These are chat archives for mithriljs/mithril.js

17th
Jan 2018
robinchew
@robinchew
Jan 17 01:32
Koganmobile
robinchew
@robinchew
Jan 17 01:40
Hey sorry, i have no bloody idea how I wrote 'koganmobile'. I did not type it, my clipboard doesn't have it. Im on android using gitter on the browser. Weird
Scotty Simpson
@CreaturesInUnitards
Jan 17 02:28
Covfefe
Isiah Meadows
@isiahmeadows
Jan 17 02:41
@robinchew Gitter on Android does get finicky at times.
It also has this unintuitive interface where hitting enter sends the message instead of a dedicated button, so it's annoyingly hard to type anything at length without being very careful.
Patrik Johnson
@orbitbot
Jan 17 08:54
@spacejack so, installed the cube and added the stats.js thingie, that gives "7-16", by eye I'd say it's more or less 16 ...
Barney Carroll
@barneycarroll
Jan 17 11:23
@CreaturesInUnitards I think you've hit the nail on the head: most frameworks aspire to control everything.
Angular started as an idle experiment in making it easy fordesigners to write forms
React was the tool they wrote for making embeddable Facebook comments
This is another reason corporate backing is bad
Those tools did not evolve naturally, out of ingenuity or necessity
Barney Carroll
@barneycarroll
Jan 17 11:40
But the corporate suspicion bit is besides the point really
Isiah Meadows
@isiahmeadows
Jan 17 11:41

Those tools did not evolve naturally, out of ingenuity or necessity

Isn't that what you just described? They started as one thing, and turned into another because of growing needs?

