tonylampada · September 12, 2012 05:41
diff --git a/spiv suggestions on iirc b/spiv suggestions on iirc
 [02:01] == tonylampada [bb026e64@gateway/web/freenode/ip.187.2.110.100] has joined #twisted
 [02:01] <tonylampada> Oh boy, do people still use Trac??
 [02:04] == Jayrays [~jayrays@unaffiliated/jayrays] has quit [Read error: Connection timed out]
 [02:04] <tonylampada> ok, Trac works too :-)
 [02:04] <tonylampada> I've made this website for open source projects crowdfunding - http://www.freedomsponsors.org/. Please have a look at it.
 [02:04] <tonylampada> Right now I'm trying to get it know by as many developers as I can, so, if you like it, please help spread the word
 [02:04] <tonylampada> I think it has a great potential to help free software projects, specially popular ones like twisted
 [02:05] == Jayrays [~jayrays@unaffiliated/jayrays] has joined #twisted
 [02:06] <spiv> tonylampada: well, good luck, but I haven't seen any other bounty system come even vaguely close to working
 [02:07] <spiv> tonylampada: and even though at a glance yours looks like an decent effort I can't see that it's sufficiently brilliant to succeed where others have failed.  So I'm not hopeful.
 [02:07] <tonylampada> spiv: Thanks for the honest feedback. Do you have any insight about what the ones you know of have in common?
 [02:07] <spiv> (I'm not sure there *is* a sufficiently brilliant solution)
 [02:08] <spiv> I like that your specification of the work items are just issues on the regular issue trackers.
 [02:09] <tonylampada> Yes, there is that and the fact that there are no strings attached
 [02:09] <tonylampada> (no escrow, no pre-paid credits)
 [02:10] <tonylampada> pay when youre happy with the solution
 [02:10] <tonylampada> :-)
 [02:10] <spiv> So if a project manages to have a culture where the issue tracker isn't just full of nebulous, impossible issues, then maybe sometimes the work to do is reasonably well defined.
 [02:10] <spiv> But imagine I'm a developer looking at the bounties to decide what to do next.
 [02:11] <spiv> If there's say $10 bounties it's not even really worth my time *looking* at the bounty list for that kind of money
 [02:11] <tonylampada> yes. but the idea is the many people can chip on the same issue
 [02:11] == oubiwann [[email protected]] has quit [Ping timeout: 255 seconds]
 [02:12] <tonylampada> Like JIRA's voting system
 [02:12] <tonylampada> only votes come with a money offer attached
 [02:12] <spiv> If it's $500 maybe it is… but then you need to pick your issue carefully so that you are confident what you consider "resolved" is what the N potential payers would
 [02:12] <spiv> Otherwise you've done a bunch of work only to end up with an argument rather than to end up with money.\
 [02:13] <spiv> Finally, in the case of Twisted, you can donate to a tax-deductible non-profit to support Twisted development already.
 [02:13] <tonylampada> That might happen in some cases
 [02:13] <spiv> Which makes "we take a 3% cut" intermediaries less attractive :)
 [02:13] <tonylampada> But from my experience it's not what I see happen in most FOSS projects
 [02:14] <tonylampada> Yeah, but if I just "donate to Twisted", I have no influence on the next issues that will be worked on :-)
 [02:15] <tonylampada> Are you a developer on this project?
 [02:15] <spiv> The only other FOSS bounty system I recall many details about off the top of my head was Launchpad's, and it pretty quickly hit the problem that the bounties were for massively under-specified tasks.
 [02:15] == robbyoconnor [~wakawaka@guifications/user/r0bby] has joined #twisted
 [02:15] <spiv> I'm a mostly dormant one, but yes :)
 [02:16] <tonylampada> You might've noticed that most of the issues registered belong to a project called Jenkins
 [02:16] <spiv> I know the project
 [02:17] <tonylampada> That's the case because they installed the freedomsponsors JIRA plugin, which adds a "sponsor this issue" link
 [02:17] == ltsstar [[email protected]] has quit [Quit: ltsstar]
 [02:17] <tonylampada> like this one... https://issues.jenkins-ci.org/browse/JENKINS-123
 [02:18] <spiv> Also, random feedback: your front page is kind of weird.  Who's the audience for the part of the page with the list of recent bounties?  Potential devs?  Folks wanting to donate?
 [02:18] <tonylampada> Well both. But I agree it kinda sucks the way it is now
 [02:18] <tonylampada> :-)
 [02:18] <spiv> If its developers looking for issues with attached bounties, then a list from all projects ever isn't useful to me
 [02:19] <spiv> (At least, not without a really easy and efficient filtering widget)
 [02:19] <spiv> If it
 [02:19] <spiv> 's for folks wanting to donate, well, they probably don't really want to pile onto one of the last 10, so again not a great fit.
 [02:20] <tonylampada> Any thoughts on how do I make it better?
 [02:20] <spiv> If I'm a developer, obvious filters (and/or sort orders) are project name, amount.
 [02:21] <spiv> Actually, for people giving money, possibly the same.
 [02:22] <tonylampada> I do have something like this, just not on the front page, here
 [02:22] <tonylampada> http://www.freedomsponsors.org/core/issue/?s=
 [02:22] <spiv> And folks tend to like "$XXXXX in bounties for YYY issues across ZZ projects" things to let them know how big the pool is
 [02:23] == khorn [[email protected]] has joined #twisted
 [02:23] <tonylampada> That's a great suggestion, thanks
 [02:23] <spiv> Plus maybe a prominent "Add a bounty!  Issue URL: [     ]  Amount $[  ] [Go!]" on the front page.
 [02:24] == khorn1 [[email protected]] has quit [Ping timeout: 246 seconds]
 [02:24] == evil_gordita [[email protected]] has quit [Quit: (╯°□°）╯︵ ┻━┻]
 [02:24] <spiv> (or whatever fields you need, or lead into a more complete form, or "hey we don't have this project on-board yet, want to tell them about us?" or whateveR)
 [02:25] <spiv> You're welcome.
 [02:25] <tonylampada> right. Those are actually great ideas, thanks :-)
 [02:25] <spiv> I'd love for this to work!  I'm just sceptical that it can :)
 [02:26] <tonylampada> Please go ahead, create your account, and give it a chance! :-)
 [02:26] <spiv> I'm not really in the market on either side atm :)
 [02:26] == Jayrays [~jayrays@unaffiliated/jayrays] has quit [Read error: Connection timed out]
 [02:26] <tonylampada> I'll do my best to make it work
 [02:27] <tonylampada> I see.
 [02:27] <tonylampada> Well, if you do come up with other ideas that you want to share, please don't hesitate to drop a feedback like at http://www.freedomsponsors.org/core/feedback
 [02:27] == Jayrays [~jayrays@unaffiliated/jayrays] has joined #twisted
 [02:28] <tonylampada> or maybe create an issue on Github https://github.com/freedomsponsors/www.freedomsponsors.org
 [02:30] <spiv> Have you thought at all about what to do about issues that split into smaller issues as the details become better understood/defined?
 [02:30] <spiv> (And other cases like issues merging, or marked as dupes…)
 [02:30] <spiv> But the first case seems pretty key to me
 [02:31] <tonylampada> added: https://github.com/freedomsponsors/www.freedomsponsors.org/issues/33
 [02:31] <spiv> As it gives you a way to handle "well, this is a big problem, but the first milestone is X, and if that is reached then here's some money)
 [02:32] <tonylampada> Well, I think the one who can solve this problem is the developer
 [02:32] <spiv> Quite likely.
 [02:32] <spiv> The answer on your side may be "add a FAQ entry" :)
 [02:33] <tonylampada> The developer would have to come up with a clever way to split the issue into smaller *deliverable* ones, and then ask for smaller sponsorships for those.
 [02:33] <tonylampada> of course, sponsors could still choose to only pay for the whole thing.
 [02:33] == Jayrays [~jayrays@unaffiliated/jayrays] has quit [Read error: Connection reset by peer]
 [02:34] <spiv> Right.  But presumably the sponsors would like an easy way to transfer their one big sponsorship into N smaller ones (if they agree with the developer's plan)
 [02:34] <spiv> Or maybe it's already sufficiently easy.
 [02:34] <tonylampada> but I'm betting that "customer collaboration over contract negotiation" will be the common practice here
 [02:35] <tonylampada> But you're right, that deserves a FAQ entry
 [02:35] <spiv> I wonder a little about cases where a project wants to reject sponsorship, e.g. someone offering to pay for something the the project leads consider "not a bug"
 [02:36] <spiv> Obviously in their own issue tracker they close the bug, but what happens to the sponsorship offer
 [02:36] <spiv> ?
 [02:36] <tonylampada> Well, they can comment on it, and that's it
 [02:36] <tonylampada> (and I think that's as much as they should be able to do)
 [02:37] <spiv> I'm thinking that if the project leads are saying "we'll never accept a patch for this" they may be annoyed if people keep getting attracted to some large bounty for it.
 [02:37] <spiv> (I'd expect that case will be rare, but if you're successful it will definitely occur)
 [02:38] <tonylampada> Yeah, if that happens we (I and the project leads) will think of something.
