Markdown (and what do you do when a community outgrows your contribution)

[wraith808]

I love markdown.  I use it when writing everything from fiction, to game books, to documentation at work.  There's something so freeing about being able to write and format in plain text.  However, the specification really sucks.  I can take the same markdown document, and run it through X parsers and get X results.  That's a problem, especially with something that's become so prevalent.  People have played around with making a standard or forking it or several different things... but the biggest holdup has been the creator of it.  He has explicitly said that he didn't really see the need for a specification.  And has consistently not wanted to be at the forefront of development on it.

A few older articles on this issue:

http://blog.codinghorror.com/responsible-open-source-code-parenting/

http://blog.codinghorror.com/the-future-of-markdown/

http://www.fastcolabs.com/3015487/john-gruber-and-jeff-atwood-explain-why-markdown-is-still-growing-up

http://www.garron.me/en/blog/on-the-future-of-markdown.html

So finally, a lot of people got together on the spec... and released a standardized version of markdown.

Then the fireworks started.

I'll let the articles speak for themselves...

http://blog.codinghorror.com/standard-flavored-markdown/

http://blog.codinghorror.com/standard-markdown-is-now-common-markdown/

Before I chime in, I'll ask for thoughts... because I'm sort of split on the issue... and I don't think its over.

Oh... and if you want to see the spec, check http://standardmarkdown.com/. ; It will be going away... and not redirected, which I don't understand.  But I'm assuming it will be at http://commonmarkdown.com/

[mouser]

Seems like having a common standard is a good idea to me, and it seems pretty obvious to me that having an unambiguous specification is a good thing.
I can understand people vying for different versions of what should be the standard, but i can't see a good argument for not having one.

[40hz]

Microsoft built an entire industry based on the proposition that, in the absence of an industry standard, any standard is better than none.

Microsoft provided a missing standard for the personal computer desktop (not a good standard - just a missing one) and went onto unbelievable success and industry dominance.

The wild and woolly frontier is fine as long as it's still a frontier. But once you have towns and cities, and people trying to collaborate and share, standards become desirable for pragmatic reasons if nothing else.

Trying to avoid a standard when it becomes obvious one is both wanted and needed by a large number of users makes zero sense. And I can't follow what Gruber's argument or point seems to be - other than him saying "this is my child - leave it as it is."

In many respects his behavior reminds me of Aesop's "dog in the hayloft." The people who want to extend and standardize markdown seem to be bending over backwards to acknowledge and show respect for Gruber's contribution when they could have just as easily forked and been done with it. But they very much wanted to keep "markdown" in the name and involve Mr.Gruber in the process. However, for some reason, Mr. Gruber seems to be highly offended by their overtures.

Dunno...guess you had to be there or something. I just don't get it.

[attach]

[rgdot]

Standards are ok and needed. It shouldn't be one person's idea though and is probably better for it to be a minimum set that is extensible for certain applications

[mwb1100]

The people who want to extend and standardize markdown seem to be bending over backwards to acknowledge and show respect for Gruber's contribution when they could have just as easily forked and been done with it. But they very much wanted to keep "markdown" in the name and involve Mr.Gruber in the process. However, for some reason, Mr. Gruber seems to be highly offended by their overtures.

Personally, I think it would have been nice for this standardization process to be allowed to use the name Markdown.

However, Markdown was created by Gruber and the Markdown project belongs to him. He therefore has the right to decide how the name for that project should be used.  I think this 'fork' should simply use a different name.  I have no doubt it will become successful (since it'll be used on StackExchange, Github, and Reddit).  And it appears that this is the tack that they've taken - the fork is now called CommonMark.

I actually think that in the long run, Gruber's Markdown will largely become a note in the history of CommonMark.  But that doesn't give them the right to the project name.

[wraith808]

The people who want to extend and standardize markdown seem to be bending over backwards to acknowledge and show respect for Gruber's contribution when they could have just as easily forked and been done with it. But they very much wanted to keep "markdown" in the name and involve Mr.Gruber in the process. However, for some reason, Mr. Gruber seems to be highly offended by their overtures.

Personally, I think it would have been nice for this standardization process to be allowed to use the name Markdown.

