Skip to content

Instantly share code, notes, and snippets.

@urcades
Last active July 11, 2017 16:36
Show Gist options
  • Save urcades/7cbc3ea4c27068fb4381fb02c939393a to your computer and use it in GitHub Desktop.
Save urcades/7cbc3ea4c27068fb4381fb02c939393a to your computer and use it in GitHub Desktop.
Product Design & Systems Conversation
amardeep

[10:05] how much is a design system just a digital usage guide?

[10:06] after going to the design sys meetup yesterday

[10:06] which i thought was pretty bad

[10:07] the difference is muddy

[10:07] and it now feels like

[10:07] #jargon

edouard

[10:11] bah

[10:11] I couldn't make it

[10:11] why did it end up being bad?

[10:11] fwiw, I thought the first one was a little ... iffy, to say the least

[10:12] not bad per se, but still had an aura of 'not being defined'

[10:12] also it was very oriented towards 'visual design'

[10:12] which is something I have a personal vendetta against

amardeep

[10:14] well maybe bad is overly harsh, but i was the first to leave

[10:14] the topic was documenting design systems

edouard

[10:14] ah

amardeep

[10:15] and all the presenters seemed to define a design sys as what i call a style guide

[10:15] for digital usage

[10:15] very uninformative and haphazard

[10:16] but maybe i’m being overly crit

edouard

[10:16] I see

[10:16] perhaps not tho ... I've seen the same crit levied against 'systems' that are much more 'styleguide' as to how they're meant to be consumed

[10:16] and applied*

amardeep

[10:16] it feels like a lot of people spending a lot of time thinking about something no one can actually explain

edouard

[10:17] lol

amardeep

[10:17] esp when you’re at a company with 3 designers

[10:18] i feel like you prob don’t need a “system” for 3 people to get on the same page

[10:18] you just have to talk

speckledwords

[10:18] @amardeep agreeed

edouard

[10:18] talking 👏:skin-tone-3:

[10:18] basic communication 👏:skin-tone-3: (edited)

speckledwords

[10:18] last two startups i’ve been at, there’s never been more than 3 of us

[10:18] so we share components in a single sketch file (and now craft plug ins) and communicate regularly

[10:19] i guess design systems are more of a struggle for tech companies with dozens and dozens of designers?

amardeep

[10:20] yeah I agree with that

[10:20] one guy was a designer @ Nasdaq

[10:20] and I thought damn dog good luck w your sys

speckledwords

[10:20] i’m still not sold its not just the latest conference topic schtick

edouard

[10:20] :sun_with_face:

speckledwords

[10:20] haha eep, nasdaq

edouard

[10:21] it's certainly not a schtick in large companies, for sure, as you said

[10:21] as @ikasliwal was mentioning

[10:21] it's like ... so necessary (edited)

amardeep

[10:21] i feels a good bit like an outlet for designers to nerd out further on design and make nice outward facing web pages about their system to get buy-in

edouard [10:21] for large companies

[10:21] also that ^^

[10:22] systems as pr ... systems as rallying ...

amardeep

[10:22] i don’t like that part

[10:22] it makes me feel

[10:22] like i work in marketing

edouard

[10:23] looool

[10:23] worst outcome

[10:23] something you got me thinking about above

[10:23] made me make a new channel to store some thoughts

[10:23] https://www.are.na/edouard-u/thoughts-i-need-to-expand-on (edited)

[10:24] but yeah

[10:24] triggered a thought above about a hypothetical 'parametric ds'

[10:24] something that is not visual in any respect

[10:24] maybe more of a 'logic template' for designing interfaces

amardeep

[10:24] #designsystemthoughts

edouard

[10:24] ya

amardeep

[10:24] #designthots

speckledwords

[10:25] that sounds interesting to me, when i first heard the term “design system” i actually thought they were referencing cybernetics

edouard

[10:26] :eyes:

[10:26] me too, tbh

[10:26] feedback loops

speckledwords

[10:26] yes exactly!

