Open Source Obligations

December 30th, 2009

One of the beautiful aspects of open source software development is that individual contributors are, generally speaking, under no obligation to contribute their work. People who find time on their hands, and an inspiration to do something great, write some code and share it with the world. End of story. Or at least, that’s the way I see it.

Jeff Atwood is pissed because John Gruber hasn’t been a “good mommy” to Markdown.

Disregarding whether the accusations are true or not, it pisses me off when somebody is criticized for giving something away, yet somehow not giving enough. What part of FREE don’t you understand? Somebody, in this case John Gruber, wrote something great, and gave it to you for free. It’s a technology you would not have access to without his generosity, and which you could not even obtain commercially, had he chosen to keep it private. A completely new, compelling solution which is available to you because one man had the community-serving idea that it should be open. Are you getting my drift?

I grew up in Santa Cruz, California. A city famous for its hippies, vagrants, and college students. (At some point in my youth I aspired to fit into each of these classes of people.) One day I was walking down the main downtown strip, Pacific Avenue, with a half-finished paper cup of coffee in my hands. A panhandler yelled out to me: “Can I have your coffee!?” Feeling generous and well-caffeinated, I surrendered my cup and strolled off. Moments later, the man shrieked after me: “There’s no whisky in this coffee!”

You can’t please everybody. In the open source community, and in the larger community that consumes open source goods, there will always be complainers. For some people, free simply isn’t good enough.

I believe that anybody who gives away the results of their hard labor for free should be praised. By no means should they be expected to contribute more than they already have, or to bend to the screeching whims of their consumers. If you don’t like all of what you’ve received, take what you do like, and modify it to make it perfect. Take the free cup of coffee, and add your own damned whisky.

50 Responses to “Open Source Obligations”

  1. bbum Says:

    Agreed in full and especially when something is given away under a truly free license such as the MIT or BSD license (of which, Markdown uses such a license).

    I’ve given away a lot of software over the years (under the MIT license generally).

    And I have had a small handful of folks ask me for bug fixes or support, receive a polite “sorry, no time and not a priority” and then get offended. My response is the same: “You get what you pay for. If you want me to care, pay me more than I’m getting paid to do what I’m doing now. Otherwise, seek other software.”

    Entitlement complexes are weird.

  2. Geoffrey Grosenbach Says:

    For the record, I paid good money for my copy of FastScripts and it didn’t come with any whisky either.

  3. Rafe Says:

    The dilemma in this case is that Markdown could use some tidying up, and people feel insecure about going forward without the blessing of its creator. I don’t think Atwood did a great job of making his point, but that’s really what it boils down to. It’s not about entitlement but about trying to respect John as the creator of Markdown while still trying to move Markdown forward.

  4. Danilo Campos Says:

    I’m going to agree with this on principle. Bill Bumgarner takes it a step further with this tweet:

    “@codinghorror Hell– Markdown is even under a BSD style license! The ultimate “Fork me if I’m not paying attention” license! Thanks @gruber!”

    (http://twitter.com/bbum/status/7186674484)

    It’s a fair point: if Gruber’s stewardship of Markdown were so truly awful as to be as untenable as Jeff says, the invisible hand of OSS would have produced a new, more authoritative fork, right?

    Indulge a moment of devil’s advocacy, though: Gruber’s extraordinary high standards for, well, everything, from usability to design to artistic expression are notorious. I would say they define him and his blog. As they should, since I find he’s very often right. With that in mind, is there not a touch of hypocrisy in reviling mediocrity everywhere — except in the presentation and progress of your own project?

    In the end, though, even this may be a Randian, non-contradictory contradiction. If we re-examine our premises, it may simply be that Gruber’s perfectionism abhors release of less than perfect product or collaboration with less-than-perfect developers. It’ll take as long as it takes and he won’t facilitate collaboration because it doesn’t fit his vision. In which case…

    We’re right back where we started from, with a guy giving away a chunk of his brain for free and someone else asking why he didn’t include a collectable jar to keep it in.

    Giving this one to Gruber.

  5. Ryan Says:

    Who’s uncomfortable? Last I checked there are several versions of Markdown and some drift off spec. Isn’t that the entire point of open source development?

