Markdown Drama Part III - The Revenge of the CommonMark
-
Oh sorry, how could I forget. The obligatory angry Hitler dub of the drama.
-
You can't abbreviate that thing as SM anymore?
He should have chosen "Strict Markdown" instead.
-
I'm done for tonight. peace out!
-
I had "no repro" to a bug even though he tried it on a different device and version of android. I wonder how long I would last at work if I took that approach.
-
Well, this was a minor technical / meta-y thing. But it just got way more interesting.
Starting with the name. In his email John graciously indicated that he would "probably" approve a name like "Strict Markdown" or "Pedantic Markdown".
Oh man, this guy makes Jeff look good for once. Aside from the obvious "complying with this idiocy, instead of sticking out a middle finger and starting to write BBCode spec" thing.
First you make the most totally insane markup language for which your spec is roughly "do whatever the fuck you want, I don't care", then when people actually try to make things easier for them by drafting up a standard spec, you throw a hissy fit, call them "pedants" and pretty much estabilish that there is no standard, and the inconsistency, insanity and having every parser on the planet produce different result is a feature.
-
In his email John graciously indicated that he would "probably" approve a name like "Strict Markdown" or "Pedantic Markdown". [...] We replied with the following suggestions:
Previous exposure to Jeff has made me read that as Unicorn Markdown.
- Compatible Markdown - Regular Markdown - Community Markdown - Common Markdown - Uniform Markdown - Vanilla MarkdownQuite frankly, I would have liked that name. It appropriately suggests "wishful thinking".
-
Someone suggested building a state machine specification, like HTML5 has.
Protip: that's called a “parser”.
-
I went and read the thread, but I don't want to create an account over there to throw in my thoughts.
I would have thought "its a problem in an external library, report it to them" is a bit of a shitty way to respond to a bug. It is extant in your own application. To draw a comparison from the general case of products bought in a shop, If I have a manufacturing fault in something I buy, my contract is with the shop, not the manufacturer, regardless of their own control over said fault. It's the shop's responsibility to sort it out.
I would put forth that bugs reported in this manner against discourse vs the standard should probably trickle up from you to the libraries you use. We are your customers, we come to you with a problem, you can't fix the problem in discourse code, but have a pre-existing relationship with those who can (if you are making money from their code, which is your plan, you should probably have a pre-existing relationship of some kind - contributary or monetary -, thats how OSS is supposed to work right?)
if you want there to be a standard (you do), and you are trying to pioneer a revolution in forum software (you are) then you should probably go ahead and pioneer the standard that you want all other forums to adopt.
-
I should note - I came here because of sexy, but I got involved in the drama instead, truly then, the drama can indeed be said to be "sexy drama"
Filed Under: I am become algorythmics, arbiter of sexy
-
Sexy Markdown?
-
I would have thought "its a problem in an external library, report it to them" is a bit of a shitty way to respond to a bug. It is extant in your own application.
Maybe. But at least acknowledging that it's in an external dependency, so it's not a simple fix would have been a major step up from passive aggressive moderation of the bug ticket.
-
true - but we all know from negotiating school that we should never start from the minimum we are willing to accept. start with the best possible outcome and then negotiate to an acceptable level for everyone.
unfortunately, I would argue that the minimum is a bit higher than "just don't be a dick about it", particularly since so many things are disregarded with a "submit a PR". Counter argument: Just submit a PR to your external library?
Filed Under: Fighting with boomzilla as if he is in charge because he is the only one responding to me
-
unfortunately, I would argue that the minimum is a bit higher than "just don't be a dick about it", particularly since so many things are disregarded with a "submit a PR". Counter argument: Just submit a PR to your external library?
Interestingly, evit trout was the guy who first engaged and said that they'd fix this at some point in the future, but it's not a priority. So, first impression: it's a Discothing. Once I learned it was external, I discovered that evil trout was a contributor, which maybe explains why he didn't think of it as totally external. Or something. Who knows.
Expectations are already pretty low. Even for something as embarrassing as the way ordered lists are handled.