However, Markdown was created by Gruber and the Markdown project belongs to him. He therefore has the right to decide how the name for that project should be used.  I think this 'fork' should simply use a different name.  I have no doubt it will become successful (since it'll be used on StackExchange, Github, and Reddit).  And it appears that this is the tack that they've taken - the fork is now called CommonMark.

I actually think that in the long run, Gruber's Markdown will largely become a note in the history of CommonMark.  But that doesn't give them the right to the project name.

While what you say is true- which is the reason for my split feelings on the subject- it's still very petty and small to both hold on to something and prohibit growth.  Especially when there are several things using the markdown name that Gruber neither has given permission to, nor endorsed.  Look at his scathing comments on MultiMarkdown to see that.  He really, I think, objected to the word Standard.  Which when replaced with Pedantic in his reply really shows his true colors, IMO, i.e.

Edit: after a long and thoughtful email from John Gruber – which is greatly appreciated – he indicated that no form of the word "Markdown" is acceptable to him in this case. We are now using the name CommonMark.

from http://blog.codinghorror.com/standard-markdown-is-now-common-markdown/

[wraith808]

Oh... and another quote that shows why a standard was needed... my example was only a bit pessimistic.

It isn't a specialized dialect.

It's a deep introspection on supporting exactly what Markdown was intended to do, but failed to specify with enough accuracy, leaving us to the current situation where this...

Code

# Hello there

This is a paragraph.

- one
- two
- three
- four

1. pirate
2. ninja
3. zombie

… renders out as 15 different outputs from 22 different Markdown parsers.

In fact, one of the design goals was to render "most" Markdown as close to the original intent of the authors as possible.

On that note... when will we get md code formatting?

[mouser]

They seem to be bending over backwards to respect the wishes of Gruber, and naming it "CommonMark" seems like a nice thing to do.

I do think, as an outsider, this episode would be sufficient for me to decide to move to CommonMark and to completely switch away from any other form of markdown other than CommonMark.  By nature I begin by deferring to the original authors (in this case Gruber) in terms of who I think should get the benefit of deference when it comes to establishing a standard.  And if this was a case of COMPETING standards, I would no doubt line up behind Gruber in terms of opposing the efforts of other groups to establish an alternative standard using the name "Markdown".

But if the choice is between two well-meaning groups, one attempting to define a standard to solve the very real problem of incompatible and inconsistent variants and the lack of a specification and standard, and the other resisting such efforts -- I'm going to line up behind those trying to establish a standard.

[Renegade]

CommonMark looks pretty good. The specification uses LOTS of examples. Oh god, is that refreshing to see! Far, far too often specs have some kind of rule that is never demonstrated. They really make the specification a lot easier to read and understand with the countless examples.

"Example 441" 

For anyone that has ever implemented any part of an RFC or standard, that just makes life so much easier. All too often rules have some bit of ambiguity, and examples help clarify the intent.

As for the standard there, I only glossed over it, but it seems consistent. They took 2 years to create it, so I'm pretty certain that they've got it down right. High marks for that.

As for picking standards, I suppose that I'm in favour of whatever standard is open, unencumbered, works, and well-supported or in common usage. This looks like it would fit that.

[phitsc]

Wow! I'm a big Markdown fan myself. Didn't know about this dispute though. Would be interesting to know what goes on inside Gruber's head. Seems like he missed the chance to be part of something important. I'm actually surprised how much effort the team behind CommonMark put into getting him on board. With the attitude he demonstrates, I think I would have just ignore him much sooner.

[wraith808]

Wow! I'm a big Markdown fan myself. Didn't know about this dispute though. Would be interesting to know what goes on inside Gruber's head. Seems like he missed the chance to be part of something important. I'm actually surprised how much effort the team behind CommonMark put into getting him on board. With the attitude he demonstrates, I think I would have just ignore him much sooner.

Well, he's aptly stated what goes on inside of his head.  I can't find the interview quote right now*, but he's outright said that markdown doesn't need a standard, when it's obvious that it does... especially to the end user.  I was using three different markdown tools... one to edit online so I could do it anywhere and sync via dropbox... one locally to take that sync and format it (a piece of software I'm working on and hoping to have for NANY) and one to take the formatted version and output a word document for distribution.  The amount of effort I had to do to make sure that the final product was good for distribution was painful.

