loading...

Name a tech you recently tried, but wasn't fully a fan of?

madza profile image Madza ・1 min read

Consider all the languages, frameworks, libraries, editors, extensions, etc you have recently used.

Leaving IE out of equation, what was something web-related you tried, but wasn't fully a fan of?

It would be Jade/Pug to me, due to it's weird syntax, white-space indentations and HTML context switch.

Discussion

markdown guide
 

Sass, I like Scss but not Sass, I never disliked brackets so I'll stay with Scss.

Also GraphQL if that counts, I tried it but I don't see myself using it since I feel like it doesn't solve anything SQL doesn't already solve.

 

+1 for GraphQL

It's so much boilerplate and obscurity that it's not even remotely enjoyable to work with.

 
 

Would love to hear the "why" answer πŸ˜ƒ

 

Here are 4 "why" :

  • don't really see what problem(s) they solve
  • demo codes (most of the time) are not convincing, not focusing on real use cases, so it's not something you can use to teach code and as a result it's more difficult to justify CSS-In-Js for big projects with money at stake
  • don't see a lot of best practices
  • when dealing with components, CSS Modules are much better imho (maintenance, readability)

So most of the time it consists in pretty cool features but there are better techniques that bring better results and solve more problems.

Would love to know your point of view on those specific points. I'm not pretending that I have all the answers but I've some experience (perso and pro) with these libraries.

 

Probably because it is unnecessary overhead and it has no additional value, but it is "fancy".

Cannot disagree more πŸ˜ƒ Have you been working with these on a bigger-scale project? CSS-in-JS are a breeze in terms of development speed, code readability and maintenance, and are also a lot easier to optimize. Creating clean CSS, without much duplicate code and with some kind of organisation is a pretty hard task.

Also, when the - almost unnoticeable - overhead is a problem for you, you can always use some of the zero-runtime CSS-in-JS libraries.

 

Just saved an article on it to explore the downsides more ;)

 

Figma : I started a design with it, but I found it hard to learn. Then I tried Adobe XD and it was a match <3

 

I would recommend to try Sketch and InVision as well, to understand the best fit ;)

 

Maybe I will try InVision for a next project.
But not Sketch, because I'm on Windows (and I don't like Apple)

Hahah, that escalated pretty quickly πŸ˜‚ πŸ˜‚

πŸ˜‚ At least it's clear ;) but well, it's a true problem for Sketch. I don't want to have to buy a mac just for a software (a friend actually did that)

Straight outta series 'How to loose a friend in web dev 101s' πŸ˜‚

Well she's a UX designer, so... maybe it's legit ?

Good for her, plus she can still run Win or Linux on that if she later decides to work on any of those two ;)

 

It competes with REST not SQL, you query data form APIs in a by SQL inspired way, for example

@madarsbiss , @ivanjeremic89
Huh. To be precise, GraphQL not competes with REST either, because REST is not about querying data, REST is about web pages with hypertext(usually HTML).

If you're interested, here are some links:
Roy Fielding(REST author) about it - youtube.com/watch?v=w5j2KwzzB-0
His blog posts:
roy.gbiv.com/untangled/2008/rest-a...
roy.gbiv.com/untangled/2008/no-res...

Original paper - ics.uci.edu/~fielding/pubs/dissert...

Also:
twitter.com/fielding/status/105297...
twitter.com/fielding/status/110809...
Short and clear: stackoverflow.com/questions/198842...
etc.

GraphQL is just another way to do build API's instead of common way's to do HTTP API's.

 

Webpack with Rails is disappointing when you don't want to use JavaScript frontend frameworks. I spent hours on figuring out why some of the JavaScripts not loading.

Finally gave up and chose to use assets pipeline/sprockets over webpacker.

 

KeystoneJS framework. It has lot of magic to build your own CMS but it doesn’t provide graphql subscriptions capabilities on top of their models

 
 

Have seen that name pop up couple of times :) Since Node is not among primary technologies in current projects, haven't really tried KeystoneJS xdd

 

Nameko - python microservices framework. That framework is supposed to speed up development of microservices. However it's documentation was missing too many key pieces of information to make it worthwhile learning. With the docs not being up to date, there's a worry that the support wouldn't be up to scratch either.

 

I might have to read more about it then. πŸ€”

 

Most of the devs called GraphQL the new REST, when it first came out ;)