What's Wrong With the FOSS Community? 348
An anonymous reader writes "Patrick McFarland, one of the major Free Software Magazine authors, has completed his second article on whats wrong with the Free/Open Source Software (FOSS) community, and what we face in this world. He touches on ESR's Cathedral and the Bazaar essay briefly, and warns against cherry-picking style software development."
Common sense says (Score:4, Insightful)
Nothing in this piece convinces that common sense is wrong.
In my opinion (Score:5, Insightful)
When did the community become an entity (Score:5, Insightful)
Hmm.. (Score:3, Insightful)
we need artists and bug hunters (Score:1, Insightful)
Nothing really is wrong except one thing. (Score:5, Insightful)
Not every software need will be be solved with FOSS.
There needs to be freedom to write Open and Closed source software. That is what bugs me are people that think selling a closed source package is evil. I just don't think that the FOSS model can work for every program.
F(L)OSS (Score:3, Insightful)
no leadership? (Score:3, Insightful)
Unlike in the Cathedral, the Bazaar has no official leadership.
Isn't this what enables FOSS? Most of the FOSS don't have official leadership (other than the creator of course
Of course... (Score:5, Insightful)
Here's my rimshot: (Score:5, Insightful)
Interesting bit about XFree86 (Score:5, Insightful)
> way of everyone who did have vision.
That's rather well said. If you're the author of a successful open source project and you find yourself unable to keep working on it, do you have a duty to turn it over to the other developers for continued maintenance? I can't think of a reason not to, and if you don't, it'll either die or get forked, both of which aren't pleasant outcomes.
Coming along just fine (Score:5, Insightful)
Perhaps the real problem is the plethora of side-liners, pundits, philosophers, and magazine authors who have nothing better to do than sit around and draw erroneous conclusions. I call these people OSS arm-chair experts. We don't need 'em. Seems the people with most to say write the least amount of code. Maybe they should learn to program and get involved rather than digging too deeply into what's wrong. Be positive.
waste of space (Score:2, Insightful)
Re:Developers who ignore users (Score:3, Insightful)
But they aren't selling anything ... FOSS developers code because it's fun; that's their compensation, not money. It's no excuse to be an ass, but I don't really see why they should necessarily cater to anyone that isn't contributing in a tangible way.
Too violent? (Score:5, Insightful)
Re:Developers who ignore users (Score:1, Insightful)
Don't look a gift horse in the mouth...
Re:In my opinion (Score:5, Insightful)
Ooops, I think you got that wrong.
There are a small number of very vocal people who are total assholes towards people.
Does it matter what the subject is?
Re:disagree (Score:3, Insightful)
Frankly, the Zealots and the Software (Score:1, Insightful)
this half-@ssed written software. I'm not saying that I can write better software. I am saying, however, that it should at least work as well as the (God, if you exist please forgive me for what I am about to say.) MS piece that it is probably trying to emulate.
I don't consider myself an FOSS zealot by any stretch, but I am rooting for its success.
actually yeah you can (Score:1, Insightful)
Say I give you a free tire for your car saying "it's a good tire" and you use it
That said, I do understand the point you were making
That's all.
Re:Developers who ignore users (Score:5, Insightful)
While this may be OSS development in the sense that people work on OSS code, it isn't about this topic - the "FLOSS Community" and the coders that form part of this community. Those coders tend to fall in the second camp. They tend to work for reasons other then direct cash. They do it for fun, peer recognition, whatever. For the majority of these people, "non-coding end-users" are the same bunch of clusterfucks they deal with everyday during the dayjob, and tend to not feature very prominently in the motivation chain. The things that drive them are project that are "fun" to code, "pet projects" and all that kind of stuff. They have little motivation to work on projects that are "boring", "seen as difficult" or "of no interest to the developers". This camp of OSS developers "must" do nothing, and more importantly, owe you nothing.
You then bring in some muddled argumentations about the "market" and "running out of business". Unless the OSS coder in question is pretty incompetent, and gets fired from his (quite possibly non-OSS related) dayjob, there is no "business" to be run out of. Most of these projects *are* pet projects, and they only reason you can use them for free is because the coder in question has an urge to tell the world: "Look what I can do!!"