	[02:01] == tonylampada [bb026e64@gateway/web/freenode/ip.187.2.110.100] has joined #twisted
	[02:01] <tonylampada> Oh boy, do people still use Trac??
	[02:04] == Jayrays [~jayrays@unaffiliated/jayrays] has quit [Read error: Connection timed out]
	[02:04] <tonylampada> ok, Trac works too :-)
	[02:04] <tonylampada> I've made this website for open source projects crowdfunding - http://www.freedomsponsors.org/. Please have a look at it.
	[02:04] <tonylampada> Right now I'm trying to get it know by as many developers as I can, so, if you like it, please help spread the word
	[02:04] <tonylampada> I think it has a great potential to help free software projects, specially popular ones like twisted
	[02:05] == Jayrays [~jayrays@unaffiliated/jayrays] has joined #twisted
	[02:06] <spiv> tonylampada: well, good luck, but I haven't seen any other bounty system come even vaguely close to working
	[02:07] <spiv> tonylampada: and even though at a glance yours looks like an decent effort I can't see that it's sufficiently brilliant to succeed where others have failed. So I'm not hopeful.
	[02:07] <tonylampada> spiv: Thanks for the honest feedback. Do you have any insight about what the ones you know of have in common?
	[02:07] <spiv> (I'm not sure there is a sufficiently brilliant solution)
	[02:08] <spiv> I like that your specification of the work items are just issues on the regular issue trackers.
	[02:09] <tonylampada> Yes, there is that and the fact that there are no strings attached
	[02:09] <tonylampada> (no escrow, no pre-paid credits)
	[02:10] <tonylampada> pay when youre happy with the solution
	[02:10] <tonylampada> :-)
	[02:10] <spiv> So if a project manages to have a culture where the issue tracker isn't just full of nebulous, impossible issues, then maybe sometimes the work to do is reasonably well defined.
	[02:10] <spiv> But imagine I'm a developer looking at the bounties to decide what to do next.
	[02:11] <spiv> If there's say $10 bounties it's not even really worth my time looking at the bounty list for that kind of money
	[02:11] <tonylampada> yes. but the idea is the many people can chip on the same issue
	[02:11] == oubiwann [[email protected]] has quit [Ping timeout: 255 seconds]
	[02:12] <tonylampada> Like JIRA's voting system
	[02:12] <tonylampada> only votes come with a money offer attached
	[02:12] <spiv> If it's $500 maybe it is… but then you need to pick your issue carefully so that you are confident what you consider "resolved" is what the N potential payers would
	[02:12] <spiv> Otherwise you've done a bunch of work only to end up with an argument rather than to end up with money.\
	[02:13] <spiv> Finally, in the case of Twisted, you can donate to a tax-deductible non-profit to support Twisted development already.
	[02:13] <tonylampada> That might happen in some cases
	[02:13] <spiv> Which makes "we take a 3% cut" intermediaries less attractive :)
	[02:13] <tonylampada> But from my experience it's not what I see happen in most FOSS projects
	[02:14] <tonylampada> Yeah, but if I just "donate to Twisted", I have no influence on the next issues that will be worked on :-)
	[02:15] <tonylampada> Are you a developer on this project?
	[02:15] <spiv> The only other FOSS bounty system I recall many details about off the top of my head was Launchpad's, and it pretty quickly hit the problem that the bounties were for massively under-specified tasks.
	[02:15] == robbyoconnor [~wakawaka@guifications/user/r0bby] has joined #twisted
	[02:15] <spiv> I'm a mostly dormant one, but yes :)
	[02:16] <tonylampada> You might've noticed that most of the issues registered belong to a project called Jenkins
	[02:16] <spiv> I know the project
	[02:17] <tonylampada> That's the case because they installed the freedomsponsors JIRA plugin, which adds a "sponsor this issue" link
	[02:17] == ltsstar [[email protected]] has quit [Quit: ltsstar]
	[02:17] <tonylampada> like this one... https://issues.jenkins-ci.org/browse/JENKINS-123
	[02:18] <spiv> Also, random feedback: your front page is kind of weird. Who's the audience for the part of the page with the list of recent bounties? Potential devs? Folks wanting to donate?
	[02:18] <tonylampada> Well both. But I agree it kinda sucks the way it is now
	[02:18] <tonylampada> :-)
	[02:18] <spiv> If its developers looking for issues with attached bounties, then a list from all projects ever isn't useful to me
	[02:19] <spiv> (At least, not without a really easy and efficient filtering widget)
	[02:19] <spiv> If it
	[02:19] <spiv> 's for folks wanting to donate, well, they probably don't really want to pile onto one of the last 10, so again not a great fit.
