The Linux User Group HOWTO is intended to serve as a guide to founding, maintaining, and growing a GNU/Linux user group.
GNU/Linux is a freely-distributable implementation of Unix for personal computers, servers, workstations, PDAs, and embedded systems. It was developed on the i386 and now supports a huge range of processors from tiny to colossal:
Note: The following supported-platforms list is not serious documentation. The point is merely to illustrate the breadth of Linux's reach. If seriously interested in the subject of Linux ports, please see also Xose Vazquez Perez's Linux ports page and Jerome Pinot's Linux architectures list (static mirrors, as both pages vanished in 2005), if only because hardware support is more complex than just generic CPU functionality, encompassing support for myriad bus variations and other subtle hardware issues (especially for Linux PDA / embedded / microcontroller / router ports).
Note that some items listed were probably one-time forks, little or not at all maintained since creation. On some of the rarer architectures, NetBSD may be more practical. (The Debian GNU/kFreeBSD port should also be solid enough to serve as a compromise option, furnishing GNU/Linux userspace code on the high performance / high stability FreeBSD kernel, and Dyson or another Illumos distribution can provide something similar on the OpenSolaris kernel.)
If you want to learn more, the Linux Documentation Project is a good place to start.
For general information about computer user groups, please see the Association of PC Users Groups.
To fully appreciate LUGs' (Linux User Groups') role in the GNU/Linux movement, it helps to understand what makes GNU/Linux unique.
GNU/Linux as an operating system is powerful -- but GNU/Linux as an idea about software development is even more so. GNU/Linux is a free operating system: It's licensed under the GNU General Public Licence (and other open source / free software licences -- though proprietary application software is sometimes also included in particular packagings). Thus, source code is freely available in perpetuity to anyone. It's maintained by a unstructured group of programmers world-wide, under technical direction from Linus Torvalds and other key developers. GNU/Linux as a movement has no central structure, bureaucracy, or other entity to direct its affairs. While this situation has advantages, it poses challenges for allocation of human resources, effective advocacy, public relations, user education, and training.
(This HOWTO credits the Free Software Foundation's GNU Project as the crucial motive force behind creating and furthering a free aka open source integrated system. Thus, it refers to "distributions" comprising the GNU operating system atop the Linux kernel as "GNU/Linux". Yes, the term is awkward, and FSF's request for credit isn't widely honoured; but the justice of FSF's claim is obvious.)
(This HOWTO's maintainer is also fully aware that the world at large will never adopt this usage, justice notwithstanding. If it seems mannered, please indulge him, and respect the gesture.)
GNU/Linux's loose structure is unlikely to change. That's a good thing: It works precisely because people are free to come and go as they please: Free programmers are happy programmers are effective programmers.
However, this loose structure can disorient the new user: Whom does she call for support, training, or education? How does she know what GNU/Linux is suitable for?
In part, LUGs provide the answers, which is why LUGs have been vital to the movement: Because your town, village, or metropolis sports no Linux Corporation "regional office", the LUG takes on many of the same roles a regional office does for a large multi-national corporation.
GNU/Linux is unusual in neither having nor being burdened by central structures or bureaucracies to allocate its resources, train its users, and support its products. These jobs get done through diverse means: the Internet, consultants, VARs, support companies, colleges, and universities. However, increasingly, in many places around the globe, they are done by a LUG.
Computer user groups are not new. In fact, they were central to the personal computer's history: Microcomputers arose in large part to satisfy demand for affordable, personal access to computing resources from electronics, ham radio, and other hobbyist user groups. Giants like IBM eventually discovered the PC to be a good and profitable thing, but initial impetus came from the grassroots, leading to groundbreaking efforts like SHARE (1955-present) and DECUS (1961-2008).
In the USA, user groups have changed -- many for the worse -- with the times. The financial woes and dissolution of the largest user group ever, the Boston Computer Society, were well-reported; but, all over the USA, most PC user groups have seen memberships decline. American user groups in their heyday produced newsletters, maintained shareware and diskette libraries, held meetings and social events, and, sometimes, even ran electronic bulletin board systems (BBSes). With the advent of the Internet, however, many services that user groups once provided migrated to things like CompuServe and the Web.
GNU/Linux's rise, however, coincided with and was intensified by the general public "discovering" the Internet. As the Internet grew more popular, so did GNU/Linux: The Internet brought new users, developers, and vendors. So, the same force that sent traditional user groups into decline propelled GNU/Linux forward, and inspired new groups concerned exclusively with it.
To give just one indication of how LUGs differ from traditional user groups: Traditional groups must closely monitor what software users redistribute at meetings. While illegal copying of restricted proprietary software certainly occurred, it was officially discouraged -- for good reason. At LUG meetings, however, that entire mindset simply does not apply: Far from being forbidden, unrestricted copying of GNU/Linux should be among a LUG's primary goals. In fact, there is anecdotal evidence of traditional user groups having difficulty adapting to GNU/Linux's ability to be lawfully copied at will.
(Caveat: A few distributions bundle GNU/Linux with proprietary software packages whose terms don't permit public redistribution. Check licence terms, if in doubt. Offers or requests to copy distribution-restricted proprietary software of any sort should be heavily discouraged anywhere in LUGs, and declared off-topic for all GNU/Linux user group on-line forums, for legal reasons.)
Since around 2003, LUGs in developed countries have seen a decline similar to that of traditional user groups. The causes can be debated, and might include:
A few time-tested tips for averting LUG flameout:
For the GNU/Linux movement to grow, among other requirements, LUGs must proliferate and succeed. Because of GNU/Linux's unusual nature, LUGs must provide some of the same functions a "regional office" provides for large computer corporations like IBM, Microsoft, and Sun. LUGs can and must train, support, and educate users, coordinate consultants, advocate GNU/Linux as a computing solution, and even serve as liaison to local news outlets.