  6. Andrew Shebanow Says:

    I agree with Ryan. Gruber may not have provided a lot of active leadership but so what? As long as there are supported implementations out there, the open source community is doing its job no matter how much stewardship Gruber is or isn’t providing.

    The sense of entitlement some people have about open source projects is truly amazing.

    To Atwood’s credit, though, at least he has open sourced his own C# markdown project.

  7. Jeff Atwood Says:

    > We’re right back where we started from, with a guy giving away a chunk of his brain for free and someone else asking why he didn’t include a collectable jar to keep it in.

    Specifically, I’m asking why there have been eight 1.0.2 beta releases by John, many of which fix some very serious bugs in Markdown, that have never seen the light of day — in over five years!

    > the invisible hand of OSS would have produced a new, more authoritative fork, right?

    Oh, has it ever — there are dozens of “versions” of Markdown and utter confusion as to which is which or even which is correct and authoritative. I’ll leave it as an exercise for the reader to determine how this might positively or negatively affect the future of Markdown.

    Hint: it ain’t the good one.

    > With that in mind, is there not a touch of hypocrisy in reviling mediocrity everywhere — except in the presentation and progress of your own project?

    The way to deal with this is to exert some leadership — to control the project and define it. In other words, don’t complain how your child is being raised in mediocrity by the state when you weren’t enough of a parent to contribute to properly raising that child yourself.

    > The dilemma in this case is that Markdown could use some tidying up, and people feel insecure about going forward without the blessing of its creator. It’s not about entitlement but about trying to respect John as the creator of Markdown while still trying to move Markdown forward.

    As Rafe said.

    What I find irritating about this is that it would be so, so easy for Gruber to step in and fix all this — to either publish his current bugfixes as 1.0.2 or bless another distribution as canon. It’d take like 2 hours, tops.

  8. Mike H. Says:

    It’s not about entitlement but about trying to respect John as the creator of Markdown while still trying to move Markdown forward.

    This.

    I thought Atwood’s post was respectful enough, and more respectful than many others I’ve seen in a long time around open source and free software.

    Markdown’s something I use a lot for myself, and I’d be perfectly comfortable maintaining a private fork or following someone else’s, but it’s also something I put in front of other people and recommend to less technical users on systems I don’t control, as I’d guess others do. That moves it out of the realm of something like a developer-facing library and into the territory of a user-facing standard.

    I’m not saying that to argue that Gruber now owes us W3C-like levels of shepherding, because he doesn’t *owe* us anything. On the other hand, letting 1,000 revolutionary schools of Markdown thought contend doesn’t seem like it would be healthy for Markdown as something anyone’s happy to see as an option on assorted CMS’s if it means missing or conflicting features depending on the implementation you stumble across.

    Atwood’s doing the responsible thing. If Gruber blows him off, he can say he tried then exercise his right to try to build a community around a more democratically maintained core specification.

  9. Michal Migurski Says:

    I don’t think the usual forking/derivation logic applies in this case, and I think that Jeff Atwood does have a solid point. Markdown isn’t an implementation of a protocol like Apache, it is itself a protocol like HTTP, and therefore subject to rules of social coordination and agreement different to those encoded in the GNU or BSD licenses. John Gruber is approaching this issue in a vague and deeply flawed way, disavowing responsibility for Markdown by claiming that he only ever built it for his own personal use while simultaneously complaining about alternative implementations in the wild.

    I feel Mark Atwood’s pain on this one, it’s only been two months since he had his ass absolutely raked over the coals for this post, where he complained that his users were too dumb or blind to figure out Markdown and use it correctly: http://www.codinghorror.com/blog/archives/001306.html

    The trouble with treating Markdown as just another open source free-for-all with competing implementations is that you have to re-learn the stupid thing on every website you visit. Who wants to do that? If Markdown is to be a proper spec, John Gruber needs to do one of two things: hand it off to someone who’s got the time and inclination to shepherd the community around it, or definitively call it finished like Douglas Crockford has done with JSON. Sadly, Markdown is a bit of a hash and has many more rough edges than something as perfectly spherical as JSON.

