Actually Jeff, my definition is pretty wide in scope. About as wide as it gets. True to the original spirit of Markdown, you might even say!
If anything is narrow in scope, it’s the CommonMark spec, which is clearly designed for a specific kind of writing—one that involves a lot of code blocks. It’s clearly for programmers, by programmers. Not even so much as a mention of tables or footnotes? C’mon.
The entire document reads like a nerd’s dream of what Markdown should be. If anyone else had decided to take this project on—say, a group of academics or designers or journalists—it would look very different. To me it proves Gruber’s point that there is no such thing as a “single Markdown spec for everyone.”
Anyway, I’m done arguing about this. Jeff, I wish you success with CommonMark, though I believe the idea is flawed at a fundamental level. I’m sure it will hit the mark with programming-focused writers out there, at least.
In listening to that podcast, did anyone else get the vibe that that entire conversation was rehearsed well in advance? The whole “let me guess, Jeff Atwood?” part of it pretty much sealed it for me. And from the get-go, the “did you hear about these guys trying to steal markdown” (or something along those lines) comment to start that part of the conversation already sounded rehearsed and that response from the host seemed to confirm it.
Although I respect your wish to no longer participate in this argument, I find your final post more confusing than clarifying, and would appreciate it if you’d provide some examples to help explain what you’re trying to say.
In particular, are there any particular items in the CommonMark spec that make it narrower in scope than the original Markdown spec, or that (in your opinion) make it particularly obvious that CommonMark is better suited for programmers than non-programmers? And (on a more hypothetical level) is there anything in particular that you think a group of (non-CS) academics would have done differently?
This is a whole impressive discussion
Just few words in case it may be of interest for you…
The mentioned “2 weeks” delay looked short to me, as you can happen to be in holiday, or for any reason in a place you can’t or don’t want to access emails. for 3 or 4 weeks.
I appreciated the explanation about the risk of trademark dilution which show why John MacFarlane should better refuse over-usage of the “Mardown" word
On the other end John tone looks to be a bit over agressive, even more when addressed to a project he could either take the lead on or benefit from even higher popularity for what he created. Deserving apologies doesn’t mean it can’t be notified nicely
My first thought for another name is one I didn’t saw yet in this discussion: Open Markdown or OpenMark
It would reflect the goal to have a community driven and also community accepted Markdown core base, to which any addition should be proposed as optional extensions.
I like your point about the file extensions, and for that reason - and to fit in with how the cool kids are naming their projects - I kind of like CommonMarkD. I could go any way on capitalization of the last letter, but saying text is “Common Mark’d” has semantic meaning, and dropping the ‘e’ makes it internet-y. Oh, yeah, and the ‘D’ is definitely for ‘Doge’. Definitely.
You’re overthinking the whole thing, and I believe that Grubber agrees since he suggested Pedantic Markdown.
You’re too focused on details and you’re missing the overall point of Markdown: Markdown is a way to make a document that’s simple to write and easy to read even if it’s not rendered by some parser. There may be edge cases in Markdown, but it’s likely that any document with those edge cases is probably unreadable in its raw form. The best thing for a parser to do in those cases is to spit out an error message, “Hey, this is Markdown, and not Microsoft Word. Keep it simple.”
There might be a need to create a decent Markdown to HTML converter. If that’s what this project is about, go for it. Most of the Markdown to HTML converters …what’s the technical term?.. Oh yeah, “suck”. We should have a decent Markdown to HTML converter, and a decent Markdown to PDF converter, and a decent Markdown to POD converter and a decent Markdown to Microsoft Word converter too. That’s the point of Markdown – make an easy way to render an easy to understand plain text document into another format.
You might feel a need for new formatting capabilities. The original Markdown specification didn’t have tables. Maybe we need a way to show -deletions- and +additions+ for editing purposes, or maybe footnotes^1.
Be careful about adding features. That’s a slippery slope. Know the ghastly tale of Confluence. Confluence started out as a simple Wiki rendering engine and was well liked. However, in each release Atlassian added more and more features to its wiki markup language. Soon it was impossible for most users to write a Confluence page with the plain text editor. Now, the format of the wiki markup language is so complex that most users can’t write a document even using the built in GUI editor.
I think you should have asked interested parties (meaning us) for name suggestions. I’m suggesting the name MarkdUp. Markdown is obviously a play on the word markup. My suggestion fuses Markdown with marked up, while trying, at least, to suggest Markdown in the name. Oh well. Probably too little, too late.