Since this document is meant as a guide not only to maintaining and growing LUGs but also to founding them, we should, before going further, discuss what LUGs already exist.
There are several LUG lists on the Web. If you are considering founding a LUG, your first task should be to find any nearby existing LUGs. Your best bet may be to join a LUG already established in your area, rather than founding one.
As of 2016, there are LUGs in 43 US states, seven of Canada's ten provinces, all six of Australia's states plus the Australian Capital Territory, in 76 locations in India, and over 100 other countries, including Russia, China, most of Western and Eastern Europe, and many parts of Africa, Asia, South America, Central America, Oceania, and the Caribbean. (This does not include Linux groups internal to Meetup.com.)
While (most) LUG lists on the Web are well-maintained, likely they don't list every LUG. If considering founding a LUG, I suggest, in addition to consulting these lists, posting a message to comp.os.linux.announce, comp.os.linux.misc, or an appropriate regional Usenet hierarchy, inquiring about nearby LUGs. You should also lodge a query (mailing list post, comment during a meeting) at any existing LUG you are aware of anywhere near your area, about LUGs near you. If no such (nearby) LUG exists, your postings will alert potential members to your initiative.
Carefully balance convenience against solidarity: If a LUG exists in your metropolitan area but on the other side of the city, starting a new group may be better for convenience's sake. On the other hand, joining the other group may be better for reasons of unity and solidarity. Greater numbers almost always means greater power, influence, and efficiency. While two groups of 100 members each might be nice, one with 200 has advantages. Of course, if you live in a small town or village, any group is better than none.
The point is that starting a LUG is a significant undertaking, which should be commenced with all relevant facts and some appreciation of the effect on other groups.
LUGs' goals are as varied as their locales. There is no LUG master plan, nor will this document supply one. Remember: GNU/Linux is free from bureaucracy and centralised control; so are LUGs.
It is possible, however, to identify a core set of goals for a LUG:
Each LUG combines these and other goals uniquely, according to its membership's needs.
The urge to advocate the use of GNU/Linux is widely felt. When you find something that works well, you want to tell as many people as you can. LUGs' role in advocacy cannot be overestimated, especially since wide-scale commercial acceptance is only newly underway. While it is certainly beneficial to the movement, each and every time a computer journalist writes a positive review of GNU/Linux, it is also beneficial every time satisfied GNU/Linux users brief their friends, colleagues, employees, or employers.
There is effective advocacy, and there is ineffective carping: As users, we must be constantly vigilant to advocate GNU/Linux in such a way as to reflect positively on the product, its creators and developers, and our fellow users. The Linux Advocacy HOWTO, available at the Linux Documentation Project, gives some helpful suggestions, as does Don Marti's excellent Linuxmanship essay. Suffice it to say that advocacy is important to a LUG's mission.
A time may come when advocacy is irrelevant, because GNU/Linux has more or less won the day, when the phrase "no one ever got fired for using Linux" becomes reality. Until then, LUGs play a vital role in promoting GNU/Linux use. They do so because their advocacy is free, well-intentioned, and backed up by organisational commitment. If a person encounters GNU/Linux through a LUG's efforts, then that new user's already ahead of the game: She knows of an organisation that will help her install, configure, and even maintain GNU/Linux on whatever computers she's willing to dedicate to it.
New users already in contact with a LUG are ahead of others whose interest in GNU/Linux has been piqued by a computer journalist, but who have no one to whom to turn for aid in their quest to install, run, and learn GNU/Linux.
It is, therefore, important for LUGs to advocate GNU/Linux, because their advocacy is effective, well-supported, and free.
Advocacy can be mis-aimed; advocacy can go wrong and be counterproductive; advocacy can be simply inappropriate in the first place. The matter merits careful thought, to avoid wasted time or worse.
Many attempts at advocacy fail ignominiously because the advocate fails to listen to what the other party feels she wants or needs. (As Eric S. Raymond says, "Appeal to the prospect's interests and values, not to yours.") If that person wants exactly the proprietary-OS setup she already has, then advocacy wastes your time and hers. If her stated requirements equate exactly to MS-Project, MS-Visio, and Outlook/Exchange groupware, then trying to "sell" her what she doesn't want will only annoy everyone (regardless of whether her requirements list is real or artificial). Save your effort for someone more receptive.
Along those lines, bear in mind that, for many people, perhaps most, an "advocate" is perceived as a salesman, and thus classified as someone to resist rather than listen to fairly. They've never heard of someone urging them to adopt a piece of software without benefiting materially, so they assume there must be something in it for you and will push back, and act as if they're doing you a personal favour to even listen, let alone try your recommendations.
I recommend bringing such discussions back to Earth immediately, by pointing out that software policy should be based in one's own long-term self interest, that you have zero personal stake in their choices, and that you have better uses for your time than speaking to an unreceptive audience. After that, if they're still interested, at least you won't face the same artificial obstacle.
At the same time, make sure you don't live up to the stereotype of the OS advocate, either. Just proclaiming your views at someone without invitation is downright rude and offensive. Moreover, when done concerning GNU/Linux, it's also pointless: Unlike the case with proprietary OSes, our OS will not live or die by the level of its acceptance and release/maintenance of ported applications. It and all key applications are open source: the programmer community that maintains it is self-supporting, and would keep it advancing and and healthy regardless of whether the business world and general public uses it with wild abandon, only a little, or not at all. Because of its open-source licence terms, source code is permanently available. GNU/Linux cannot be "withdrawn from the market" on account of insufficient popularity, or at the whim of some company. Accordingly, there is simply no point in arm-twisting OS advocacy -- unlike that of some OS-user communities we could mention. (Why not just make information available for those receptive to it, and stop there? That meets any reasonable person's needs.)