    My own feeling is that it’s not a battle worth fighting: Markdown is too dorky to be worth babying as a standard, which means that Jeff Atwood and the other chumps who adopt it as a semi-standard are going to continue to feel left out in the cold. There’s no there there! They should either bite the bullet and go with a real rich text editor, or (my preference) treat text as its own meaningful self-description and use something like CSS’s pre-wrap white-space as I describe here: http://mike.teczno.com/notes/comments-markdown.html

  10. Thomas Leonard Says:

    >>It’s not about entitlement but about trying to respect John as the creator of >>Markdown while still trying to move Markdown forward.

    >This.

    Not this.

    You want to know how you move markdown forward? You stop talking about it and start doing it. So what if you’re not ‘blessed as cannon.’ If your fork is good and worthwhile then the community will naturally follow and you’ll become the go-to, this might even put the pressure you want on Gruber to make your fork “official” should he continue to “ignore” his own damned project.

    Long story short: Gruber doesn’t owe anybody anything. Shut up and ship. Stop talking (or blogging) about coding and code.

  11. hawkman Says:

    First off, Daniel, I completely agree that noone owes anyone anything; but I’m still with Rafe and Mike H.

    Whether mister Gruber likes it or not, he’s still “in charge” of Markdown – largely because of his celebrity, and because to my knowledge he’s never officially disowned it. Every OSS project needs a lead, or it never moves forward. That lead might be one dude, or it might just be an IRC channel where stuff gets decided; but without one you’ve got lots of people pulling in different directions, and things become so fragmented they’re of no use in anything public-facing. Which is rather what Ryan said, although he seems to think it’s a strength. :)

  12. Augusto Says:

    Who cares about what some random Jeff blogger writes anyway? I’d never heard of him before, I’ll forget about him in 5 seconds. There.

    John Gruber, thanks for all the fish.

  13. corbin Says:

    I didn’t know you were from Santa Cruz! I grew up there too. I went to Aptos High and eventually UCSC.

    corbin

  14. Diederik Hoogenboom Says:

    As an Open Source ‘parent’ you still have some responsibility. You should at least make clear what the status of the project is and if you are not supporting it anymore, hand over its guardianship.

    I respect Gruber and I am too unfamiliar with the Markdown story to point the finger to him. Nevertheless, it looks like asking Open Source developers to take, even limited, responsibility is still taboo.

  15. Davide Says:

    The situation here is quite simple.
    Developer X builds something, and decides to give it away for free.
    Version 1.x works perfectly for him. He has never experienced a bug.
    Bugs pop out in different environments. Developer Y spots them and asks developer X to fix them. Now, why should developer X fix something that works for him? Out of kindnes?. Frankly, free time is better spent with friends, or pursuing what you’re after, rather than what a developer Y is after.
    Unless the ‘parenting code’ idea is a moral one; in that case we’re not talking logic, we’re talking religion, and religion is something you can or cannot subscribe to.

  16. Rob... Says:

    Davide,

    Just because you give something away for no monetary recompense doesn’t mean that you aren’t getting a reward. Usually this is “fame”. That is, the release of Markdown has raised John Gruber’s profile outside of the Mac community and, potentially, increased visits to his website from which he earns his living.

    Having said that, I agree that you can’t expect someone to work on code that they don’t want to if they feel that the reward isn’t big enough for the time invested.

    I’d argue that Open Source works at its best when the original developer puts in place systems that allow for the project to continue without them.

  17. Davide Says:

    Rob, I might agree with your fame argument, but the fame self regulates itself.
    If Gruber’s profile was really dependent on his markdown fame, and if the community really feels Markdown is poorly coded and/or maintained, well, he might get some monetary loss (bad reputation > less jobs > less money).
    Now the argument works, but please notice that there’s no open source ethics involved; It’s just the usual process: poor product, poor reviews, less work.
    What I cannot dig is Atwood’s theory that there’s a moral obligation to follow up things you gave away for free. It’s a present, take it or leave it. Obviously you’re also free to critique my present, although this also sounds a bit rude to me (“Don’t look a gift horse in the mouth”); more importantly, the license allows you do do whatever you want with my code, forking it is much more effective than complaining. I understand you might not have time to do it but why should I have time?

