On 12/08/2017 05:13 AM, Walter Bright wrote:
> 
> I just don't see much of any improvement of markdown over LINK2.
> 

I do.

On Sunday, 10 December 2017 at 06:20:43 UTC, Walter Bright wrote:
> On 12/9/2017 9:17 PM, meppl wrote:
>> since commonmark exists, is specified and is compatibale to many markdown-languages,  I claim there is a markdown standard: http://spec.commonmark.org/
>
> It certainly wants to be the standard, but until most everyone decides to follow it, it is not. There is no list of anyone on that site conforming to it.

I think these are wrong criterias to estimate the value of commonmark. Commonmark doesn't need to list anyone and doesn't need to be listed by anyone to be a standard. commonmark is a standard proven by following "facts":
1) whenever a language feature is used by all popular markdown languages, it is standard
2) there are markdown features who are used by all popular markdown languages
3) everyone can reveal this matter of fact - e.g. by writing it down as a specification
4) any language feature published by the commonmark-spec is used by all popular markdown languages
ergo: commonmark == standard markdown
well, at least, if the commonmark people did their homework right

>
> Besides, commonmark has a lot of stuff we don't need, like multiple ways of doing the same thing.

I am not the one who is implementing it, so I will not argue against this.

On 12/10/2017 6:22 AM, meppl wrote:
> I think these are wrong criterias to estimate the value of commonmark. Commonmark doesn't need to list anyone and doesn't need to be listed by anyone to be a standard. commonmark is a standard proven by following "facts":
> 1) whenever a language feature is used by all popular markdown languages, it is standard
> 2) there are markdown features who are used by all popular markdown languages
> 3) everyone can reveal this matter of fact - e.g. by writing it down as a specification
> 4) any language feature published by the commonmark-spec is used by all popular markdown languages
> ergo: commonmark == standard markdown
> well, at least, if the commonmark people did their homework right

I have a more pragmatic definition of a standard:

1. Products that implement it say they adhere to it and defer to it as the authority on correct behavior.

2. There's more than one such product.

3. There's more products adhering to that standard than some other competing standard.

So far as I know, commonmarkdown satisfies zero of those.

Don't get me wrong, I think commonmarkdown is a worthy effort, and is definitely in the running to be a standard. Certainly a lot more effort seems to have been put into it vs other markdowns. It is entirely reasonable to refer to it to answer questions about whether some detail should yin or yang.

But implementing commonmarkdown in Ddoc is not what we're going to do, at least for the near term.

On Monday, 11 December 2017 at 00:54:00 UTC, Walter Bright wrote:
> On 12/10/2017 6:22 AM, meppl wrote:
>> I think these are wrong criterias to estimate the value of commonmark. Commonmark doesn't need to list anyone and doesn't need to be listed by anyone to be a standard. commonmark is a standard proven by following "facts":
>> 1) whenever a language feature is used by all popular markdown languages, it is standard
>> 2) there are markdown features who are used by all popular markdown languages
>> 3) everyone can reveal this matter of fact - e.g. by writing it down as a specification
>> 4) any language feature published by the commonmark-spec is used by all popular markdown languages
>> ergo: commonmark == standard markdown
>> well, at least, if the commonmark people did their homework right
>
> I have a more pragmatic definition of a standard:
>
> 1. Products that implement it say they adhere to it and defer to it as the authority on correct behavior.
>
> 2. There's more than one such product.
>
> 3. There's more products adhering to that standard than some other competing standard.
>
> So far as I know, commonmarkdown satisfies zero of those.
>
> Don't get me wrong, I think commonmarkdown is a worthy effort, and is definitely in the running to be a standard. Certainly a lot more effort seems to have been put into it vs other markdowns. It is entirely reasonable to refer to it to answer questions about whether some detail should yin or yang.
>
> But implementing commonmarkdown in Ddoc is not what we're going to do, at least for the near term.

There are loads of implementations of CommonMark https://github.com/commonmark/CommonMark/wiki/List-of-CommonMark-Implementations the one I have written is not listed. That covers 1 and 2.

Also Markdown is not a standard, it started out as a pearl script and the a short documentation on how to write text for it. It has several ambiguities, leading to a lot of implementations do things differently. So they don't agree on the authority of Markdown. Which makes Markdown a mess because you don't know what behaviour you will get from the different implementation. So that covers 3.