Barney Carroll
@barneycarroll
Jan 17 11:43
No, Facebook & Google's interest in expanding the scope of those neat tricks does not necessarily align with any particular developer rationale
This goes back to what we were saying the other day about granular redraw
You & I aren't best placed to determine what's best there, because we're not coming at it from a position of genuine necessity
…And the RouteResolver issue is similar: the justifying example involved Webpack, devised and illustrated by people who didn't use Webpack
Expansion for the sake of it
The lack of progress on that issue, combined with collective hours of intellectual effort expended, can be attributed to the fact that the mother of invention – necessity – is absent.
Barney Carroll
@barneycarroll
Jan 17 11:50
Which isn't to say everything tends to entropy over a given lifetime – some of this thinking crept into the golden days of the original blog posts.
Isiah Meadows
@isiahmeadows
Jan 17 11:50
Yeah...I was never present in those conversations, and I personally have nothing but criticisms for the route resolver API.
MithrilJS/mithril.js#1095
Note: you were involved in that quite a bit more than I was.
Ideally, this is my preference: MithrilJS/mithril.js#2073 (Basically, it should be a third-party thing.)
Barney Carroll
@barneycarroll
Jan 17 11:55
Yeah
Isiah Meadows
@isiahmeadows
Jan 17 12:02
So to be fair, it's not like you didn't have much of a role (in #1095, you were the one who came up with onmatch).
I spent most of it on the sidelines, with most of my contributions being to the rendering and component side.
But just following the conversation, I knew why it was being introduced (I was the one who got the ball rolling on it), but I wasn't sure why it had to be designed that way. It always felt a little clunky and unnecessary to me.
And ever since #2073, I'm very in favor of deprecating/removing the whole mess outright.
Patrik Johnson
@orbitbot
Jan 17 12:11
#v3? :fire: :fire: :fire:
Isiah Meadows
@isiahmeadows
Jan 17 12:15
@orbitbot If at all practical, and if I can get @pygy / @tivac on board, I'd love to see it in v2. (Migration would be trivial given the constraints: just add the required method in the next v1 release, and remove route resolvers altogether in v2. Users could be pointed to my utility that already exists as a gist, if they still want that API.)
Patrik Johnson
@orbitbot
Jan 17 12:16
oh, thought the targets for that was more or less set
Isiah Meadows
@isiahmeadows
Jan 17 12:17
Mostly, but this would be a very small change in core. Trust me.
Patrik Johnson
@orbitbot
Jan 17 12:17
sure, why not :P
I have zero-to-no insight in to what the whole discussion is about, other than having played around with the api ofc
well, and reading the issues at the time
Dominik Dumaine
@Bondifrench
Jan 17 12:25
As long as it's easy to change to a new router and relatively intuitive, fine with me (not that my opinion counts), i do like the relative simplicity of the current one
Isiah Meadows
@isiahmeadows
Jan 17 12:26
@orbitbot I've actually used route resolvers a few times. In case you're wondering which converation, look here
@Bondifrench It's not like it's that complicated to write on top of what exists (mod the default route handling).
That particular helper would probably even pass all the router tests if the proposed m.route.reject() was added.
Isiah Meadows
@isiahmeadows
Jan 17 12:43
Basically, I just need a way to redirect to the default route. This could be done by wrapping m.route to get its second argument, and I've similarly updated the issue + gist to reflect this further development.
Barney Carroll
@barneycarroll
Jan 17 13:01
I think the default router is great. It does intuitive things simply. The problem with RouteResolver is that it acknowledges the existence of disparate routing concerns, and in hindsight Mithril is the wrong place to try and address those comprehensively. It's better to have acknowledged limitations – at which point we can tell people "ah yes that's complicated" – rather than say that in theory they might be able to mix and match different features to do all sorts of things, but we're not really sure how or why.
As with granular redraws, it sets expectations of comprehensive problem space coverage which are difficult to analyse, debunk and advise on.
Isiah Meadows
@isiahmeadows
Jan 17 13:12
Granular redraws are something you need much deeper integration to make something sensible out of. Note that very little needs to change within Mithril proper to make my helper properly integrate with it: MithrilJS/mithril.js#2074
Barney Carroll
