[Conference 2018] Making Open Source Work

rmehta · September 13, 2018, 6:36am

One of the biggest challenges for open source software today is sustainability. Everyone knows software is free but the time put in by contributors is not. A truly “resilient” open source community needs to move away from “voluntary” contributors and to a more sustainable model where there is guarantee that the product grows and prospers independent of volunteers. For this there are three models:

Foundation Model - Work is supported by a foundation
Services Supported - Work is supported by profit making companies, who give direct services on the project.
Cross Subsidy - Work is supported by profit making (but loosely related products), like React/Facebook, Chromium/Google, who do not directly monetize the open source offering. (This is somewhere between 1 and 2, but depends on a very profitable sponsor company).

Currently we are in Model #2. Primary contributions come from Frappe Technologies that is supported by services around ERPNext. Thanks to ERPNext’s popularity, there are many other service providers too, but contributions are not as many.

Very few contribute, as they see this as an additional overhead, and this is 100% legitimate use of Open Source. No one is “required” to contribute anything. A few service providers also have private modules that they sell on top of Open Source ERPNext. This is questionable practice, but there is no real way to stop this either.

Moreover, the incentive of service providers is for end-users not to directly consume ERPNext, thereby creating an survival crisis for them. We have to accept this is the way of society.

This brings us to the only true way to sustain open source → “The Foundation Model”. If we as a community are able to raise a reasonable amount, say $500k to start with, we can ensure that the product remains 100% open source. For this we expect the community to contribute directly and also help us get grants from other funds who can support this work.

During this conference, let us keep this the focus of our discussions. How can we survive as a 100% donation / grant funded organisation and how we can be truly independent from commercial interests that will, for legitimate short term gains, threaten the long term open source nature of the product.

As key contributor and maintainer of the product, I am interested in keeping ERPNext 100% open source. Lets make this happen!

arikfr · September 13, 2018, 7:54am

Hi

I’m the creator of Redash, the open source project and the founder of the Redash company. I started the company 2.5 years ago in order to make sure Redash has a sustainable future.

Due to similar goals and challenges I’m following ERPNext/Frappe from a far. This time I figured I might join the conversation.

When I started the company that will support Redash I also thought about the foundation model. But I have hard time to see this as a truly sustainable model in the long term, as you are at the “mercy” of donations givers. Also, I think that as both products are business software that creates massive value to its users, I don’t see it right that it should depend on donations.

Something that you don’t mention explicitly but might be implied by Model #2 – is the SaaS model. Where you provide a hosted version of your product. It’s a bit different to services, which usually means professional services. This is the model Redash currently employs to be sustainable.

Another thing I learned is that most people don’t have the same context as you have on your product/business, so usually their opinion and advice are worthless. The same applies to my thoughts here – obviously I don’t have the same context as you have on ERPNext’s position, community, business situation, etc.

I would love to discuss this further to get more context and maybe collaborate on ways to make both projects more sustainable. We can continue the conversation here or maybe on call?

Thanks and good luck!
Arik

rmehta · September 13, 2018, 8:09am

@arikfr thanks for writing in! Redash is awesome

Yes you make a legit point about SAAS, some things are different for ERP though.

Unlike developer tools, ERP is a business tool, hence a company wanting to use an ERP will go via a “partner” rather than go self. A small % do go self (there are close to 600 paid customers on erpnext.com)
ERPs are hard products to deploy and get started unlike point solutions like SAAS like CRM or Helpdesk, that are SAAS friendly.
ERPs are mission critical, most companies would prefer hosting their own instances if it is reasonably easy to do so (and it is!)
The ERP ecosystem is well established with many players and service providers, and most companies deploy this via a service provider, who will host it for the end user.

Also there no real “moat” we have around erpnext.com SAAS unless we have premium features (which will not make it 100% open source)

I guess the scale of the product means that there it will be very hard for someone to successfully defend SAAS and be 100% open source. Pick most other Open Source Business tools like Odoo, SugarCRM etc, most of them are not 100% open source.

Lets do a call. Will PM you

Edit:

Some more after thoughts:

Foundations work only after a scale (Linux, Mozilla, Wikipedia), but they can be very successful as well.
A lot of service providers make money on ERPNext, but not all give back, so it feels unfair for Frappe to do all the charity, while others make money.

jai_kejriwal · September 13, 2018, 11:54am

As a purely ‘business’ problem, two things need to happen. First, Frappe needs to make more money (more customers) and to me this will drive sustainability in the short to medium term and second, we need to keep growing the foundation which will likely be a slower process but snowball once there is critical mass. The foundation should also look at potentially revenue generating options like the hub. I too am extremely vested in ensuring that ErpNext remains sustainable so looking forward to hearing from people who are running successful businesses about their ideas.