And to add more, CommonMark on the other hand has a full spec written and several test that covers the difficult to get right parts of Markdown/CommonMark. I'm sure I don't need to tell you the virtues of a good test suit.

Cheers, Jakob.

On Monday, 11 December 2017 at 14:22:37 UTC, Jakob Bornecrantz wrote:
>
> And to add more, CommonMark on the other hand has a full spec written and several test that covers the difficult to get right parts of Markdown/CommonMark. I'm sure I don't need to tell you the virtues of a good test suit.
>

The CommonMark approach is to just take the union of all possible features and call it a day.
Standards without opinions don't deserve to be implemented by anyone.

On Monday, 11 December 2017 at 14:22:37 UTC, Jakob Bornecrantz wrote:
>
> There are loads of implementations of CommonMark https://github.com/commonmark/CommonMark/wiki/List-of-CommonMark-Implementations the one I have written is not listed. That covers 1 and 2.
>
> Also Markdown is not a standard, it started out as a pearl script and the a short documentation on how to write text for it. It has several ambiguities, leading to a lot of implementations do things differently. So they don't agree on the authority of Markdown. Which makes Markdown a mess because you don't know what behaviour you will get from the different implementation. So that covers 3.
>
> And to add more, CommonMark on the other hand has a full spec written and several test that covers the difficult to get right parts of Markdown/CommonMark. I'm sure I don't need to tell you the virtues of a good test suit.
>
> Cheers, Jakob.

If there is a desire to add CommonMark support, it is best to output ddoc to commonmark and call an external tool specified by the user. The proposal to add a few markdown features to ddoc is a reasonable short run goal. If you set commonmark support as a goal, it will probably never get done.

On Monday, 11 December 2017 at 15:45:07 UTC, Guillaume Piolat wrote:
> On Monday, 11 December 2017 at 14:22:37 UTC, Jakob Bornecrantz wrote:
>>
>> And to add more, CommonMark on the other hand has a full spec written and several test that covers the difficult to get right parts of Markdown/CommonMark. I'm sure I don't need to tell you the virtues of a good test suit.
>>
>
> The CommonMark approach is to just take the union of all possible features and call it a day.
> Standards without opinions don't deserve to be implemented by anyone.

It's crazy to see that the most basic HTML features still cause issues in Markdown. CommonMark can be seen as a pragmatic approach but may suffer from the "Babel tower" problem: try to assemble all the languages tend to make everybody disagree, leading exactly to the opposite of the initial goal, that is unification (it's probably not exactly that but i'm not a theologist after all, 😂).

On Monday, 11 December 2017 at 15:45:07 UTC, Guillaume Piolat wrote:
>
> The CommonMark approach is to just take the union of all possible features and call it a day.
> Standards without opinions don't deserve to be implemented by anyone.

I disagree. If anything, it's more of a subset of features common to the most widely-used markdown implementations, with corner cases ironed out (see lots of discussion on the [CommonMark forum](https://talk.commonmark.org/), and also comparisons of output from different implementations at [Babelmark2](http://johnmacfarlane.net/babelmark2/) (linked to from commonmark.org)).

On Monday, 11 December 2017 at 00:54:00 UTC, Walter Bright wrote:

> Don't get me wrong, I think commonmarkdown is a worthy effort, and is definitely in the running to be a standard. Certainly a lot more effort seems to have been put into it vs other markdowns.

Note that CommonMark isn't simply a markdown implementation. It's a:

  * spec
  * test suite
  * two reference implementations
  * collection of community implementations

and also includes an active community forum.

The people involved are dedicated. The spec is conservative, uncontroversial, and detailed.

> But implementing commonmarkdown in Ddoc is not what we're going to do, at least for the near term.

Right. That said, if you want to add a handful of markdown-ish features, I think it would be most useful to draw your very limited markdown subset from the CommonMark markdown spec. The other option takes you down the road of unintentionally creating yet another variant for users to contend with.

On 12/11/2017 11:29 AM, John Gabriele wrote:
> Right. That said, if you want to add a handful of markdown-ish features, I think it would be most useful to draw your very limited markdown subset from the CommonMark markdown spec. The other option takes you down the road of unintentionally creating yet another variant for users to contend with.

That's basically what I said :-)