@barneycarroll
Jan 17 13:14
When you say 'This is all I need', do you have a clear idea of what the code change actually is?
…or is that more in relation to the painlessness of the surface API change request?
Isiah Meadows
@isiahmeadows
Jan 17 13:16
Both.
The API change would be relatively straightforward to implement (although it's non-trivial), and it'd be easy to alter my utility to use it.
And my helper basically implements granular redraw, just in a way that abstracts over the difference between m.mount and m.render.
Barney Carroll
@barneycarroll
Jan 17 13:19
Makes sense
Isiah Meadows
@isiahmeadows
Jan 17 13:21
(Ironically, that helper started out as an experiment to see if a practical API could even be written outside of core, and now I'm basically recommending it to everyone who wants a granular redraw API. :smile:)
Barney Carroll
@barneycarroll
Jan 17 13:23
So you're saying all your code needs to work is to be able to control the auto-redraw function injection?
josiahking
@josiahking
Jan 17 13:23
How can i access a dom element by id and update using mithril, just like you do in jquery?
Barney Carroll
@barneycarroll
Jan 17 13:24
What do you mean by 'update'?
josiahking
@josiahking
Jan 17 13:27
like on click or submit change the value or attr of an element @barneycarroll
Patrik Johnson
@orbitbot
Jan 17 13:30
@josiahking is this dom element inside of something that mithril maintains, ie. it comes from a mithril view method?
or who/what controls the content of that dom element?
Barney Carroll
@barneycarroll
Jan 17 13:31
@josiahking the beauty of hyperscript as used by Mithril is that you don't to find a way to reference and query elements, you change how they're represented directly.
Isiah Meadows
@isiahmeadows
Jan 17 13:31
@barneycarroll Yes. That's all I need to get rid of state.link, which does it explicitly. (It'd make component use a little better/easier, too.)
josiahking
@josiahking
Jan 17 13:57
@barneycarroll What i mean is how do you update an element attr when an onsubmit button is clicked?
Isiah Meadows
@isiahmeadows
Jan 17 14:00
@josiahking If you're working from an event handler, Mithril passes the raw event through, so you could just use ev.currentTarget to get the form element (assuming that's where your onsubmit is).
Barney Carroll
@barneycarroll
Jan 17 14:12
The jQuery style is imperative – you write a series of statements about things to do.
  1. Create some DOM
  2. Bind an event handler on some of that DOM
  3. …And when that resolves do something to another bit of that DOM
In order for this kind of thing to work you need to know how and when to refer to things, and you need references for them. Every time I mentioned 'DOM' that needs a reference.
In the example I made use of the fact that elements with IDs are exposed in the global space as a reference of that name, but you could use document.querySelector('#id') or $('#id')
The Mithril style is declarative – you write an expression, which can contain other expressions, and describes the totality of the application at any given time
So by default, there is no difference between 'create some elements' and 'update some elements'
And we don't need IDs or other ways of referencing those elements so you can refer to them later. The beauty of Mithril is you stop writing your application as a series of references to things like DOM elements, and more as the data.
Barney Carroll
@barneycarroll
Jan 17 14:18
So in the Mithril example, we have this 'value' on the 'state' object. That's where we store the thing that changes when the event handler fires. And when that happens, Mithril processes the view again, notices that the content of the label has changed since last time, and updates it for you.
Rasmus Porsager
@porsager
Jan 17 14:39
@barneycarroll any reason you're loading mithril from cdnjs ? (it's not working because they don't allow cors from flems ;) )
Barney Carroll
@barneycarroll
Jan 17 14:44
Oops, I tried that when unpkg stopped working and forgot to revert
Oscar
@osban
Jan 17 14:44
@barneycarroll @josiahking I think this is the Flems ;)
Barney Carroll
@barneycarroll
Jan 17 14:46
@josiahking @porsager it's working again! Please follow this one, the previous made some mistakes.
Rasmus Porsager
@porsager
Jan 17 14:47
@barneycarroll ah ok, so perhaps an unpkg outage?
Barney Carroll
@barneycarroll
Jan 17 14:47
I think it was. False alarm, everyone back to bed!
Oscar
@osban
Jan 17 14:48
:zzz:
Rasmus Porsager
@porsager
Jan 17 14:48
:smile:
Patrik Johnson
@orbitbot
Jan 17 15:07
we don't have beds at the office. Though there's a sofa...
cаввидис
@panoply
Jan 17 15:10

I'm a bit late to the party regarding the discussion recently had with the popularity of Mithril. I just wanted to add that I feel as if the exposure behind other Frameworks like Vue come from its inclusion in larger backend projects, I mean if we go back far enough, Vue's popularity is largely owed to it being included in the Laravel PHP Framework imo.

For me, it's always a battle because as a freelancer and contracted developer you see companies always wanting to use the "latest" trending tech, this means they request Vue, React or Angular and this really does irritate me because 90% of the time these frameworks are just an overkill for the use case presented.

I'd love to see Mithril gain the respect and positioning it deserves. In the last 12 months I've turned down well paid contracts because senior developers didn't want to "risk" using Mithril on the front-end and this absolutely blew my mind.

I'd love to see Mithril being shipped with the Phoenix Framework (elixir) or other more rising stars of the development world, I believe if this was to happen then we would see that shift in popularity happen fast.

Patrik Johnson
@orbitbot
Jan 17 15:11
@panoply interesting, was there any explanations to what the perceived risks were?
and senior in what domain? :)
Barney Carroll
@barneycarroll
Jan 17 15:12
Right, there was a solid push to get Mithril used for Wordpress' reworked dashboard. @pdfernhout has been active on numerous fronts but his concerns are usually met with silence… followed by a breath of relief when React dropped their sinister patent clauses and everybody gleefully rushed for that.
Patrik Johnson
@orbitbot
Jan 17 15:13
I guess the problem being with any one particular backend framework is that you'd need to 1) have someone inside that project who has enough insight into frontend development, 2) is familiar enough with Mithril's implementation, and 3) actually agrees with Mithril's take on how to do things
everyone's lazy enough to not go through the complete effort of evaluating every framework out there, since there's a boatload
Barney Carroll
@barneycarroll
Jan 17 15:18
Yeah
And corporate backing means maintenance!
cаввидис
@panoply
Jan 17 15:22
@orbitbot this happened 3 times last year to me. It basically come down to the senior and lead developers not knowing of its existence and in turn making them hesitant. When I explained the benefits and presented the active repository (thanks to all you absolute poets) I couldn't sell them on it because most of their teams were hired specifically to work with Vue or React etc. I suppose you could put it down to front-end developers not really knowing what they are coding and being copy paste warriors. – I will say that Mithril in combination with a simple JSON API running on Phoenix (elixir) is a soulmate combination.
Barney Carroll
@barneycarroll
Jan 17 15:28
There's a lot of job security in this stuff.
When I was working with Ember, they hired an Ember guy (they'd decided Ember was good because the agency they hired for the proof of concept used Ember).
He loved the fact that Ember did videos, and spent about half a day watching them every working day.
He was particularly taken with their persistence library, and became convinced any serious application absolutely needed it.
I thought it had serious problems, and managed to produce an alternative model that achieved our requirements in a week. 4 weeks in he refused to deal with the code I'd written on principle because it wasn't doing it the Ember way – but he couldn't produce any working code that did things the Ember way.
Barney Carroll
@barneycarroll
Jan 17 15:33
They kept that guy on for a further 2 months.
cаввидис
@panoply
Jan 17 15:33
@orbitbot well I know the term "senior" developer is thrown around a lot these days, but from what I could make of it, they were mostly back-end developers who dabbled in front-end and wanted to stick to a front-end framework that they could easily maintain, when I explained that Mithril is basically Vanilla their argument was: "It's easier finding a Vue or React developer to handle upkeep, than a Mithril developer" – I suppose it's a simple lack of understanding. Sometimes I feel simplicity is harder to sell than complexity.
Barney Carroll
@barneycarroll
Jan 17 15:34
Getting rid of him would have humiliated everybody. And he can walk from contract to contract with the story that it's not Ember that's the problem, it's real world application requirements.
cаввидис
@panoply
Jan 17 15:34
@barneycarroll this would of given me an absolute brain aneurysm. Credit to you for keeping your cool.
Barney Carroll
@barneycarroll
Jan 17 15:35
Sadly in my experience this is by and wide the norm.
Patrik Johnson
@orbitbot
Jan 17 15:35

Sometimes I feel simplicity is harder to sell than complexity.

:+1:

Barney Carroll
@barneycarroll
Jan 17 15:35
Complexity is reassuring, especially when it tells you it's actually simple.
If you don't trust yourself to understand the problem domain, somebody else throwing 400 APIs at you and saying "trust me, I've done the legwork so you don't have to" sounds awesome.
Alternatively, somebody telling you you just need 10 APIs, but you need to think about when and why you're using them poses a risk, because at some point you may not know and that will put you in a bind.
Whereas with Ember you can say every other Ember user can't be wrong.
And if those really clever people who actually write it can't solve this problem, then it probably really is too complicated. The problems of any large enough framework are essentially unrelatable and therefore acceptable because they're inevitable and impersonal.
cаввидис
@panoply
Jan 17 15:45

If you don't trust yourself to understand the problem domain, somebody else throwing 400 APIs at you and saying "trust me, I've done the legwork so you don't have to" sounds awesome.

Preach! – I feel larger companies wanting to lock in clients practice this approach and when a freelancer comes along and says, "dude, what the fuck is this?" that scares these bigger companies and lead devs who are so deeply invested in this "one way" approach. The masses tend to control the way tech evolves. It's a cruel world we live in. a cruel, cruel world.

Scotty Simpson
@CreaturesInUnitards
Jan 17 16:22
@barneycarroll yes, thanks. And I agree about many of the various reasons why it's difficult getting traction. That said... At the risk of offending the sensibilities of everyone in this chat, I keep asking myself "What if it's the name?"
I love Tolkien as much as the next person (including sci-fi/fantasy geeks; I am one), but I can't help thinking it's a poor brand. Not only is it awkward to say, but it doesn't mean anything to those who were never into Tolkien, and to those who already know the term, it might come across as amateurish. I dunno.
I often wonder how a fork with a different name might perform in the open waters...
spacejack
@spacejack
Jan 17 16:27
@orbitbot what device are you getting such bad framerates on?
Barney Carroll
@barneycarroll
Jan 17 16:31
@panoply the funny thing is when you scale that psychology upwards, a lot of companies actually want to be locked in
In corporate structures, decision making is incredibly expensive
If you limit your possibility space – we'll go with this and that will lock us in for X years – you kill many birds
spacejack
@spacejack
Jan 17 16:32
The vast majority of people involved have no tech understanding and don't want to expend any effort thinking about it.
Barney Carroll
@barneycarroll
Jan 17 16:33
Executives can stop worrying about it and think about other aspects of the business, competing managers can't shoot down your project by spreading doubt, etc
Critical reasoning is bad for productivity
Scotty Simpson
@CreaturesInUnitards
Jan 17 16:33
@spacejack YES, that's perhaps the most salient point we can make wrt Mithril's Evangelism Problem.
We need resources available which spell out in clear and simple terms the advantages
cаввидис
@panoply
Jan 17 17:02