    Apart from the theoretical argument, the truth here is that markdown is excellent code, which works for the vast majority of his users; this incidentally explains also why Gruber hasn’t got that bad reputation so far.

    Cheers,
    Davide

  18. John C. Welch Says:

    This is the best example *ever* of why the “oh, you have the code, just fix it yourself” meme is kind of…B.S.

    People have the code. They’ve been fixing it themselves. Yet, because one guy isn’t waving his hand insouciantly at one to approve it, he sucks.

    Of course, if he did take the, what “two hours, tops” to approve one, that means he’s not looking at the code or doing any kind of quality review. That’s how long it would take him to set up the dartboard with the different versions on sticky notes and pick one at random. So, he picks one at random, because that’s all “Two hours, tops” gives him time for, and of course, that one will have problems too. Then he’ll suck for being so casual about it, with the accompanying hate mail and whiny blog posts. Those will take two hours, tops, to start up.

    Doing a quality review takes *time* and here’s one, maybe, just maybe, he doesn’t have that time, and it works fine for him. Markdown solves, has solved, and evidently will continue to solve the problem for which he created it.

    If Jeff Atwood is so dagum concerned about the stewardship of Markdown, he should man up and ask John if there’s a problem with him taking over said stewardship.

    What’s the worst that can happen? *Nothing*, it’s open source. What’s Gruber going to do, undo that? I don’t think so. About all Gruber could do is write angry emails and whiny blog posts about it. Big deal.

    But no one will actually do that, because as it turns out, it’s not a “two hours, tops” commitment, which is why Jeff is writing blog posts and blog comments about it, instead of stepping up and actually doing something.

    Hail the Open Source Community.

  19. Richard Says:

    The underlying problem here is that if someone “manned up” and took this project over he/she would run the risk of being “jackass”/snarked by Gruber and given the number of people who drool over Gruber’s every word, that’s a risk few might want to take.

    By the way, this paranoia about Gruber’s blessing (or not) can be seen in this very comment thread.

    I’ve been reading DF for many, many years but when John started calling people “jackasses” I stopped buying t-shirts and supporting the site. I still read him because he’s a clear-headed thinker and a good writer but the name calling makes me cringe.

    He’d have just as good a brand (probably better) if he left out the “jackassery.” You can call out folks who are wrong without calling them names.

    Anyway, I think the name calling is a piece of why few are willing to touch Markdown in a public way.

  20. Nikolaos Dimopoulos Says:

    > – What I find irritating about this is that it would be so, so easy for Gruber to step in and fix all this — to either publish his current bugfixes as 1.0.2 or bless another distribution as canon. It’d take like 2 hours, tops.

    I disagree with the above. Although it might take Mr. Gruber 2 hours tops to fix what is broken, one cannot assume that Mr. Gruber is obliged in any way to do that and devote 2 hours or less of his time.

    > – Mike H. Says: Atwood’s doing the responsible thing. If Gruber blows him off, he can say he tried then exercise his right to try to build a community around a more democratically maintained core specification.

    Totally agree

    > – What I cannot dig is Atwood’s theory that there’s a moral obligation to follow up things you gave away for free. It’s a present, take it or leave it. Obviously you’re also free to critique my present, although this also sounds a bit rude to me (“Don’t look a gift horse in the mouth”); more importantly, the license allows you do do whatever you want with my code, forking it is much more effective than complaining. I understand you might not have time to do it but why should I have time?

    I believe the above sums it up pretty well.

    Concluding, I believe the underlying issue here is people’s attitudes in general towards open source software. Some – and thankfully not the majority – believe that the original developer that gave the x piece of code for free is obligated morally or otherwise to maintain it forever. Sad but a reality nonetheless.

    /0.02 USD

  21. John C. Welch Says:

    “Gruber might call them names”

    I believe a poem I learned in kindergarten sums up the effectiveness of that threat:

    “Sticks and Stones may break my bones, but names can never harm me.”

    Is the open source community that…lacking in spine that the mere threat of a *blogger* calling them *names* leaves them cowering under their beds crying for mommy? Is this seriously a reason to not do something? “He might say mean things about me”?

    Seriously? The open source community is an oversensitive six-year-old?