Now its time to bring market drivers / basic economics into the picture. You, as a non-coding end-user, want an application. There are some half-way-there projects out there, but non really fit your bill. You are angry because all the selfish devs only think of their pet projects and having fun. Some entrepeneur, somewhere, will know this, and hire a bunch of devs to create a project you, and hopefully many others, will pay good money for. Only now, once renumeration has entered the picture, can you speak of a market in a meaningful way. Now you are a paying customer, and you can vote with your wallet and feet.
Unless you are a cheapskate, and don't want to pay for anything, but still want every little piece of functionality handed to you "just so". If you ain't paying the cash, either do it yourself, or STFU.
Re:actually yeah you can (Score:2, Insightful)
Off the top of my head... (Score:4, Insightful)
- Users, What Users?: Coding for yourself is nice, but if you want users to flock to your app, you might want to actually consider what they want. Don't bitch and moan at them when they offer suggestions, even if said suggestions don't fit your own personal vision, or even if they are downright stupid. That doesn't mean you have to implement them, but it means you have to be weigh them equally with your own ideas. Try to be inclusive and open to your userbase. "Go code it yourself" is a great way to keep OSS in the geek ghettos of the computing world.
- But It Looks Pretty: That's a snazzy looking interface you just whipped up, is it consistent? No? Does it follow standard UI principles? No? I'm sure people won't become frustrated and dismissive of your hard work. You can say that UI standards impinge on your freedom as a developer, but they make a user's life much easier, and makes people much more likely to actually use your software.
- Ask, Don't Beg: Asking companies and organizations to open code is nice and helpful, but be careful how you go about it. It can easily come across as "The OSS community could never dream of putting something like that together. Gimme!" Don't act like you *expect* the code, and that they are evil incarnate for withholding it. Don't make it seem as if the OSS community is incompetent and needs privately-developed projects turned over wholesale to get anything accomplished. Sure it helps a whole lot, but don't make it seem as though OSS is just mooching off the investment of others.
- Vendettas: If two projects can fight over something, no matter how petty, they will. Try coding, it's more productive and makes you appear like a mature, competent project that might help win over those hesitant to support OSS. Or you could just continue the pissing matches and flamefests over icons and licensing minutiae that could probably be settled if egos were set aside for a few moments. Public wars of words, endless forking....nothing gets accomplished but the stroking of egos. Well, except the whole "OSS developers come across as immature, childish amateurs" thing.
I think I know what's wrong... (Score:0, Insightful)
The article is all wrong. (Score:4, Insightful)
In OSS much more than anywhere else, the best floats on top. That's why Outlook mail sucks and KMail sucks considerably less. Linux works because NOBODY doubts that Linus is the chief, Blender works because NOBODY doubts that Ton is the chief, because they both do an excellent job at what they do: leading large OSS projects.
Of course there's weedy stuff in OSS that's buggier and more twisted than Autodesk Converter and Macromedia Director together, but that sinks to the lowest bottom, and does not get pushed onto the market by monopolies and marketing budgets of galactic proportions (Windows XP anyone?).
The article is bogus and has it all backwards. I want my 5 minutes back.
Re:In my opinion (Score:5, Insightful)
A knowledgeable person who is simply inexperienced in an area will generally phrase a question better than a 12 year old kid demanding attention NOW.
"Gentoo is shit, it won't install why not?"
vs
"I attempted to install Gentoo on my computer (an aging P2 on an Acer motherboard) and came up with a number of problems during the install. It spent about 20 minutes compiling before it stopped saying 'The XYX system could not be compiled: missing file xyz.c'.
I tried looking around the furum but couldn't see where I am going wrong. Can somebody give me some assistance please?"
Re:Here's my rimshot: (Score:2, Insightful)
you have absolutely no right to complain about something you got for free.
I'm sorry, but you have utterly failed to explain:
(1) why you have a right to complain about something you got in exchange for money, through barter, etc.
(2) why you do not have a right to complain about something that you got for free, stole, etc.
In fact, if you're willing to justify #1, then I can prove that #2 is not true by simply introducing that elementary concept known as opportunity cost. A free pile of crap in your yard is free, but it still costs you time and effort to deal with it that you could otherwise invest in a more valuable endeavor.
we do this for fun
Incomplete statement. You do it for fun and then claim that every rational person should do it to because it is better, more capable, and more bug free. You then refuse to fix identified defects and bugs by hiding behind the excuse that it is free. It never occurs to you that the time, effort, frustration, and cost expended by individuals who buy commercial software could be less than the time, effort, and frustration expended by individuals buying into your claims.
Why not advertise the entire deal? "Our software won't cost you a dime, and if you don't like it exactly as it is, fix it yourself or suck eggs."
Re:disagree (Score:3, Insightful)
Re:I can pay the coder to do it. (Score:2, Insightful)
Secondly, you can contact most software companies and they will finish a feature. They would more than likely take payments to customize the system to include the feature. It probably wouldn't cost too much if they can see an ongoing benefit to their current and future customer base.
Remember: Microsoft is not the only proprietary software company.
Just a description (Score:2, Insightful)
It's true that better leaders help projects produce things faster, but F/OSS has never been strong because of DEVELOPMENT SPEED, F/OSS has been strong because of diversity and the LACK of an authoritarian view. The community (warts and all) is precisely WHY F/OSS has succeeded.
The article author assumes that there is one direction we all want to go in and we should just get there as quick as possible. This is not how we got to where we are now, and it's not required for the future. Certain projects are chugging along with speed with a vision, others are meandering along to the sound of their own drums. These are all good. No need to panic, certainly no need to criticise the VERY WELLSPRING from which this world arose.
Re:what is wrong (Score:5, Insightful)
And I will wait for someone else to pay ... (Score:4, Insightful)
However progress will be slow because most of us will wait for someone else to pay for the changes we want. Most people will freeload if given the opportunity, Econ 101. Since you are reading this right now, I will thank you in advance for your future gifts to the community.
Re:People.. the same as any community (Score:5, Insightful)
Sad really.
Re:Nothing really is wrong except one thing. (Score:3, Insightful)
If you don't like it, don't patronize the people who enforce it (RIAA, MPAA, etc.). Nobody is forcing you to listen to music or watch movies either.
Re:What's wrong with Santa Claus (Score:1, Insightful)
OK, just kidding. Well sort of...I have had similar discussions with folks about what is coming. If people don't think corporations can't lock up or take over a number of the efforts that under way they have no understanding of business and IP law.
The majority of people here and in the FOSS communities are techies, geeks, programmers, engineers, hobbyists, etc. Most of us don't really have a firm understanding of business and IP law. This is the Achilles heel of FOSS.
Do I think FOSS is doomed? No as long as everyone stays vigilant and educates themselves about things not so geeky like law and business; and as long as we start to become more customer oriented. Ok, ok...stop your screaming and vomiting. Hey, I said stop it! They is one of the major reasons Linux has NOT taken off on the desktop. Linux is a beautiful piece of work, but is it human-friendly? Come on folks.
Any how that is my $0.02...or less.
Re:Nothing really is wrong except one thing. (Score:2, Insightful)
Don't confuse the wishes of society with the decisions made by our elected leaders. They're not always the same.
Re:Nothing really is wrong except one thing. (Score:3, Insightful)
Re:People.. the same as any community (Score:3, Insightful)
No... wait...
The difference between the FOSS and commercial software worlds is that in commercial enterprises, in the absense of leadership, someone will be unilaterally appointed, not to lead, but to dictate.
In the FOSS world, if there's no leader, there's no leader. People will choose their own direction until they find someone they want to follow or find others wanting to follow them.
I think that's a major reason why FOSS is winning. The smarter you are, the less you like having your actions dictated to you by a moron in a suit, and for all its faults, participation in FOSS projects doesn't generally ask you to put up with that sort of shit.
Re:People.. the same as any community (Score:5, Insightful)
Which in turn may or may not be successful. The mambo/joomla mess illustrates that some forks work and you end up with two relatively strong branches. Go the other way, and a fork splits its community, diverts resources, and eventually kills off one, the other, or both.
And while no one wants a moron in a suit yelling at them, OSS developers are notorious for chery-picking the "cool" aspects of the project and ignoring others, and generally being insensitive to things like schedules and deadlines.
As to "winning", you have some strange definitions. Get an OS with more than a percentage point or two of the average desktop, and "maybe" you can start waving that flag. Utill then...
FOS Fundamentalism, Ignoring the Customer (Score:1, Insightful)
Everytimes someone suggests something about Windows (the EULA, Vista DRM) the FOS hisses and spits that everyone should switch to Linux. When users say "but 3DSMAX" they vent and froth "Use Blender!" Photoshop? "Use GIMP you morons!"
This is not the way to win people over. Many people point this out, but some vocal members in FOS can't move beyond it. If there is another view in FOS, they're very quiet about it.
It's not unique to Linux. The Firefox developers have been incredibly arrogant when it comes to things like memory size, SaveAs filename. The so-called MySQL community refuse to accept the possibility that maybe, just maybe, MySQL is too buggy and unpredictable. I used an enterprise project written in MySQL and I found it to be beyond painful. But whenever I posted of my experience on
It's often said Microsoft "don't get it", and they don't. But FOS is equally arrogant and even more zealous.
Re:People.. the same as any community (Score:5, Insightful)
A lot of people who are given the label "self-appointed dictator" in this realm are really just people leading by doing, but leading in a direction different than those doing the labeling would prefer.
The "I'll donate some of my time to some project" developers like the "cool" features, yes, and the real leaders will take what help they can get. As they do most of the work.
Most successful FOSS projects seem to be based around a core group of people whose prime driver is their interest in fulfilling their vision of what the result should be, assisted in small ways by a large group of vaguely interested people.
This is leadership. You can tell the difference between a leader and a director with a simple comparison: If the person would eventually/theoretically get the project done even if everyone else left, they're leading, and if they wouldn't get anything done when everyone left, they're not leading.
Of course, there's no reasoning with people like that... they don't give a fuck about what you want, they're blazing trail.
As to "winning", which do you think most people care about, their desktop, or the Internet it connects to? How many people do you know these days who can't just sit down in front of any computer whatsoever, log onto whatever services they need, finish up and walk away? There are a lot of them. The services they're logging onto are the "Network is the Machine" effect Microsoft has been fearing and fighting all this time, and that network is pretty much owned by FOSS.
Linux might not be on the desktop, but the desktop is becoming more and more "That virus infested annoyance you're forced to deal with to get on the Internet", and the Internet is FOSS.
Re:In my opinion (Score:4, Insightful)
Now, it may be that you lack the experience to discern what portion of TFM is the information you are seeking. If so, say so! Say that you have looked in TFM and not found an answer. Ask for help explaining specific parts of TFM, or ask for a more specific pointer to what part of TFM you should be looking in.
Reading TFM is an important skill, and one that must be acquired. If you have that skill, then there is no call for you (or anyone else, of course this entire post is directed generally) to go demanding that other people use their energy and time to do what you are perfectly capable of. If you don't have that skill, then the greatest ROI for people responding to your question comes when they encourage you to acquire that skill. If you have trouble acquiring it on your own, then generally you can still find someone who is willing to help you acquire it. But not many people want to spend their time and energy doing something that either you can do or that you should be learning to do, unless such an expenditure will help you learn to do it yourself. If you expect someone to put down little arrows on the ground in front of you when you are lost in an unfamiliar city, then you'd better have some cash in hand. Similarly, many distros offer paid support contracts.
When you spend 5 minutes saying exactly how to do something in detail, you are often setting yourself up to spend another 5 minutes saying exactly how to do something else in detail later. If someone figures out the answer themself, even if it is with guidance and aid (think Socrates), then they are much more likely to be able to figure out the next answer as well.
Required reading (or it should be): http://catb.org/esr/faqs/smart-questions.html [catb.org]
Portability (Score:2, Insightful)
Re:what is wrong (Score:3, Insightful)
You know what the funny thing is? (Score:5, Insightful)
The bazaar model still worked when the pinnacle of software complexity were "cat" and "vi". That's it. It stopped working almost completely when complexity meant Open Office Org.
The Asperger's Syndrome kind of coder (and I'm one, so I can make fun of myself if I want to) which finds more joy in coding something cool instead of going out and flirting with a girl, also has a very narrow focus of attention and gets bored easily when he must deal with stuff either (A) outside that focus, or (B) which is basically homework instead of getting to the cool stuff. That's how we ended on the bad side of teachers in school, after all. Spending weeks understanding someone else's framework and code before you can even start on your cute "number paragraphs in Klingon" idea, is boring, and it's even more boring to understand and test all dependencies so you don't break something else.
So today in F/OSS the only ones making any progress nowadays are, sad to say, the Cathedrals.
Yes, everyone likes to use the Linux kernel and such as an example of why the Bazaar is strong, but have a look at the actual contributors some day. It's _not_ bored nerds like you and me working in their free time. Most of them are paid employees of Red Hat, IBM, etc. Linux as the work of bored nerds in their free time was a security shithole until Red Hat spent some real money doing a code and security review. And it was a joke in the enterprise arena until IBM started pumping some real money and formerly Cathedral-developped closed-source code into it. There's a reason why IBM looked like a believable target to SCO (as opposed to just a tempting target, by having deep pockets), and that's the sheer quantity of Aix code that IBM donated.
The same goes for OOo: practically all development is paid for by Sun, and it's bleeding Sun a ton of money. The same goes for Apache, which everyone uses as an example of why OSS is better than MS's software on a server: it, and most other Apache projects for that matter, is mostly IBM work. Go figure. IDE's? Both Eclipse and Netbeans are paid work by respectively IBM and Sun and a number of other corporate contributors. Compilers? You'd be surprised how much in GCC actually comes from Intel and the like. Browser? Mozilla was mostly paid work by Netscape, then AOL, and now it's mostly sponsored by Google. Etc.
So yes, as you aptly put it:
And that's why most of F/OSS nowadays is nothing more than a way for various corporate Cathedrals to pool their resources against MS. Sure, it's a good goal and I have nothing against benefitting from it. But let's stop pretending that ESR's Bazaar is anywhere _near_ relevant any more. The actual "Bazaar" projects are the thousands of unfinishet things on Source Forge that noone gives a damn about, either to help develop/debug or to use seriously or to pay the developper for features.
Re:People.. the same as any community (Score:3, Insightful)
Lord Vetinari, in "The Truth" by Terry Pratchett.
I think he's got a point.
Re:People.. the same as any community (Score:2, Insightful)
This is a pretty flawed definition of leadership... While getting my sociology degree I took a few classes in small group theory and leadership theory and the definition I'd submit would go more like, "If people are following you, you're leading; if you have to pull them constantly, you're not leading.
Leaders aren't supposed to be heroes who can do everything themselves, they're people who can get everyone on the same page so that the group can get it done.
Re:In my opinion (Score:2, Insightful)
In a word, amateurism (Score:3, Insightful)
Let's starts with how much I *loathe* OpenLDAP, and the literally weeks I spent getting it working, and have yet to have it work with autofs, and I'm fighting it, right now, with Samba. I tried to find a GUI editor. The one that seemed best for my situation installed from an rpm... and had *no* useful sample configuration files, and even when I managed, using google, to set up some, it gave errors.
PHP4 (we have our reasons for not going to 5 yet) is a royal pain, and a *mess*. I mean, php.ini in */lib?!, and not in */etc? Why? Why scatter files all hither and yon? Oh, and then there's where I have to hand-edit the Makefile to add
On the other hand, Webmin was a literal no-brainer, and Nagios was only a bit harder. AND it came with working minimal configuration files. Even setting up virtual hosts with Apache were not *that* big a deal.
The amateurism covers things like inadequate testing, absolute requirements of a specific library (and not allowing a *later* version of the library), and not having an easy uninstall method.
It seems to me that a lot of folks push the envelope ->on their own system-, and don't try to meet standards that might run on nearly *everyone's* system. It doesn't have to be tested on *everything*, just follow standards. And to look at commonly-accepted practice, if not best practice.
mark, with more than two dozen years of software development experience, and half a
dozen with sysadmin
Re:Nothing really is wrong except one thing. (Score:1, Insightful)
I liked your parent post very much, and this paragraph of yours made me like it even more. He made a very important point about the need to think outside of what you are told, and you actually confirmed it with your post.
Just so you know, I live in Cuba. Yes, that "communist" Cuba. I wont argue if it is true or not, but we are also told that "The people have the ultimate power here." I belive you will consider that claim to be false, as much as I consider that statement to be false when it comes from you.