5 ms·

do you think that earning money or having a viable business model is a bad thing? look at our actions since the inception of the company, as well as our "big t

by nopzor 6y ago

do you think that earning money or having a viable business model is a bad thing? look at our actions since the inception of the company, as well as our "big tent" philosophy; i think they speak for ourselves. we built a sustainable business over the course of 4 years, prior to raising a dollar of VC money. we raised VC money to go even faster. most of our VC money goes to paying our employees. most of our employees are engineers. most of what our engineers do is open source. if we don't build a sustainable business, who will develop the software? the overwhelming majority of development on things like grafana, loki, tempo are done by grafana labs employees. plus, we are the main contributors behind other OSS projects that we didn't create and don't own (like prometheus and cortex). [note: am co-founder/ceo at grafana labs]

andrewstuart2 6y ago
Neither earning money nor having a viable business model are bad things, but that's about as straw as a strawman argument comes. A business model existed before these changes. As far as I can tell, though, it's not _captured_ the value that the founders and investors seem to think ought to be their wealth. Grafana Labs and all these other companies have recieved both value add and goodwill free marketing from open source contributions and communities, and I don't see these companies being at all interested in figuring out who/what they've gained value from, beyond the currently existing company, in order to make a profit and benefit the founders and shareholders. I'm not against AGPL. I'm against the relicensing of software made in good faith against one model of OSS being discontinued and basically forked by the most prominent for-profit company that have hitherto received the significant benefits of both contributions and goodwill that being more freely open source brings. By relicensing you've probably significantly reduced the number of companies willing to run future versions of your software, and the number of companies willing to build businesses off of extending it and offering that value add, while still potentially contributing changes upstream. You're going to slowly eat away at that good will and the number of people asking for your product internally, since fewer and fewer people will remember the benefit from their last gig. I could be wrong here, only time will tell, but I honestly don't think this is a move that increases the long term sustainability of your business model. If anything, I think it's going to be detrimental in the long term. Perhaps even in the short term.
tshaddox 6y ago
> do you think that earning money or having a viable business model is a bad thing? Nope, but I kinda thought the whole idea behind open source is that no one company or group of companies/licensees gets privileged access to or control over the software, including any company that happens to have the same name as the piece of software. I realize that “open source” and “free software” and other related labels mean many things to many people, and some people have very strong opinions about what does and doesn’t qualify, so I’m not trying to be prescriptive about what these terms “really” mean, but to me this is a pretty important almost self-evident part of it. If you want to sell proprietary software just do that!
- ankmathur96 6y ago
  I don't believe this is a realistic take on open-source. There's not a single successful modern OSS project that's not basically controlled by a single organization (or was not at some point in its conception). Control can take multiple forms: most of the time, it's directional to the project, since the roadmap of the company managing the project dictates what gets done. Sometimes, it's in the form of how you can use the project. The great thing is that you can always fork the project and create a version for which the new rules/direction does not apply. The challenging part is making that a project where active development is actually happening. Certainly, there's things like the Linux Foundation that serve as places where projects can live without formal control by a single organization, but in practice, orgs that commit most of the code control things (for better or for worse - I'm not saying this is a good thing)
  - tshaddox 6y ago
    > I don't believe this is a realistic take on open-source. ... The challenging part is making that a project where active development is actually happening. Well, yeah, that’s just the most obvious gut reaction everyone has when they first hear about open source software: Why in the world would anyone put work into software that’s free for everyone to use? My impression is that open source advocates have put tons of work into explaining why this can and does actually work and can result in software that’s better than proprietary alternatives.
DetroitThrow 6y ago
I'm glad you were able to find a FOSS license that's able to support your business. I think grafana lab's use of AGPL will help encourage its adoption by other businesses wanting to make money in free software as well. For other's who are interested, here's a faq from grafana including a discussion about "Why AGPL and not SSPL?": https://grafana.com/blog/2021/04/20/qa-with-our-ceo-on-relicensing/ https://grafana.com/blog/2021/04/20/qa-with-our-ceo-on-relic...
- xbar 6y ago
  Very much this. As a commercial consumer of FOSS, AGPL has been troubling for me. As a non-commercial hacker, AGPL has been fine for me: GPL is fine here, and distribution as defined by SaaS or tarball is equivalent; I share back. My opinions are that non-commercial FOSS projects probably shouldn't choose AGPL but that small-ish commercial FOSS projects (like Grafana, et al) probably should so that big commercial entities don't eat their lunch. AGPL seemed like a tool that was mostly ahead of its time but is nice to have now.
sdesol 6y ago
Here is an analysis of the investment that grafana labs is putting into grafana: https://public-001.gitsense.com/insights/github/repos?r=github%3Agrafana%2Fgrafana&v=repo https://public-001.gitsense.com/insights/github/repos?r=gith... Switch to the impacts view and sort by "days active". I think it is safe to say, any developer with days active >= 25% is a grafana labs employee or somebody that really loves to contribute to grafana because they are being paid to.
valyala 5y ago
Could you name companies that forced you changing the license from Apache 2 to AGPLv3? It looks like this isn't Amazon because Grafana labs has good partnership with them [1] [2]. The license change can hurt Prometheus ecosystem [3] [4]. It looks like projects such as VictoriaMetrics [5] benefit the most from the Grafana license change. They don't use source code from Grafana projects and have no plans to change Apache 2 license :) [1] https://news.ycombinator.com/item?id=25432729 https://news.ycombinator.com/item?id=25432729 [2] https://news.ycombinator.com/item?id=25432630 https://news.ycombinator.com/item?id=25432630 [3] https://twitter.com/juliusvolz/status/1384565189249572866 https://twitter.com/juliusvolz/status/1384565189249572866 [4] https://twitter.com/fredbrancz/status/1384571338552184834 https://twitter.com/fredbrancz/status/1384571338552184834 [5] https://github.com/VictoriaMetrics/VictoriaMetrics/ https://github.com/VictoriaMetrics/VictoriaMetrics/