-
The "resolution" was: "I personally don't use that page, so it's not a bug. CLOSED."
It still appears to be exactly as broken as when I reported it all those months back.
I've seen some shit on here, but that's just fucking ridiculous.
-
Expectations are already pretty low. Even for something as embarrassing as the way ordered lists are handled.
Jeff seems to be stepping up his A-game with these topic closures, and coming across badly in general.
They certainly don't care about how they are perceived by their users, but they really fucking should!
-
We are your customers, we come to you with a problem, you can't fix the problem in discourse code
Given that it's coming from a forum full of programmers, I'm okay with being asked if we can kindly file the bug over yonder, since we're presumably capable of doing so more so than the average customer. I'm generally amenable to making people's lives easier if I'm asked nicely.
It's the "asking nicely" that seems to be totally lost. I rarely post on meta.d but because I come from TDWTF I'm treated badly when I do. It's like some gang fight someplace. Nevermind that I also belong to several of the StackExchange family of gangs, which are presumably allied with the Discogang, I came over wearing TDWTF colours instead of SE colours and got knifed for my trouble.
-
I would have thought "its a problem in an external library, report it to them" is a bit of a shitty way to respond to a bug. It is extant in your own application. To draw a comparison from the general case of products bought in a shop, If I have a manufacturing fault in something I buy, my contract is with the shop, not the manufacturer, regardless of their own control over said fault. It's the shop's responsibility to sort it out.
Yeah, but sometimes it's a complete pain to get it sorted inside your app when it's a simple bug in the dependency. It's fine having a bug report that says that this a fault deep in code that you don't control, that you've reported it (with link to the upstream report) and that the whole lot is in some sort of parked state. That lets people downstream of you have a chance to work around it if necessary or to decide that it isn't critical.
It's great to have lots of abstractions that simplify things, except when the abstractions leak…
-
I should go dig up the email I sent yesterday, as an example... hang on...
I got this report:
From: [Team lead I don't report to]
Sent: Thursday, September 04, 2014 2:09 PM
To: [Me]
Subject: [new automated test runner for his Node.js product that plays nice with our existing runners for frontend javascript]Any idea why it is saying zero ms for tests?
[image snipped, basically showed all tests passing in 0ms]And I replied:
The output from nodeunit in junit format doesn’t list the time it took per test case. When run with console output, [the test] reports 1059ms to the console.
I filed a bug report against the library (https://github.com/caolan/nodeunit/issues/275), but they’re very cautious about introducing change so I suspect it won’t get answered for a bit. Is this important to you? I could try to fix it and issue a PR…
My hope is he doesn't really care, but if it's important for some reason, I can prioritize it with my other work, let him know where he is in the queue, and see if I have the js skills to fix it. I might, but I'll be sure to make no promises.
-
In my case, I've got a weird problem where the first thing to connect to my webapp sets the address that the webapp thinks it is running on for all time (or at least until we restart the container); run a quick test after deployment? Boom, stuck with localhost. Now, my code just delegates all that stuff straight to a framework, so something in there is causing the problem. No idea what (and the framework is damn complicated under the hood).
Fortunately, normal deployments run this all behind a portal and that can trivially hide all the nasty shit that's going on.
-
I discovered that evil trout was a contributor, which maybe explains why he didn't think of it as totally external
He's not a mere "contributor" to Discobiscuits. He's a co-founder of TCoCDCK.
-
He's not a mere "contributor" to Discobiscuits. He's a co-founder of TCoCDCK.
No, I meant contributor to the external markdown-js library. He's obviously a discodev.
-
Downfall is such a great movie, it really kind of cheese me off when people use it for these cheap jokes.
-
Downfall is such a great movie, it really kind of cheese me off when people use it for these cheap jokes.
I bet blind people are pissed off that one scene keeps getting posted for no reason.
But yeah, fantastic movie.
-
Ah, right. My bad.
