool stuff out there aimed at kids, anyone have any experience to share and / or recommendations?
Nintendo Network ID: AzraelRose
DropBox invite link - get 500MB extra free.
There is a bit of learning curve before your brain remembers to put the function between the parentheses, especially if you switch alot, but the overall conventions of immutability, lazyness, and data structure recastability makes for some really succinct and understandable code.
The fun part is that whether you are in clojurescript for web or clojure proper, generally your entire project is loaded into the repl, and you just mess around in there with instant feedback until you get it right, then copy that and tests into where it should go. Fortunately for me, I got into clojure in college after viewing Rich Hickey's somewhat seminal "simple made easy" talk, even used it for my grad school project, kind of fell off till now, but if you want to learn Clojure I would recommend getting IntelliJ Cursive going (for breakpoints), going to 4clojure.com and googling your way through all the docs. Also get the Leinigen build tool because everyone uses it and it is awesome.
Yes, I've had the chance to work with it on a daily basis for slightly more than an year now(was involved with the language for longer). I'm loving it and productivity across the team is great too. Where are you based, Europe or NA? It's been kind of fun to see the number of growing jobs with Clojure over the last 18 months, but North America is still far, far ahead on this front. The European community feels more made of people interested in the language, than doing it professionally (with some exceptions).
I went to a local elementary school for Hour of Code last year, and the kids seemed to mostly stay entertained by it. They've got a bunch of courses here: https://studio.code.org/
My girlfriend's niece (10) has been playing with Code Combat for a few weeks, and I think she's enjoying it.
Good news on the getting paid to program front, I accepted another offer today.
Company was just a better fit for where I am right now. Though I am kind of bummed about not being able to try out Lisp professionally.
The reason, "lack of tests". I just double checked the code base, it's like 100 lines of code being tested by 400 lines of tests, include some fairly clever generative tests. I did the entire thing TDD style. At this point I have no ideas what these guys were looking for. I wish I could post it for review but unfortunately that could get me into trouble.
I finally got feedback on my previous take home that got rejected. They apparently didn't like the style of tests I wrote, too many mocks apparently.
I am starting to think people who give take home tests expect people to be mind readers.
from the sounds of it, it doesn't seem like you would actually want to work there, to be honest.
PARKER, YOU'RE FIRED! <-- My comic book podcast! Satan look here!
I could also just really suck at take home tests, I was 0/2 this interview round.
Anyway, interviewing for engineering positions is pretty much the worst. So glad to be trading the full time job of finding a job for a job that will hopefully be fun and actually pay me.
It was amusing to see people from big corporations fetishize startup culture after getting out of a startup that was one of the most poorly run / disorganized companies I've worked for. I get the sense that for a lot of engineers dropping "we have a startup culture" sounds sexy and hip. But after multiple years of working erratic hours and having no work life balanace I am ready to run when I hear that.
The 6-month lifecycle on best practices/libraries has done a good job of keeping me away from the frontend.
I have a set of data that is returned from a web query i want to pipe into google visualisations.
All of the google code samples I see use static values whether in an external source or in a client side source.
I am pretty certain the following should be possible:
Request the data using basic authentication
Transform the data to be in the format google expects for a datatable
Display graph
I think I am running into an issue with the request the data step and don't really know where to test it. Can I do all that in jsfiddle? What's the best way to see whether my request grabbed data or not?
The idea is to shore up a deficiency in a dashboard of a system I can make only minimal changes to that won't display what I want it to how I want it to. The web request uses its REST API to pull the right info in almost the right format.
Is fiddler an extension or a separate frogram?
Is there a recommended way to do a request? Some of the stuff I see suggests $.ajax() but samples I see on jsfiddle use New <something>.Request. I need to use basic authentication over SSL and probably ignore a self signed certificate.
I realise this is probably basic dumbs.
I am filled with the rage of 10 million suns.
http://www.telerik.com/fiddler
See the "learning javascript in 2016" page; this stuff is getting built fairly quickly, though you can still use the older ways of doing things, of course.
Request is part of the new Fetch APIs that wrap up XMLHttpRequest in a better way. You can still use raw XMLHttpRequest if you don't like libraries, or $.ajax() if you like older libraries, or a polyfill that will make your browser pretend it supports those APIs even if it doesn't.
(why would you want to use fetch in the first place? here's an explanation of sorts)
a preview:
Added a stopwatch to monitor my performance for the generative part creating the lines.
My biggest thing is that in order to find each note I want to be below 5ms per note. I've updated my plans a bit after meeting with my supervisor and it'll end up generating phrases at a time instead of notes at a time, which frees up a little bit of performance, and I can cheat some of it if it ends up being crazy. Right now it's pretty simple, but it does have to do several list look-ups, and as the corpus expands, that might run into problems if I start doing all the fudging and
At first I thought the stopwatch was malfunctioning or I was using it wrong, because it kept reporting back 0.
Turns out when I went to elapsedticks instead of elapsedmilliseconds, I'm averaging at about 0.5 ms to generate each 32-note phrase of 5 lines.
And my next step is to actually simplify some of the generation.
Yeah I think I don't have to worry about my performance ceiling for a while.
http://stateofjs.com/2016/introduction/ recently came out
It was interesting to get an idea of currently leading and newly emerging frameworks and tools. TLDR: React is killin' it, JS development is overly complicated, but it's getting better.