> Although rich, interactive contents can be made with HTML5/CSS/JavaScript technologies, most Internet users do not possess related programming skills.
Great! Finally someone in the open source forum community designing a UI for the average user!
> On twincl.com, a piece of rich content (which we called a twinclet, pronounced as “twink-lit”) is written with a Markdown fenced code block.
And... they've failed right out of the gate. Face-down in the dirt.
I will never understand the obsession some developers have with Markdown.
1. Plain text only interfaces aren't feasible in a lot of contexts. Rich text formatting and/or hyperlinking is far too useful, if not actually mandatory, in most contexts.
2. Raw HTML isn't feasible either. Even ignoring the cognitive load it imposes on non-developers, it's too easy for malicious users to exploit unless you validate and whitelist aggressively, and that invites human error in even the most technical users.
3. WYSIWYG editors aren't great. They're slow and cumbersome and frequently have major variations in behavior between different libraries, or even between the same library on different platforms. Most of the implementations I've used are buggy. And they actively fight against the reality that an HTML snippet can and does appear very different in different page contexts or on different screen sizes. What you see in an editor is almost never actually what you get on the final rendered page, and unless the site designer is extremely careful that can easily cause more confusion than it prevents.
So we're left with simplified domain-specific markup languages as the least bad option. Of those, Markdown is probably the best. Its pave-the-cowpaths approach to how people pseudoformatted text in old plain-text forums and emails means it's probably as close to intuitive as a markup language could ever get. Which is, admittedly, still not very intuitive; but in my experience it can be learned quickly by an intelligent but not technical user, provided they have some motivation to do so.
And the fact that it's becoming a de-facto standard on much of the internet means that the percentage of users who have some experience with it is only going to go up over time. That, combined with the relative ease of implementation, makes it a very attractive option for developers.
> And... they've failed right out of the gate. Face-down in the dirt.
Loved that comment. The difference between engineers and average users is that the average user isn't an engineer. The very idea that someone thinks using Markdown is a reasonable idea shows the disconnect.
I agree. All too often people are surprised by it on Reddit. Whenever someone replies with a number followed by a period (eg "42."), it's always converted into "1." Among other problems (escaping of backslashes, etc)
People seem ok with a lot of awful things. I don't see that as an argument for Markdown.
The question is: has Reddit (or other Markdown users) actually studied Markdown's usability? Even a basic comparative study against a WYSIWYG equivalent?
It seems like "techy" people just kind of assume Markdown is great and easy and awesome without ever bothering to actually make any effort to question that assumption. That's really the part that bugs me most.
And the entire pitch of Markdown is that you can style your text the way you did back when you were using text-only stuff. Ok. So _this should be underlined, then_. And /this should be in italics, right?/ And naturally this is boldfaced. Oh wait. Markdown is ZERO FOR THREE.
(Strangely, Markdown doesn't even have underline. Go figure. It's not even remotely close to feature-parity with the WYSIWYG textboxes we were all using on the web in 2008.)
I'm quite confused with a lot of your dislike for Markdown.
I have a reverse question, what is better than Markdown in your mind? I'm starting a project that is heavily based around text, and for simplicity i just wanted plain text. Ie, i have no editor for the project. So for markup, i need something. What do you feel is the right markup language?
Personally i feel Markdown is easiest markup language i've found. Compared to some, it's quite intuitive. Combined with the fact that it is becoming quite prominent means users are also learning it. The only downfall with Markdown in my mind is a lack of a standard, but i think that's being worked on (there are some attempts at that, i believe).
Anyway.. i would really appreciate your feedback, if you feel there are better pure text Markup languages :)
If an end-user is using a markup language at all, you've already screwed up as a developer of usable software.
If you're going to use a markup language, at least use one that supports basic shit like underlining text, or typing the sentence "21." without it thinking you're trying to start a list.
No matter how users are interacting with the text on your site, do usability testing. If you do usability testing in good faith, and Markdown comes out as the winner, then use it, with my blessing.
Don't base your usability decisions on gut instinct or knee-jerk reactions. Which is pretty much exactly what every site using Markdown currently is doing. AFAIK it's never been studied in front of a fair selection of users.
I believe OP means a WYSIWYG editor is much better than Markdown.
In terms of a markup language in pure text, I think Markdown is your best bet. Folks in CommonMark.org tries to "standardize" Markdown. Our implementation in Twincl.com follows that standard as much as we can.
Appreciate the reply. Honestly though, this confuses me even more. Many sites have both Markdown and a rich editor.. isn't that sort of required for users anyway? No matter what language powers your Markup, at the end of the day many users want an editor.
The median user would rather click a "add a poll" link and select options (with interactive help) rather than read a guide page on exactly how to punctuate and arrange whitespace to get a poll that works properly
Yup. I'm working on a sort of forum / collaboration space for my nonprofit (we're within a few months of open-sourcing it), and one of the big reasons we started from scratch is allowing users to work natively with rich text editors.
Well, we ended up using quill.js in our own system (it's rad, and pretty easy to wire into an operational transform system to handle concurrent edits). But at least at the time we were looking, discourse was markdown only.
