Response to comments by Issac Roth of StrongLoop

gistfile1.md

This is rather out of date and we've all moved past it. Me and Issac are cool now.

http://venturebeat.com/2013/09/17/is-corporate-open-source-an-oxymoron-not-where-node-is-concerned

"We believe that the right model for Node.js is a win for the community, a win for the project, and a win for the steward. Those who contribute to Node, use Node, or are evaluating Node can take comfort in knowing that we will never be the Red Hat of Node!"

@benzenwen that is a great quote :)

Thanks for publishing your thoughts @mikeal. Spot on.

Might be worth clarifying that it is TJ Fontaine working on node core at Joyent and not TJ Holowaychuk as many might get the two confused.

@sintaxi too many TJ's!

I work at StrongLoop, so I get to see things from behind the scenes. We definitely aren't trying to "steal" node from Joyent. In fact, we've been trying to partner with them from the very start.

The comment about us being the "largest corporate sponsor" is maybe a bit misunderstood. We have the largest commit activity to node core. What that really means is hard to say. We added or changed a bunch of lines, are they GOOD lines? We hope so.

We also have many initiatives to experiment with ideas on how to improve the future of node. This is userland stuff, where possible. Some aspects of node core, in it's current state, simply do not meet the requirements of some of our clients; so development in that area is required also. We hope what we change will be accepted back into node core.

As for our statement that Joyent "doesn't carry the project forward", I believe that is a legitimate concern, although their approach is completely valid. They got the platform to the state they needed to, and now are focused entirely on stability. That is good. Stability is very important, with enterprise wanting to get in on this. Our wording may have been a bit harsh there, in my opinion.

As long as they are focused on stability though, they aren't investing much into exploring possibilities for the future of node. They believe the community can run that for them. I mostly agree, but community AND corporate sponsorship seems even better, does it not? That's where we hope to help out.

I mostly agree, but community AND corporate sponsorship seems even better, does it not?

@Qard I think the attitude that there is no corporate sponsorship could use some adjusting.

@sintaxi I don't mean that there isn't corporate sponsorship currently. Just that there could be more.

I see nothing wrong with having more companies willing to put forward resources to contribute to node and ensure it has a good future, as long as they are virtuous. We all invest a lot of our time into node, we don't want to see it railroaded to conform to some corporate agenda. I can relate.

I think it's completely justified that people are wary of the StrongLoop announcement. We have a lot of involvement in node. We contribute a lot to node core, so we have a lot of work ahead of us to earn your trust. Ben and Bert have been very visible in the node community and have earned their trust, but the rest of us aren't so familiar.

Believe me, if StrongLoop somehow morphed into this horrible monster everyone fears, I'd be at the front of the lines with my pitchfork.

@Qard agreed. hoping we can all move forward united from this experience. I certainly don't see StrongLoop as a monster or anything like that but I think peoples worst fears surfaced momentarily based on perhaps taking some words out of context.

The health of a platform is reliant heavily on the health of the community itself. Im glad we have one that takes care of itself. Cheers.

Does the estimate on 'largest commit activity to node core' include commits that happened before the company existed?

@Qard thank you for your response.

I think I understand your position well because in the early 2000s I felt very much the same way and, in that environment, I think that position was reasonable and valid. But, again, this is the present, not the past, and open source is very different with differing incentives and investment.

As I've written about before, the Open Source of the 90s and early 2000s basically won a war for defining the foundational technology of the world and only lost the war for defining products, particularly consumer products.

The world after you win is very different. There is not a company that has been newly funded in the last 7 years that is not built on top of open source technology. A lot of this has to do with economics, free as in beer is free as in it doesn't burn my runway, but this also has to do with the fact that open source technology got easier to use and, web technology in particular, lead to mass amatuerization.

Then GitHub happened. The new open source landscape is so different from what it was that we literally have to throw the entire book away on how we organize, incentivize, and ultimately build open source technology. This is doubly so in the node.js community where we enjoy a vast number of tiny highly compatible components, very few vertical stacks, and relatively small core/stdlib.

In this landscape, I fail to see the value of increased corporate "sponsorship" in the way we used to have it. People should be paid to write open source when it is in the interest of their companies business goals to do so as a user/contributor. Enough people are using open source, really using it, that the investment of individuals on the dime of those that need the technology to work and to be better and more reliable is enough for those projects to succeed.

Projects that fail tend to be those that enjoy a disproportionate amount of monied investment for the sake of the technology rather than a product or service the technology addresses.

Joyent is a great user of node.js but, more importantly, they are currently host to even bigger users of node.js they actively support and their continued investment is in service of the products those companies produce which are bending node.js to its current breaking point. This is actually the best way for any company to invest in the project.

Paying people to dick around in open source at their own discretion is, in my opinion, a poisonous thing for the projects that investment is intended to improve. Investment should be motivated, either by individual contributors taking interest in their own time, or by the underlying business interests (usually a product) of a monied company.

On rare occasion an individual contributor might enjoy a salary with little direction and contribute in a "sponsored" manor to a project in the same way they would in their own time. Trevor Norris enjoys such a role at the moment, sponsored by Mozilla to work on node.js, and his deeply ingrained desire to improve performance has been a win for the project. At the same time, similar sponsorships like that of Larry Wall to work on Perl full time have been cited as reasons that the project became increasingly academic and divorced from its users.

But, even in these cases of "sponsored" contributions the direction and potential leadership of the corporation have been removed. The company is not playing a role in leading the project to in any way other than paying someone who is may or may not be leading at his/her own discretion.

Some of the worst open source projects, the last decades largest failures, have been those that were "lead" by corporations. In nearly all cases this is because that corporation was trying to productize the technology rather than a service or product enabled by that technology.

The biggest apprehension I have to both StrongLoop's comments and the implied motivation to "sponsor" the project is that an obvious attempt to co-opt the project is only worth this level of effort and PR if the project is node.js and if successful I have no doubt that it would destroy the project long term. I'm pessimistic that StrongLoop will succeed in its effort to co-opt node.js, or productize it, so I'm not worried enough to freak out, I'm only motivated enough to write a gist in response to miss-leading and borderline offensive comments by your CEO.

How about I handle this for all parties involved in the most democratic possible fashion...

I WILL TAKE OVER ALL THE NODES, ALL THE TIMES. NODE NODE.JS NODE[CLOUD]JS #NODEJS SLONODE

This message is only slightly spam, but closer to Spam Musubi.

Node takeover benefit all party, party with keg?

@voodootikigod i would expect that the person with the most robots would be the one to try and take us over by overt force.

well played.

We (me and the drones and bots) just let you all think you have control over the nodes...

youdont

+1 to @voodootikigod commencing the node singularity.

@mikeal Well, don't have to worry about my commits getting too "academic". That'd be pretty hard to do since I didn't even make it though the system. :P

But seriously, complexity and abstraction bother me. Probably because I'm dumb and it just confuses me. I'm hoping in v0.13 development we can build out a lower level API, like what I did with smalloc and what I'm currently doing with asyncListener, so no one feels the need to take Node in "another direction". Instead they'll have enough flexibility to create whatever crazy complex and abstract user-land modules they want. Allowing me to just make all the things faster.

My vendetta against all those who have said "Node isn't as fast as X" is still going strong, and I will win!

The fact that the message isn't just "we're a company that sponsors people that work on Node.js" and rather has superlatives like "biggest" and "most" makes me very uncomfortable. We are riding a movement that was started by Linus and Stallman and at no point should there be a bottom line lest they chase after us for royalties.

Why does there need to be a competition? Why even bring up Joyent? Is there not enough work to go around?

What promises were made to raise a 8M seed round? What are they going to profitize?

VC + open source contributions leaves me with a bad taste. The motives are no longer about altruism and contributing to a flourishing community. I don't like where this is headed, but I think the community will adjust appropriately (I hope).

As a member of @thenodefirm I receive emails daily from large cap multinational corporations, startups and indie developers all proclaiming their adoption or serious interest in Node. This is happening at a breakneck and unprecedented pace.

The value behind Node as a technology solution is not hype. The value of Node as recruiting differentiator is not understated. The ability to unlock creative potential in technologists with Node is real (see Arduino, Nodebots, Raspberry Pi, LG TVs, etc.).

As this rapid adoption and explosion of innovation happen at the same time it makes sense that companies will form and attempt to either profit, contribute or do both. This is actually healthy for the long term sustainability of the project, for the more businesses that are created around Node to support companies using Node, fuels the ecosystem(s) of technologists writing Node apps.

The inflection point happens, IMO, with the community at large. We have an immense amount of responsibility as good stewards of the community to continue to allow creative potential to be unlocked while simultaneously building and allowing to be built, sustainable and profitable businesses. It's a positive feedback loop.

This is paramount to our mission at The Node Firm and we are convinced that this model works, or is at least working, and we hope that more companies share in our vision. Community first.

@joemccann when we started The Node Firm what we always said of the mission was "to help companies be successful with node.js" and I think that still holds true. The Firm has never moved to own parts of the community or even to dictate them but always to gather the expertise necessary to help companies that lack the experience with node.js they need to succeed.

@mikeal agreed and the business has evolved and grown, we still share those core values as ultimately a means to an end to support the community at large. As more enterprises adopt Node, there are more jobs for Node developers. All boats rise.

@mikeal I think we're saying the same thing. =)

@joemccann we are definitely agreeing :) the mission is the same but the needs people have in order to be successful with node have changed over time and so the firm's offerings have needed to evolve. While I haven't had the time to be actively involved I have appreciated that evolution.

i hope @bnoordhuis and @piscisaureus are successful with what ever they do, they deserve it!

as long as there is an accessible git-repository with a cool LICENSE in it, everything is fine :D

My response is here: https://gist.github.com/piscisaureus/6707356

@mikeal I think of our relationship to node being rather close to Voxer.

We build a product (our mBaaS) that is completely tied to node. We're trying to sell this to big companies that are going to be pushing a lot of data. If node is performing badly, or we get irregular behavior, we need to be able to understand why, and fix it quickly. Employing Ben and Bert, who understand the internals of libuv and node very intimately, makes that possible. If node is performing poorly, a product on top of it is going to be a hard sell.

Let me stress that we are NOT trying to sell node. We are trying to sell our mBaaS, which happens to use node because we found it to be most suitable for the needs of the product. We are taking one particular common use-case of node, and trying to make it easier to build, monitor, and scale.

As for the "corporate sponsorship" statement, you're probably interpreting that a bit differently from how it really is in our case. We are not paid to sit around and make random stuff in node. We all have a whole pile of tasks that are critical to making our product competitive. Sometimes those tasks happen to be improving libuv or node itself to handle the capacity a customer requires.

@Qard sure, but the comments made by Roth don't echo the tone, content, or motivation your statements here have. I'm sure that we could talk this all out and come to an agreement but it wouldn't change the statements your CEO is making in public to the press.

@Qard Certainly the (frequently edited) StrongLoop website often contains somewhat misleading text, like:

(grabbed at Thu Sep 26 02:56:37 UTC 2013 from http://strongloop.com/node-republic/)

@nomdeveloper fixed

@mikeal @nomdeveloper I'm just a dev, so I don't have much say in the PR and wording of things. I agree that some of what has been said might smell a bit of corporate positioning, but we definitely are good intentioned. We're just a new startup, trying to get attention. Apparently it's working, albeit not in quite the way we intended.

When node started to get attention from big companies, Joyent would just bring Ryan into meetings with customers, sit him down, and say "we have that guy". We're basically doing the same thing with Bert and Ben. Perhaps that's a bit arrogant, but it's also true. Members of our team have made significant contributions to node. There's no denying that.

I'll be in SF this weekend mentoring at Startup Weekend, if you have any desire to speak in person. Feel free to reply here or DM me on twitter @stephenbelanger. Isaac will also be around, not sure what his schedule is though.

@mikeal Hi this is IssacR the person you're complaining about. :)

We could both be more precise in our commentary but I get the sense we both really care. Sorry for any harm done. I am adjusting and learning. Let's get together and talk it out. I live in SF so should be easy. I'll hit you on email.