5 ms·

Like it or not, Google is probably paying around a million $$ a year to keep senior full-time developers around that want to work on the language. That could be

Like it or not, Google is probably paying around a million $$ a year to keep senior full-time developers around that want to work on the language. That could be used as a benchmark to calculate how much of an investment is required to have a healthy development cycle. If a community-maintained fork is created, it would need time and monetary investment similar to what google is doing just to maintain and develop non-controversial features. Question is: Is this assessment sensible and if so, is the community able or willing to make this kind of investment?

bradfitz 7y ago
$10M per year would be closer, actually. (Source: public git logs for all core Go repos in past year, look for google.com or golang.org emails, multiply by typical Google salaries, which you can find on various sites.)
- jdm2212 7y ago
  10MM might be on the low side. I'd be surprised if Ian Lance Taylor, Russ Cox and Rob Pike are making less than a million dollars a year each. They're all L8+, and even by that standard they're notably famous/distinguished/respected.
- phillipcarter 7y ago
  Another one to add is the cost of evangelism. Even if someone isn't evangelizing a language directly, they're helping further it so long as they demonstrate how to do stuff with it.
geodel 7y ago
I read somewhere on mailing list that there are about ~30 people inside Google working full time on Go. And that is different from engineers working on products built using Go. With infrastructure cost added it would be easily $10 million a year. So these Linkedin/Twitter style thought leaders demanding some kinda OpenGo foundation need to ask if they have capacity to raise this amount consistently over years to start and run this foundation.
- smt88 7y ago
  A foundation doesn't need to replace all the corporate money. At minimum, it's a guarantee that the language will be maintained if/when the corporate entity loses interest. Maintaining a language is not nearly as cumbersome as improving it. At best, a foundation is also going to provide more neutral governance, as well as a nexus for multiple corporations to work on the language together. Other companies are more likely to adopt Go if they see it as not just a Google project. I personally don't care whether my favorite languages have a foundation, as long as they're used for large projects by large companies. That alone is a good guarantee that they'll stick around.
- harimau777 7y ago
  Would it be viable to setup some sort of trust at the beginning so that it wasn't necessary to raise that money repeatedly?
  - mrunkel 7y ago
    Sure, but that means raising 10x+ since you'd really want the trust to be untouched and just have the foundation run off the interest.
jeffalyanak 7y ago
I think, to a degree, the growth in popularity in the language — coupled with the perceived value of whichever changes the current Go team won't make — will determine the viability of a fork. If the popularity of the language grows enough, there will inevitably be new companies and ecosystems reliant on it and those interests may eventually band together to create forks they think bring value to their businesses. Outside of that formula, forks will almost always remain niche and without significant investment.

bradfitz 7y ago

$10M per year would be closer, actually. (Source: public git logs for all core Go repos in past year, look for google.com or golang.org emails, multiply by typical Google salaries, which you can find on various sites.)

jdm2212 7y ago
10MM might be on the low side. I'd be surprised if Ian Lance Taylor, Russ Cox and Rob Pike are making less than a million dollars a year each. They're all L8+, and even by that standard they're notably famous/distinguished/respected.
phillipcarter 7y ago
Another one to add is the cost of evangelism. Even if someone isn't evangelizing a language directly, they're helping further it so long as they demonstrate how to do stuff with it.

geodel 7y ago

I read somewhere on mailing list that there are about ~30 people inside Google working full time on Go. And that is different from engineers working on products built using Go. With infrastructure cost added it would be easily $10 million a year. So these Linkedin/Twitter style thought leaders demanding some kinda OpenGo foundation need to ask if they have capacity to raise this amount consistently over years to start and run this foundation.

smt88 7y ago
A foundation doesn't need to replace all the corporate money. At minimum, it's a guarantee that the language will be maintained if/when the corporate entity loses interest. Maintaining a language is not nearly as cumbersome as improving it. At best, a foundation is also going to provide more neutral governance, as well as a nexus for multiple corporations to work on the language together. Other companies are more likely to adopt Go if they see it as not just a Google project. I personally don't care whether my favorite languages have a foundation, as long as they're used for large projects by large companies. That alone is a good guarantee that they'll stick around.
harimau777 7y ago
Would it be viable to setup some sort of trust at the beginning so that it wasn't necessary to raise that money repeatedly?
- mrunkel 7y ago
  Sure, but that means raising 10x+ since you'd really want the trust to be untouched and just have the foundation run off the interest.

jeffalyanak 7y ago

I think, to a degree, the growth in popularity in the language — coupled with the perceived value of whichever changes the current Go team won't make — will determine the viability of a fork. If the popularity of the language grows enough, there will inevitably be new companies and ecosystems reliant on it and those interests may eventually band together to create forks they think bring value to their businesses. Outside of that formula, forks will almost always remain niche and without significant investment.