Also, I think you pismelled "dickhead"
-
I think we are all agreed, and I think @Yamikuronue's example is exactly what I would expect from a dev in that situation, the problem is in an OSS dependency (we have links to it's github in the thread) and as I mentioned, there is a fondness for "Submit a PR" when we mention stuff with discourse. I am just thinking - why doesn't the same thing apply to this bug? Maybe it is being, but the way it's been handled over on meta certainly doesn't imply that.
-
I am just thinking - why doesn't the same thing apply to this bug? Maybe it is being, but the way it's been handled over on meta certainly doesn't imply that.
I think it probably does, or will. The original discoreplier has done it in the past, after all. What I find particularly amusing is that we quoted (effectively) Jeff to Jeff as to why he's wrong and he chose no NANANANANANANA CAN'T HEAR YOU.
-
Sexy Markdown?
I'd prefer something I little more to the point like Markdown-I-Would-Have-Sex-With or Markdown-That-I-Want-To-Have-His-Children or Markdown-I-Will-Devote-My-Life-To.
-
- Sexy markdown
-
Here's the new year's eve countdown!!!
hey they fixed it so a single number doesn't automatically get screwed!
And so if you have no text after the number you can actually get it to work!
- but
- if
- you
- try
- text......
-
no chance
-
afd
-
aklsdf
-
alksdf
This format for the win.
-
Well, html fubars it too, why no autofixing!?
10. [spoiler]eat it![/spoiler]
9. eat it!
8. eat it!
7. eat it!
6. eat it!
5. sfa
4. adf
3. asf
2. asf
1. Markdown you have been defeated.
-
DRAMA UPDATE DRAMA UPDATE DRAMA UPDATE
EDIT: new url: http://commonmark.org/
So, this is now called CommonMark.
Notice one thing: they are NOT using the word "Markdown" in the name of the spec. That was the main point of contention and the means through which the Markdown author Gruber held them by the balls.
Jeff and Co had at first tried to appease this guy, but now they feel somewhat bitter and are snipping back a bit.
The last bit is from the site. Those who visited the version 1.0 will remember the last question in the FAQ was an open suck up to Gruber through the Yankees baseball team, which he apparently loves. But now...
Are the Yankees the best team in baseball?
We used to think so.Aww, broken heart.
Hopefully, now they can move forward without having to suck up to this manchild holding the standard back.
-
UUh interesting. The site went down. Could be they are just switching domains? This is how it looked 10 minutes ago.
-
And on the http://standardmarkdown.com/ site right now:
This domain was disabled at the request of John Gruber.
HA! Good old passive-agressive Jeff. All it needs is a meme with a sad panda or something.
-
codinghorror
"At some point in the future, Markdown may support starting ordered lists at an arbitrary number." This was written in 2004.
2:10 PM - 5 Sep 2014FORK FORK FORK FORK FORK
-
UUh interesting. The site went down. Could be they are just switching domains? This is how it looked 10 minutes ago.
Looks pretty much like that to me now. I'm not doing a detailed diff here...but it's definitely getting served to me.
-
I bet he's IMing John Gruber right now and going, "hey buddy, hey pal, wanna play some Battlefield: Hardline? Buddy pal?"
-
Right now, the old site is "fuck you" to Gruber and the new link is 400 Bad Request.
Is this the worst specification launch in history or what? Everything that could go wrong went wrong.
-
Someone is bitter :-)
-
-
Hey, at least no one suggested a couple prefix words on top of that to make it match another acronym...
Filed under: But I did suddenly think it was slightly close to that acronym as soon as I saw @aliceif's post
-
Sexy Markdown Funtimes?
sorry, I'll get my coat.
-
That acronym (assuming we're both thinking of the same one) would be more appropriate for the experience of using markdown.
Filed under: It hurts so
goodmuch!
-
Except for all the wrong reasons.
-
Personally, I don't think there are any right reasons. I'm not into pain.
-
Fair enough. It isn't for everyone.
-
Apparently my godaddy hosting is promoting SMF.
I was confused.
-
Heh.
-
Badly Designed Sexy Markdown?
-
-
https://twitter.com/codinghorror/status/508027568839479297
Someone is bitter :-)
It's been renamed to
http://t.co/hdDpdUK5Ug
? Who came up with that name?