Pawan · September 13, 2018, 12:57pm

Some random thoughts/ideas

To raise more funds, the foundation should classify incoming funds into ‘donations’ (voluntary contributions which can be used by foundation as deemed fit by them) and ‘grants’ (funds for a particular project eg: UI improvement, specific feature building, hub etc.)
This will give funds clear purpose, which today is absent.
I think service providers should morally pledge certain % of their earnings to the foundation as ‘donation’ to the foundation.
Frappe should start a pure hosting plan at a very low cost, there should be just standard email support given with just latest stable version installed with certain no of users and nothing else. No upgrades, no on call support etc. This could help increase adoption and move users away from self-hosting. Whatever revenue comes from this plan could go to the foundation.
Foundation should engage more with educational institutes, government organizations to raise funds by working with them.

dhananjay · September 13, 2018, 1:40pm

Some ideas for fund raising, to help the Foundation Model:

Merchandise Sale
Foundation membership fee based on the revenue generate by ERPNext related business for Service Providers
Donation campaigns targeted towards the end-user(Admins or business owners), like Mozilla, Wikipedia

snv · September 13, 2018, 2:19pm

I like the landing page of Elementary OS.

The default option provided by them is “Purchase Elementary OS”.
It’s pay what you want, but to some extent, I think it makes the user think about the value he/she is going to derive from the software.

If the user decides to pay nothing, he/she has to enter 0$ in the custom field.

Maybe, something similar can be done on erpnext.org, where bench install instructions may be migrated.

rpol · September 13, 2018, 11:59pm

Hello, I recently started to follow this great project and as a service provider, i’m considering to make some money out of it, by customizing it and selling it to customers. It seems that a lot have changed since you first wrote this article: How to Build A Sustainable ERPNext Business, where i could interpret that there’s nothing wrong with selling the software. As a service provider, I can tell you that you are right by saying that most customers are unlikely to implement an ERP by themselves and much less if it is an Open Source ERP. So I think you should create a plan or membership with premium support, implementation & customization services, aimed at the folks who are making money out of it, which happens to be the service providers. If it is reasonably and somehow proportional in it’s price, to the earnings we make, I would be more than happy to enroll to this product.

rmehta · September 14, 2018, 12:54am

@jai_kejriwal I think we must look beyond Frappe. Why should Frappe make money to create value for everyone? The community should be independent and responsible for its own health and future. Those who are invested must be open to “donate” not get a service!

@Pawan appreciate your thoughts! But how many of us actually contribute money to say Python or Linux or mariadb and many of the libs we use? Agree about a clear roadmap. Already working on one, since it seems no one else is up to taking this up.

@dhananjay @snv I agree we must look at all options. Maybe a bold statement here on discuss too

rmehta · September 14, 2018, 12:58am

Yes memberships are one option (we already have them). But the money we raise barely covers the effort needed. We need 10x of our current income.

jai_kejriwal · September 14, 2018, 2:48am

@rmehta I was not talking about what’s right or wrong. On that front, I am aligned with your thinking. Practically, however, I believe there will be a path to getting where we want to be and a strong, fast-growing, money-making Frappe will keep things moving in the right direction. Simultaneously, the foundation has to brainstorm how it can be a revenue generating machine which means it’s got to build revenue-oriented service models of its own. And lastly, we all have to ideate about fund raising to make a substantial corpus for the foundation. Off the top of my head, Make in India, Incubators, Crowdfunding, Approaching HNIs etc. Come to mind. However, a coordinated strategy needs to be put in place for fund-raising. I’m happy to take a jab at starting a document for this.

Pawan · September 14, 2018, 3:00am

It depends on person to person or company to company is all I can say about this. Hence, I said it is a moral pledge you can take to give back but it is not necessarily binding on anyone.

yogeshvachhani · September 14, 2018, 6:49am

@rmehta please look at LiveCode’s business model (https://livecode.com/). One thing that they have implement is raising fund for developing a particular feature.

I also liked this article of a Norwegians software developer (HexLicense, Patreon and all that | Jon L. Aasenden). He is using Patreon (https://www.patreon.com/) for funding.

Personally a better option is to use SaaS. But from my experience I can say that no business will be able to use ERPNext out of the box. Almost all businesses require customization of varying proportions. So if you give SaaS based access to latest version of ERPNext then it is written in stone that the user cannot customize ERPNext as per their requirements. This will deter users from going for it…

Regards,

Yogesh

becht_robert · September 14, 2018, 11:41am

I am following this type of discussion on financial sustainability covering the period that the number of paying customers went from less then 100 to the actual 600…and the number of Frappe staff (Webnotes Technologies in those days) grew from 6 to some 15…

This resulted in many experiments but so far none has been an overwhelming success.
And over the years may people, including myself, aired there opinion…also not yielding the right answer.