Of course, it doesn't help that he is sort of acting passive aggressive about the whole thing.  He's not personally insulting CommonMark or Jeff Atwood... he's letting his followers doing it an retweeting them.



Staying classy and above the fray there...

* It was a tweet, as most things in this seem to be: https://twitter.com/gruber/status/507364924340060160

[wraith808]

And it gets worse...

https://twitter.com/gruber/status/507647288471470080

and it appears that Jeff Atwood has learned a lesson... even to douchey responses (though admittedly, he could have left the finally off his original tweet or used another word), he's responding better now...

https://twitter.com/codinghorror/status/507847788592246785

Some historical context also:

https://twitter.com/gruber/status/262287246953164800

https://twitter.com/gruber/status/261650083689426945

[Renegade]

Fight! Fight! 

I want to see blood a crash a null pointer exception!

[phitsc]

The same thing just starts happening with Scala. Even has some reference to the Markdown quarrel in the comments

[mouser]

So I finally had a chance to read through the comments in the second codinghorror/stackoverflow post (here), and I guess I can see part of the argument being made on behalf of Gruber, which I interpret something like this: "There is a person who clearly invented this thing called Markdown, and you can't just publish something that you name in a way that makes it sound like the official standard version of it, against the will of the creator, even if that person is being an asshole.  You should have tried harder and earlier to contact the original inventor, and you certainly shouldn't have tried to be cheeky on day 2 by changing the name from 'standard markdown' to 'common markdown' and act like that was the end of the issue."

As someone who thinks the original creator is entitled to some real deference, and as someone who has always been uncomfortable about the possibility of forks of opensource projects being used to take control away from the creator of a project, even if they are unreasonable bastards, I have some natural sympathy for the position of the original author..

Anyway, I think the lessons are pretty clear for the rest of us:

• Work harder to get the original author to be part of your project.
• At the very least, determine if the original author is going to be a jerk about improving their creation, or is against your idea.
• If the original author is going to make it impossible for you to use the name of their project, for whatever reason, find another starting point and let go of the idea of using that product as a starting point.

[wraith808]

But mouser, they did try to get him involved.  They've been trying for the whole two years that they worked on it.  And he either (a) ignored them or (b) denigrated them on social media.  Look at some of the links involved.

And truthfully, I think if they started out with the commonmark idea, they would have gotten the same reception from him, and others- that they stole it, etc.  By starting out in the Standard Markdown space... they shed a lot of that... then moved to a better space where they could say that they tried to bend over backwards to make the spec part of markdown... but were put off at every turn.

When *finally* (and I use that word for a reason) he stopped being passive-aggressive, and said that he'd not approve any permutation of Markdown with their project, they moved.  And the having to get a blessing from him?  there are so many uses of markdown in a name that he had nothing to do with.  I'm sorry... if you want control, don't make it open source, trademark your name, and protect your trademark.  

He used a bastardization of an open source license... and no one is calling him on it is the other thing that's crazy.

But I think Jeff says it best in this particular response:

I do not want stewardship over Markdown. I want it to be a viable, community maintained open source project since millions of people rely on it. That is what open source is supposed to mean.

[mouser]

They've been trying for the whole two years that they worked on it.

I didn't quite get that impression.  I noted that there is a mention of talking with him when they started (2 years ago) -- not clear how serious that attempt to talk was -- and then 2 weeks before release.. But I couldn't tell if they had talked to him any time in between.

Essentially I think we are on the same page -- that when something like this is released as open source, the natural interpretation is that it's going to be a project that the community can improve and evolve forward -- either led by the original creator, or by others entrusted by the community.

And I've already made clear that I believe standardization -- especially in the form it was done here -- is a good and important thing.

So the only real question in my mind is what you do when the original inventor becomes an obstacle to the continuing use and development of the project.

And I think my position is that when this happens, and attempts to convince the original inventor to join the team fails, that perhaps the best solution is to make a clean break from the original name of the invention and come up with a clearly different name -- even though that could lead to some (initial or continuing) confusion for users.