	[02:20] <tonylampada> Any thoughts on how do I make it better?
	[02:20] <spiv> If I'm a developer, obvious filters (and/or sort orders) are project name, amount.
	[02:21] <spiv> Actually, for people giving money, possibly the same.
	[02:22] <tonylampada> I do have something like this, just not on the front page, here
	[02:22] <tonylampada> http://www.freedomsponsors.org/core/issue/?s=
	[02:22] <spiv> And folks tend to like "$XXXXX in bounties for YYY issues across ZZ projects" things to let them know how big the pool is
	[02:23] == khorn [[email protected]] has joined #twisted
	[02:23] <tonylampada> That's a great suggestion, thanks
	[02:23] <spiv> Plus maybe a prominent "Add a bounty! Issue URL: [ ] Amount $[ ] [Go!]" on the front page.
	[02:24] == khorn1 [[email protected]] has quit [Ping timeout: 246 seconds]
	[02:24] == evil_gordita [[email protected]] has quit [Quit: (╯°□°）╯︵ ┻━┻]
	[02:24] <spiv> (or whatever fields you need, or lead into a more complete form, or "hey we don't have this project on-board yet, want to tell them about us?" or whateveR)
	[02:25] <spiv> You're welcome.
	[02:25] <tonylampada> right. Those are actually great ideas, thanks :-)
	[02:25] <spiv> I'd love for this to work! I'm just sceptical that it can :)
	[02:26] <tonylampada> Please go ahead, create your account, and give it a chance! :-)
	[02:26] <spiv> I'm not really in the market on either side atm :)
	[02:26] == Jayrays [~jayrays@unaffiliated/jayrays] has quit [Read error: Connection timed out]
	[02:26] <tonylampada> I'll do my best to make it work
	[02:27] <tonylampada> I see.
	[02:27] <tonylampada> Well, if you do come up with other ideas that you want to share, please don't hesitate to drop a feedback like at http://www.freedomsponsors.org/core/feedback
	[02:27] == Jayrays [~jayrays@unaffiliated/jayrays] has joined #twisted
	[02:28] <tonylampada> or maybe create an issue on Github https://github.com/freedomsponsors/www.freedomsponsors.org
	[02:30] <spiv> Have you thought at all about what to do about issues that split into smaller issues as the details become better understood/defined?
	[02:30] <spiv> (And other cases like issues merging, or marked as dupes…)
	[02:30] <spiv> But the first case seems pretty key to me
	[02:31] <tonylampada> added: https://github.com/freedomsponsors/www.freedomsponsors.org/issues/33
	[02:31] <spiv> As it gives you a way to handle "well, this is a big problem, but the first milestone is X, and if that is reached then here's some money)
	[02:32] <tonylampada> Well, I think the one who can solve this problem is the developer
	[02:32] <spiv> Quite likely.
	[02:32] <spiv> The answer on your side may be "add a FAQ entry" :)
	[02:33] <tonylampada> The developer would have to come up with a clever way to split the issue into smaller deliverable ones, and then ask for smaller sponsorships for those.
	[02:33] <tonylampada> of course, sponsors could still choose to only pay for the whole thing.
	[02:33] == Jayrays [~jayrays@unaffiliated/jayrays] has quit [Read error: Connection reset by peer]
	[02:34] <spiv> Right. But presumably the sponsors would like an easy way to transfer their one big sponsorship into N smaller ones (if they agree with the developer's plan)
	[02:34] <spiv> Or maybe it's already sufficiently easy.
	[02:34] <tonylampada> but I'm betting that "customer collaboration over contract negotiation" will be the common practice here
	[02:35] <tonylampada> But you're right, that deserves a FAQ entry
	[02:35] <spiv> I wonder a little about cases where a project wants to reject sponsorship, e.g. someone offering to pay for something the the project leads consider "not a bug"
	[02:36] <spiv> Obviously in their own issue tracker they close the bug, but what happens to the sponsorship offer
	[02:36] <spiv> ?
	[02:36] <tonylampada> Well, they can comment on it, and that's it
	[02:36] <tonylampada> (and I think that's as much as they should be able to do)
	[02:37] <spiv> I'm thinking that if the project leads are saying "we'll never accept a patch for this" they may be annoyed if people keep getting attracted to some large bounty for it.
	[02:37] <spiv> (I'd expect that case will be rare, but if you're successful it will definitely occur)
	[02:38] <tonylampada> Yeah, if that happens we (I and the project leads) will think of something.