Snoyberg's The Do's and Don't's of Running an Open Source Project:
Real title should be: how to get members of any open source community to be interested in helping you. But the given title is catchier.
There's an old "ha ha, only serious" joke. If you go to a Linux forum and ask for help fixing your WiFi driver, everyone will ignore you. If, instead, you say "Linux sucks, you can't even get a f*&$ing WiFi driver working!" thousands of people will solve the problem for you.
This story is a great example of manipulating people, but it's obviously a negative take on it. I'd like to share some thoughts on this from a much more positive standpoint, which will help you get people to pay more attention, be more helpful, and—perhaps most importantly—create a healthier open source community over all.
These items will appear in no particular order, and will almost all fall into either the attractor or obstacle category. An attractor is something you can do to make people want to participate with you. An obstacle is something you should not do, which would prevent people from interacting with you.
And it should go without saying, but I'll say it anyway: this is an opinionated list, written by one guy. I'm including in here things that I personally care about, and things which friends and colleagues have shared with me. No example is specific to any individual, so don't think I'm calling you out: I'm most certainly not. And some people may disagree, or have other items for this list. Sharing such differing thoughts would be very healthy.
The list:
- Don't waste people's time
- Demonstrate you've tried
- Help other people
- Don't be rude
- Say thank you
- Admit if you're new
- Offer to help
- Give money
Saying, "Lennart Poettering sucks" is not on the list of recommendations.
(Score: 2, Interesting) by Anonymous Coward on Wednesday October 25 2017, @05:21AM (7 children)
It turns out that it's pretty hard to be a maintainer.
People just aren't good at leading, or organizing themselves well enough to respond appropriately. A lot of times, egos really get in the way.
The best maintainers are people who make suggestions that lead would-be contributors towards an intended goal without contributors realizing that they are the ones being manipulated into doing the grunt work. The worst maintainers insist on doing things a certain way, finish off contributions and then take credit for them, and just ignore others' input.
Ain't nobody got time for bad maintainers; just fork that shit and move on.
(Score: -1, Troll) by Anonymous Coward on Wednesday October 25 2017, @05:41AM (6 children)
Your comment is out of context because TFA is about asking for help, not about trying to offer help.
You fail at reading comprehension, you are a retard, and your attempted contribution to the discussion is unwelcome. Fuck off.
(Score: 0) by Anonymous Coward on Wednesday October 25 2017, @05:53AM (5 children)
You should have ended with "Fork off."
(Score: 0) by Anonymous Coward on Wednesday October 25 2017, @06:10AM (4 children)
Firstly, maintainers tell you to fuck off and die in a fire. Second, you fork the project. Third, you come back with a pull request that upstages the original author. Fourth and finally, you get ignored for eternity.
Here in the real world, that is the lifecycle of a contribution.
(Score: 3, Insightful) by tangomargarine on Wednesday October 25 2017, @03:26PM (3 children)
Maybe it has something to do with your attitude.
"Is that really true?" "I just spent the last hour telling you to think for yourself! Didn't you hear anything I said?"
(Score: -1, Flamebait) by Anonymous Coward on Wednesday October 25 2017, @07:22PM (2 children)
You are a two-faced moronic hypocrite, and here is why.
Your sig says:
Say you spend a month tracking down the root cause of a bug, and you fix it, and you contribute a fix. The maintainer rejects it without comment. You respond, "I just spent a whole month fixing this bug, I have proof of a bug, and I have proof that the fix works. Did you even look at anything I submitted?"
Maybe the maintainer is a moron, just like you are a moron.
(Score: 2) by tangomargarine on Wednesday October 25 2017, @07:55PM
How is my sig either twofaced, or hypocritical? Trust, but verify.
It's not like I changed my sig for this exact comment tree, dude. It's been the same thing for the last year or two at least.
"Is that really true?" "I just spent the last hour telling you to think for yourself! Didn't you hear anything I said?"
(Score: 0) by Anonymous Coward on Wednesday October 25 2017, @08:10PM
I love how simplified the stories are when told by the "victims". Given your attitude in a simple article comment section it is highly likely you piss people off on a regular basis. Not saying that is reason to reject a good pull request, but as someone said egos can often be a problem.
(Score: 4, Insightful) by jelizondo on Wednesday October 25 2017, @05:34AM (5 children)
In summary, be polite and people will help you. Who would have known? (Thanks mom!)
(Score: 3, Informative) by frojack on Wednesday October 25 2017, @06:28AM (2 children)
Stuff I don't do when asking for help is stuff that bugs me when I'm trying to Help:
I Don't threaten (or even hint about) going "back to windows" or a different distro/package if I can't get help.
I don't provide 37 reams of copy/paste of mostly useless information in my first request. Just basics.
I don't give long winded explanations starting from the day I first ran into linux, hoping I can drift into the problem by and by.
I don't ask for help and google later.
I don't post a question, and 5 minutes later tack on an "Anybody?" post.
I don't write direct to the maintainer/packager without first looking for a bug report site specific to the package/distro involved.
I don't use a problem statement as thin as "it doesn't work".
No, you are mistaken. I've always had this sig.
(Score: -1, Flamebait) by Anonymous Coward on Wednesday October 25 2017, @08:35AM (1 child)
Cool story bro, next time we'll call the story "What Frojack does NOT do when manipulating maintainers".
PS. Are you the same person who, every time there's a story about intelligence, posts their life story about how hard it is living with average people? Real interesting stuff.
(Score: 5, Funny) by The Mighty Buzzard on Wednesday October 25 2017, @11:15AM
No, that's me. I think you bastards make it intentionally difficult too.
My rights don't end where your fear begins.
(Score: 2, Troll) by turgid on Wednesday October 25 2017, @09:15AM
As Mrs Turgid once famously observed about us computer geek types, we're all, "Smelly, autistic people with poor social skills.'
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].
(Score: 4, Insightful) by TheRaven on Wednesday October 25 2017, @01:41PM
sudo mod me up
(Score: 2) by bradley13 on Wednesday October 25 2017, @05:45AM (18 children)
I keep thinking I ought to get into some open-source project. I'm a programmer, I'd like to think a good one. Just as an example, a couple of weekends ago I got bored, so I implemented a (very simple) digital currency, just for fun. Not a huge project, but not bad for a weekend's work.
The question I can't seem to get past is: how and where? You have big projects (Linux, LibreOffice) where it would take years of effort to really work one's way into the project. Or you have projects so specialized that maybe 10 people in the whole world will ever care about them, and I'm not one of them. Or you have projects where - when you look into the community - you realize that you really don't want anything to do with them.
Any tips on finding an interesting project, not massively huge, but still meaningful, where the community is focused on the tech and not on politics?
Everyone is somebody else's weirdo.
(Score: 5, Informative) by Anonymous Coward on Wednesday October 25 2017, @05:50AM (4 children)
Open Source software is about scratching your own itches; you're asking people where you should be scratching.
Also, you sound like more of a hacker than a programmer. Getting something cobbled together isn't that interesting; script kiddies can do that. What matters is whether your work is fundamentally sound—whether it's beautiful not only now, but for the rest of time.
(Score: -1, Flamebait) by Anonymous Coward on Wednesday October 25 2017, @06:02AM (1 child)
Cool story, bro. You are fucking hilarious. According to your unrealistically high standard of code beauty, Randall Munroe is Not A Real Programmer, and we all know that is simply horseshit.
Here is proof, conveniently linked to explainxkcd for you because you are an idiot.
https://www.explainxkcd.com/wiki/index.php/1296:_Git_Commit [explainxkcd.com]
(Score: 2, Insightful) by Anonymous Coward on Wednesday October 25 2017, @06:07AM
Chill out, bro.
Nobody has yet made it to the stars, but it's important to aim for them; it's not enough to fly Cessnas on the weekend.
(Score: 5, Insightful) by bradley13 on Wednesday October 25 2017, @02:31PM (1 child)
In a different comment thread someone wrote: "most support forums have a nasty streak that's really, really offputting"
I waited several hours to reply to this comment, because I was curious to see what responses it might generate. The comment I'm replying to is a perfect example of that. Rather than addressing my question, AC says: "you sound like more of a hacker than a programmer"
In fact, I am accustomed to doing solo projects, or at most projects with one other person. However, this includes things like complete ERP systems (yes, plural. Yes, sold to real customers, and supported over many years). So the offhand insult "you're just a hacker" is annoying, but it absolutely typifies what I said about some projects I have looked into: Who wants to deal people like that?
At least if I'm programming solo, there's only one jerk involved :-/
Everyone is somebody else's weirdo.
(Score: -1, Troll) by Anonymous Coward on Wednesday October 25 2017, @03:54PM
So what? I bet your work does stink, like most proprietary crap.
(Score: -1, Troll) by Anonymous Coward on Wednesday October 25 2017, @05:57AM
Since you say you are a programmer you are supposed to have a natural uncontrollable urge to fix bugs in software you use. Pick some bug fixes that you surely have written by now and try to contribute your fixes back to the maintainers. Find out your software is maintained by an exclusive club of elitists who will ignore you. Give up on programming and retire to a life of trolling the shit out of stupid fuckheads on news-for-troll forums.
(Score: 5, Interesting) by c0lo on Wednesday October 25 2017, @06:43AM
Start with tools/utils/libraries you know had helped you before (but there was an itch along the way, could be made smoother or have some extended functionality) or one that you'd love to have used but it wasn't usable when you needed it.
(without info on prog-langs, I certainly can't offer examples. With that info, I may be able to offer examples but again I may not be able)
https://www.youtube.com/watch?v=aoFiw2jMy-0 https://soylentnews.org/~MichaelDavidCrawford
(Score: 0) by Anonymous Coward on Wednesday October 25 2017, @06:57AM
It's actually simple. Pick a piece of software that you actually use (there is more open source software on your system than you actually are aware of). E.g. lots of libraries used today are open source. Check out their project site and sign up to their mailing list (if any), and contact the people in that project asking them where you can help. They will point you in the right direction.
Some projects also have summer of code projects, which might have not been all assigned. These are entry level sub-projects that are great to get you started (they often have also a mentor, who can help to get you started).
(Score: 0) by Anonymous Coward on Wednesday October 25 2017, @07:28AM (1 child)
look at scipy (www.scipy.org) and children (numpy, matplotlib, etc).
those are python with a lot of C and possibly fortran in the back.
extremely useful.
with numpy for instance, I see a lot of warnings when I compile it, I guess that's something that anyone can try to fix.
I don't know if they enter the "massively huge" category.
if you like C++, see if you can help with vtk (www.vtk.org).
this one is, probably, massively huge.
but they need better examples and documentation.
in fact vtk also has wrappers in various scripting languages, maybe you can help with converting examples from C++ to the scripting languages.
(Score: 2) by bradley13 on Wednesday October 25 2017, @02:37PM
Good suggestion, thanks...
Everyone is somebody else's weirdo.
(Score: 0) by Anonymous Coward on Wednesday October 25 2017, @08:38AM (1 child)
Hi bradley13 - please can you FIX the cross referencing bug in LibreOffice. If somebody doesn't do it THEN I AM GOING BACK TO MICR$$$OFT Word. Need any more ideas?? I got plenty.
(Score: 2) by bradley13 on Wednesday October 25 2017, @02:39PM
LibreOffice is one candidate that I am seriously considering. It is something I use, and has problems I would like to see fixed. The only reason I haven't jumped in is: it is definitely "massively huge", and will take a long time to work into...
Everyone is somebody else's weirdo.
(Score: 3, Interesting) by The Mighty Buzzard on Wednesday October 25 2017, @11:52AM (1 child)
You're always welcome to show off your chops here until you find something that tickles your fancy a bit more. We may do politics out on the site but they're utterly irrelevant to coding. No political opinion is going to turn an O(2N) perl sub into O(N) or vice versa.
My rights don't end where your fear begins.
(Score: 2) by bradley13 on Wednesday October 25 2017, @02:41PM
Thanks for the offer! I haven't done any sort of web-oriented development for a long time, but I'll certainly consider it.
Everyone is somebody else's weirdo.
(Score: 2) by bzipitidoo on Wednesday October 25 2017, @12:14PM
What holds me back is pay. Contributing is hard, but easier than getting paid to contribute.
What little free time I have I prefer to do research. Which I'm not getting paid for either, but rather write snippets of code for that than monkey with someone else's tedious and boring code, if it's all charity work on my part anyway.
(Score: 2, Insightful) by Anonymous Coward on Wednesday October 25 2017, @01:10PM
"Or you have projects so specialized that maybe 10 people in the whole world will ever care about them, and I'm not one of them"
That just means you haven't found the right project. I've been of the same view for several years. Finally I found a guy who wrote a paper on a software project where I read it and said, "Yes! Somebody else gets it!".
He's gone off-line since and I was never able to get in touch with him. Which probably has something to do with my government bombing the fuck out of his country. I remain hopeful he will surface again.
God save us from the righteous.
(Score: 2) by Fluffeh on Wednesday October 25 2017, @09:11PM
THe things I get involved in are always software that I use. Sometimes (actually, most times) the projects I get involved in are just the small applications that I think would be great if they just had this "one extra bit" or if they already have something, but it's hard or non-intuitive to use. I try to make things more user friendly. That's my little contribution.
It's surprising how often the original developers don't go back and make sure new additions to their code are integrated well for someone who is installing it for the first time.
(Score: 2) by hendrikboom on Thursday October 26 2017, @12:54AM
What is most lacking with most open-source/libre software is documentation. So take some software that you figured out how to use despite the documentation or lack thereof, and start to write the documentation you wish had been available.
(Score: 5, Insightful) by MrGuy on Wednesday October 25 2017, @11:24AM (1 child)
OK, if you're running a project that's only expected to be used by professional programmers, a lot of this is good advice. And don't get me wrong - despite what I'm putting in the rest of this post, I agree the world would be a better place if people asking for help on forums followed all of these pieces of advice.
But if you're building a tool that's intended for a non-professional audience (linux would be a great example, but there are plenty of allegedly "consumer-grade" open source products), then the whole approach is well-meaning but backwards. It's trying to fix the customers, instead of acknowledging the level and functionality of support that's available is a problem. Volunteer support through forums is a really poor option to have be your best support option if your product is aimed at non-techies. And most support forums have a nasty streak that's really, really offputting.
You need to look at changing how you offer support before expecting to "fix" your customers.
Most non-professional-techie consumers just want things to work. If they don't work, they want it to be as easy as possible to get to a working state. They want to simply ask someone and be told the answer. That is not an inherently unreasonable expectation.
Have you ever heard what support calls to professional software companies sound like? They violate a heck of a lot of these rules. You can get support even if you express frustration. They will help you even if you haven't exhausted every resource available to you and spent an hour trying to fix the problem on your own. They will help you even if you're new to the software. Scratch that - ESPECIALLY if you're new to the software - they won't condescend to the inexperienced.
Are people sometimes less helpful than they could be to professional support organizations? Sure. Is that a good thing? No. But not helping people because they don't fit into a box that indicated that they're "worthy" of help means you'll never rise above being a niche product.
Having to ask on a forum is already a frustrating experience - there's nobody who can help me fix the problem right now, but if I leave a message here MAYBE someone will read it and help me, but no guarantees. That is a massively frustrating thing to find is your only support option. Sure, expressing that frustration in your post is not a great idea, but it's understandable.
"Don't waste people's time?" What's "wasting their time?" When a person has a problem, it's the only problem they care about. They don't think asking for help is wasting someone's time. Should they read the FAQ first and search the forums? Sure. But it's illustrative of the problem that helping someone is seen as "wasting time" if the person in theory could have solve it themselves. The example above with the "my wifi driver doesn't work" is a great example. Hey, your computer has a Broadcom model 27b/6 wifi adapter, and there's a previous forum post from 9 months ago that said there's a known bug with those and gave steps to fix it. Oh, you didn't know your wifi adapter was a 27b/6? There's a pinned topic that tells you how to check with make and model your adapter is. You didn't read the entire forum to find those two posts, or didn't know how to search for it? You didn't try the manual workaround from 9 months ago that has you hack a bunch of deep magic settings that you don't understand and wouldn't be able to change back if you make a mistake (and possibly trash your system) on the hope that the workaround is still valid? Why are you wasting my time by daring to ask the question again?
Someone earlier mentioned "don't tell me your life story," but also said you shouldn't stop at "it doesn't work." OK, but where the stopping point is between those two extremes isn't obvious to everyone. Expecting non-technical folks to understand what's relevant and what isn't for support is basically expecting them all to BE technical. It's not clear how many details they should expect to include about what their system is and what they're trying to do. "Hi, folks, I'm hoping someone can help me with a WiFi problem. I have a two-year-old Dell Smallbook that I bought at a garage sale that had Microsoft Bob on it when I bought it. I downloaded the iso images from SuperSlack's website and I installed SuperSlack 3.0 from the CD drive. The machine booted and I successfully created a root login and personal logins for me, my girlfriend, and my cat. I can get FreeOffice and other installed apps to run. I can connect to the internet with an ethernet cable. But I can't get the wifi to work - I can see the available networks, but it won't connect to my home network."
The above almost certainly contains some irrelevant details (it used to have Microsoft Bob installed, I installed from the CD drive), and it's certainly missing some relevant ones (what's the error? Have you tried any other wifi networks?) I get why this would be a frustrating bug report. But this is how the user thinks about their problem, and expressing frustration at THEM for not editing out the "irrelevant parts" as well as not recognizing the "important parts" will 1.) turn them off from your support channel as useful, and 2.) likely turn them off from your software.
Again, offering support on forums would be a lot better experience for both the person with a problem and the person offering help if the customers were more savvy, more helpful, and knew what to ask. But EXPECTING them to do so as a precondition for them being able to get help for real problems they're having is a sure-fire way to ensure you never have a lot of non-technical users.
I'm lifting the following from Shamus Young's [shamusyoung.com] website - I think it's a pitch-perfect example of the kind of experience many people have on support forums:
ALLEN: Hi, I’m new to driving and I need to move my car back around 5 meters. How can I move the car backwards?
(2 days later.)
ALLEN: Hello? This is still a problem. I’m sure someone knows how to do this.
BOB: I can’t believe you didn’t figure this out yourself. Just take your foot off the gas and let the car roll backwards down the hill. Tap the bake when you get to where you want to be. Boom. Done.
ALLEN: But I’m not on a hill. I’m in my driveway and it’s completely flat.
CARL: Dude, I don’t know what you’re trying to accomplish, but you should never be driving backwards. It’s dangerous and will confuse the other drivers. See the big window in FRONT of you? That’s your first clue. Don’t drive backwards.
ALLEN: I’m not trying to drive backwards. I just need to move back a little bit so I can get out of my driveway and start driving forwards.
CARL: So just drive in circle until you’re pointed the right way.
ALLEN: I don’t have enough room to turn around like that. I only need to move back a few meters. I don’t understand why this has to be so hard.
CARL: Sounds like your “driveway” isn’t compatible with cars. It’s probably made for bikes. Call a contractor and have them convert some of your yard into driveway to be standards-compliant with the turning radius of a car. Either way, you’re doing something wrong.
DAVE: I see your problem. You can adjust your car to move backwards by using the shifter. It’s a stick located right between the passenger and driver seats. Apply the clutch and move the stick to the “R” position.
ALLEN: But.. I don’t have a clutch. And there isn’t a stick between the seats.
CARL: Sounds like you’re trying to drive in Europe or something.
ALLEN: Ah. Nevermind. I figured it out.
(Score: 4, Insightful) by Anonymous Coward on Wednesday October 25 2017, @05:40PM
The worst part of that exchange for me is the "ALLEN: Ah. Nevermind. I figured it out." Sure, Allen got frustrated because no one was offering him a solution so was short to kill the thread. But now when someone (maybe even him) have the same problem later, the solution isn't there. And, more frustrating sometimes, the new person with the same problem also knows that there is apparently a solution out there but they might not ever know what it is.
(Score: 5, Funny) by Anonymous Coward on Wednesday October 25 2017, @01:43PM (4 children)
Post a question.
Use another account to post an elaborate, but completely wrong answer.
Let the indignant mob pile on.
(Score: 1, Funny) by Anonymous Coward on Wednesday October 25 2017, @03:27PM (3 children)
Ah, TMB and khallow are sock puppets of Aristarchus. It all makes sense now...
(Score: 0) by Anonymous Coward on Wednesday October 25 2017, @04:46PM (2 children)
While I could see TMB and khallow being smart enough to pull of sock puppets, Aristarchus seems to be genuinely mentally unbalanced and is obviously well practiced at bringing people down to his level and then beating them with experience (to steal a phrase). And all three seem to suffer the need to have the last word (at least until it gets late enough for TMB to stop caring). With that in mind I can't imagine either of the two sock puppeting the third.
That said, I am curious to see how the three respond to both this post and its ancestors.
(Score: 1, Funny) by Anonymous Coward on Wednesday October 25 2017, @08:13PM (1 child)
Well it isn't like I'll log on to dignify this garbage post with a reply, so I guess there you have it.
(Score: 0) by Anonymous Coward on Wednesday October 25 2017, @08:32PM
I have no dog in this fight, but aristachu is a pederast, just so you know, but you knew that already.
Still, you didn't hear that from me.
(Score: 2) by Aiwendil on Wednesday October 25 2017, @10:37PM
Just split the post/question into four parts
1) Brief description/title
2) Describe what you wanted to do
3) Try to summarize what you think is relevant
4) Kitchen sink - provide everything you can think off that might be even remotely relevant (preferably as an attachment or link).
Funnily enough this is how you want manuals.
1) Brief description
2) Summary of what item is intended to do
3) Common usage and setup
4) The fun stuff including every quirk.
It allows people to skim and drop out as soon as they either found it outside their interest or when all relevant info found.
(Score: 2) by gidds on Thursday October 26 2017, @07:56AM
So, basically, someone's reinvented (or plagiarised) Eric Raymond's long-standing article [catb.org]?
[sig redacted]