We had a lot of needs that weren't met by other FOSS platforms, so it wasn't just rich text, but rich text was a big driver there.
It's an interesting concept, but what's stopping people from making this as a plugin for any other forum script? Because nothing in the page seems like it'd be hard to replicate for a good developer (read, someone who's likely coding add ons for existing platforms).
Seems like it'd be best just to wait for a version for my existing forum script. Or get someone to program one.
The use of markdown is questionable too. As mentioned, normal users find it somewhat awkward. Might work much better with a button/modal driven system to set up the quiz or Tic Tac Toe game...
To post or comment on Twincl site, just login with a verified Facebook account. Alternatively, you may register
a new Twincl account, and subscribe to the membership with
a minimal fee. (Our point is to make the account less
disposable, which we believe is a key to pleasant community
experiences.)
And you're gonna struggle here too.
Many people who use forums do not care for Facebook. They especially do not care for trying to tie every single account to an identity with their real name (and likely photo) attached to it.
This is especially true of less 'savoury' subjects, which are some of the subject areas that forums can work really well for. A forum on say, psychological issues or relationship problems is pretty much a non starter with this script. As is any forum in a legally grey area. Or heck, any forum where you just want a different username for whatever reason or don't want the staff/members to know every little thing about you.
Or any forum where themed usernames are a big part of the community. No one's going to pay $5 a year to call themselves say, Mario. Or Master Chief. Or Pikachu.
So it seems like the registration system basically makes it a non starter for any forum where people want some anonymity/privacy, any forum about a fictional universe or series (or where role playing is a thing) or any forum relating to anything remotely illicit or legally grey.
It has some interesting concepts, but the membership stuff is a problem and the UI stuff seems like it's both a tad complicated and easily copyable by your competitors.
Forum anonymity (or more precisely, disposable account) is the source of moderation pain and the reason of suboptimal forum content quality. It's one of the main issues I want to address from day 1. If FB account puts a lot of people off, I may consider adopting SMS verification sooner, or change the yearly subscription to a one-time small fee.
It seems Markdown editing is a red flag for broader adoption. I've thought about a dual approach before: allow casual users to edit with GUI, and provide the Markdown editing capability for pro users.
I do not have a fool-proof way to prevent plugin copying. If somehow it's widely copied, it's not a bad thing (to most people), too.
As someone who's run a ton of forums, I disagree it's the source of suboptimal forum content quality. If I had forced real names and identities, I would have lost a significant amount of users and activity. And not just trolls and spammers either.
A lot of people like to keep identities separate. Any system that tries to remove the anonymity site of things will end up stifling quite a few people who worry about their personal information being misused or who don't want everyone to know about their every activity online. That's not a bad thing, especially not in a world where intelligence agencies want to spy on everything you do, you could lose your job because of some comments on Facebook or Twitter and being swatted is sadly quite common.
Not everyone wants their friends and family to know they're an active member on say, Donald Trump Fan Forums. Or that they're posted leaked content from an upcoming game, like quite a few people on 4chan or Reddit. Or that they're interested in setting up a Tor node.
As for markdown, yeah, your solution there is exactly the solution. A GUI for new users, a raw editing text box for pro users. It's like how a lot of forum editors switch between BB code/raw text and a WYSIWYG editor.
Plugin copying is a good thing here (because the features mentioned could be good in other scripts), but I'm just warning you'll need quite a bit more if you want to differentiate themself enough from competitors in the market (of which there are thousands).
> To post or comment on Twincl site, just login with a
> verified Facebook account. Alternatively, you may register
> a new Twincl account, and subscribe to the membership with
> a minimal fee. (Our point is to make the account less
> disposable, which we believe is a key to pleasant community
> experiences.)
$5/year or $2/three months.
If you live forever, that's an infinite amount of money.
cf. metafilter, which charges $5 one time to prevent disposable accounts.
You can use a verified Facebook account to join Twincl site for free. Ideally, we should use our own SMS verification process to prevent disposable accounts - maybe when we have money. Since Facebook does a decent job of fraud account prevention, we leverage their work for now.
Thanks for your comment. We'll take that into account.
...Not that I'm a fan of G+, but I am happy to use that google account identity to sign into other services.
It's because that google account identity is tied to a work email account, too... You know what doesn't go on there? What I do on the weekends. That ends up on facebook even if I don't put it there... So leave that account alone.
It's worth noting that I am permitted more than one Google account, but not more than one Facebook account.
...There MAY be some rule about not having more than one G+ account, right? But, I don't really use that service. It was only included here because it's the only "profile" page I know of for that account.
At least you must be careful when using multiple Facebook accounts from the same computer. It might cause your main Facebook account locked, and take you a while to have Facebook unlock it for you.
Great! Finally someone in the open source forum community designing a UI for the average user!
> On twincl.com, a piece of rich content (which we called a twinclet, pronounced as “twink-lit”) is written with a Markdown fenced code block.
And... they've failed right out of the gate. Face-down in the dirt.
I will never understand the obsession some developers have with Markdown.