GitHub feature: “Request Service” button

Am I the only one on the planet thinking a new “Request [Paid] Service” button on GitHub to send requests for help to all interested/signed-up parties would generate a lot of $$$ for open source, for developers and for Microsoft AND would help users of open source too?

Way more valuable than Facebook’s “invention” of the “Like” button in my opinion!

1 Like

That is a gig economy. I don’t like gig economy. I like long term, well-budgeted projects.

That is going to be Upwork all over again, with prices and quality taking a dive.

1 Like

True in some sense but [done the right way] I look at it from a different perspective:
(and Microsoft are welcome to DM me to findout what “right” means exactly)

1- It can generate developers of hosted open source projects some side income

2- Often I need help with different aspects of an open project (because my clients use them) and I really hate going to Upwork/PPH/etc. or anything the like (99.9999% of the times you get spam bids or really unskilled responses - those platforms are good for finding an App developer not someone who knows how to fix X bug in Y subsystem of Z project)

3- The greatest benefit of all is the “marketing” point you are missing, if you engage on a small gig economy work with a company who are willing to spend money then you are way more likely to get a real long term project from them - and boy have I got that experience - what better platform to tell the world what that specific X+Y+Z is which you are skilled at!?

4- Look at this, it happens and is needed on a daily basis even amongst big organisations:

I can’t call this “gig” economy, whether I like or not this feels like “the” economy these days.

I simply expect the same happen on Github. Lots of low-quality coders sign up and spam any request for service with their offers.

1 Like

That comes to the “right” design and philosophy behind the design of the system.
I don’t want to make this about how to fix all those other platforms but I give you one of dozens of ideas that comes to my mind that you can build into this to fix it: Restrict it to people who have actually contributed to a project with a minimum number of commits and over a certain amount of time.

You can make the exact same argument about people who answer questions on Stackoverflow but they are managing it pretty well, wouldn’t you say so!?

That would reduce the value of that button:

Suppose the people who actually work on the project are busy and don’t have time or inclination to do the paid services.
At the same time, a person who I know would be able to do the work and is willing, can’t do it because of “commits and time”.

But if you remove the “commits and time” requirement, you get a swarm of low-quality bids.

The difference is that in SO the mob evaluates concurrent answers, and the good ones drift up. In GH, one person - the requestor - would have to evaluate dozens of bids, and once they accepted one - there is no more competition, you get what you get.

The comparison would be closer if the bidders provided their implementations, and other developers evaluated them and voted for the best in their opinion. But that would drive the prices and quality even lower, to the rock-bottom level, because only the very cheap coders can afford to do coding with only a probability to get paid. However, this model does exists in the wild, - where the designers make logos effectively for free until the client choose one. The quality is not too good, IMHO.

1 Like

Why can’t this person just put in the effort to contribute a bit? Nothing is free in this world, why should the right to bid come free!?
If they can’t put in that initial effort then maybe they are not good enough for this platform to bid either :smiley:

So this is just fine in my view. Indeed it’s a matter of opinions now. But if you are good at it, and really care about entering the market then you put in the effort (This is like the interview to land a job. We don’t get a job in a company without an interview and doing the homework for the interview so why should we get that job on this platform if can’t bother with a few commits!?).

It would also encourage contributions back into an open souce project that is a source of revenue for you (if you don’t contribute you don’t get to earn from them).

Even SO doesn’t let you answer when you initially join until you leave a couple of “comments” to enable access to the main answers.

How do we define value? If value is quality then a method like the above would increase the quality and as a result the value. Of course, not everyone who has the skills will be allowed to bid until they put in the effort (too bad for them) but the ones who do get to bid are guranteed to not be spammers (good for the payer).

Hmm, I’m not following you.

Suppose I have an urgent problem with a specific project on Github - say a library. None of the project team is interested to help me on urgent terms despite the offered money. Then I bring my own developer … and he can’t take on the task because commits.

Of course, he can create a patch and ask the project team to accept it. Once it is accepted, he should be paid. Github could be an escrow service for such “service patches” - but here we again have an unrelated party (the project team) who can become a bottleneck.

You talk like it is some unbelievable privilege to get to commit something to Github. :face_with_raised_eyebrow: You do realize that there are enough markets out there where I can offer my services with fewer hurdles?

In this case, I defined value as “as a client, can I get a solution here quickly?”. Turns out I can’t, because there are some artificial obstacles which use only a remote approximation for developer’s quality.

1 Like

Perhpas I dind’t explain well.

This is the issue. You can’t find your own developer. If you could you wouldn’t need GitHub.

The actors in my mind are:

  1. The main team behind the library (too busy to provide a service)

  2. Third parties who like to provide services on demand (they must first contrinute to the library to earn the bidding rights, these contributions are from the traditional routes - not via this feature)

  3. Clients (they may or may not choose to contribute the results back into the project, they are paying after all - all they get is bids from the persons who had the right to bid)

It’s not an unbelievable privilege, but you can only contribute and merge your code if the team behind the library actually takes the time to review and approve your pull request. That works perfect as a monitoring system. That is the privilage.

The way I see it is that firstly people won’t go to GitHub to ask for or bid on something general like “Build an App”, “Make a Website” or “I need a plugin”. Instead they would go to ask and bid for very specific needs. If a bidder is clueless about a project or has never taken the time to learn its structure to get a few PRs merged then the bidder would not be able to bid - simple.

Of course everyone and anyone can do it, it is “open” to everyone after all.
But not everyone actually does it on every project and every library, right?
Out of millions of libraries and projecs, you go and focus on the ones you really understand and use and care about.

If I take myself as an example, I have contributed to a few GitHub projects. For example the mono compiler platform. I came across bugs to fix as a result of using the projects so much, but even then that was not enough, the mono PR took a while to be reviewed by Xamarin for instance, improved by me and finally merged.

  1. Xamarin don’t want to be service providers in this case
  2. I am the third party who took the time to learn the systems and actually get PRs merged
  3. So when someone (a client) needs help with the mono compiler platform not every random bidder can bid - the platform gives the right to someone whose work was scrutinised and reviewed (me in this case)
  4. I would not be able to go and randomly bid on something like, ZLib or XmlSec or any other lib, because I never took the time to learn their system and contribute…

Does this still not make sense?

It does, as long as we assume that there are enough people in the world who are not the project team members and yet contributed to a project and at the same time are willing to do one-off jobs.

I feel this intersection is not big enough for seeding the marketplace.

Let’s estimate the projected revenue.

  • Github has 40 mln of users.
  • Based on some 2017 statistics I found, about 5% of users contribute to other projects.
  • That’s 2 mln.
  • Let’s assume that on average those folks can do 2 hrs/mo (some more, some none).
  • That’s 4 mln w-h.
  • Let’s assume the average rate $50/hr (CA would charge $100, India - $10).
  • That’s $200mln/mo
  • Let’s assume Github would take a cut, how big? Let’s assume 15%.
  • That’s $30mln/mo revenue for GH.

That feels too small to bother, if you take into account MS’s size and Github worth.

And note that the number is inflated because I assumed that there is a demand for 2 mln paid patches a month - that is most certainly not true.

1 Like