[phitsc]

I agree mouser that choosing a different name and just stating that one of the main goals was Markdown compatibility would probably have been better.

[wraith808]

And I've already made clear that I believe standardization -- especially in the form it was done here -- is a good and important thing.

So the only real question in my mind is what you do when the original inventor becomes an obstacle to the continuing use and development of the project.

And I think my position is that when this happens, and attempts to convince the original inventor to join the team fails, that perhaps the best solution is to make a clean break from the original name of the invention and come up with a clearly different name -- even though that could lead to some (initial or continuing) confusion for users.

Look at the links in my post above.  Specifically:

https://twitter.com/gruber/status/262287246953164800

https://twitter.com/gruber/status/261650083689426945

He does make it quite clear that he doesn't want to be involved.  He's also very passive aggressive about it.  The later one is after he's said he doesn't want to be involved, they as a courtesy gave it to him to make suggestions and or say anything about it.  That's where the two weeks came from.  It seems a lot like the old if I don't see it and don't acknowledge it, then maybe it will just go away approach.

I agree mouser that choosing a different name and just stating that one of the main goals was Markdown compatibility would probably have been better.

And that's what they did in the end after he finally made it clear that he wouldn't allow the name to be used under any circumstances.  That's after telling them that he might allow pedantic or strict markdown... which both don't convey what is being done at all.  What makes it even worse is that this is instructions for a parser- but because there are a lot of non-technical people interested in markdown, they don't get that this is what it is for, and that even non CommonMark Markdown will still be able to be parsed just as normal.

[mouser]

The good thing that should come out of all of this is that developers should run, not walk, to CommonMark -- and abandon Markdown as rapidly as possible.

[wraith808]

The good thing that should come out of all of this is that developers should run, not walk, to CommonMark -- and abandon Markdown as rapidly as possible.

That's one thing I'm afraid of.  A fragmentation.  But I don't think that CommonMark should really be hit by that as it doesn't include in the spec not to render other markdown.  Just to make sure that you render this markdown.

[phitsc]

...
And that's what they did in the end after he finally made it clear that he wouldn't allow the name to be used under any circumstances.  That's after telling them that he might allow pedantic or strict markdown... which both don't convey what is being done at all. 
...

Well, the Markdown license states that the name “Markdown” may only be used with specific prior written permission. So choosing a different name right from the start, while most certainly not changing Gruber's opinion about the whole endeavor, would maybe have led to less negative reactions from some parts of the community.

[DeVamp]

Personally I think it is a good thing to have a standardized reference somewhere, because all the different kind of markdown forks, are hard to know.

But it all depends on what the developers who implement some kind of markdown, will use of course.

[wraith808]

...
And that's what they did in the end after he finally made it clear that he wouldn't allow the name to be used under any circumstances.  That's after telling them that he might allow pedantic or strict markdown... which both don't convey what is being done at all. 
...

Well, the Markdown license states that the name “Markdown” may only be used with specific prior written permission. So choosing a different name right from the start, while most certainly not changing Gruber's opinion about the whole endeavor, would maybe have led to less negative reactions from some parts of the community.

It's not a license... and several people have used it without prior permission.  And if you follow the thread of posts, they asked.  And he never gave a response.  It's also open source... and he inserted closed permission type license into an open source license.  It just doesn't work that way.  Though while he attempted to wield it as such... and has indeed tried to negatively affect certain initiatives as if it were such, he in all probability (never can tell with judges) would have no legal ground to stand on.  You might argue moral or ethical grounds.  But since the licensed code itself had ambiguities and bugs... and when pressed on them as part of this initiative he said there were no bugs- then after CommonMark was released, he said he might go back to address some 'problems'- I'd argue that he already gave up the high ground on that one.

I even think that if they had branded as they are now... he would have still phrased it as they were taking his work.

But, that's neither here nor there.  They did change it, and the historical bit will very likely fade into the ether, probably faster than if they'd taken this approach to begin with.

[Renegade]

I've got the distinct feeling that Atwood & co. had an understanding that Gruber was pretty much a capital dick before they did much, and tried to deal with it as nicely as possible. But, that's just my own impression. I could be off base there.