  22. goundoulf Says:

    There are two different things:

    - publishing a software under an open-source license

    - having an open development model

  23. Warren Says:

    Let Jeff Atwood make his own fork. For pity’s sake, rename the thing. The project is dead (aka stable) if it hasn’t had a release in 5 years.

    Whine whine whine. It’s open source. Just fix it. Instead of blogging about code as if it was a child. WTF?

    W

  24. Jeff LaMarche Says:

    I find this really strange. I keep looking for the piece of the puzzle I’m missing. Jeff Atwood is one of the few bloggers outside of the Mac/iPhone dev world whose blog I consistently read. Although I often disagree with him, he’s always struck me as intelligent and fair-minded.

    But unless I’m completely missing something, I’m in total agreement with Daniel here. Having released and eventually abandoned more than one project to the Open Source world in my day, I have seen this same attitude many times, but usually it comes from people without an understanding of what goes into software development. It’s primarily people who don’t necessarily realize just what they’re asking for.

    But that’s not Jeff. He’s been involved in software development far too long to claim that kind of ignorance. It seems to me that there’s go to be more to this story that we’re not seeing.

    As for comments about people not forking Markdown because they’re scared of Gruber’s popularity, that’s a stupid argument. I’ve been fireball’ed on a few occasions, both when John liked something I wrote and when he really hated and disagreed with something I wrote. I survived both and, in fact, saw huge increases in my page hits and advertising revenue both times. It’s like the old saw about “any publicity is good publicity”. A newly forked version of Markdown could only benefit from getting mentioned on John’s blog.

  25. Bill Brown Says:

    As several have said, the problem is mostly that conflicting implementations dilute the viability of the brand. If you advertise to users that they can use Markdown in some text area, they *might* come with some predetermined notions of what that means. If they come from Site A that had one style of Markdown and where they participated heavily to Site B that has another, they’re going to be frustrated when their carefully crafted text ends up completely borked.

    We see that scenario play out perfectly with WikiMarkup, which means very different things depending on the underlying Wiki software. What Jeff wants is for Markdown to not end up like that, where his users’ expectations are all over the map. I gather that he’d like something more along the lines of BBCode, which, in my experience, is fairly standardized.

    When he exhorts Gruber to put up 1.0.2 and thereby bless it as canon, he’s arguing for something of a definitive reference. I can’t blame him, but I wouldn’t put any duty on Gruber either. In the end, perhaps Markdown isn’t the solution for Jeff’s problems or he’ll have to live with the ambiguity.

  26. Michel Fortin Says:

    My own reaction was worth a blog post: Markdown Sustainability.

  27. jonathan Says:

    @goundoulf
    +1

  28. Michael Langford Says:

    John G. Has a history of “going after” forks in flame fests (from what I’ve read), and that was what they were avoiding…

  29. John C. Welch Says:

    So what.

    So Gruber yells at you. Big deal. What, he’s going to take away your birthday? Make you give back your Xmas presents?

    No. The worst he can do is grump at you, and if he does, grump back at him, and point out that if he’s not going to be an active leader for Markdown development, then he can just deal with someone else stepping up.

    Sheesh, how did some of you survive childhood or high school if someone calling you names on a blog paralyzes you so.

  30. Colin Mattson Says:

    “As several have said, the problem is mostly that conflicting implementations dilute the viability of the brand.”

    How so? Gruber’s got an entire spec for Markdown, including edge cases, available to freely read.

    You can have ten thousand *implementations*—if they all follow the damn spec, the core functionality is always going to work (a) properly and (b) identically.

    If some idiot then goes and fundamentally breaks Markdown in his own implementation, that’s not “diluting the viability of the brand” or John Gruber being a bad den mother or anything else; it’s one idiot failing to actually make a *Markdown* parser.

    HTML isn’t deeply flawed because Microsoft can’t be bothered to implement it properly.

  31. Glenn Fleishman Says:

    So I’ll offer a different interpretation, since I use Markdown constantly, and count Gruber as an acquaintance I like.

    Markdown comes with the baggage of both immense respect for John (from some quarters) and immense hatred (from other, I believe fewer quarters).