Accepting that @arikfr bases this statement on his experience building an OS based software company I am wasting my time

I find it quite difficult following ErpNext to find clear strategies in their development path and in their search for a financial sustainable model. It leaves the impression of ad hoc and trial and error actions. Saying this I also realize that I do not have the inside knowledge Frappe staff has, and can they vary well justify their actions.

Let me express a few observations:

There has never been a concerted, structural, long-term/sustained effort to really understand the needs of the customer base in terms of technical functionality, service requirement, deployment, configuration. The information what the user wants is passed to Frappe thru this platform, direct mails and/or discussions, the user conferences, during demo sessions and so on. I believe there are ample IT solutions to address this. In the present model it are the loudest voices who set direction…not the majority.

If I study the four (4) points @rmehta is making in this thread I feel there is some kind of mismatch between these statements and the development path of ErpNext. V11 has many options only interesting for larger companies: Shareholders, Multi-Companies and so on. This observation is actually confirmed by @umair in the first line of his latest (aug18) Blog "In last few months, we at Frappe received decent traction from the large enterprises. " Large companies are for various reasons not so much interested in cloud subscriptions offered by Frappe…and as a consequence one may question how far the developments targeting the needs of larger companies actually contributes to the profitability of Frappe… As stated by Rushabh “A lot of service providers make money on ERPNext, but not all give back, so it feels unfair for Frappe to do all the charity, while others make money…”

Maybe Frappe should consolidate the rapid developments of new functionalities for a while…and fully concentrate on money generating activities…that is the SAAS users. What are their needs in terms of functionalities, service level, willingness to pay , what are the constraints to subscribe to SAAS and so on…

anon-forker · September 15, 2018, 4:37am

If foundation memberships are a significant piece of ERPNext’s model of financial sustainability, I think it would be both helpful and strategic if there were a fair bit more information on erpnext.org about what foundation membership actually looks like. What does it involve, and what are the benefits? How does the foundation influence the direction of ERPNext development? Is any of that spelled out anywhere?

clarkej · September 15, 2018, 6:36am

“Is any of that spelled out anywhere?”

Where to look for that is Community Foundation as per the screenshot

anon-forker · September 16, 2018, 4:08am

Thanks @clarkej. In reality, though, even after browsing through many of the forum posts that come up in those categories, the structure, function, and strategic scope of the foundation is still not entirely clear to me. I might just be missing it. In any case, I suspect that a more consolidated and more explicit statement of the foundation’s role would help those not already active members of the community evaluate the benefits of joining.

The Mozilla foundation’s page strikes me as a particularly good example of how an open source foundation can define itself.

rmehta · September 17, 2018, 3:35am

@becht_robert thanks for sharing your views. You have seen us for a long time. I am also observing how this discussion is tending towards, how can Frappe make more money. I would like to argue that this is the wrong approach.

Maybe the reason we don’t make a lot of money at Frappe is because we are not sure if we are a charitable company or a profit making one. This has led me to the conclusion that it is time, we at Frappe stop thinking ourselves as a charitable organisation and put our energies in building a profitable services and hosting business. That does not mean we stop volunteering. It means we continue “volunteering”, and stop taking “responsibility” for everything.

That brings us to the foundation. I was wrong about the foundation being a community-led entity. Unlike Frappe, most service providers in the community have clear vision that they are here to make money. So it was naive to assume that they will bring some leadership here.

The way forward will be that we will be “moving” a whole bunch of Frappe ops that are related with supporting the open source project, and raise independent funds for it. We would still love community volunteers for the foundation, but we won’t depend on them for going forward.

@anon-forker, already posted that we are working on a clear roadmap and vision.

anon-forker · September 17, 2018, 5:13am

Great! I wasn’t talking about a roadmap but rather a simple description of what the foundation is and how it operates, but if that’s included in the roadmap you’re talking about I’m glad that it’s a priority. I think it will be a big help to attracting more involvement.

I’d argue that this is a false opposition. With the exception of Wikimedia, perhaps, I can’t think of a single FOSS foundation that doesn’t have substantial support and leadership from for-profit companies. There’s no contradiction to that. Supporting open source is often just smart business.

If the foundation isn’t getting the support it needs from service providers, I hope you’ll reach out to see why not. In many other projects I’ve used, business support companies are a major source of manpower, code, and cash.