I think most people are missing the point of what Jeff’s trying to do. He’s not trying to expand Markdown or add features. He’s trying to unify it. Github, SO, Discourse and any number of sites use their own interpretations of the Markdown spec. This means that any Markdown written on Github may or may not work on Discourse, and vice versa. What he and the team are trying to do is create a more concrete, unambiguous spec that all sites can use so we have consistent implementations of Markdown across the internet.
I see nothing but good intentions here, and also nothing but good things can come out of this.
Well, with the exception of this little spat, of course.
What gives you the impression that Jeff doesn’t completely and totally understand this and agree with it?
…did you actually look at the example of ambiguity that Jeff posted? It’s hardly “unreadable” in its raw form.
That would be extremely pedantic, and much, much more contrary to the spirit of Markdown than anything Jeff is doing. The best thing for a Markdown parser to do in edge cases would be to follow the standard procedure for such edge cases as clarified by the spec, which of course would only work if the spec actually clarifies such cases. This way, either the text will be rendered as the user intended without them needing to adjust anything, or the user would see that it’s not what they expected and re-write it. And if the user copies-and-pastes the source onto a site using a different parser, the output will be the same, as expected.
You know what would help with that? Having a spec, so that implementers of these converters know exactly what to implement.
These statements are contradictory and also miss the point of standardization. Standardization is not the time to add a bunch of new features. Jeff isn’t creating a new “flavor” of Markdown with cool new formatting tricks. He’s eliminating edge cases to ensure that CommonMark really is the simple and extensible markup language Markdown was always intended to be.
Heh, yes, this little spat. Sadly, my feelings are mostly negative towards John due to his handling of the situation. I understand he wants protect what is his, but the intent here is to better his creation, and it is done out of love for that creation, not in spite of it or him. For him to be so against it is beyond me, especially when he’s neglected Markdown for so many years.
You must be factually wrong here. Gruber has access to lawyers. If there was a case, they would have done something.
Instead, we have to remember the following facts:
Gruber’s contribution was an actual code implementation.
CommonMark is functionally a spec, and code with a totally separate behavior (it validates an implementation as conforming to a spec, rather than converting text to marked up text). There is no work-product crossover. Gruber never defined a spec. CommonMark is not providing a Markdown implementation.
CommonMark is not materially hurting Gruber or the Markdown brand. If anything, it is increasing the Markdown brand’s value. IANAL, but without harm, there isn’t a legal case.
He might have some legal claim if the name were exactly the same, due to work-product confusion (which is which?), but creating a product that is ‘a specification of the common understanding of Markdown, called CommonMark’ clearly points to Markdown being something separate. I’m not even sure he has a moral claim here: it’s sort of like saying Henry Ford should have some say as to what the National Highway Traffic Safety Administration calls their policies as to what defines an ‘Automobile’.
Honestly, I was really surprised that Gruber was so passionate about it. I kind of doubt that many people realize he’s the one who started it. Also, as it’s been pointed out he’s been pretty silent about it for a long time. With an idea that attempts to make things really open it almost appears to be put in place to spur on more development and evolution.
It’s a sad thing that something so useful and adopted is now tarnished with this stifling of continuation by the creator(Gruber).
It would be easy for him to throw his spec and implementation up on github/bitbucket/theinternet an monitor pull requests to continue it’s evolution.
This definitely has some affect on my opinion of John Gruber and I hope he attempts to provide a better solution in the future.
I recently noticed about your work at one of Russian’s biggest IT community. Frankly speaking, I was excited that you guys organized such project. I am creating a web project, based on m******n syntax and I’d be glad if format specification comes out, you are right, it’s time has come.
If Gruber claims to that word, leave him in 2004 with it, but go on with your idea. I hope SE, GH, Reddit and Meteor guys are the right company to go with, touchy grumbler is not.
I’ll wrote with big letters in future project that markup language is CommonMark. Not the-one-whose-name-can-not-be-called
I’d be pleased to help with python/js in your project.
Wow, I’m surprised Gruber would respond in such a hostile manner to the efforts of Atwood and Macfarlane to improve Markdown. My impression is that their intent was for the greater good of the community around a technology that they know and love, and that is in apparent need of ongoing stewardship. I’d be honored if A & M approached me about something I’d invented. -10 for Gruber
This thread makes me a bit sad… it’s as if there was something important going on, but actually, it is just people on the internet fighting over a name.
Markdown is Markdown and will be Markdown, no matter what some people with way to much time on their hands may think is “right” on the internet. You can call it Common Mark as much as you want, it still is Markdown.
Having that out of the way, I think everyone should just forget about how the spec should be called and just use it for writing more important stuff.