"What if it's the name?"

Maybe. I don't think many people are aware that the name "mithril" comes from Tolkien, I for one had no idea until i was a month deep in the framework.

I work mostly with fashion brands and have had my fair share of experience with startups seeking catchy names. The most important aspect to take from the name is the 2 syllable pronunciation. Two syllable names are the preferred approach to take if you want a name that will stick. Mithril ticks that box but its definition may be a deterrent. Not too sure, would be good to get actual feedback on it.

Barney Carroll
@barneycarroll
Jan 17 17:04
I think the name's fine
It's just never been marketed heavily
cаввидис
@panoply
Jan 17 17:04
@barneycarroll I'd have to agree here.
Barney Carroll
@barneycarroll
Jan 17 17:05
Arguably it's confusing that the fictional metal (as referenced in video games) comes up in search results
But React, Angular, Vue are far worse in that regard – it's only because of the marketing they surface better
cаввидис
@panoply
Jan 17 17:05
that's where the ".js" extension comes into play.
Kevin Fiol
@kevinfiol
Jan 17 17:05
my coworker has joked about the name before. "whats that thing you use min-minsthril?"
Gilbert
@gilbert
Jan 17 17:14
hi chat! Long time no see. I hear there's a 2.0 version in the works. Is this true? What's new?
Barney Carroll
@barneycarroll
Jan 17 17:47
Hi @gilbert ! We missed you
Gilbert
@gilbert
Jan 17 17:51
Good to see you're still around @barneycarroll :)
Barney Carroll
@barneycarroll
Jan 17 17:53
I haven't been here the whole time :)
What's new with you?
Gilbert
@gilbert
Jan 17 18:00
Not too much. I'm no longer teaching, but I'm still working at the same company as a developer now
lots of react and salesforce, heh
jsx has become palatable, ignoring react's common annoyances like className and camelCase events
still, I hate setting up build systems.
Gilbert
@gilbert
Jan 17 18:07
Nice commit logs. The mithril community looks quite strong!
is Leo still around?
Barney Carroll
@barneycarroll
Jan 17 18:07
He's long since left these shores
Gilbert
@gilbert
Jan 17 18:08
lol
Barney Carroll
@barneycarroll
Jan 17 18:08
Somebody recently spotted him trying to introduce TC-39 to critical reasoning
Barney Carroll
@barneycarroll
Jan 17 18:17
Great work on pipeline BTW 👍
Gilbert
@gilbert
Jan 17 18:18
Thanks :) I'm surprised it's gotten this far tbh
Barney Carroll
@barneycarroll
Jan 17 19:11
What's it like moving from teaching back to programming?
Gilbert
@gilbert
Jan 17 19:18
it's easier, to be honest
Patrik Johnson
@orbitbot
Jan 17 19:18
@spacejack one 2017 smart TV, the karvalakkimalli of the model line, though. If I'd venture a guess I'd assume that the WebGL implementation isn't actually hooked up to any hardware so it's a software emulation or something, or then the driver is borked...
spacejack
@spacejack
Jan 17 19:19
Can you run my test on it? That should indicate if there's a "majorPerformanceCaveat"
Gilbert
@gilbert
Jan 17 19:20
@barneycarroll I miss teaching, but programming (with a good manager) is a nice change of pace
spacejack
@spacejack
Jan 17 19:21
not that the implementation necessarily reports it...
but the spinning cube should be smooth on just about anything unless something is very wrong
Patrik Johnson
@orbitbot
Jan 17 19:23

but the spinning cube should be smooth on just about anything unless something is very wrong

