IRC channel logs

2017-08-19.log

back to list of logs

<manumanumanu>Happy saturday, guilers!
<amz3>\\o/
<stis>yey!
<amz3>time for some guile hacking!
<amz3>I am wondering what to do next
<amz3>maybe hack on my toy editor
<amz3>or improve guile-wiredtiger
<amz3>or create some webapp with guile-squee
<amz3>to use something that is standard and show that it's possible to do regular webdev in guile
<amz3>s/regular/casual/ ;)
<amz3>or improve somehow forward.scm
<manumanumanu>amz3: i'd like to be able to write web stuff, but I am too stupid for web development.
<amz3>manumanumanu: oh!
<amz3>first time I read, web dev is complicated
<amz3>I was under the impression that webdev was 'programming for the dummies' kind of thing..
<amz3>maybe i have impostor syndrom or something
<amz3>well the problem with webdev, is that you need to know: HTML, CSS, JS and Guile
<amz3>manumanumanu: what do you find complicated in webdev?
<manumanumanu>I can't understand front end stuff
<amz3>then it's prolly GUI development that's is the culprit
<amz3>maybe start with desktop GUI dev and later try to understand webdev frontend
<amz3>I don't remember how I jumped into frontend dev, I learned webdev as thing evolved, so in the beginning it was very easy since there was not a lot to know about
<manumanumanu>amz3: mostly it is the web development that itches. I hate html/css and js. I can't wrap my head around it. Give me qt any day of the week. :)
<rekado>wingo: do you happen to know a fix for the memory bloat when compiling large files with Guile 2.2? I vaguely remember that you traced it to the garbage collector.
<ArneBab_>Amynka: well-done webdev is pretty complicated these days: You have to know visual and interaction design, the browser, the server (from the libraries to the hardware), the connection between both, details of TCP and HTTP, often also databases and distributing a system between several physical or virtual systems, and you need to combine all that to build a system which has an initial load time *at the client* of less than 1 second — while you
<ArneBab_>already lose 100ms per round trip. And you have to keep the complexity low enough to be managable for humans.
<Amynka>ArneBab_: was it for me?
<Amynka>is*
<ArneBab_>Amynka: argl, sorry, no, it was for amz3, but you were first on the tab-completion list …
<Amynka>haha no worries :)
<ArneBab_>manumanumanu: I’d start by noting the data that needs to be shown, than doodling how it should look and how people can navigate, and only then realizing that in HTML+CSS: HTML is the content, CSS positions it. Put stuff which the user needs right away (the main content) earlier into the HTML file.
<ArneBab_>JS comes in for anything which you can’t do easily with HTML+CSS, mainly updating only parts of the site.
<ArneBab_>(what would have been done with frames earlier)
<manumanumanu>ArneBab_: nah. I have no interest. Managed to scrape koketteriet.se together, which is good enough for my needs.
<ArneBab_>no probs :)
<manumanumanu>every time I try web dev I just realise that life is too valuable :)
<manumanumanu>So, I have realised that cut and and-let* are my two favourite srfis. Are there any neat ones I have missed?
<amz3> http://i.imgur.com/xm49sQK.jpg
<amz3>create a persistent datastructure library
<amz3>compile it to javascript
<amz3>...
<amz3>profit
<amz3> https://github.com/swannodette/mori