by nopzor 6y ago

andrewstuart2 6y ago

Neither earning money nor having a viable business model are bad things, but that's about as straw as a strawman argument comes. A business model existed before these changes. As far as I can tell, though, it's not _captured_ the value that the founders and investors seem to think ought to be their wealth. Grafana Labs and all these other companies have recieved both value add and goodwill free marketing from open source contributions and communities, and I don't see these companies being at all interested in figuring out who/what they've gained value from, beyond the currently existing company, in order to make a profit and benefit the founders and shareholders. I'm not against AGPL. I'm against the relicensing of software made in good faith against one model of OSS being discontinued and basically forked by the most prominent for-profit company that have hitherto received the significant benefits of both contributions and goodwill that being more freely open source brings. By relicensing you've probably significantly reduced the number of companies willing to run future versions of your software, and the number of companies willing to build businesses off of extending it and offering that value add, while still potentially contributing changes upstream. You're going to slowly eat away at that good will and the number of people asking for your product internally, since fewer and fewer people will remember the benefit from their last gig. I could be wrong here, only time will tell, but I honestly don't think this is a move that increases the long term sustainability of your business model. If anything, I think it's going to be detrimental in the long term. Perhaps even in the short term.

tshaddox 6y ago

> do you think that earning money or having a viable business model is a bad thing? Nope, but I kinda thought the whole idea behind open source is that no one company or group of companies/licensees gets privileged access to or control over the software, including any company that happens to have the same name as the piece of software. I realize that “open source” and “free software” and other related labels mean many things to many people, and some people have very strong opinions about what does and doesn’t qualify, so I’m not trying to be prescriptive about what these terms “really” mean, but to me this is a pretty important almost self-evident part of it. If you want to sell proprietary software just do that!

ankmathur96 6y ago
I don't believe this is a realistic take on open-source. There's not a single successful modern OSS project that's not basically controlled by a single organization (or was not at some point in its conception). Control can take multiple forms: most of the time, it's directional to the project, since the roadmap of the company managing the project dictates what gets done. Sometimes, it's in the form of how you can use the project. The great thing is that you can always fork the project and create a version for which the new rules/direction does not apply. The challenging part is making that a project where active development is actually happening. Certainly, there's things like the Linux Foundation that serve as places where projects can live without formal control by a single organization, but in practice, orgs that commit most of the code control things (for better or for worse - I'm not saying this is a good thing)
- tshaddox 6y ago
  > I don't believe this is a realistic take on open-source. ... The challenging part is making that a project where active development is actually happening. Well, yeah, that’s just the most obvious gut reaction everyone has when they first hear about open source software: Why in the world would anyone put work into software that’s free for everyone to use? My impression is that open source advocates have put tons of work into explaining why this can and does actually work and can result in software that’s better than proprietary alternatives.

DetroitThrow 6y ago

I'm glad you were able to find a FOSS license that's able to support your business. I think grafana lab's use of AGPL will help encourage its adoption by other businesses wanting to make money in free software as well. For other's who are interested, here's a faq from grafana including a discussion about "Why AGPL and not SSPL?": https://grafana.com/blog/2021/04/20/qa-with-our-ceo-on-relicensing/ https://grafana.com/blog/2021/04/20/qa-with-our-ceo-on-relic...

xbar 6y ago
Very much this. As a commercial consumer of FOSS, AGPL has been troubling for me. As a non-commercial hacker, AGPL has been fine for me: GPL is fine here, and distribution as defined by SaaS or tarball is equivalent; I share back. My opinions are that non-commercial FOSS projects probably shouldn't choose AGPL but that small-ish commercial FOSS projects (like Grafana, et al) probably should so that big commercial entities don't eat their lunch. AGPL seemed like a tool that was mostly ahead of its time but is nice to have now.

sdesol 6y ago

Here is an analysis of the investment that grafana labs is putting into grafana: https://public-001.gitsense.com/insights/github/repos?r=github%3Agrafana%2Fgrafana&v=repo https://public-001.gitsense.com/insights/github/repos?r=gith... Switch to the impacts view and sort by "days active". I think it is safe to say, any developer with days active >= 25% is a grafana labs employee or somebody that really loves to contribute to grafana because they are being paid to.

valyala 5y ago

Could you name companies that forced you changing the license from Apache 2 to AGPLv3? It looks like this isn't Amazon because Grafana labs has good partnership with them [1] [2]. The license change can hurt Prometheus ecosystem [3] [4]. It looks like projects such as VictoriaMetrics [5] benefit the most from the Grafana license change. They don't use source code from Grafana projects and have no plans to change Apache 2 license :) [1] https://news.ycombinator.com/item?id=25432729 https://news.ycombinator.com/item?id=25432729 [2] https://news.ycombinator.com/item?id=25432630 https://news.ycombinator.com/item?id=25432630 [3] https://twitter.com/juliusvolz/status/1384565189249572866 https://twitter.com/juliusvolz/status/1384565189249572866 [4] https://twitter.com/fredbrancz/status/1384571338552184834 https://twitter.com/fredbrancz/status/1384571338552184834 [5] https://github.com/VictoriaMetrics/VictoriaMetrics/ https://github.com/VictoriaMetrics/VictoriaMetrics/