TBH wouldn't be too surprised. Basically the only apps these kind of things support reasonably are video-on-demand things, and that basically just requires them to implement a decent H.264 & MP4 decoder in practice

again, no longer at the office, but I can try the test tomorrow
though it seems that at least some 2D css animations are quite smooth, so perhaps an approach around that could solve the issues...
spacejack
@spacejack
Jan 17 19:26
Yep, CSS is always your best bet across the hardware/browser spectrum
FYI I find transform3d smoother than transform2d. 2d doesn't always seem to promote to a texture. at least not without some extra fiddling.
On another topic... is there some dev tool that lists all running timeOuts or requestAnimationFrames?
Andrea Coiutti
@ACXgit
Jan 17 19:30
@spacejack if you use Chrome there's this extension
spacejack
@spacejack
Jan 17 19:31
Ah yeah, I've seen that hack... I suppose I could just add in my own monkey-patch. I need timeOut and RAF rather than setInterval.
Kinda surprised browsers don't have this built in
Andrea Coiutti
@ACXgit
Jan 17 19:31
yep, strange
Carl Mungazi
@CarlMungazi
Jan 17 19:54
Re: earlier discussion. I’m currently searching for a new job and I’ve lost count of the number of times I’ve had to explain what Mithril is
Gilbert
@gilbert
Jan 17 19:57
I usually tell people mithril is like react but leaner
Carl Mungazi
@CarlMungazi
Jan 17 20:07
That’s what I’ve said
But I don’t think it sinks in :laughing:
Fred Daoud
@foxdonut
Jan 17 20:24
The older mithril web site was kind of, er, not attractive, to be honest. Looked like something out of the 90s to me.
The current one is clean and simple, I like it, but not sure if it's effective, marketing-wise.
Gilbert
@gilbert
Jan 17 20:35
it's great as a documentation website
but I agree it's not a flashy, modern, "look how cool we are" sort of website
with css animations, fancy logos, parallax, and all that crap
(people eat that stuff up though)
Fred Daoud
@foxdonut
Jan 17 20:36
I don't like flashy, markety myself -- but then again, I am not the target, since I made it here on my own :)
Kevin Fiol
@kevinfiol
Jan 17 20:43
i love the minimal style of the mithril site
if marketing is an important, maybe a a pretty splash screen that leads to the docs would suffice
at least that's what Vue does :eyes:
Gilbert
@gilbert
Jan 17 20:45
it's a good strat
Patrik Johnson
@orbitbot
Jan 17 20:50
:guitar:
Pat Cavit
@tivac
Jan 17 20:52
As a community of mostly programmers I'm sure that will be simple to accomplish :laughing:
Patrik Johnson
@orbitbot
Jan 17 20:54
maybe @spacejack or @JAForbes or someone can make the documentation a 3d webgl shooter. That'd stand out, at least
spacejack
@spacejack
Jan 17 20:56
I think I floated a design concept some time ago
Would probably spread that into 3 columns on a wide display...
Gilbert
@gilbert
Jan 17 21:13
I like the typography
needs a working demo. Angular 1 skyrocketed during its day due to its in-page demo IMO
cаввидис
@panoply
Jan 17 21:27

with css animations, fancy logos, parallax, and all that crap

kids these days need the visuals.