ikasliwal

[12:42] I think if your design team is applying systems thinking to its design process then it's worth calling it that as a reminder that the design process does work as a system

[12:42] My last team was 2 designers (including myself) and 1 frontend (me) and we had a design system

[12:43] The customers of a design system are other designers and developers, not your team

[12:44] So for a design team that doesn't really apply systems thinking I can see how this whole thing might just seem like another name for a styleguidr

edouard

[12:51] yeah! re: what i was bumbling about earlier, I don't think 'systems' are completely useless on small teams

[12:51] i'm just interested in the working definition of 'systems thinking' at that scale

[12:51] actually, in general

[12:52] I like to hear about what people's definitions for 'systems thinking' even is (edited)

edouard

[12:52] because it tends to veer wildly from team to team 1 reply 5 days ago View thread

edouard

[12:52] i mean, i have 'systems' for myself as well, re: how I go about things

[12:53] but it's a funny way of collecting how I work about myself/personal projects/etc.

callil

[14:43] damn I missed the convo yesterday

[14:44] I have a lot of thoughts about msft stuff as I was there as this whole thing was forming and worked on some of this stuff

[14:45] it was such a complex issue but comes down to design as org becoming more prevalent and consolidated at microsoft - as a result, many teams with conflicting visions are being put together for the first time

callil

[14:45] the first C level design exec position was formed last year 1 reply 2 days ago View thread

callil

[14:46] its totally true that there is not a lot of confidence or cohension in design teams there, but not because design didn't or couldn't, but because just now (yes in 2017) microsoft is treating design as its own distinct discipline. 4 replies Last reply 4 days ago View thread

callil

[14:46] Much of that is driven by the holo lens team

[14:47] Fluent is really the holo-lens design team and visual system applied to the larger microsoft product portfolio

[14:48] the reason this could happen was because the holo lens team was seen as the future and making real forward progress - whereas the design systems that came out of windows (metro etc.) are from what is essentially a dead/old product now

edouard

[14:48] dang

goretexsuit

[14:48] :thinking_face:

callil

[14:48] thats not to say windows designers didn't work on this (some of my friends ans super talented designers are there)

[14:49] but what you're seeing is the business 'success' of the creative org that worked on holo lens flowing outward across the other products

[14:49] the same way the metro design system flowed across other microsoft products after it was seen as a 'success' and new

edouard

[14:50] that's incredible that one product division is having such a radical effect/projecting such a unifying force

callil

[14:51] The other very complex part is that this design system is inherently tied to the concept of UWP - but has a ton of tech that ahs to be built from the ground up

[14:51] for instance, I worked on the teams that were prototyping effects that the compositor of windows couldn't do!!

[14:51] through framer!!

[14:51] lol

[14:52] so this is also a platform play where they are investing a lot in unifying the tech across many platforms that enable these effects

[14:52] for a company this big its a yuuuge investment

[14:53] for instance to get that glass effect in the video above

[14:53] but I wouldn't call this a design system

[14:54] its not yet so prescriptive or functional in the same way that there is really robust tooling (will be made over the next year or 2)

[14:54] its more like a unifying art direction

[14:54] @neue had it spot on

edouard

[14:55] yuge yuge design investments

callil [14:56] everything you see in that video was essentially concept cars

[14:56] after effects ideas of what they want to get to

[14:57] and the space between those concept cars and the real thing is pretty huge

[14:59] probably one of the most interesting things here is that a lot of the work in this is coming from the gaming industry

[14:59] the holo lens teams hired from production studios (cause their origin was in the gaming space through the xbox team)

edouard [15:00] yeesh

callil [15:00] and they have been bringing the design and production systems to the product design space in microsoft

[15:00] and its fascinating

edouard [15:00] yeah, from an external perspective it's pretty nuts

[15:00] to witness

callil [15:00] cause game studios are LIGHTYEARS ahead of product designers imo (edited)

[15:01] their tools are so so so much more robust for building parametric systems and dealing with visual effects

[15:01] it just hammers home to me the idea that the tool influences the result

callil [15:02] you can see how metro design came from print (flat simple) 1 reply 5 days ago View thread

callil [15:02] and fluent is coming from game design (volumetric + lighting systems)

[15:03] in game design - designers have a workflow where everything is inherently connected - unity auto imports from cinema maya etc, and allows you to play your game while you design

edouard [15:03]

cause game studios are LIGHTYEARS ahead of product designers imo yeah, this

callil [15:03] the turnaround for testing and modifying your game environment is so so so fast

[15:04] compared to the turnaround time from sketch - code to building your app is mindblogglingly slow

edouard [15:04] yeah =|

[15:04] and it's incredible to map out how long game dev has been as fast as it is

[15:05] product design as it's implemented now is so primitive..

callil [15:05] so for me the most interesting part is how these two disciplines are crashing inside mircorosft and the game studio type of workflow is winning

[15:05] yes!

[15:05] exactly

[15:05] it made me feel like I was 50 yrs behind when I was first seeing this stuff

edouard [15:05] yeah..

callil [15:05] even things like jon golds work is primitive compared

[15:06] I have thought a lot about why this is so different and what lead to the split

[15:06] and I believe the difference has to do with 'worldbuilding' (edited)

[15:06] when you are doing game design - you inherently have to consider the whole environement

edouard [15:06] i tend to try and be soft w/ how i levy criticism against contemp prod design

callil [15:06] to even achieve anything remotely realistic, you need to start from first principles

amardeep [15:06] prod design vs game design is a really interesting thing to think about that I never even considered before you wrote them in the same sentence

[15:07] !

callil [15:07] gravity, lighting etc.

[15:07] as a result

edouard [15:07] but i 100% believe the grounding of prod design in graphic design/print methods

[15:07] as being the reason why it's so primitive

callil [15:07] you are being forced to think parametrically and set up systems for yourself to work within

[15:07] in traditional product design you are not forced to make constraints for yourself in such a fundamental way

[15:08] so the constraints of game design have forced them to innovate so much farther beyond

[15:09] so this worldbuilding is like the key aspect of what helped game design just do design systems by default (edited)

[15:09] there is no other way

[15:10] where as in photoshop or sketch, you are not forced to set up a system for yourself

edouard [15:10] yeah, it'd be impossible

callil [15:11] in conclusion, fluent isn't really interesting for the marketing side of it, but because of the way the design is being made and how its changing within the soft

[15:11] but its really fractured and who knows what will really happen

edouard

[15:14] yeah, suppose we'll see..

[15:15] my v limited experience w/ large organizations attempting to cohere their overall approach to building stuff

[15:15] leads me to believe uniting efforts across software/hardware/mixed-space-ar-vr will be ...

[15:15] ... very difficult

callil

[15:16] ha yeah

ikasliwal

[15:17] omg

callil

[15:17] I think that part of the reason why this is really an art direction release

ikasliwal

[15:17] @callil yes to everything

[15:17] can more people think like this

[15:17] lol

[15:17] :broken_heart:

callil

[15:17] lol

[15:18] @ikasliwal its really important I think

[15:18] most people walking around with blinders kinda

ikasliwal

[15:18] yeah, it's insane

[15:18] i think dreamweaver was a concept

[15:18] that tried to assuage the distance between visual design and code

[15:19] but the technology just wasn't there yet

[15:19] or, idk, the software failed

[15:19] figma seems like it's a promising concept that could eventually get to holistic design thinking

[15:19] i think a big issue is just how easy and fast it is to ship web design

[15:20] like, you can ship a buggy design and it's still usable

edouard

[15:20] i'm so excited for figma! it's 'of the web'

[15:20] like onshape...

[15:20] or fusion360...

ikasliwal

[15:20] same

[15:20] v excited for where it'll go

[15:20] sketch is gonna die

[15:20] really soon i predict

edouard

[15:20] :smiling_imp:

amardeep

[15:21] i actually think sketch is about to pop!

douard

[15:21] the only thing it has going for it is the plugin architecture

amardeep

[15:21] they’re using JSON architecture now

[15:21] in 43

callil

[15:21] maybe they will put sketch on the web

amardeep

[15:21] which for me is really cool

callil

[15:21] or in electron

[15:21] in the future

ikasliwal

[15:21] the plugin system is so so dope, but with figma, imagine building chrome extensions for figma!

callil [15:21] when it gets mroe efficeint

amardeep

[15:21] since then it can be programmed

callil

[15:22] I don't think figma has it so,ved yet

amardeep

[15:22] https://medium.com/sketch-app-sources/sketch-43-is-coming-to-town-with-a-new-game-an-open-file-format-ae62e7e7c223 (edited)

edouard

[15:22] would be very cool to see them glorp and merge

[15:22] imo...

callil

[15:23] its more like the problem space will not lead to good design systems - the constraint of world building is a problem that still forces better systems thinking than any product/app I've ever seen

[15:23] so I don't think sketch on the web solves that

ikasliwal

[15:23] ^

edouard

[15:23] w/r/t what cal was saying tho, i like the idea of new tooling just being generated that encourages building w/ physics (as in rulesets/logic) (edited)

ikasliwal

[15:23] sketch isn't designed in a way to encourage systems thinking

[15:24] it's designed in a way to encourage design

callil

[15:24] which is good! its like cinema or maya in that regard

[15:24] we really need the unity of product design

ikasliwal

[15:24] yeah

[15:24] the issue is frontend codeeeee....

callil [15:24] like framer/figma/chrome inspector in 1

ikasliwal

[15:24] is ever changing and poorly defined

callil

[15:24] so you are directly inspecting modifying the real thing (edited)

edouard

[15:25] i've harped on twitter about this before

[15:25] but truly parametric design tooling

[15:25] is something i rly want to see

[15:25] 'relational' component building

ikasliwal

[15:25] it's just relaly hard to map designs to html structures

[15:25] that work

[15:25] because yu hve to know about everything on the page to write good html

callil

[15:25] its why react and the work jon gold is doing is getting there

edouard

[15:25] yeah, html/css was never meant to be tortured into app-forms (edited)

callil

[15:26] right now the gui for his system is sketch

[15:26] I don't think thats the best personally

ikasliwal

[15:26] whatever comes after react is going to solve this i think

[15:26] i think the GUI for this is something like dreamweaver, like they were on to something

[15:26] and someone just needs to make something that isn't html or css

callil

[15:26] I'll have to look at that I never really used it (edited)

ikasliwal

[15:26] omg it's so terrible

[15:27] but the UI was really smart

edouard

[15:27] lol it's the worrrst

[15:27] as-is (edited)

[15:27] the results that is..

ikasliwal

[15:27] also like old af software so keep that in mind lol

callil

[15:27] yeah

ikasliwal

[15:27] but yeah html and css need to die

edouard

[15:27] this whole convo got me :thinking-intensifies:

ikasliwal

[15:27] so that we can have a unity for web design

[15:27] and stop dealing with making design systems

callil

[15:27] like imagine if sketch was like ok you're doing a new design?

[15:27] here is a page where you define colors

[15:28] now here is a page where you define type

[15:28] etc

edouard

[15:28] !

callil

[15:28] without you having to create that all from scratch

ikasliwal

[15:28] figma is doing something like that i believe, they're component-izing those visual guidelines

callil

[15:28] but somehow in a way that wasn't perscriptive

edouard

[15:28] an app forcing you into systematizing an environment is interesting

[15:28] or providing that as a path

[15:28] that one can take

[15:29] reminds me of how 'dimensioning' is a required aspect of using something like solidworks

[15:29] you can't build anything without defining it

callil

[15:29] I don't actually think thats a good idea - but that idea is that the tool encourages and promotes the systematic mindset

[15:29] yes!

ikasliwal

[15:29] ok imagine

[15:29] OKOK

edouard

[15:29] yeah, i meant forcing as in the solidworks method

ikasliwal

[15:29] if you were to open a new sketch file

edouard [15:29] not a wizrd

ikasliwal

[15:29] and a modal popped up that said

[15:30] "What is this design going to be used for? How will people continue to iterate on this or use this in the future?"

[15:31] "Is this a component?"

edouard

[15:31] oo

[15:31] is this a component

[15:31] 'will this be reused?'

callil

[15:31] in solidworks and other tools they have the concepts of parts and assemblies

[15:31] when you make a new design you are inherently making a part

edouard

[15:31] ugh, solidworks has so much right

[15:31] imo

callil

[15:32] and then when you are ready with the parts you make assemblies

[15:32] this is a perfect example of the tool enforcing good design patterns

ikasliwal

[15:32] ya, so design teams, even if it's a 1 person design team, if the whole team isn't thinking in parts and assemblies then your whole design system crumbles

callil

[15:32] the assemblies are what move and actuate

[15:32] the parts are static but defined parametrically with history

edouard

[15:32] even stress testing individual parts vs. stress testing entire assemblies

[15:32] is like

[15:32] SUCH a good model

callil

[15:32] maannnn

ikasliwal

[15:33] :sob:

edouard

[15:33] this convo makes me want 2 die

[15:33] in a good way

ikasliwal

[15:33] same

[15:33] imagine the web we're gonna work on in like 5 years

[15:33] i'm optimistic we'll have good tooling eventually

[15:33] and then VR will take over and no one will use web sites anymore

[15:33] :slightly_smiling_face:

[15:33] ^_^

[15:33] :coffin:

edouard

[15:34] i'm optimistic too isha

[15:34] because of folks like yall

ikasliwal

[15:34] :heartbeat:

callil

[15:35] :seedling:

edouard

[15:35] in general, this would be such a fun research topic..

[15:35] to work on

[15:35] i wish we could constitute an r&d dept

ikasliwal

[15:35] I JUST WANNA DO RESEARCH

callil

[15:36] I'm starting a company! doing design R&D 1 reply 2 days ago View thread

callil

[15:36] hope to do this ALL THE TIME

[15:36] lol

ikasliwal

[15:36] how do i apply

callil

[15:37] uh check back in a few months

[15:37] I'm not kidding though - and I'd love to work with you on it (edited)

edouard

[15:38] :innocent:

[15:38] will check back

ikasliwal

[15:38] doing reserach is my dream is so yes keep me in mind

[15:38] (and getting paid to do it)

[15:38] (i like money)

edouard

[15:38] (money is sick)

ikasliwal

[15:39] (my favorite system of them all)

[15:39] (just kidding but i am a slave to it)

edouard

[15:39] (hell ya)

[15:40] there's so much to be said for the tools used in human-spatial industries ... so much that isn't drawn from in digital prod design tooling (edited)

[15:40] i hear many similarly good things about revit

[15:40] i mean ... architects and interior designers define irl buildings w/ this software

[15:40] they aren't such well-integrated systems as gamedev environs

[15:41] since arch-ppl don't * build * and produce their works

[15:41] with the same software

[15:41] but there's a lot to be learned from how the components standard fixtures

[15:41] like walls/stairs/electrical/etc

callil

[15:42] its funny that even within creative world

[15:42] we don't know eachothers tools that well (edited)

[15:42] imo the best things come when people who don't have a lot in common come together

[15:43] and you can see how their process can be used for your problem

[15:43] there is something super powerful about domain transfer of processes

[15:43] I don't know exactly why or if there is a name for it

[15:43] but its something I am fascinated by

edouard

[15:44] s ... a me

ikasliwal

[15:44] definitely

goretexsuit

[15:45] learnin so much rn 👏:skin-tone-3:

andre [15:58]

@goretexsuit reading this whole thread throughout the day, I feel the same, good good stuff

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment