- Analysts are not employed by SDxCentral.
- Views and opinions expressed in analyst content belong solely to the author and do not reflect the views of SDxCentral, LLC.
- SDxCentral does not fact check analyst content. If you believe there is a factual error in analyst content, please notify email@example.com. Should we find factual irregularities, that article will be unpublished from the SDxCentral website.
Effective April 18th, 2019, the SDxCentral analyst blog syndication program has been terminated.
SDxCentral Statement about AvidThink, LLC
- Roy Chua, the founder of AvidThink, was a co-founder of SDxCentral. As of September 30, 2018, Roy is no longer affiliated with SDxCentral.
- The views expressed by AvidThink and Roy Chua are independent of SDxCentral and do not represent the views or journalistic principles of SDxCentral.
- As of April 18th, 2019, SDxCentral is no longer publishing AvidThink analyst blogs on the SDxCentral website.
Late last year, at the ONF Connect event in Santa Clara, California, I had the pleasure of moderating a panel on open source in telecommunications with some of the thought leaders in the space, including David Bainbridge from Ciena, Bill Carter from OCP, Sangho Shin from SK Telecom, and Bill Snow of the ONF. We discussed the building of open source communities and the different open source models.
Last week, I attended a launch event held by QCT, Intel, and many of their ecosystem partners on their new edge products targeted at the next-generation central office (NGCO). There I was able to hear presentations around open source from the Linux Foundation and from Timon Sloane of the ONF. (As an aside, I’ll try to explain the differences between NGCO, VCO, and CORD another day.)
At both events, the topic of curated open source came up. Some audience members whom I chatted with after the QCT event brought up the same discussion points that we had addressed during the ONF Connect panel. Their concerns revolve around whether curated open source is still open source.
At the ONF Connect panel last month, I was a little surprised at the limited number of raised hands in the audience when I asked who had read Eric Raymond’s essay, “The Cathedral and the Bazaar,” also known as “CatB” and first published about 20 years ago. Perhaps I had many shy members in the audience, but given that many of us are operating in the open source world, it was strange to me that a minority of the audience were familiar with Raymond’s work. Regardless, for those readers who haven’t yet had a chance to check out the essay, I’d encourage you to do so.
Essentially, the essay describes two types of software development philosophy. The “cathedral” represents a closed source model where developers sit in their ivory tower, building an application hidden from the world and unveil it when ready. While in the “bazaar,” which represents the open source model, everything is visible, transparent, and much more free-flowing. Not quite a free-for-all, but somewhat akin to that. The implication back then was that Microsoft represented the cathedral while Linux was the flag-bearer of the Bazaar.
In the 20 years since, much has changed. Microsoft has embraced products of the bazaar, especially on its fast-growing Microsoft Azure platform, and development philosophies have gotten more hybrid. Developers and businesses have become more intelligent about how to incorporate open source into their models when appropriate and benefit from that. Open source libraries and software pervade cloud computing services – Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP) are all powered by open source – and appear in numerous commercial products as well. Just about every enterprise software maker targeted at data center installs leverages Linux and other open source components: Oracle, IBM, Microsoft, SAP, and many others.
And with this as a backdrop, when the term “curated” open source comes up some in the telco community worry that it’s not true open source. Instead of trying to build consensus within the community and have projects bubble up then promoted when they reach certain critical mass – something more aligned with the Linux Foundation and Apache Foundation models – the ONF curated model sees their carrier partners select certain projects that ONF seeds with dedicated engineering. They pick select go-to-market ecosystem partners and open up the source only after a certain milestone is reached. The philosophy here is that even open source projects need to aggregate critical mass before the community will join and contribute.
I had jokingly referred to this curated model as the Whole Foods of open source: not quite a bazaar, but one in which the prettier and better apples are picked for you and displayed in a comfortable setting. Now, you might end up paying a higher price for those apples, and I’m not sure how best to apply the higher-price analogy to curated open source. Perhaps the price here is that consumers have less ability to participate and influence the project roadmap. Though I want to point out the management of the Linux kernel wasn’t entirely democratic either. For the longest time, Linus was the benevolent autocrat who decided what went in. The kernel could have been chaotic otherwise.
In any case, my view is that whether curated open source is open source is a red herring. When I hear critique about this approach it has to do more with the openness of governance and control of the roadmap. In some sense while open source is open in terms of everyone’s ability to pull down a copy of it and use it, the real value for vendors and service providers in our telco ecosystem is when it can be leveraged for business. And the quid pro quo of open source in our world is that if you upstream and contribute your changes back, you get to benefit from others’ contributions and save your development dollars, plus a whole host of other benefits that I won’t go into here, but think recruiting and retention of talent.
However, if a business or service provider feels that their upstream contributions will not be accepted, or that they have limited influence over the roadmap, or that the governance model isn’t based on meritocracy – it seldom is in real life – then they disengage. And they have every right to take their toys to whichever sandbox they wish to play in, or build a new sandbox — it’s a free market, after all.
The main danger for our community is that we end up with fragmented efforts carrying insufficient momentum and the “let a hundred flowers blossom” analogy ends up with lots of dead flowers because we simply don’t have enough water, fertilizer, and sun to grow them all.
Again, that may be just fine because open source should embody Darwinism. And we can accelerate the process by harmonizing them as the Linux Foundation did on LFN projects and will continue to have to do so – culling isn’t necessarily bad.
Longer term, instead of looking at Darwinism at the open source project level, we should be looking at it one level higher – at the organization level. We should watch and see which governance and business model yields greater success over time: ONF versus Apache Foundation versus Linux Foundation. The reality, though, is that each organization has slightly different business goals so they might all attain or claim success at the end of the day. And to that, I say, to each its own.