Last, understand that the notion of "use value" for software is quite foreign to most people -- the notion of measuring software's value by what you can do with it. The habit of valuing everything at acquisition cost is deeply ingrained. In 1996, I heard a young fellow from Caldera Systems speak at a Berkeley, California LUG about the origins of Caldera Network Desktop (the initial name of their GNU/Linux distribution) in Novell, Inc.'s "Corsair" desktop-OS project: In surveying corporate CEOs and CTOs, they found corporate officers to be inherently unhappy with anything they could get for free. So, Caldera offered them a solution -- by charging money.
Seen from this perspective, being conservative about the costs and difficulties of GNU/Linux deployments helps make them positively attractive -- and protects your credibility as a speaker. Even better would be to frame the discussion of costs in terms of the cost of functionality (e.g., 1000-seat Internet-capable company e-mail with offline-user capability and webmail) as opposed to listing software as a retail-style line-item with pricing: After all, any software project has costs, even if the acquisition price tag is zero, and the real point of open source isn't initial cost but rather long-term control over IT -- a key part of one's operations: With proprietary systems, the user (or business) has lost control of IT, and is on the wrong side of a monopoly relationship with one's vendor. With open source, the user is in control, and nobody can take that away. Explained that way (as opportunity to reduce and control IT risk), people readily understand the difference -- especially CEOs -- and it's much more significant over the long term than acquisition cost.
Not only is it the business of a LUG to advocate GNU/Linux usage, but also to train members, as well as the nearby computing public, to use our OS and associated components -- a goal that can make a huge real-world difference in one's local area. While universities and colleges are increasingly including GNU/Linux in their curricula, for sundry reasons, this won't reach some users. For those, a LUG can give basic or advanced help in system administration, programming, Internet and intranet technologies, etc.
In an ironic twist, many LUGs have turned out to be a backbone of corporate support: Every worker expanding her computer skills through LUG participation is one fewer the company must train. Though home GNU/Linux administration doesn't exactly scale to running corporate data warehouses, call centres, or similar high-availability facilities, it's light years better preparation than MS-Windows experience. As Linux has advanced into journaling filesystems, high availability, real-time extensions, and other high-end Unix features, the already blurry line between GNU/Linux and "real" Unixes has been increasingly vanishing.
Not only is such education a form of worker training, but it will also serve, as information technology becomes increasingly vital to the global economy, as community service: In the USA's metropolitan areas, for example, LUGs have taken GNU/Linux into local schools, small businesses, community and social organisations, and other non-corporate environments. This accomplishes the goal of advocacy and also educates the general public. As more such organisations seek Internet presence, provide their personnel dial-in access, or other GNU/Linux-relevant functions, LUGs gain opportunities for community participation, through awareness and education efforts -- extending to the community the same generous spirit characteristic of GNU/Linux and the free software / open source community from its very beginning. Most users can't program like Torvalds, but we can all give time and effort to other users, the GNU/Linux community, and the broader surrounding community.
GNU/Linux is a natural fit for these organisations, because deployments don't commit them to expensive licence, upgrade, or maintenance fees. Being technically elegant and economical, it also runs very well on cast-off corporate hardware that non-profit organisations are only too happy to use: The unused Pentium III in the closet can do real work, if someone installs GNU/Linux on it.
In addition, education assists other LUG goals over time, in particular that of support: Better education means better support, which in turn facilitates education, and eases the community's growth. Thus, education forms the entire effort's keystone: If only two or three percent of a LUG assume the remainder's support burden, that LUG's growth will be stifled. One thing you can count on: If new and inexperienced users don't get needed help from their LUG, they won't participate there for long. If a larger percentage of members support the rest, the LUG will not face that limitation. education -- and, equally, support for allied projects such as the Apache Web server, X.org, Freedesktop.org, TeX, LaTeX, etc. -- is key to this dynamic: Education turns new users into experienced ones.
Finally, GNU/Linux is a self-documenting operating environment: In other
words, writing and publicising our community's documentation is up to
us. Therefore, make sure LUG members know of the
Linux Documentation Project and its worldwide
mirrors. Consider operating an LDP mirror site. Also, make sure to
publicise -- through
comp.os.linux.announce, the LDP, and other
pertinent sources of information -- any relevant documentation the LUG
develops: technical presentations, tutorials, local FAQs, etc. LUGs'
documentation often fails to benefit the worldwide community for no
better reason than not notifying the outside world. Don't let that
happen: It is highly probable that if someone at one LUG had a question
or problem with something, then others elsewhere will have it, too.
Of course, for the newcomer, the primary role of a LUG is GNU/Linux support -- but it is a mistake to suppose that support means only technical support for new users. It should mean much more.
LUGs have the opportunity to support:
New users' most frequent complaint, once they have GNU/Linux installed, is the steep learning curve characteristic of all modern Unixes. (That sentence was true in 1997 when this HOWTO's first maintainer wrote it, but happily not much any more.) With that learning curve, however, comes the power and flexibility of a real operating system. A LUG is often the a new user's main resource to flatten the learning curve.
During GNU/Linux's first decade, it gained some first-class journalistic resources, which should not be neglected: The main (surviving) monthly magazine of longest standing is Linux Journal (USA). More recently, they've been joined by Linux Format (UK), LinuxUser and Developer (UK), Linux Magazine (German publishing firm; publishes in English, German, Polish, Brazilian Portuguese, and Spanish; North American edition is named Linux Pro Magazine), Open Source For You (India; formerly LINUX For You), Full Circle (international; covers Ubuntu family distributions), Linux Voice (UK), easyLinux (German), LinuxUser (German), and Ubuntu User (German publishing firm; in English).
Standout on-line magazines and news sites with weekly or better publication cycles include Linux Weekly News, DistroWatch Weekly, Linux Today, and FreshNews.
All of these resources have eased LUGs' job of spreading essential news and information -- about bug fixes, security problems, patches, new kernels, etc., but new users must still be made aware of them, and taught that the newest kernels are always available from kernel.org, that the Linux Documentation Project has newer versions of Linux HOWTOs than do DVD/CD-based GNU/Linux distributions, and so on.
Intermediate and advanced users also benefit from proliferation of
timely and useful tips, facts, and secrets. Because of the GNU/Linux
world's manifold aspects, even advanced users often learn new tricks or
techniques simply by participating in a LUG. Sometimes, they learn of
software packages they didn't know existed; sometimes, they just
vi command sequences they've not used since
LUGs can help consultants find their customers and vice-versa, by providing a forum where they can come together. Consultants also aid LUGs by providing experienced leadership. New and inexperienced users gain benefit from both LUGs and consultants, since their routine or simple requests for support are handled by LUGs gratis, while their complex needs and problems -- the kind requiring paid services -- can be fielded by consultants found through the LUG.
The line between support requests needing a consultant and those that don't is sometimes indistinct; but, in most cases, the difference is clear. While a LUG doesn't want to gain the reputation for pawning new users off unnecessarily on consultants -- as this is simply rude and very anti-GNU/Linux behaviour -- there is no reason for LUGs not to help broker contacts between users needing consulting services and professionals offering them.
Caveat: While "the difference is clear" to intelligent people of goodwill, the Inevitable Ones are also always with us, who act willfully dense about the limits of free support when they have pushed those limits too far. Remember, too, my earlier point about the vast majority of the population valuing everything at acquisition cost (instead of use value), including what they receive for free. This leads some, especially some in the corporate world, to use (and abuse) LUG technical support with wild abandon, while simultaneously complaining bitterly of its inadequate detail, insufficient promptness, supposedly unfair expectations that the user learn and not re-ask minor variations on the same question endlessly, etc. In other words, they treat relations with LUG volunteers the way they would a paid support vendor, but one they treat with zero respect because of its zero acquisition cost.
In the consulting world, there's a saying about applying "invoice therapy" to such behaviour: Because of the value system alluded to above, if your consulting advice is poorly heeded and poorly used, it just might be the case that you need to charge more. By contrast, the technical community has often been characterised as a "gift culture", with a radically different value system: Members gain status through enhanced reputation among peers, which in turn they improve through visible participation: code, documentation, technical assistance to the public, etc.
Clash between the two very different value-based cultures is inevitable and can become a bit ugly. LUG activists should be prepared to intercede before the ingrate newcomer is handed her head on a platter, and politely suggest that her needs would be better served by paid (consultant-based) services. There will always be judgement calls; the borderline is inherently debatable and a likely source of controversy.
Telltale signs that a questioner may need to be transitioned to consulting-based assistance include:
In general, LUG members are especially delighted to help, on a volunteer basis, members who seem likely to participate in the "gift culture" by picking up its body of lore and, in turn, perpetuating it by teaching others in their turn. Certainly, there's nothing wrong with having other priorities and values, but such folk may in some cases be best referred to paid assistance, as a better fit for their needs.
An additional observation that may or may not be useful, at this point: There are things one may be willing to do for free, to assist others in the community, that one will refuse to do for money: Shifting from assisting someone as a volunteer fundamentally changes the relationship. A fellow computerist who suddenly becomes a customer is a very different person; one's responsibilities are quite different, and greater. You're advised to be aware, if not wary, of this distinction.
Please see Joshua Drake's Linux Consultants Guide for an international list of GNU/Linux consultants. (Note: As of 2016, Drake's list was last updated in 2006.)
LUGs also have the opportunity to support local businesses and organisations. This support has two aspects: First, LUGs can support businesses and organisations wanting to use our OS (and its applications) as a part of their computing and IT efforts. Second, LUGs can support local businesses and organisations developing software for GNU/Linux, cater to users, support or install distributions, etc.
The support LUGs can provide to local businesses wanting to use GNU/Linux as a part of their computing operations differs little from the help LUGs give individuals trying GNU/Linux at home. For example, compiling the Linux kernel doesn't really differ. Supporting businesses, however, may require supporting proprietary software -- e.g., the Oracle, Sybase, and DB2 databases (or VMware, CrossOver Linux, and such things). Some LUG expertise in these areas may help businesses make the leap into GNU/Linux deployments.
This leads us directly to the second kind of support a LUG can give to local businesses: LUGs can serve as a clearinghouse for information available in few other places. For example:
Maintaining and making this kind of information public not only helps the LUG members, but also helps friendly businesses and encourages them to continue to be GNU/Linux-friendly. It may even, in some cases, help further a competitive environment in which other businesses are encouraged to follow suit.
Finally, LUGs may also support the movement by soliciting and organising charitable giving. Chris Browne has thought about this issue as much as anyone I know, and he contributes the following:
A further involvement can be to encourage sponsorship of various GNU/Linux-related organisations in a financial way. With the multiple millions of users, it would be entirely plausible for grateful users to individually contribute a little. Given millions of users, and the not-unreasonable sum of a hundred dollars of "gratitude" per user ($100 being roughly the sum not spent this year upgrading a Microsoft OS), that could add up to hundreds of millions of dollars towards development of improved GNU/Linux tools and applications.
A user group can encourage members to contribute to various "development projects". Having some form of "charitable tax exemption" status can encourage members to contribute directly to the group, getting tax deductions as appropriate, with contributions flowing on to other organisations.
It is appropriate, in any case, to encourage LUG members to direct contributions to organisations with projects and goals they individually wish to support.
This section lists possible candidates. None is being explicitly recommended here, but the list represents useful food for thought. Many are registered as charities in the USA, thus making US contributions tax-deductible.
Here are organisations with activities particularly directed towards development of software working with GNU/Linux:
Contributions to these organisations have the direct effect of supporting creation of freely redistributable software usable with GNU/Linux. Dollar for dollar, such contributions almost certainly yield greater benefit to the community than any other kind of spending.
There are also organisations less directly associated with GNU/Linux, that may nonetheless be worthy of assistance, such as:
Based in San Francisco, EFF is a donor-supported membership organization working to protect our fundamental rights regardless of technology; to educate the press, policy-makers, and the general public about civil liberties issues related to technology; and to act as a defender of those liberties. Among our various activities, EFF opposes misguided legislation, initiates and defends court cases preserving individuals' rights, launches global public campaigns, introduces leading edge proposals and papers, hosts frequent educational events, engages the press regularly, and publishes a comprehensive archive of digital civil liberties information at one of the most linked-to Web sites in the world.
The TeX Users Group (TUG) is working on the "next generation" version of the LaTeX publishing system, known as LaTeX3. GNU/Linux is one of the platforms on which TeX and LaTeX are best supported.
Donations for the project can be sent to:
TeX Users Group c/o Robin Laakso, executive director TeX Users Group PO Box 2311 Portland, OR 97208-2311
Alternatively, donations can be made online.
Project Gutenberg's purpose is to make freely available in electronic form the texts of public-domain books. This isn't directly a "Linux thing", but seems fairly worthy, and they actively encourage platform independence, which means their "products" are quite usable with GNU/Linux.
Project Runeberg is similar to Project Gutenberg, except concentrating on making editions of classic Nordic (Scandinavian) literature openly available over the Internet.
The Open Source Education Foundation's purpose to enhance K-12 education through the use of technologies and concepts derived from The Open Source and Free Software movement. In conjunction with Tux4Kids, OSEF created a bootable distribution of GNU/Linux (Knoppix for Kids) based on Klaus Knopper's Knoppix, aimed at kids, parents, teachers, and other school officials. OSEF installs and supports school computer labs, and has developed a "K12 Box" as a compact Plug and Play workstation computer for student computer labs.
(Please note that suggested additions to the above list of GNU/Linux-relevant charities are most welcome.)
I have referred throughout this HOWTO to what I call the GNU/Linux movement. There really is no better way to describe the international GNU/Linux phenomenon: It isn't a bureaucracy, but is organised. It isn't a corporation, but is important to businesses everywhere. The best way for a LUG to support the international GNU/Linux movement is to keep the local community robust, vibrant, and growing. GNU/Linux is developed internationally, which is easy enough to see by reading the kernel source code's MAINTAINERS file -- but GNU/Linux is also used internationally. This ever-expanding user base is key to GNU/Linux's continued success, and is where the LUGs are vital.
The movement's strength internationally lies in offering unprecedented computing power and sophistication for its cost and freedom. The keys are value and independence from proprietary control. Every time a new person, group, business, or organisation experiences GNU/Linux's inherent value, the movement grows. LUGs help that happen.
The last goal of a LUG we'll cover is socialising -- in some ways, the most difficult goal to discuss, because it isn't clear how many or to what degree LUGs do it. While it would be strange to have a LUG that didn't engage in the other goals, there may be LUGs for which socialising isn't a factor.
It seems, however, that whenever two or three GNU/Linux users get together, fun, hijinks, and, often, beer follow. Linus Torvalds has always had one enduring goal for Linux: to have more fun. For hackers, kernel developers, and GNU/Linux users, there's nothing quite like downloading a new kernel, recompiling an old one, fooling with a window manager or desktop environment, hacking some code, or experimenting with an innovative embedded Linux computer. GNU/Linux's sheer fun keeps many LUGs together, and leads LUGs naturally to socialising.
By "socialising", here I mean primarily sharing experiences, forming friendships, and mutually-shared admiration and respect. There is another meaning, however -- one social scientists call acculturation. In any movement, institution, or human community, there is the need for some process or pattern of events in and by which, to put it in GNU/Linux terms, newcomers are turned into hackers. In other words, acculturation turns you from "one of them" to "one of us".
It is important that new users come to learn GNU/Linux culture, concepts, traditions, and vocabulary. GNU/Linux acculturation, unlike "real world" acculturation, can occur on mailing lists, Web forums, and Usenet, although the latter's efficacy is challenged by poorly acculturated users and by spam. LUGs are often much more efficient at this task than are mailing lists, Web forums, or newsgroups, precisely because of LUGs' greater interactivity and personal focus.
In the previous section I focused exclusively on what LUGs do and should do. This section's focus shifts to practical strategies.
There are, despite permutations of form, two basic things LUGs do: First, members meet in physical space; second, they communicate in cyberspace. Nearly everything LUGs do can be seen in terms of meetings and online resources.
As I said above, physical meetings are synonymous with LUGs (and most user groups). LUGs have these kinds of meetings:
What do LUGs do at these meetings?
The commercial rise of the Internet coincided roughly with that of GNU/Linux; the latter owes something to the former. The Net has always been important to development. LUGs are no different: Most have Web pages, if not whole Web sites. In fact, I'm not sure how else to find a LUG, but to check the Web.
It makes sense, then, for a LUG to make use of whatever Internet technologies they can: Web sites, mailing lists, wikis, e-mail, Web discussion forums, netnews, etc. As the world of commerce is discovering, the Net is an effective way to advertise, inform, educate, and even sell. The other reason LUGs make extensive use of Internet technology is that the very essence of GNU/Linux is to provide a stable and rich platform to deploy these technologies. So, not only do LUGs benefit from, say, establishment of a Web site, because it advertises their existence and helps organise members, but, in deploying these technologies, LUG members learn about them and see GNU/Linux at work.
Arguably, a well-maintained Web site is the one must-have, among those Internet resources. My essay Recipe for a Successful Linux User Group, for that reason, spends considerable time discussing Web issues. Quoting it (in outline form):
That essay partly supplements (and partly overlaps) this HOWTO.
Some LUGs using the Internet effectively:
Please let me know if your LUG uses the Internet in an important or interesting way; I'd like this list to include your group.
Finally, I want to make some very practical, even mundane, suggestions for anyone wanting to found, maintain, or grow a LUG.
There once were numerous organisations offering assistance to LUGs. One of the long-time ones remains active:
Tux.Org is an umbrella organisation for LUGs and open-source software development projects, providing a corporate entity, Web hosting, mailing lists, mirrors of popular software, and expertise and funding in planning special LUG events. More information can be found at the http://www.tux.org/ Web site.
comp.os.linux.announceand on an appropriate regional hierarchy.
The case for formal LUG organisation can be debated:
Pro: Incorporation and recognised tax-exemption limits liability and helps the group carry insurance. It aids fundraising. It avoids claims for tax on group income.
Con: Liability shouldn't be a problem for modestly careful people. (You're not doing skydiving, after all.) Also, even incorporated technical groups seldom carry liability insurance, and that insurance is typically so narrow in coverage that almost nothing a LUG does would be covered. A corporate liability shield is little use for such needs, either (limiting only the group's potential losses to the equity stake of the owners, but conferring no immunity to anyone for deeds that person carries out). Fundraising isn't needed for a group whose activities needn't involve significant expenses. (Dead-tree newsletters are so 1980.) Not needing a treasury, you avoid needing to argue over it, file reports about it, or fear it being taxed away. Meeting space can usually be gotten for free at ISPs, colleges, pizza parlours, brewpubs, coffeehouses, computer-training firms, GNU/Linux-oriented companies, hackerspaces, or other friendly institutions, and can therefore be free of charge to the public. No revenues and no expenses means less need for organisation and concomitant hassles.
For whatever it's worth, this HOWTO's originator and second maintainer lean, respectively, towards the pro and con sides of the issue -- but choose your own poison: If interested in formally organising your LUG, this section will introduce you to some relevant issues.
Note: this section should not be construed as competent legal counsel. These issues require the expertise of competent legal counsel; you should, before acting on any of the statements made in this section, consult an attorney.
Thanks to Chris Browne for the following comments about the Canadian situation.
The Canadian tax environment strongly parallels the US environment (for which, see below), in that the "charitable organisation" status confers similar tax advantages for donors over mere "not for profit" status, while requiring that similar sorts of added paperwork be filed by the "charity" with the tax authorities in order to attain and maintain certified charity status.
Correspondent Thomas Kappler warns that the process of founding a non-profit entity in Germany is a bit complicated, but comprehensively covered at http://www.buergergesellschaft.de/?id=106947.
In Sweden, LUGs are not required to register, but then are regarded as clubs. Registration with Skatteverket (national tax authority) offers two classification options: non-profit organisation or "economical association". The latter is an organisation where the goal is to benefit its members economically, and as such is probably unsuitable, being traditionally used for collectives of companies, or building societies / co-operative tenant-owners, and such).
Non-profit organisations in Sweden doesn't have specific laws to follow. Rather, general Swedish law applies: They can hire people and they can make profit. Generally they don't pay tax on their profits. (Profits stay in the organisation; unlike the case with "economical associations", members don't receive business proceeds.) To be able to do business, you must register with Skatteverket to get an "organisation number", allowing the group to pay and get paid. Otherwise you will probably have to arrange business through a member in his/her individual capacity. It may then also be possible, after securing an organisation number to apply for government financial support.
There are at least two different legal statuses a LUG in the USA may attain:
Although relevant statutes differ among states, most states allow user groups to incorporate as non-profit entities. Benefits of incorporation for a LUG include limitations of liability of LUG members and volunteers (but only in their passive roles as member/shareholders, not as participants), as well as limitation or even exemption from state corporate franchise taxes (which, however, is highly unlikely to be a real concern -- see "Common Misconceptions Debunked", below).
While you should consult competent legal counsel before incorporating your LUG as a non-profit, you can probably reduce your legal fees by being acquainted with relevant issues before consulting with an attorney. I recommend the Non-Lawyers' Non-Profit Corporation Kit.
As for the second status, tax-exemption, this is not a legal status, so much as an Internal Revenue Service judgement. It's important to realise non-profit incorporation does not ensure that IRS will rule your LUG tax-exempt. It is quite possible for a non-profit corporation to not be tax-exempt.
IRS has a relatively simple document explaining the criteria and process for tax-exemption. It is Publication 557: Tax-Exempt Status for Your Organization, available as an Acrobat file from the IRS's Web site. I strongly recommend you read this document before filing for non-profit incorporation. While becoming a non-profit corporation cannot ensure your LUG will be declared tax-exempt, some incorporation methods will prevent IRS from declaring your LUG tax-exempt. Tax-Exempt Status for Your Organization clearly sets out necessary conditions for your LUG to be declared tax-exempt.
Finally, there are resources available on the Internet for non-profit and tax-exempt organisations. Some of the material is probably relevant to your LUG.
Common Misconceptions Debunked:
Incorporation and tax-exempt status are separate issues. You don't have to be incorporated to get recognition of tax-exempt status (except it's required for one tax-exempt category, 501(c)(3)). You don't have to be tax-exempt to be incorporated. (Odds are, you honestly won't want either. You just probably assume you do.)
The "liability shield" one can get from incorporating doesn't protect volunteers from legal liability. All it does is prevent any plaintiffs from suing individual shareholders (LUG members, in this case) for tort damages merely because they own the corporation, if the corporation itself is alleged to have wronged the plaintiff. Plaintiff's maximum haul in damages from suing the corporation is limited to the corporate net assets, in that one case. However, volunteers are still fully liable for any personal involvement they're alleged to have had.
Umbrella insurance coverage against tort liability (i.e., against civil litigation) for your volunteers almost certainly costs far too much for your group to afford (think $2,500 each and every year in premium payouts, give or take, to buy $1M in general liability insurance coverage -- which generally would cover only the corporation as a whole and its directors in the strict performance of their defined duties), if you can find it at all.
IRS recognition as a tax-exempt group doesn't mean donations to your group necessarily become tax-deductible: Automatic deductibility is reserved to charities only, IRS category 501(c)(3), which must obey extremely stifling restrictions on group activities (e.g., it would then become illegal to host anti-DMCA events or support any other political activity), and must meet exacting paperwork and auditing standards. It's difficult to envision 501(c)(3) charity status actually making functional sense for any Linux group -- though one continually hears it recommended by those who imagine being able to tell people their donations will be guaranteed tax deductible must justify any accompanying disadvantages. Most LUGs would more logically file (if at all) for recognition as a "social and recreation club", category 501(c)(7).
In any event, unless one wishes to become a registered charity to render incoming donations tax-deductible, there is literally no point in applying for IRS recognition of your small, informal Linux group under any of the Internal Revenue Code section 501(c) tax-exempt statuses, because IRS simply doesn't care about groups with annual gross revenues less than $25,000, and doesn't want to hear from them (2010 update: IRS now does require a very simple annual e-Postcard informational filing from all small non-profits, to keep their 501(c) certifications, but still doesn't want tax from them).
The Federal Volunteer Protection Act of 1997 does not, in fact, shield volunteers of Internal Revenue Code section 501(c)(3) charities from tort lawsuits. At most, it furnishes some legal defences that can be raised during (expensive) civil litigation, with a large number of holes and limitations, and in most states will be denied unless the group also carries large amounts of (very expensive -- see above) liability insurance. Also, unless the volunteer's duties are not very meticulously defined and monitored, and the alleged tort occurs strictly in the scope of those duties, there's no shield at all -- plus the litigated action must not involve a motor vehicle / aircraft / vessel requiring an operator's licence, nor may the volunteer be in violation of any state or Federal law, else again there's no shield at all. (On the bright side, it's completely false, as often alleged, that the volunteer must be a member of the group, to be covered: In fact, the Act clearly states that a volunteer may be anyone who performs defined services for a qualifying group and receives no compensation for that labour.)
As may be apparent from the above, a number of groups have, in the past, talked themselves into unjustifiable levels of bureaucratic strait-jacketing with no real benefit and serious ongoing disadvantages to their groups, because of misconceptions, careless errors, and tragically bad advice in the above areas. In general, you should be slow to heed the counsel of amateur financial and tax advisors. (This HOWTO's maintainer had past experience during his first career as a professional finance and tax advisor, but, if you need competent advice tailored to your situation, please have a consultation with someone currently working in that field.)
As a reminder, it's vital that offers or requests to copy distribution-restricted proprietary software of any sort be heavily discouraged anywhere in LUGs, and banned as off-topic from all GNU/Linux user group on-line forums. This is not generally even an issue -- much less so than among proprietary-OS users -- but (e.g.) one LUG of my acquaintance briefly used a single LUG-owned copy of PowerQuest's Partition Magic on all NTFS-formatted machines brought to its installfests for dual-boot OS installations, on a very dubious theory of legality.
If it smells unlawful, it almost certainly is. Beware.
It's healthy to discuss the consulting business in general in user group forums, but for antitrust legal reasons it's a bad idea to get into "How much do you charge to do [foo]?" discussions, there.
Chris Browne has the following to say about the kinds of intra-LUG political dynamics that often crop up (lightly edited and expanded by the HOWTO maintainer):
GNU/Linux users are a diverse bunch. As soon as you try to put a lot of them together, some problem issues can arise. Some, who are nearly political radicals, believe all software, always, should be "free". Because Caldera charges quite a lot of money for its distribution, and doesn't give all profits over to (pick favorite advocacy organisation), it must be "evil". Ditto Red Hat or SUSE. Keep in mind that all three of these companies have made and continue to make significant contributions to free / open-source software.
(HOWTO maintainer's note: The above was a 1998 note, from before Caldera Systems exited the GNU/Linux business, renamed itself to The SCO Group, Inc., and launched a major copyright / contract / patent / trade-secret lawsuit and PR campaign against GNU/Linux users. My, those times do change. Still, we're grateful to the Caldera Systems that was , for its gracious donation of hardware to help Alan Cox develop SMP kernel support, for funding the development of RPM, and for its extensive past kernel source contributions and work to combine the GNU/Linux and historical Unix codebases.)
Others may figure they can find some way to highly exploit the "freeness" of the GNU/Linux platform for fun and profit. Be aware that many users of the BSD Unix variants consider their licences that do permit companies to build "privatised" custom versions of their kernels and C libraries preferable to the "enforced permanent freeness" of the GPL as applied to the Linux kernel and GNU libc. Do not presume that all people promoting this sort of view are necessarily greedy leeches.
If/when these people gather, disagreements can occur.
Leaders should be clear on the following facts:
The main principle can be extended well beyond this; computer "holy wars" have long been waged over endless battlegrounds, including GNU/Linux vs. other Unix variants vs. Microsoft OSes, the "IBM PC" vs. sundry Motorola 68000-based systems, the 1970s' varied 8-bit systems against each other, KDE versus GNOME....
A wise LUG leader will seek to move past such differences, if only because they're tedious. LUG leaders ideally therefore will have thick skins.
It is important to be careful with finances in any sort of non-profit. In businesses, which focus on substantive profit, people are not typically too worried about minor details such as alleged misspending of immaterial sums. The same cannot be said of non-profit organisations. Some people are involved for reasons of principle, and devote inordinate attention to otherwise minor issues, an example of C. Northcote Parkinson's Bike Shed Effect. LUG business meetings' potential for wide participation correspondingly expands the potential for exactly such inordinate attention.
As a result, it is probably preferable for there to not be any LUG membership fee, as that provides a specific thing for which people can reasonably demand accountability. Fees not collected can't be misused -- or squabbled over.
If there is a lot of money and/or other substantive property, the user group must be accountable to members.
Any vital, growing group should have more than one active person. In troubled nonprofits, financial information is often tightly held by someone who will not willingly relinquish monetary control. Ideally, there should be some LUG duty rotation, including duties involving financial control.
Regular useful financial reports should be made available to those who wish them. A LUG maintaining official "charitable status" for tax purposes must file at least annual financial reports with the local tax authorities, which would represent a minimum financial disclosure to members.
With the growth of GNU/Linux-based financial software, regular reports are now quite practical. With the growth of the Internet, it should even be possible to publish these on the World-Wide Web.
Governing your LUG democratically is absolutely vital -- if and only if you believe it is. I intend that remark somewhat less cynically than it probably sounds, as I shall explain.
Tangible stakes at issue in LUG politics tend to be minuscule to the point of comic opera: There are typically no real assets. Differences of view can be resolved by either engineering around them with technology (the GNU/Linux-ey solution) or by letting each camp run efforts in parallel. Moreover, even the most militantly "democratic" LUGs typically field, like clockwork, exactly as many candidates as there are offices to be filled -- not a soul more.
It's tempting to mock such exercises as empty posturing, but such is not (much) my intent. Rather, I mention them to point out something more significant: Attracting and retaining key volunteers is vital to the group's success. Anything that makes that happen is good. It seems likely that the "democratic" exercise stressed in some groups, substantive or not, encourages participation, and gives those elected a sense of status, legitimacy, and involvement. Those are Good Things.
Thus, if elections and formal structure help attract key participants, use them. If those deter participants, lose them. If door-prizes and garage sales bring people in, do door-prizes and garage sales. Participation, as much as software, is the lifeblood of your LUG.
The reason I spoke of "key" volunteers, above, is because, inevitably, a very few people will do almost all of the needed work. It's just the way things go, in volunteer groups. An anecdote may help illustrate my point: Towards the end of my long tenure as editor and typesetter of San Francisco PC User Group's 40-page monthly magazine, I was repeatedly urged to make magazine management more "democratic". I finally replied to the club president, "See that guy over there? That's Ed, one of my editorial staff. Ed just proofread twelve articles for the current issue. So, I figure he gets twelve votes." The president and other club politicos were dismayed by my work-based recasting of their democratic ideals: Their notion was that each biped should have an equal say in editorial policy, regardless of ability to typeset or proofread, or whether they had ever done anything to assist magazine production. Although he looked quite unhappy about doing so, the president dropped the subject. I figured that, when it came right down to it, he'd decide that the club needed people who got work done more than they needed his brand of "democracy".
But we weren't quite done: A month or so later, I was introduced to a "Publications Committee", who arrived with the intent of doing nothing but vote on matters of newsletter policy (i.e., issue "executive" orders to the volunteer production staff). Their first shock came when I listened politely to their advice but then applied my editorial judgement as usual. Much worse, though: I also assigned them work, as part of my staff. Almost all immediately lost interest. (Bossing around other people seemed likely to be fun; doing actual work was not.)
The point is that the widespread urge to vote on everything is at best orthogonal to any desire to perform needed work; at worst, the former serves as an excuse to compulsively meddle in others' performance of the latter.
To sum up: Have all the "democracy" that makes you happy, but watching after the well-being of your key volunteers is what matters. (To quote Candide, "We must cultivate our garden.")
Last, plan for your replacement: If your LUG is a college student group, and must go through a paperwork deathmarch every year to stay accredited, make sure that and all other vital processes are documented, so new LUG officers needn't figure everything out from scratch. Think of it as a systems-engineering problem: You're trying to eliminate single points of failure.
And what works for the guys in the next town may not work for your crowd: Surprise! The keys to this puzzle are still being sought. So, please experiment, and let me know what works for you, so I can tell others. Have fun!
Copyright (C) 2003-2016, Rick Moen. Copyright (C) 1997-1998 by Kendall Grant Clark. This document may be distributed under the terms set forth in the Creative Commons Attribution-ShareAlike 4.0 International licence at http://creativecommons.org/licenses/by-sa/4.0/, or, at your option, any later version.
New versions of the Linux User Group HOWTO will be periodically uploaded to various GNU/Linux Web/ftp sites, principally http://linuxmafia.com/lug/ and the Linux Documentation Project.
I welcome questions about and feedback on this document. Please send them to me at email@example.com. I am especially interested in hearing from LUG leaders around the world, especially outside the USA. Please let me know of innovative or noteworthy things your group does, meriting description here.
I would like to give a big thank-you to Kendall Grant Clark for the initial versions of this document in 1997-1998, and for trusting me to take over and renovate his creation starting in 2003.
Warm regards and thanks to Chris Browne for describing the situation with non-profit and charitable groups in Canada, his thoughts on financial donations as a way to participate in GNU/Linux and the free software and open-source software movements, and his ideas about the kinds of political issues likely to arise within LUGs.
In addition, the following people have made helpful comments and suggestions: