How Long Should an Open Source Project Support Users?

Ubuntu Kitten writes "Since October the community-generated database of cards known to work with Ndiswrapper has been down. This is apparently due to an on-going site redesign, but right now the usual URL simply directs to a stock Sourceforge page. Without the database, the software's usability is severely diminished but this raises an interesting question: Is an open source project obliged to provide support for its users? If so, for how long should the support last? Web servers cost money, especially for popular sites. While developers can sometimes find sponsorship, is it possible to get sponsorship simply for infrastructure and user services?"

Comments

No Obligations, Take What You Can Get

[eldavojohn]

Disclaimer: These are my experiences & opinions only.

It seems you are looking for a list of cards supported by Ndiswrapper, nothing else? Is the software development not keeping up with cards or something? I'm more concerned that I can no longer access their wiki. I'm not sure how the lack of a database of cards it works with would cause its functionality to "diminish" but you are right that this raises an interesting question.

Without the database, the software's usability is severely diminished but this raises an interesting question: Is an open source project obliged to provide support for its users? If so, for how long should the support last?

No. Although from time to time I notice that Maven2's repo1 is sometimes down which irks me a bit when I'm using new packages. And that's why I have a local repository on my list--in case the bandwidth I steal from Jason van Zyl of Codehaus ever dries up. And if it should, I realize there's not a lot I can do about it ... although I can always keep downloading packages (or even building them myself) and installing them on my local network albeit tedious. I am lucky though as Maven2 is well thought out in this respect, always defaulting through a whole list of repos (indeed if repo1 went down, there are others).

I appreciate Mr. van Zyl's work and efforts but he and I have signed no prior contract guaranteeing the length of time his service should be available to me. And I, of course, expect nothing from him. He's doing me a great service at the moment but the service--though rarely spotty--doesn't have to last past this second.

Say, where's your local repository of Ndiswrapper's database?

Web servers cost money, especially for popular sites.

This is correct. And by that logic, it may benefit you to send the sourceforge developers a simple message asking them if a modest donation of funds could ail this predicament? Every so often I anonymously throw $10-$20 at a project that I use heavily, I really wish others would do the same.

While developers can sometimes find sponsorship, is it possible to get sponsorship simply for infrastructure and user services?

I'm really not sure although I do realize that if Ndiswrapper is talking to this database on the backend, there's probably no eyeballs looking at ads to the left and right of this database. Which makes it kind of hard for magical ad revenue to come in (similar to the codehaus repo1 scenario listed above). I think you'd be better off appealing to some distribution that may hinge heavily on Ndiswrapper but I'm pretty sure the developers would have exhausted these resources before letting this site lapse into oblivion.

Re:No Obligations, Take What You Can Get

[Chyeld]

Every so often I anonymously throw $10-$20 at a project that I use heavily, I really wish others would do the same.

I don't know how anything this informal and erratic can be made to work long-term.

Snowfall is informal and erratic. Chaotic and unplanned. And yet every year I manage to wake at least once to an entire world covered in snow.

Random simply means you need a large number of participants.

Re:No Obligations, Take What You Can Get

[Austerity Empowers]

Even if you did, sending money in no way puts the author into your debt such that he needs to offer support. You do it as a reward, not a binding contract.

Fundamentally, open source is centered around the design, not support. In the long run, you will need to pay for support from one of many people capable of doing so. If you see a program so many people use, which lacks the support you think you need, I hear business opportunity knocking at your door.

How Long Should Open Source Project Support Users

[Ethanol-fueled]

As long as its users support it, duh.

Project not required to provide support

[Anonymous Coward]

No, an open source project is not obliged to provide support for its users. They're giving you the software (and sometimes documentation) for free. They weren't even required to do that (even if you use GPL components you can keep your modifications to yourself as long as you don't go handing out binaries to the rest of the world).

The people responsible for the project have absolutely zero obligation to help you with anything. If they want to help, good for them (and you). If not, you have the source - read through that to figure out what it does. Or pay somebody else to do that for you.

There are companies that provide support for open source software, but unless you're paying them for it, they have no obligation to help you.

Re:Project not required to provide support

[Actually, I do RTFA]

No, an open source project is not obliged to provide support for its users. They're giving you the software (and sometimes documentation) for free.

It depends on what you mean by obligated. Certainly, legally they are under no obligation. But if you want a userbase (again, the something that other people develop with their time), testing, and relevency outside a small circle of people willing to do their own debugging, then yes, they do.

OSS cannot just take that libertarian attitude and be expected to be taken seriously. Call it one example of the "RTFM n00b"-type mentality. For most people outside of school, time is more valuble than money. So, feel free to rest on the "free as in beer" mentality, and don't get surprised when Microsoft continues to dominate the OS/Office space.

I guess, what I'm saying is, if you want users, you have to support them.

When beer is free...

[RulerOf]

When your beer is free, someone still paid for it.

The difference between purchasing software and choosing whether or not to donate to a F/OSS organization is that you choose how much the software (or service) is worth to you, should you actually decide to pay for it.

Disclaimer: I'm a huge advocate of F/OSS, just not Linux... I honestly wish my interests aligned with reality :P

Re:When beer is free...

[TheRaven64]

If I need the software, and I don't plan on selling it, then giving away copies costs me nothing. If I give away the code too, then it may help you and if you send me patches then it might benefit me too. Supporting your use of my code, however, does not benefit me in anyway, and does cost me in terms of time. If you want me to do this, then you should provide something in return, whether it's code, beer, or even bug reports or documentation.

Users of software are not automatically entitled to free support. This is true of all software, free or proprietary, although if you bought software then you may have had a small - or large, if you paid enough - amount of support included in the price you paid (again, this applies to free and proprietary software).

My biggest pet peeve...

[Blakey Rat]

Ugh. People who take down the existing page because they're redesigning the site.

Generally you only see this mistake from 14-year-old "web developers" whose qualifications all come from adding animated GIF background images to MySpace profiles. Of course, these "web developers" always severely doubt the amount of time it'll take to finish the page and put it back online, so "check back in a couple days" typically turns into months, years, or "kiss that page goodbye, sucker!" Saying the term "staging server" to these type of people will usually garner the response: "caging what? I was too busy picking my nose to listen."

If you're lucky, it was actually a hostile admin pulling down the site and holding it hostage to the project for (pinky-in-mouth) one-hundred-billion-dollars! and they didn't just recruit an incompetent idiot to run it. In the former case, at least the pages will come back once the FBI breaks down his door and holds an assault rifle to his head, in the latter case they'll be "under construction" until the end of time.

So, uh, yeah. The question here isn't "how should open source projects support users?" But more along the lines of, "should open source projects do intensely retarded things with their websites?" (The answer is no.)

Answer: no

[mcvos]

Is an open source project obliged to provide support for its users?

No.

A project itself is not obliged to do anything. In the case of non-commercial volunteer projects (which not all open source projects are), the people working on the projects aren't obliged to do anything either. And by the very nature of Open Source, even the users of the project aren't obliged to do anything (except when it's GPL and they want to distribute their own changes to the project).

Ofcourse successful Open Source projects are often very well supported. But that's because the people working on it want it to be big and not because they're under any kind of obligation.

Re:Answer: no

[Jeff Hornby]

But if that's the case, then OSS becomes worse than useless for businesses. If the software is a key component of my business it's got to continue to be available. Attitudes like "it's not my problem if my software no longer works" can only hamper the uptake of OSS. This attitude is fine if you believe that OSS should be relegated to hobbyists but the Slashdot community tends to trumpet OSS as a business solution. If there are people depending on your software, then you've made a commitment.

Re:Answer: no

[david.gilbert]

If there are people depending on your software, then you've made a commitment.

No. THEY'VE made a commitment.

Re:Answer: no

[Jeff Hornby]

When an open source project is first released, there is rarely any talk of cost. In fact the reason businesses go with open source is because the cost is lower. Now you're saying that there is a cost, and potentially a huge one. The difference being that in open source you don't know the cost until after you've been using the software. I know we joke about "the first one is always free", but is that really the sort of business model that we want? The same business model used by drug dealers and payday loans?