    Those who respect John would prefer that his occasional comments that he’s going to release an updated 1.x spec or a 2.0 version would rather wait to see what he does, since he’s the daddy of the spec, and there’s not enough wrong with it for most of us to not want to wait.

    Those who hate John apparently don’t have enough moxie to create a github repository or Google Wave or whatever the hell, and start writing a revised spec, naming it GruberBugger or whatever, fork it, and fork off. It’s possible they would come up with some good, we’d coalesce around it, and there would be both Markdown 1 and GruberBugger 1 support in various products. Or Gruber, who is not known to ignore good ideas, could fold in GruberBuggers improvements into 2.0. Or whatever. Whatever!

    As it stands, we lightly forked Markdown internally at TidBITS, adding what we call “lazylinks,” in which a [linked text][reference] style link can have [*] instead of reference, and then the next [*]: URL is used. Doing this doesn’t break Markdown, and it’s made it more useful to us.

    I asked Gruber 3 or 4 years ago to consider adding lazylinks to Markdown, and he said he would think about it. Do you see tears on my face today?

  32. Glenn Fleishman Says:

    Colin, I’ve got it figured out. You wrote: “You can have ten thousand *implementations*—if they all follow the damn spec, the core functionality is always going to work (a) properly and (b) identically.”

    Right, but only Gruber’s is being blessed for inclusion in Movable Type and other platforms. So what the folks complaining want is their changes (or any improvements) implementing in Markdown which becomes an official spec which is then updated in various Markdown-supported applications.

    That’s where Gruber does have a kind of lock, and rightly so. The platforms trust Gruber’s version, and since no one has forked anything in a broad way, there’s little interest in forked support for a Markdown variant.

  33. Watts Says:

    Hm. I think Atwood’s basic point — to rephrase it — is that it’s good to have a canonical reference implementation of something like Markdown out there, and that it’s a little peculiar that Gruber hasn’t maintained his implementation to be that. This makes sense.

    However, John Welch is also right on target: “if [Gruber]‘s not going to be an active leader for Markdown development, then he can just deal with someone else stepping up.” For Pete’s sake, we’ve seen the de facto reference implementation of *X11* change from XFree86 to X.org a few years ago and the world didn’t end — I’m pretty sure the world can also handle the de facto reference implementation of Markdown ending up being Michel Fortin’s, Fletcher Penney’s or someone else’s.

  34. warp Says:

    > Concluding, I believe the underlying issue here is people’s attitudes in general towards open source software. Some – and thankfully not the majority – believe that the original developer that gave the x piece of code for free is obligated morally or otherwise to maintain it forever. Sad but a reality nonetheless.

    It’s not about maintaining it forever, but if you know you’re no longer interested in maintaining it — or if you don’t have the time — or whatever the reason, allow someone else to step in and take over from you.

    I’ve been in this situation, it doesn’t take 2 hours to solve it. I just updated the download page with the simple message that my software was no longer supported. About one or two years later someone came along and offered to take over, I gave them my blessing. The software is now much better than It ever would have been under my guidance.

  35. Michel Fortin Says:

    Wrap, there’s a big difference here: the biggest problem isn’t that the software is unmaintained, it’s that the language is unmaintained. For someone else to successfully take over he’ll need to make consensus among the implementers.

  36. Jessica Allan Says:

    I have to say here that I agree with Atwood on the whole — if you can’t continue giving to an open source project — and there are quite a few people who start out dabbling in open source and never finish — you should leave the project in a state that a new maintainer can resuscitate it and re-established the shared code and patch it and the like.

    On the other hand, what you said is equally valid. If it’s free, you don’t get to demand how it’s presented. But it’s good etiquette, I think, to make it possible to revive an open source project and share code.

  37. Kevin Walzer Says:

    I have to agree with those who say, “If you don’t like it, fork it.” I’ve written lots of open-source code, mostly for my own use. Some of it has been used by other developers, some of whom have requested bug fixes, and/or some of whom have offered patches. If it makes sense for my use of the software to accept the patch or fix the bug, I’l do so; otherwise I won’t, but others are free to do so with my blessing.

  38. Jeff Atwood Says:

    Hell, I’d just be happy if Gruber would do something — *ANYTHING*.

    At some point (and I would argue that we have reached that point), the lack of *any* kind of leadership is actively harming the future of Markdown.

    I don’t buy the too much effort argument. How long would it take Gruber to:

    - publish his current, already written “secret” bugfixes as an official 1.0.2

    - update his Markdown page to indicate this is the final release

    - bless another distribution as official and worthy of the “Markdown” name so the standard can be taken forward without him

    That is — and I’m being generous here — at most 8 hours of effort. TOPS.

    I do not think asking someone to spend 8 hours over the last *five years*, on a project *they themselves created*, is such an onerous request.

  39. Daniel Jalkut Says:

    Jeff – I see your point – and I empathize from the start in the sense that I appreciate how frustrating it is when something you love and admire is not being managed in a way that you see fit.

    But your efforts to reduce this to a “N hour job” are overlooking the fact that it would only take N hours to “relinquish and move on.” It’s possible that John doesn’t want to relinquish ownership, even if he is a “bad steward.” What you’re asking is more than an “N hour job”, you’re asking him to decide to give up control, AND to go through the steps to see that through.

    I have found the comments here very interesting, and I definitely appreciate the minority of comments that have helped clarify what I think your position is. Basically, you’d like to see John take some action: either embrace his role as leader, or officially relinquish it.

    While I can his choosing one of those options and following through with it would be a relief to you and many others, it still shouldn’t be required or expected of him.

    Based on your stated position, I think you could get some good momentum if you simply started a fork and called it “MarkdownUnity” or something. You can take the 2-8 hour job and go forward with it. That’s the best protest I can think of.

    By the way, have you see this post from John earlier this year:

    http://daringfireball.net/linked/2009/10/23/github-flavored-markdown

    So, it would seem you have your hint as to which flavor of enhanced Markdown John tends to favor.

  40. Glenn Fleishman Says:

    Daniel wrote that Jeff was probably saying, “Basically, you’d like to see John take some action: either embrace his role as leader, or officially relinquish it.”

    My rejoinder would be that the whole point of licensing something so that the world can use it is that if there’s enough of a demand for something different to emerge, it forks and moves on.

    Markdown isn’t unique, but it’s rare in that it’s a description of a thing, not the implementation of it, and people refer to it as if it’s an implementation. If John had released Markdown in several module formats (perl, PHP, Ruby, JavaScript, etc.) as canonical modify-and-distribute (or whatever) licenses along with a written spec, then perhaps those modules would have forked.

    But he didn’t. There are some quasi-canonical and anointed libraries in repositories, but I don’t know that the major implementations of Markdown (in Movable Type, say) rely on these libraries or libraries that were written in house.

    Thus those that want Markdown evolution also want Markdown-as-implemented to be changed everywhere (as I wrote above) to be updated.

    If it were a library issue, Movable Type could just support Markdown-1.2.8.pl or whatever or allow drop-in substitution Markdown-plus-1.0.pl. Right?

  41. John C. Welch Says:

    Ah, so much for the myth of “source code sets you free”. The source code is out there, it’s under an open license, and yet, everyone complaining the loudest just sits on their hands, weeping because Gruber won’t do anything they wish, yet too timid to do it themselves.

    It must be truly awesome to have the power Gruber does. By doing nothing, he roadblocks every programmer on the planet, forcing them to wait for his approval. That’s pretty cool…to have that kind of power.

  42. Jeff Atwood Says:

    > It must be truly awesome to have the power Gruber does. By doing nothing, he roadblocks every programmer on the planet, forcing them to wait for his approval. That’s pretty cool…to have that kind of power.

    So I can come over to your house and tell you how to raise your child, because you’ve clearly abandoned it? You’re totally cool with that?

    Let me see, what’s the #1 Google link for Markdown.. the one I have no control over updating. The same one it’d take Gruber all of two hours (and, again, that’s being generous) to update.

  43. Drew Says:

    Love how people forget John’s whole deal is being a flame artist. I’m a huge fan but lets not tell fibs. Anyone who trusts the “spec” for markdown he released 5 years ago enough to use it on their site deserves the burn marks they get. If Gruber doesn’t care about his baby to feed it or give it up then it dies. If I were Atwood (ie I had skin in this) I’d put out some feelers to the other various non-spec Markdown implementations out there and see if they can’t reach an accord on a draft 1.0 spec under a new name. How about Markover?, Textup? Anything but Gruberbuger since the whole point is to give this thing a future.

  44. Danilo Campos Says:

    John C. Welch Says: “Ah, so much for the myth of “source code sets you free”. The source code is out there, it’s under an open license, and yet, everyone complaining the loudest just sits on their hands, weeping because Gruber won’t do anything they wish, yet too timid to do it themselves.

    It must be truly awesome to have the power Gruber does. By doing nothing, he roadblocks every programmer on the planet, forcing them to wait for his approval. That’s pretty cool…to have that kind of power.”

    I laughed so hard at this.

    I think an important lesson from all of this, regardless of where you come down on the main subject, is that Gruber somehow scares the shit out of people. I mean, there he is, with his two-tone blog, writing with casual authority on whatever comes into his head to bitch about or admire, sometimes venturing out into the greater world to bitch some more. Perhaps he can’t roadblock every programmer on the planet, but he seems to have anyone interested in taking Markdown in new directions entirely in his thrall.

    Clearly his force of personality, whether or not you buy the full extent of the no forking excuses, has had some effect on the direction of Markdown. There’s no replicating Gruber, and I can’t begin to guess at his motivations, but if you want see a possible case study in the auteur theory of open source software leadership, there he is.

  45. Danilo Campos Says:

    Jeff Atwood Says: “Let me see, what’s the #1 Google link for Markdown.. the one I have no control over updating. The same one it’d take Gruber all of two hours (and, again, that’s being generous) to update.”

    Come on, man, you run one of the best developer sites in the history of the world. You know how this works. Search engine results aren’t immutable. They’re a meritocracy. If Markdown were truly desperate for new life, wherever that life took hold, the links would point accordingly.

  46. Glenn Fleishman Says:

    Geebus Christmas. This has devolved into whining.

    No one can tell John Gruber what to do (not even if his wife, if you follow them on Twitter).

    Someone needs to man- or woman-up and just take a goddamn step in making a fork that does what people want.

    Anyone who doesn’t take that step and continues to whine about the inability to set Gruber’s agenda should switch from Markdown to something else.

  47. Hyperbolist Says:

    John Gruber praised [GitHub Flavored Markdown][1] back in October. The three main additional features (autolinking, escaping intra-word underscores and respecting line endings) sound like a reasonable beginning for a Markdown For Comments spec.

    [1]: http://daringfireball.net/linked/2009/10/23/github-flavored-markdown

  48. John C. Welch Says:

    “So I can come over to your house and tell you how to raise your child, because you’ve clearly abandoned it? You’re totally cool with that?”

    Yes. Because *source code* and *children* are *exactly the same*. Oh, wait, no they aren’t, that’s an inane comparison. And now it’s GOOGLE JUICE? Gruber might yell at you and GOOGLE JUICE?

    Oh come on, just admit that you never were going to do any of the work on Markdown, and are just sitting there whining at Gruber to do something you’re not going to do yourself, because while your time has infinite value, his time, not being your time, has zero value.

    It’d be more honest than the litany of silly excuses. I have a child, and he comes up with better bad excuses than those. Takes him two hours, tops.

  49. scott Says:

    Bill Brown said:

    “As several have said, the problem is mostly that conflicting implementations dilute the viability of the brand. If you advertise to users that they can use Markdown in some text area, they *might* come with some predetermined notions of what that means. If they come from Site A that had one style of Markdown and where they participated heavily to Site B that has another, they’re going to be frustrated when their carefully crafted text ends up completely borked.”

    As far as I can tell, the potential of dilution of brand viability hasn’t stopped open source forks in the past. Typically there’s more active forks in any given open source project than at the local buffet on “seniors eat free” day.

    Is it a valid concern? Yes. Has it stopped a ton of other forked projects in the past? No. I therefore have a hard time believing it’s the reason for avoiding just forking it and moving on.

  50. simon Says:

    So why don’t you all use multimarkdown, which is being actively developed, does everything markdown does plus some nifty extra stuff?

Comments are closed.

Follow the Conversation

Stay up-to-date by subscribing to the Comments RSS Feed for this entry.