Gilbert
@gilbert
Jan 17 21:29
the public needs it. Humans love shiny stuff
cаввидис
@panoply
Jan 17 21:30
hahaha! well said.
Dominik Dumaine
@Bondifrench
Jan 17 22:06
I am in total agreement here, Mithril's website needs more visuals, for instance, in the page, who uses Mithril, we could have, with the links of the different websites, screenshots.
Images, screenshots are appealing to people, while links are kind of conceptual, images makes them more concrete in people's minds.
I would look at Dribbble or Hyperpixel for inspiration.
Maybe we also need to be more active on Twitter, and have a Twitter feed on Mithril's page, that shows everytime it is mentionned, with copy to #js #javascript etc..
Kevin Fiol
@kevinfiol
Jan 17 22:11
those two sites are oriented toward design though. i think trying to imitate them would just end up looking superfluous. screenshots aren't needed, just a short (and interactive) embedded demo that shows mithril's strengths
Dominik Dumaine
@Bondifrench
Jan 17 22:12
Bulma have an exposition page and a hashtag #madewithbulma
Rasmus Porsager
@porsager
Jan 17 22:37
Heh.. That reminds me of the logo suggestion I made back then. http://porsager.com/mithril/
3D Flashy enough?
Patrik Johnson
@orbitbot
Jan 17 22:39
I'd think perhaps something like a collective (moderated) blog, with f.e. twitter poll voting on what topics to write on could perhaps work to re-kindle some interest
say that you'd aim for ~1 post a month or something
public vote rules, maybe pick 2-3 authors at a time and write articles on whatever people might be interested in
eg. integration with bulma, how to use tachyons with mithril, using together with jquery widgets, how to integrate react, whateva
so if that was somehow put into place, there'd be some continuous content being generated on the twitter feed, if it's published to the home page
Scotty Simpson
@CreaturesInUnitards
Jan 17 22:46
@orbitbot :clap:
Carl Mungazi
@CarlMungazi
Jan 17 22:47
I’d be up for helping with @orbitbot’s idea
Scotty Simpson
@CreaturesInUnitards
Jan 17 22:47
yup me too
I'm also in the process of rebooting Mithril 0-60; this time NOT on scrimba
I'll be hunting feedback sooner than later
Patrik Johnson
@orbitbot
Jan 17 22:52
ofc, keep posting any new stuff to the appropriate offsite places (HN etc)
Barney Carroll
@barneycarroll
Jan 17 23:09
@spacejack I absolutely loved & continue your design. It was never going to work out because of the leadership structure at the time. We could change that now with @orbitbot's program
Speaking holistically about the Mithril website, I think there are some medium-is-the-message problems.
The minimalism, the lack of bravado, these are consistent with some of the spirit
But what's terrible IMO… is that the site has very bad UI, especially on mobile
Scotty Simpson
@CreaturesInUnitards
Jan 17 23:13
@barneycarroll tell me more about medium-is-the-message wrt Mithril
... and I agree, the mobile site is broken
Barney Carroll
@barneycarroll
Jan 17 23:14
BRB
Patrik Johnson
@orbitbot
Jan 17 23:15
I'm not sure if I just got declared a mutineer or what
Carl Mungazi
@CarlMungazi
Jan 17 23:19
From a content perspective, I think mithril has good mileage as an educational tool because of the lack of friction between itself and the language
It forces you to think more about what you’re doing, I’ve found anyway
Barney Carroll
@barneycarroll
Jan 17 23:50
@CarlMungazi not sure when you dropped into chat, but we were discussing the idea earlier that what you're describing there is actually a negative from a corporate perspective… like, having to think about things, or even the opportunity to think about things, is bad for some people
@CreaturesInUnitards WRT medium / message, the reason I've continuously evaluated Mithril above other solutions is that I know realising good UIs can be complicated in all sorts of ways, and Mithril opens up incredible flexibility of expression in ways other libraries don't, while making other concerns trivial. Hyperscript / vdom is seen as an implementation detail or burdensome concept by others, but it's thinking about that as an opportunity that allows otherwise really difficult UI design concepts to flow.
When we were getting to version 1 and thinking of mottos, I was insistent on the 'ambitious' part
Carl Mungazi
@CarlMungazi
Jan 17 23:55
@barneycarroll In the sense that there is limited time to think when working on a commercial product or the person doing the thinking might not follow the "wrong" train of thought?
Barney Carroll
@barneycarroll
Jan 17 23:56
Whereas React, Ember, etc race to the bottom. They want to show you how to do very basic stuff in a completely different way.
So for me what's galling about the new Mithril website is that it's completely unambitious in terms of UI and it doesn't use Mithril
Worse, it uses ad-hoc precompiled includes in a way that really restricts UI niceties
There's a lot of 'false minimalism' there, and it's totally anathema to the practice of what Mithril affords
I noticed ages ago that the navigation, especially on mobile, was buggy in all sorts of data
*ways
Ways that would be trivial to solve using a library like… er… Mithril