And before you bring out the tired old argument that the sopurce is available, you can just hire somebody, think about how much that costs. Maintaining software is expensive. Very expensive. Forking your own version of a major open source project would cost in the millions of a dollars per year. It's ludicrous to expect any commercial enterprise to do that.

Given your and many other arguments regarding lack of support for OSS, I would have to say that OSS is still far too risky for any commercial uptake. Commercial software is still the better way to go. Enterprises that have critical systems depending on OSS really need to rethink their strategies if there is such a big risk that a key component of their systems will just evaporate overnight.

Re:Answer: no

[mea37]

No, other people deciding to depend on my software does not create a commitment on my part.

It would be more correct to say, if someone decides to depend on my software without first securing a commitment from me (or from a third party capable of providing support based on the source code), that someone probably isn't a very good business-person.

So, if an individual project wants to be commercially viable, that project would be wise to think about a support model that offers its business users some assurances; but the answer to the general question "is an OSS project obligated to provide support" is still no.

Re:Answer: no

[Jeff Hornby]

And how much does this cost?

I've said it before and I'll say it again: anybody who says that a business can just hire someone to work on open source software if it becomes unsupported does not understand the first thing about the nature of business.

Re:Answer: no

[maxume]

You say, somewhere above in the thread, that people adopt open source because of cost. I don't think this is the case. I think they adopt if because of quality and value. I think this is the case for all software (That is, when using a computer saves $15,000 over not using a computer, $1,000 or even $10,000 of licensing costs will not impede the decision). So projects like the Linux kernel, Apache, Perl, Python, etc., are adopted because they deliver a great deal of value, not simply because they are cheaper than the alternatives.

In that context, OSS is a development model, not a cost model, and any decision to use software is going to include examination of the quality of the software and the quality of the support available for the software, and so on. Grouping software by the license it happens to be available under is a false argument (because no one would buy proprietary software that came with shitty support).

Re:Answer: no

[mcvos]

And how much does this cost?

Depends. When you rely completely on proprietary software that suddenly becomes unsupported, switching can become very expensive. When it's Open Source, switching is a lot less expensive, since the software isn't a black box. And in fact, it may not even be necessary to switch, because someone else can take over support of the software.

I've said it before and I'll say it again: anybody who says that a business can just hire someone to work on open source software if it becomes unsupported does not understand the first thing about the nature of business.

It's a lot easier than hiring someone to work on proprietary software that isn't yours.

OSS provides fallback solutions that proprietary software simply doesn't.

Re:Answer: no

[mcvos]

I agree. But don't argue that Redhat is not a commercial company.

Why would anyone argue that?

I see no difference from a commercial perspective between deploying RedHat Linux to my servers and deploying Windows to my servers.

Then I'll explain the difference to you. If Microsoft ever decides not to support the Windows version on your servers, you've got a problem. You're at their mercy. And don't think they'll never do that; Microsoft has abandoned services that people paid for. Those people are now out of luck, lost their money, and don't have any alternative.

If RedHat ever decides not to support the OS on your servers, chances are someone else is willing to do it. They might not be as good as RedHat, or they might be more expensive. Or maybe your own server administrator knows enough about it to do it. In any case, you've got options.

Re:Answer: no

[bankman]

And how much does this cost?

I've said it before and I'll say it again: anybody who says that a business can just hire someone to work on open source software if it becomes unsupported does not understand the first thing about the nature of business.

...or software development.

Re: How Long Should Open Source Project Support Us

[morgan_greywolf]

Exactly. If you want support from an open source project, you need to help that project out. Whether that's in the form of development work, testing, documentation writing, helping uses in the forums or lists out, or good old fashioned cash depends on what the project needs. Most projects are more than happy to list what they need, and if they don't, e-mail the project's lead(s) or e-mail their support list -- they'll be very happy to hear from you.

You get out of it what you put into it. Like anything else in life.

Re:Uh...No.

[mlwmohawk]

One of the hazards of the trade is that some software may cease to be supported. This goes double for OSS, where the developers are often unpaid.

I think your premise is faulty. It has been my experience that commercial products become unsupported far faster than open source projects. Of course, with all aggregate generalizations there are specific instances that counter the general trend, but I think it is safe to say that you are safer banking on open source support than you are commercial support for a few reasons:

(1) As mentioned, generally speaking, support is longer term with open source.
(2) Unlike proprietary solutions, the code is generally available, it is less likely that a useful project will ever *really* become unsupported.
(3) If it is a marginal project, you have the source, you can pay someone to support you.

What kind of question is that?

[qoncept]

"Is an open source project obliged to provide support for its users?"

Of course not. If I give you a car, are you going to expect me to change the oil in it every time its due? Sure, people that spend their time developing software for free may be inclined to help you out to an extent, but they don't owe you anything.

Take the issue I found in Pidgin. It was crashing seemingly randomly, and debugging showed it had something to do with playing sounds. I opened a ticket, someone marked it as an actual defect, and 14 days later, since no one had looked at the ticket again, it automatically closed. Annoying, but I still have a Windows XP disc laying around somewhere (for which there are a number of IM clients that run just fine for me).

One of the major OSS problems

[abigsmurf]

There is a view with OSS that "you should be greatful with what others have done otherwise code it yourself" Which to some extent is true. You should be thankful that these hubs and support sites are provided or supported by the authors.

Unfortunately this can only run so far. If you're a business and you've spent 100 hours installing a piece of software across a network only to find updates and support drops a week later, that can work out to be very expensive.

Likewise if you're a student and a paper is due but you can't complete it due to a bug/error and the support section for the program you've used no longer exists, it's a big issue.

This is even more of a problem if there is a leading OSS solution that is so well known, no one wants to write competing software for it so when development and support stops, there's a gaping vaccuum in that area.

Open Source has to compete with commercial software and usually commercial companies will give you support for the lifespan of a product or until it becomes obsolete (not always, companies go bust, get taken over etc.). It's no good software being free if lack of support means you waste a fortune on wages trying to fix issues.

Two possible solutions: OSS developers give in and run ads on their sites (it's not hard to find unobstrusive ads with acceptable rates nowadays) or owners of sites are given incentives to hand over control of their sites to a central OSS archive where you can at least get snapshots of support forums and wikis, as well as the downloads and source.

Re:One of the major OSS problems

[JoeMerchant]

If this old guy who enjoys piloting his boat provides free ferry rides across the river, you can certainly ride the ferry for free.

You could even take a job on the other side of the river and take out a mortgage on a new house based on that job, that is incidentally dependent on the old guy and his boat.

If the old guy decides that he's going to take the month off and do some traveling in his boat, is it really his problem that you have become dependent upon his generosity? Or that you don't have the resources or skill to purchase and operate your own boat? Even if you chipped in for gas once in awhile, I don't think he really has an obligation to ferry you across the river twice a day, five days a week, just because he used to do it and you are incapable without him.

I can see some judges attempting to press the old guy and his boat into service because he has become a vital component of the local economy, but true justice would be if the users of the service were made to pay the whole cost of providing the service, including providing the necessary labor.

And... when the old guy dies, you truly are up the creek....

Re:One of the major OSS problems

[serviscope_minor]

There is a view with proprietary software that "you can buy it and all your problems will be easily solved" Which to some extent is true. You should be thankful that these hubs and support sites are provided or supported by the software companies.

Unfortunately this can only run so far. If you're a business and you've spent 100 hours installing a piece of software across a network only to find updates and support drops a week later, that can work out to be very expensive.

Likewise if you're a student and a paper is due but you can't complete it due to a bug/error and the support section for the program you've used no longer exists, it's a big issue.

This is even more of a problem if there is a leading program solution that is so well known, no one wants to write competing software for it so when development and support stops, there's a gaping vaccuum in that area.

Commercial software has to compete with OSS and usually the authors and for-profit support companies will give you support for the lifespan of a product or until it becomes obsolete (not always, companies go bust, get taken over, authors move on etc.). It's no good paying for software if lack of support means you waste a fortune on software and cannot fix issues.

I was going to rewrite your post to hilight the downsides of non OSS. I expected it to be much more work. I needed to to little more than swap OSS for proprietary software. Frankly yaor arguments for the downsides of OSS apply for the most part equally to CSS. Basically, the packages are cheap/free but worthwhile support is hard to get and/or very expensive.

Re:One of the major OSS problems

[Eil]

There is a view with OSS that "you should be greatful with what others have done otherwise code it yourself" Which to some extent is true. You should be thankful that these hubs and support sites are provided or supported by the authors.

This isn't "to some extent true," it's all the way true. This is the problem I have with people bashing Linux distributions and other open source software because they contain some bugs or don't have some ability that they think they need. I've said it before and I'll say it again:

The defining difference between open source and commercial software is that you pay for it with effort rather than money. There are lots of open source programs out there that you can install on your computer for free and they just work and that's the end of the story. But this doesn't mean you're entitled to perfect support or a bug-free experience every time. If you run across a bug or a problem somewhere, you're expected to help fix the problem even if it means submitting a bug report or simply asking for help on a mailing list or forum. And that's just the minimum requirement. A thriving OSS project also needs people to contribute new features, non-critical bugfixes, and documentation so that others can have an even better experience with the software.

Unfortunately as OSS creeps further into the mainstream there seem to be an increasing number of users who expect everything for free. If the open source community should be doing anything differently, it's communicating the message to the public that yes, the software you are getting is free of charge but if you encounter a problem, you need to help us help you solve it, not write some whiny post on your blog (or a Slashdot story submission like this one) bashing the project publicly to gain attention.

Unfortunately this can only run so far. If you're a business and you've spent 100 hours installing a piece of software across a network only to find updates and support drops a week later, that can work out to be very expensive.

There are a number of things wrong with this example. The first is that you have to carefully evaluate the software you use as the base of your system regardless of whether it's OSS or proprietary. The second is that an OSS project gives you or anyone else the ability to pick up where the original developers left off because the code is open and free. With a proprietary vendor, you don't get this option and I've seen plenty of businesses left in the lurch because the vendor decided to suddenly stop supporting a piece of very expensive proprietary software that the company had formed their entire business around. Third, no software that I'm aware of requires weekly updates in order to run correctly.

Likewise if you're a student and a paper is due but you can't complete it due to a bug/error and the support section for the program you've used no longer exists, it's a big issue.

Again it comes down to choosing the right software for the job. Any worthwhile piece of open source software has a community around it that takes over if the main developer decides to move on to something else or lock up the code. Open source software support almost never vanishes into thin air. Proprietary software support often does.

This is even more of a problem if there is a leading OSS solution that is so well known, no one wants to write competing software for it so when development and support stops, there's a gaping vaccuum in that area.

Yet again, support doesn't stop for an OSS project as long as there's a community around it. This goes double if it's a "leading" OSS solution. You're manufacturing doomsday scenarios where none are even remotely likely to occur.

Open Source has to compete with commercial software and usually commercial companies will give you support for the lifespan of a product or until it becomes obsole

Re:One of the major OSS problems

[aukset]

If you're a business and you've spent 100 hours installing a piece of software across a network only to find updates and support drops a week later, that can work out to be very expensive.

You're also a very stupid business. It would be the exact same thing if you're a business and purchase proprietary software without a support contract in place. When that developer goes bust next week or decides to end of life that product, there isn't a single thing you can do about it.

Likewise if you're a student and a paper is due but you can't complete it due to a bug/error and the support section for the program you've used no longer exists, it's a big issue.

That is rather far-fetched. There are so many varieties of word processing and typesetting programs out there, many of them using open formats, that you will still have choices. What if Word starts crashing? You might find a workaround, you might be able to open the proprietary file in another word processor, but MS is under no obligation to provide you with support unless you pay for it.

Simply, F/OSS is not a business model, it is a development model. It is used to create a software product, a product which is provided as-is (source included), and is no different from proprietary software products provided as-is. Support is secondary in both cases, and paid for separately. Lack of support is not inherent or unique to OSS products.

Re: How Long Should Open Source Project Support Us

[Binder]

Open source projects don't support users... they are the users.

If the main groups no longer wishes to participate in the project then other users need to step up.
This is one of the greatest things about OSS.

Software for the Users, by the Users!

Re: How Long Should Open Source Project Support Us

[westlake]

Corporations certainly do their bit, but open source is about bazaars, not cathedrals.

Is it really? I sometimes wonder. The marque projects of open source - OpenOffice.org and Firefox, for example - look corporate to my eyes. The Dirac video codec emerged from the BBC, and you can't get more high church than that.

The Question Is Wrong

[Rantastic]

This has nothing to do with Open Source. The question should be "Should projects that give software away for free be obligated to provide support?"

There is plenty of closed source software that can be downloaded for free. There is plenty of open source software that can be purchased with support.

The answer, by the way, is no. Just because software is free does not mean that the makes of it are obliged to give you support. Support costs money. Businesses who use software (open or closed source) pay for support, either through a support vendor or in house talent.

Re: How Long Should Open Source Project Support Us

[SkunkPussy]

I'm not 100% sure what "marque" means but I would suggest that gcc is the marque project of open source and has been for about 20 years.

Re: How Long Should Open Source Project Support Us

[wisty]

Mozilla is kind of corporate, but may have ended up crappy abandonware it were not for Dave Hyatt and Blake Ross forking it (and all the wonderful bazaar of contributors adding spell checkers, pimped skins and other wonderful ... crap). Forks only happen in open source. OpenOffice is the best open source office suite, and I personally like it better than MS office (yey for math type), but it's not that great.

If you want to talk about marque projects, lets stick to ones that blow the competition out of the water. SVN, emacs, git, gcc, Python, perl, Ruby, Apache, LaTex, OpenSSH, and so on are certainly good projects.

If you want consumer products, look at BitTorrent (which is no longer open, but there are forks). Look at RSS readers. Look at web forums and chat servers and that those DVD players that don't make you watch the stupid anti-piracy ads (as a way of thanking you for buying a real copy...).

Re: How Long Should Open Source Project Support Us

[wisty]

Problem is that ego-boo works better with programmers showing off to programmers. Spanish speakers don't necessarily crave the praise of programmers, and too many programmers don't want to thank the technical writers.

Re: How Long Should Open Source Project Support Us

[grizdog]

I don't agree. I think there is an obligation to do something, but not necessarily for that project. If you are putting in 30 hours a week, or more, to one widely used open source project, there is no need for you to support other ones, no matter how much you use them

Some people would go further and say if you are donating your time to any worthy cause, it absolves you of the responsibility to provide any support to F/OSS. That water is a little murky for my taste, especially when deciding what causes are "worthy", e.g., a political campaign, or converting people to your chosen religion. But I view F/OSS as one community, and while we all have to be involved, we don't have to be involved in every project that we use.

Exactly. If you want support from an open source project, you need to help that project out. Whether that's in the form of development work, testing, documentation writing, helping uses in the forums or lists out, or good old fashioned cash depends on what the project needs. Most projects are more than happy to list what they need, and if they don't, e-mail the project's lead(s) or e-mail their support list -- they'll be very happy to hear from you.

You get out of it what you put into it. Like anything else in life.

Re: How Long Should Open Source Project Support Us

[morgan_greywolf]

Python is a consumer product, as well as a product for seasoned developers. One of Python's design goals is to make coding easier for newbie coders. I have to say that I've seen more non-coders pick up Python easily and readily -- more than any other language.

Re: How Long Should Open Source Project Support Us

[serviscope_minor]

s it really? I sometimes wonder. The marque projects of open source - OpenOffice.org and Firefox, for example - look corporate to my eyes.

What about Linux itself?

Re: How Long Should Open Source Project Support Us

[Blakey Rat]

I used to help out via entering bug reports for bugs I found. Invariably, the bug report would either get a gruff, unhelpful reply (like: http://sourceforge.net/tracker/?func=detail&aid=1865630&group_id=95717&atid=612382 ), or it would simply be ignored for months and months until the project either closed it due to inactivity or switching bug trackers (like this ex-bug: http://sourceforge.net/tracker/index.php?func=detail&aid=1609779&group_id=93438&atid=604306 ).

I don't bother anymore.