So I guess the difference between me and an OSS developer is that I just clutter my own hard drive with a bunch of unfinished projects, instead of publishing them to SourceForge.

I recently had a meeting with a "senior manager" at work and he asked me a very interviewish question: what motivates you? The answer was easy: solving problems with software. Whenever I encounter a problem, I immediately start thinking about how I could solve it with a computer. This may sound like a good quality for a developer - but it has its downside. I'm pretty much ONLY interesting in solving the problem. Once I write enough code that proves to myself that I CAN solve the problem, I completely lose interest. No usable UI, documentation, robust features, etc. My hard drive is filled with those projects.

Joshua -

It's one thing to litter your own drive with unfinished projects - we all do that.

It's another to act self-important about it, and spend more time fretting about licensing issues for the project you haven't (and will never) complete (because you spend too much time fretting about licensing issues).

This is one of the reasons why I haven't pushed out my own code as "open source" yet. I don't really see what the benefit would be, aside from trying to recapture that warm and fuzzy feeling I had when I got a PowerBook a while ago.

I'd rather keep the source code and let the apps go out for free. Keep the project alive and evolving, but controlled.

Rory,

I'm a big loser. I spend a lot of time on IRC. I think the two are related. In any case, we all wanted a simple IRC bot that would do certain things. This one guy on the channel kept stating about how we was going to make this really cool bot one day, and I got tired of listening to him and decided to show him up by writing my own bot. So I did. It has existed as a project on SourceForge (The URL is http://sf.net/projects/jackbot if you're curious.) for the last year or so. I haven't really touched the source code in eight months, though, since I got it to do what we needed. To parallel Joshua: I have no more problems to solve on it, so I've stopped working on it.

My stupid little IRC bot gets maybe 50 downloads per month. Now, I could have just released the binaries and let people download those, but if I release the source code, then maybe if somebody else has a problem to be solved, they can use my bot as a jumping off point. I released it under the GPL because I want to ensure that those changes will come back to me so that I can benefit from the work they've done by building on my base.

As you observed, there are thousands of projects just like mine on SourceForge. And nobody has made any contributions to my project, beyond a friend who helped fix a couple of bugs. But I'm hoping, and at least there's a /possibility/ that I'll get some benefit from it. If it just sits on my hard drive, there's not even a chance.

Brian

Randy -I tend to agree with you...
The hell if I am going to let you come along and have access to the source safe. It's my project, I am going to finish my dream.

This is what most Open Source is about. After the product has been completed, they will either release all of the source code, or the majority of it, excluding the clever components (that differentiate the product).

Now-a-days, Open Source is just a facade for a different type of Marketing, i.e. IBM. IBM won't make a dime off of Linux, but they will make a lot off the value-added services and components on top of it. The truth is that IBM as a whole cares very little about Open Source. Let's be straight, though. If I were IBM, I would do the same thing. They lost the war to Microsoft, so they have to find a new way to compete. The problem is that IBM is going to get caught in a version war (The number of Linux variants is growing too fast to build products that support them, but this goes back to the point of no one opening the source safe, because they want to finish their own dream).

Joshua -I feel the same way. As soon as I feel I have found a solution, it is very hard to keep myself motivated. So, execution is the real key to development. Problem solving (for people like us) is the natural progression of our mind. That is just how we think, so the key is fine-tuning other aspects of software.

Brian -

Your project actually released files, so you're a step *up* from what I'm talking about.

I'm talking about the projects for which very little coding is done, but plenty of license-switching and self-important task assignment takes place.

The reason that I release all of my stuff as open source is so that I may be able to help someone with a problem down the line (similar to what Brian was saying). If I can help someone get over that problem they've been working on for who knows how long, I'm glad to be able to help.

My God, are you saying that Sourceforge is full of MANAGERS?? ;)

The pontification and self importance sounds awfully like the "brain banks" who spend their days in meetings and in their private offices except when checking up on we low-life coders who actually write the damn stuff.

"My God, are you saying that Sourceforge is full of MANAGERS??"

No - but if there *were* a few, then some of those projects might be further a long.

There are *good* managers, too :) They're not that common, but they exist, and they can help move a project along.

I agree, though, that there's some major bullshit going on with *some* of the people who do nothing but attend meetings...

In Open Source, all bugs are shallow.

Unfortunately, so are the developers.

I don't think that this has anything to do with free software qua free software really. As previous commenters pointed out, if there weren't all these project false starts on Sourceforge, they'd just be on people's hard drives anyways. How many people have tried to start write their own commercial software on the side while working full time and ran out of time or motivation?

The grandiose posturing is a reflection of the importance given to the owners, managers, and leads; many coders haven't had the luxury of these position's power and covet it. Personally I think the whole power hierarchy is a crock of shit. I'm all for anarchism in the work place and I think free software development is an awesome opportunity for people to free their minds of the capitalist/proletariat owner/worker class mindset. Unfortunately, most coders are not prepared to make that jump.

The license issue really reminds me of the hundreds of business decisions that come up when people begin a start-up. Hours and hours are spent talking about this and that decision, with all this mindfulness of how the business may grow huge or how it will look to outside people, with very little focus on the product itself. In fact, the false starts at Sourceforge remind a lot of boom-era start-ups.

Until people are prepared to seriously explore the social possibilities of free software development we're stuck in the same crappy occupational relationships that we have now. Free software is the antidote, not the poison.

"... if there weren't all these project false starts on Sourceforge, they'd just be on people's hard drives anyways."

That's not what I'm talking about, though.

It's not the "false starts" that bother me - it's that plenty of attention is paid to all the "meta" stuff - the time during which people get excited about a project, which is often the time that precedes the ditching of a project (at least in my experience, people disappear when any sign of real work arrives - for pay or not).

"Personally I think the whole power hierarchy is a crock of shit. I'm all for anarchism in the work place and I think free software development is an awesome opportunity for people to free their minds of the capitalist/proletariat owner/worker class mindset."

The "power hierarchy" exists throughout nature, and probably for a good reason. The reality is that *not* everybody is cut out to lead a group through an effort, whatever that effort may be. Some people don't *want* to lead, and don't *want* to be on the same level as everybody else in terms of power.

Design by committee isn't practical. It might be a nice dream, but application of the idea can be disastrous. Even Linus would agree with the idea that control has to be applied to the process.

Even if a leader isn't declared in a group, one will emerge and become the de-facto lead. The distinction between the people in the hierarchy might be subtle, but they're there.

"The license issue really reminds me of the hundreds of business decisions that come up when people begin a start-up. Hours and hours are spent talking about this and that decision, with all this mindfulness of how the business may grow huge or how it will look to outside people, with very little focus on the product itself. In fact, the false starts at Sourceforge remind a lot of boom-era start-ups."

I was thinking the exact same thing, and this has a direct link to the distaste I have for what I consider to be "make-believe" business, such as what I've seen on SourceForge (and experienced directly).

I was asked to join a startup a few years ago - they wanted a JSP site - and 95% of the whole deal was just sitting around and talking about "this deal" and "that deal." It was a bunch of crap, and irritating pseudo titles popped up all over the place ("director of blah blah blah" and "head of blabbity blah blah").

I left the "company" after a meeting with a lawyer who did nothing to dispel the make-believe feel of the whole thing. The lawyer had drawn up a bunch of "documents," and I got the feeling that they could have been done better by three year olds with crayons, so I bolted.

It was all *talk*. Talk, talk, talk. Talk, talk, talk, talk, talk, talk, talkity-talk. No "doing" at all.

Just like sitting around, discussing the importance of which license to choose for your unreleased vaporware - talking about how you're going to change the world with your software, and then walking away because there's probably something easier to do someplace else, like paying with the PS2.

"Until people are prepared to seriously explore the social possibilities of free software development we're stuck in the same crappy occupational relationships that we have now. Free software is the antidote, not the poison."

What would the purpose of this software *be*? Is there going to be funding? How are people going to survive while they make all this software? Who's going to lead them? If there's no leader, then how is anything going to get done? Is there just going to be an anonymous "vision statement" to which people will be expected to fashion bits of code? How are they going to know which bit of the code "belongs" to them? Isn't it going to be problematic if several people are working on the same code at the same time, but don't know it because they're either lacking the tools or the leadership to coordinate the work? If there's no team leader, then who's going to choose the tools? What technologies will be used? Is there going to be a design doc? Even written anonymously, a document describing the software would still function as a source of leadership. Updates to the document would also act as a proxy for a leader.

There's no question - there's nothing wrong with free software development. However, making claims about how free software development is going to solve "this" problem and "that" problem reminds me of the medical claims made by sellers of totally unproven "herbal remedies" - it's just claims.

The major difference between free software and commercial software, as far as I can tell, is that free software is free. Is there *really* a difference in how it's produced aside from a few superficial issues (like office buildings/"real" companies/etc.)?

Rory,

One of the big differences between Free software and commcercial software is the driving motivator of the people doing the work. The very first Free software projects were done for fun, on the side, with the goal of just doing some cool shit. The idea of the GPL license is to ensure that the community remains vibrant by requiring derivative works to be released back to the community, thus creating a feedback loop.

Of course some people get paid full-time to work on the software, and while it certainly helps get some things done faster, the community doesn't depend on it. If IBM decided tomorrow that Linux was a losing proposition and pulled out all their money; if Redhat went out of business; if Linus got hit by a bus, the software would continue to grow and evolve because it's simply a source of entertainment and growth for a lot of people. Sure the projects would change form, and there might be different versions and branches and pactches, but that's not anything different than exists today.

As a corollary to the community idea, Free software is produced differently than commercial software in that it tends to evolve through simple natural selection. Sure, Linus decides what goes into the kernel he maintains; but if there is a good idea out there that Linus decides he doesn't like, people will create independent patches for the kernel and begin to distribute them. In time, if the demand is great enough, Linus will either adopt that feature, or people will stop using his kernel. In commercial software, the entire system is driven from the top-down. The customers have ultimate power of the managers, who have ultimate power of the developer leads, who have ultimate control over the developers. Developers can voice their opinions to the dev leads, who can voice their opinions to the managers, who can voice their opinions to the customers. But in the end, the higher-up always has the final say. There is no room for evolutionary progress caused by the differing opinions of those involved, and there can never be because the product is owned and controlled tightly.

It is precisely that evolutionary community that has allowed Linux to advance so quickly from a school hobby into a big money-maker and a real Microsoft competitor. At the end of the day, there is no ego involved; there is no potential lost investment; there are no politics. At the end of the day, the actual users can choose what they want to use, and whatever works best is going to win.

Brian

Brian -

Thanks for answering my questions about the difference between free software/commercial software development.

For the most part, I agree with and understand what you're saying.

However, when it comes to this:

"There is no room for evolutionary progress caused by the differing opinions of those involved, and there can never be because the product is owned and controlled tightly."

I don't agree at all.

The need for software that works with the users is why we have testers - it's why Microsoft is giving Longhorn away *years* before its release.

Although there might not be any competition for Longhorn, Microsoft is still able to refine the product based on customer feedback. This isn't evolutionary in the sense that the best *product* wins, but it *is* evolutionary in the sense that the best features will.

All zealotry aside, Windows XP, let's be honest with ourselves, is extremely easy to use. Unlike Linux, users aren't having to choose between various desktop managers, and then have to be confused as their apps pop on screen using different GUI toolkits. It's pretty obvious that Microsoft went through a process to ensure that Windows would be easy to use.

Microsoft has, as you pointed out, tight control over its product. However, Microsoft listens to its customers, listens to its testers, and forces its employees to constantly work with the products they're developing in order to figure out where they fall short. It isn't a perfect system, but that's no big deal since there *isn't* a perfect system. In the end, it's perfectly good enough, and has resulted in *good* software.

The same could be said for Apple - OS X isn't particularly difficult to use, and it's developed (at least the user experience) in much the same way Windows is. Just because a company seems like a black box doesn't mean that it can't engage in the sort of activities that will lead to a good product.

"At the end of the day, the actual users can choose what they want to use, and whatever works best is going to win."

At the end of the day, the actual users are often lucky if they can even find the "On" switch.

Giving users too much choice is just asking for trouble. I don't think that users are a good example of demonstrating the success of OSS.

Rather, the success of OSS has been in areas where large amounts of cheap hardware have been needed to engage in digital ditch digging. Szulik was smart in advising that home users stick with Windows - the desktop is *not* one of the strengths of Linux.

At the end of the day, users use what you put in front of them.

'The "power hierarchy" exists throughout nature, and probably for a good reason.'

Do you truly know that? How? Is it just a projection of the current social mores as being "Natural" (tm)? I mean, arguments towards the Naturalness of patriarchy, white 'superiority', nationalism, etc have been made for centuries based on the circular reasoning "that's the way it is now because it's the right way". Our very ideas of competition, fitness, evolution, progress, power and hierarchy are derived from Christian, patriarchal, nationalistic, racist, and oppressive philosophical foundations. Compare Charles Darwin with Peter Kropotkin. Look at many other cultures' views of Nature. Traditional Western culture does not have the mandate on Truth.

Looking at a forest could you tell me who the leader is? What leading is going on? If the development of geological strata what power hierarchy is happening? How about a traditional example, an ant nest. The "queen" clearly has all the power, non? Hold on a minute -- upon what scientific, objective or empirical grounds do we lay the credence for this anthropomorphization of such an entirely different creature than our own? What is this 'hierarchal power' all about? Perhaps in a few other large mammals do we see this kind of 'hierarchy' take place, but can we at all stake this crumb of evidence as a totality of Natural order??

How about the "food chain"? There at the top, the alpha, the head of the predators isn't there a logical lie being perputuated? I mean, after all, wouldn't the bacteria and parasites that feed off that 'alpha' really be 'superior'? If you go into a factory and see 100 people working and 1 work master who says 'stop' and 'start' who really has the power? Is it really the work master? Or is it the workers?

When *I* look at complex natural systems I see the orderly anarchy of things coming into natural equilbrilium, the natural increase of things which are sustainable, and the very proofs against the artificiality of nominal power that the justifiers of social imbalance, the peddlers of oppression and tyranny, the very ones who see Nature as just another competitor would seek to indoctrinate the true possessors of work (the workers) as "Natural".

'Design by committee isn't practical. It might be a nice dream, but application of the idea can be disastrous. Even Linus would agree with the idea that control has to be applied to the process.'

Linus is a corporate shill who has no ideological or philosophical commitment to the principles of free software. This person who thinks that he 'owns' the development of Linux and therefore can sell out to one of the most fatuous and arrogant personages I've ever read (Larry McVoy) is of only coincidental importance to the development of free software ideals, in my opinion. If you seriously would like to know how development in an anarcho-syndicate or otherwise anarchist industry works, I'd be happy to refer you to a large number of published works, both on the web and in print.

'I was asked to join a startup a few years ago - they wanted a JSP site - and 95% of the whole deal was just sitting around and talking about "this deal" and "that deal." It was a bunch of crap, and irritating pseudo titles popped up all over the place ("director of blah blah blah" and "head of blabbity blah blah").'

But why do we even care about those titles? Let's say someone is VP of Marketing for a company of 150 people -- if this 'power' is not supposedly a fraud then they deserve our respect? To hell with all of the pompousness of titles. The emulation of a disgusting practice can hardly be viewed with more contempt than the disgusting practice per se.

'What would the purpose of this software *be*?'
What do you mean? If it isn't to make money or increase the power of the creators is it purposeless? Those shallow purposes are just corruptions of the true purpose of a software program; a word processor's purpose is more essentially to create and edit documents, not to create profit for the program's authors.

'Is there going to be funding?'
When a person needs a program that isn't available to them and they write it, what does funding add to it? If we weren't in a constant push-or-perish mode due to privation (in other words, if the natural resources and means of works were in common, equal possession then there would be enough for everyone) why would funding be of any importance?

'How are people going to survive while they make all this software?'
That question doesn't make sense until you take away all of the land, all of the means of work, and threaten me with imprisonment if I try to use any of these things without participating in usury. It's like going up to a bear and saying, "how are you going to eat all these berries around you if you don't have any money to pay my tax?" There is plenty for everyone! How true this is with software were copying complete functionality is virtually without labor or resources.

'Who's going to lead them? If there's no leader, then how is anything going to get done?'
Eh? Who leads you in tying your shoes? Who leads you in reading a book you're reading? Who leads you in having a conversation? Who leads you in living you life? If you start conversing with people do you tell one of them, "okay, I see you are the leader, tell us all what to talk about"? If you are hanging out with a group of your peers do you all just menially follow the will of one? I would judge true leadership to be guiding people, teaching them in a socratic way, in other words: helping them; not taking credit for their work, or casting them as subserviant or lesser in any way.

'Is there just going to be an anonymous "vision statement" to which people will be expected to fashion bits of code?'
The key part is mutual, equal respect and responsibility. If you write a document and I'm to code to it or vis versa, there should be a basic of mutual trust and understanding, not some hierarchal power bagatelle. The necessity in making anarchy work socially is nothing more or less than virtue. Earnestness, integrity, consideration, even love play a role in being a good person.

If one person has more experience with something than another than sharing that experience is truly a beautiful thing. That is one of the absolutely most wonderful things that free software projects can do.

'How are they going to know which bit of the code "belongs" to them?'
For what effect? Vanity? Fame? Natural fame is something to be admired, but the pursuit of the fame is folly.

'Isn't it going to be problematic if several people are working on the same code at the same time, but don't know it because they're either lacking the tools or the leadership to coordinate the work?'
I'm sorry but I don't understand what this has to do with anarchist work philosophy. I believe this is begging the question.

'If there's no team leader, then who's going to choose the tools?'
Workers know their work better than those who do not work, in my experience and belief.

'What technologies will be used? Is there going to be a design doc?'
These questions should be decided as well as possible using the *collective* experience of the group, particularly those who will be using the tools as the tools should be considered in conjuction with the workers' skills.

'Even written anonymously, a document describing the software would still function as a source of leadership. Updates to the document would also act as a proxy for a leader.'
How on earth would you define leadership? Any thing that guides? Any thing that instructs, codifies, describes? I don't get it. If you write down on a piece of paper, "I will do what is right" and then you do what is right who is responsible for doing what is right, you or the piece of paper?

In a playful mood,
chroootstrap

'All zealotry aside, Windows XP, let's be honest with ourselves, is extremely easy to use.'
In all honesty, Rory, if a person does not have prior experience with Windows 95+, Windows XP is not easy to use, much less extremely easy to use. Have you ever thought it strange that SO MANY people out there who have used computers for years don't know how to use the copy-cut-paste shortcuts or how to find files with the Explorer? So many people are actually afraid of the learning process because the have to learn by trial and error (or memorize their tutor/friend/family's example) in order to do anything new on their Windows machine. Go into a typical window in control panel and how much context help do you get? Almost none! A pithy one-three word name and value pair.

We need to get over the arrogance that Joe/Jane computer user is stupid and can't figure out how to cancel their print job because their dumb and come to terms with the fact that the real reason they can't figure it out is because we're stupid in how we design programs.

Right-click on the misnamed 'Start' button in order to Explore -- oh, of course, I must be a total idiot for not knowing that. Of course, if I was smart I would have clicked on Start then All Programs then Accessories then Windows Explorer then My computer, D:(...) (at least there's a CD icon) -- in order to run the startup program on the CD when another program quietly turned off CD autorun. I'm sorry, I'm just an idiot. Tell me, "Are you sure you want to open this directory????" and make me feel like a total idiot. It is completely intimidating and instead of feeling for these people, instead of sympathizing and truly trying to better our products to make things more logical, consistent, and intuitive we deride, ridicule, and patronize them (just look at your local customer support water-cooler chat).

It just doesn't have to be that way. People SHOULD feel like they can accomplish everything their computer has to offer. They should feel like they can explore and click on new things and try things and learn by themselves without a three volume manual and an arrogant computer nerd to point out their ignobility for not being familiar with CTRL-ALT-DEL. But they don't and it basically boils down to the arrogant, patronizing way software companies (including, perhaps especially including Microsoft) treat their customers.

'Unlike Linux, users aren't having to choose between various desktop managers, and then have to be confused as their apps pop on screen using different GUI toolkits.'
That isn't the problem and who cares if it looks different. Look at Windows Media Player for goodness' sake. Fundamentally (oh and I know lots of people refuse to admit this) people are darn smart creatures. Give them a fair shake, honesty, and patience and they'll figure out anything we ever did.

'It's pretty obvious that Microsoft went through a process to ensure that Windows would be easy to use.'
By copying the previous Windows design and adding some eye candy (further confusing many rote-learned users)?

'Microsoft listens to its customers, listens to its testers, and forces its employees to constantly work with the products they're developing in order to figure out where they fall short.'
Which is why they've stopped releasing new Internet Explorer versions (surely that doesn't have anything to do with them no longer competing with Netscape). And, why months after serious security holes are widely known about in IE and other MS products they still remain unpatched?

'It isn't a perfect system, but that's no big deal since there *isn't* a perfect system. In the end, it's perfectly good enough, and has resulted in *good* software.'
But, what if there are systems that are far better? Isn't that a big deal? Wouldn't it be better if people could actually improve it, perfect it if you will? What has caused the good software is the illegaly gained monopoly dominance. And they had to battle through completely (and in many cases willfully) non-standard libraries and tools creating almost completely unportable code because if it's made for Microsoft, than rewriting it is almost a complete port in itself (for the majority of native apps).

'At the end of the day, the actual users are often lucky if they can even find the "On" switch.'
*Sigh*

'Giving users too much choice is just asking for trouble. I don't think that users are a good example of demonstrating the success of OSS.'
Imagine a computer that you can look inside and out and actually figure out what's going on. A computer where almost all configuration is done in universally readable text files. A system where knowledge of a few basic conventions like 'man' pages and user/group/all bits is truly reusable knowledge. A system that is highly aware of portability and therefore follows frequently follows standards. A system where people are invited to create and modify their own software using tools as basic as shell scripts and as advanced as low level languages.

Most of Unix's design has a solid rationale that is consistent and that, in itself, is the single greatest aid in helping someone learn how to use their computer that I know of.

>> "'The "power hierarchy" exists throughout nature, and probably for a good reason.'"

> "Do you truly know that? How? Is it just a projection of the current social mores as being "Natural" (tm)? I mean, arguments towards the Naturalness of patriarchy, white 'superiority', nationalism, etc have been made for centuries based on the circular reasoning "that's the way it is now because it's the right way"."

I didn't say anything about patriarchy or white superiority - I said that there's a power hierarchy in nature, and there *is*. No amount of wishing otherwise is going to change that.

How many places are there in nature where there isn't a pecking order?

What about primates? Are you going to tell me that primates don't have power hierarchies?

"How about a traditional example, an ant nest. The "queen" clearly has all the power, non? Hold on a minute -- upon what scientific, objective or empirical grounds do we lay the credence for this anthropomorphization of such an entirely different creature than our own?"

If the other ants are *serving* her, then there's a hierarchy. I think you'd agree that not all ants in the nest are treated equally.

"When *I* look at complex natural systems I see the orderly anarchy of things coming into natural equilbrilium"

That "natural equilibrium" involves quite a bit of killing, and killing involves having power over something else - unless you think that gazelles are running away from predators for the fun of it, or for the exercise.

And, when the kill is made, why do certain lions get to eat before the others?

"But why do we even care about those titles? Let's say someone is VP of Marketing for a company of 150 people -- if this 'power' is not supposedly a fraud then they deserve our respect? To hell with all of the pompousness of titles. The emulation of a disgusting practice can hardly be viewed with more contempt than the disgusting practice per se."

I care about those titles in cases in which I believe the titles *mean* something. Have I been in situations where I was working with people who didn't deserve their titles? Yes! Have I been in situations where they *did*? Yes!

I want to agree with you - It would be really nice if we didn't have to set up these funny little offices and give some people more responsibilities than others, but when things need to get done in a reasonable time frame, people need someone to whom they can defer judgment on an issue.

We *could* all come to the table with equal stakes, but I'll tell you this: Most people *are not like you*, and the situation will get nasty. Whether you like it or not, a de-facto hierarchy will be established, and you will either accept it or leave. Think about pagdig - at every meeting there was at least one person who seemed to be trying to become leader of the group - we didn't ask for it, but it happened anyway.

"...if the natural resources and means of works were in common, equal possession then there would be enough for everyone) why would funding be of any importance?"

Good luck making that work.

It sounds ideal, but people *are not equal* - some are more talented than others, some more power-hungry, and they *will* mess up your system. We *all* have a portion of the brain dedicated to territoriality, and some among us are quite skilled at listening to it.

"It's like going up to a bear and saying, 'how are you going to eat all these berries around you if you don't have any money to pay my tax?'"

I don't know if you'd have a lot of time to ask questions while the bear feasts on your body, demonstrating that there may, in fact, be a hierarchy of power in nature.

"Eh? Who leads you in tying your shoes? Who leads you in reading a book you're reading?"

These are solitary activities - creating software, and especially complex software like word processors (which you mentioned a while back), would be collaborative, which changes the rules.

"Who leads you in having a conversation?"

I spend most weekends in Coffee People, conversing with whoever happens to be there, and can tell you that every conversation has its leaders. There will *always* be one person who speaks slightly more than the others, or who leads the conversation in new directions more than the others. The difference in participation might be very subtle, but it exists.

"The key part is mutual, equal respect and responsibility."

Turn on the news, Dominic - There isn't much "mutual, equal respect and responsibility" going around.

"If one person has more experience with something than another than sharing that experience is truly a beautiful thing."

"Shareing that experience" is just another way of saying "teaching" which is another way of *leading*.

>> 'How are they going to know which bit of the code "belongs" to them?'
> For what effect? Vanity? Fame? Natural fame is something to be admired, but the pursuit of the fame is folly.

People need to know what code they "own" so that they can *code* and *maintain* it.

>> 'If there's no team leader, then who's going to choose the tools?'
> Workers know their work better than those who do not work, in my experience and belief.

That clearly doesn't answer my question - given what you've just said, I have to assume that each person will be choosing his own tools. What's going to happen when different people want to use different build systems? Different languages? Different frameworks? Who decides what's to be used?

>> 'What technologies will be used? Is there going to be a design doc?'
> These questions should be decided as well as possible using the *collective* experience of the group

So the less experienced people in the group will have to defer to those who know more about the work being done.

I sense a de-facto hierarchy building here...

"How on earth would you define leadership? Any thing that guides?"

Let's just say for the sake of this argument that I think leadership involves centralizing decision making, authority, and guidance. The idea is to provide everybody on a team with the help they need to move a project toward a common goal.

If someone writes down the design for a piece of software, and if that design is *followed* by others, then I would say someone, somewhere, is leading.

"In all honesty, Rory, if a person does not have prior experience with Windows 95+, Windows XP is not easy to use, much less extremely easy to use."

If someone has no prior experience with a computer, then *nothing* out there is going to be easy to use.

Don't even try to tell me that someone's going to have an easier time with KDE or Gnome (even though common configurations of both try to mimic the layout of Windows).

Having used Linux/OS X/Windows XP on the desktop, I can honestly say, *with blinders on*, that XP has some very nice things going for it, and consistency is one of them. I've found meta-key combos for common tasks to change depending on what software I'm using in OS X, and there doesn't seem to be just one common clipboard, which has sometimes resulted in frustration when trying to copy and paste between apps - XP, for the most part, is devoid of these problems.

That's actually very important.

"So many people are actually afraid of the learning process because the have to learn by trial and error (or memorize their tutor/friend/family's example) in order to do anything new on their Windows machine."

What are you going to tell me next? That they just sit down at a Mac/Linux machine and just "get it?"

I knew some "non-technical" users who made the switch to Linux back when it was the Next Big Thing (I had to install it for them, of course - not one of them could figure out the process on his own), and they all eventually ditched it.

Common complaints? "It doesn't do what I want." "It doesn't work with [X] piece of hardware." "My video looks weird."

I suppose that I could have helped with the video stuff, but I don't really want to be the guy running around and fiddling with manually entering refresh rates for all my friends. Maybe this makes me a selfish bastard who thinks life is too short to have to manually configure a monitor, but whatever.

"We need to get over the arrogance that Joe/Jane computer user is stupid and can't figure out how to cancel their print job because their dumb and come to terms with the fact that the real reason they can't figure it out is because we're stupid in how we design programs."

Woo-hoo! At least Joe/Jane computer user can get something printing on a Windows system!

Other common complaint? "I can't print."

My answer? "I don't know how, either."

I'm not claiming that they're stupid - Put me in front of the french-fry contraption at McDonalds, and I'll destroy several tons of potatoes before I've figured the thing out *on my own*.

Get me to operate a car wash - a lathe - a loom, and I'll show you a small disaster.

Am I stupid for it? No. Did I call anybody else stupid? No. I know it's hard, but I think that some systems make things easier.

"Right-click on the misnamed 'Start' button in order to Explore -- oh, of course, I must be a total idiot for not knowing that."

Or double-click the aptly named (and often imitated) "My Computer."

"Of course, if I was smart I would have clicked on Start then All Programs then Accessories then Windows Explorer then My computer, D:(...) (at least there's a CD icon) -- in order to run the startup program on the CD when another program quietly turned off CD autorun."

I don't know how "My Computer" was removed from the default view of your Start menu, but it's there on all of my (vanilla) installs.

"I'm sorry, I'm just an idiot. Tell me, "Are you sure you want to open this directory????" and make me feel like a total idiot."

You get the "Are you sure" dialogs when you try to open folders that might contain data/apps that really shouldn't be touched if you don't know what you're doing. It isn't to make you feel like an idiot - it's to make you think twice before going in and possibly mangling things.

"They should feel like they can explore and click on new things and try things and learn by themselves without a three volume manual and an arrogant computer nerd to point out their ignobility for not being familiar with CTRL-ALT-DEL."

"Arrogant computer nerd?" Reminds me of one of the reasons I couldn't stand Linux. Just try to get help in a common forum without getting flamed to hell and back for not knowing that you should have compiled "x" with the "xyz" switch.

A common answer to questions was "RTFM." *That* sure made me feel welcome. Too bad there *wasn't* a three-volume manual with which to begin, and I *didn't* (oddly enough) instinctively know about the existence of "man."

The arrogance in the *nix world pretty much forced me to live your ideal - clicking around until I figured out how to make things work. Would have been nice to have had the option, though.

>> 'Unlike Linux, users aren't having to choose between various desktop managers, and then have to be confused as their apps pop on screen using different GUI toolkits.'
> That isn't the problem and who cares if it looks different.

Oh, OK - So, if it isn't a problem for *you*, then it isn't a problem at all, right? The answer to "who cares" is *me*.

I'm not the only one who thinks it's a problem, either. If you *truly* believe that it isn't a problem, then how do you explain this: http://slashdot.org/articles/04/01/04/1614219.shtml?tid=121

It's confusing as hell When every third app looks completely different. And, when I don't have "x" GUI library installed, it's a total pain in the ass to go through the process to get it on the box. "You need to get 'x" GUI library, and 'x' is dependent on the 'y' library, which is dependent on the 'z' library, which you need to get, but it isn't widely available, and there aren't any instructions, and............"

Blah. Give me OS X or XP any day.

They aren't perfect, but they have some problems nailed.

"Which is why they've stopped releasing new Internet Explorer versions"

They've stopped releasing on the Mac because it's obvious that Safari is getting the job done quite well.

They *haven't* stopped releasing for Windows, and there's a very substantial service pack coming out that includes fixes/new features (pop-up blocking/tabbed browsing (I think), etc..).

They're also furthering the system for Longhorn.

>> 'It isn't a perfect system, but that's no big deal since there *isn't* a perfect system. In the end, it's perfectly good enough, and has resulted in *good* software.'
> But, what if there are systems that are far better? Isn't that a big deal?

The hypothetical tone of your questions is very appropriate.

*Are* there systems that are far better? *How* are they better? For someone who doesn't seem to think that having every next app look completely different than the previous is a problem, I'm not sure that you and I would agree on what's "better."

>> 'At the end of the day, the actual users are often lucky if they can even find the "On" switch.'
> *Sigh*

Oh, please.

It's *true*. Maybe not for the hax0rz, but it's true.

It isn't to say that anybody is *stupid* (see previous explanation), but not everybody, intelligent or not, has a natural talent with computers, even as a user.

My father, by all accounts brilliant, *still* calls me with pretty rudimentary computer questions. The same is true for all of my non-techie friends.

Do I think they're stupid? No. Do I think they really know what they're doing with *any* system? No.

My girlfriend doesn't know how to drive. Do I think she's stupid? No.

"Most of Unix's design has a solid rationale that is consistent and that, in itself, is the single greatest aid in helping someone learn how to use their computer that I know of."

Of course. I'm sure that most computer users find "ls" easier to understand than popping open a folder and looking at the contents. Or that "cp" makes more sense than simply dragging files from one place to another. What about pipes and redirects? Totally intuitive, right? What about grep?

The "design" seems to have been centered around an economy of vowels.

Rory,

I'll avoid the little flame war going on and just respond to what you said:

You have a completely valid point. Micorosft *is* putting a lot of time and energy into listening feedback, and that is resulting in a greater number of good changes. However, it is still Microsoft's prerogative whether to accept such feedback or to disregard it. My point is that there is no such fundamental controlling right in the Free software arena. The best idea /always/ wins. But there is a caveat.

The evolutionary process of Free software can only apply to ideas that the community has yet produced. You brought up the user interface, and you're totally right. I, too, can't stand X, and I despise the Linux desktop. (I run XP on my desktop, and I use Linux sans X for my servers.) Microsoft's user interface is far superior to anything that the open source community has been able to produce. I believe the reason has to do with point of origin for the two products. Why is that? It's because Microsoft has complete control over the process. They can make arbitrary decisision that affect every person who uses their software, and the people using their software cannot change it. The de-facto standard becomes the definition of "easy to use" because everybody knows how to use it. Would there be such a standard if Microsoft hadn't had the extremely good business sense to create a monopoly for themselves? Or would the marketplace for operating systems look similar to the competition that occurs between the various X window managers?

At the end of the day, there is a place for both types of software. Commercial software has an advantage in that it can be very focused and high-energy all of the time. If there is money to be made, even if it's not cool or interesting, somebody will produce the software for it. But because it is driven by dollars instead of ideas, it might not be the best ideas out there. Conversely, Free software is driven by ideas rather than money. The net result is that it only continues to the extent that people pour their energy into it. It's tough to make a living hacking on my IRC bot. But you'll always get the best idea in the end.

Which you consider more valuable depends on you. Most geeks will probably end up balancing the two because they need both.

Brian

Wow, this has to be one of the more in-depth comment discussions I've seen in a blog.

I just wanted to chime in and say that was one of the most amusing descriptions of peeing I've seen.

We now return you to the debate in progress.

Brian -

First of all, let me thank you for not quoting me out of context or doing any of the other things that people *usually* do when talking with me about OSS. I'm not being facetious when I say it's really refreshing.

"My point is that there is no such fundamental controlling right in the Free software arena. The best idea /always/ wins."

Is that true, though? How is it decided which feature/project will be used? Is there a vote, or do a few people "at the top" make the decision?

From my experience, Gnome was the more "interesting" desktop while KDE was easier to use. However, it seems that the community has been head away from KDE and its various K* technologies, in spite of the fact that some of them really rock.

In other words, *who* decides what's best? A single decision has to be made somewhere. It's even more complex than that, though - different distros might choose to package different technologies, resulting in one standardizing on KDE while another standardizes on Gnome. In a case like this, which distro is correct? And, aren't the people putting together the distros acting as authorities by making the final call on what to use?

How is any single distro company different from MS in a case like this?

"The de-facto standard becomes the definition of "easy to use" because everybody knows how to use it."

I had considered that this argument might pop up, and I really should have qualified my claim that XP's easy to use by saying that what I really mean is that it isn't any harder than anything else already out there, and typically better in many areas.

It doesn't mean that it's "correct."

"Would there be such a standard if Microsoft hadn't had the extremely good business sense to create a monopoly for themselves?"

Well, did Microsoft create the monopoly, or did the market create it?

When I was growing up, the common deciding factor for choosing between a Mac and a PC was this: Do you want to be able to do anything with it?

Mac owners bought Macs because they loved Macs. PC owners bought PCs because they wanted a diverse selection of software.

We still see this to some extent today. I sat down with a Mac *fiend* the other day, and he told me that he was going to switch to Windows because he couldn't find the software he needed for the Mac. Software developers are free to choose which platforms they're going to support - it isn't MS's fault that my friend is now moving to Windows from the Mac.

I'm not certain, but I think that what I'm trying to say is that a great deal of the success of Linux/OSS is that, at least on the surface, it's *cheap*, but not necessarily *better*.

Kevin -

"I just wanted to chime in and say that was one of the most amusing descriptions of peeing I've seen."

Thanks :) I found myself about to type "So, I was peeing," and then decided that I wanted something different...

I bet your bladder is pleased that you relieved it *before* this all kicked off.

>And, why months after serious security holes are widely known about in IE and other MS products they still remain unpatched?

They have been fixed. Check the latest security update for IE. Unlike OSS, which doesn't particularly care who gets affected by a change, MS had legitimate reasons to careful structure a fix in this case that provided minimal impact. OSS zealots typically ignore the fact that MS often has security fixes far before problems actually happen, and often find and announce security holes themselves before CERT organizations find them these days.

> Imagine a computer that you can look inside and out and actually figure out what's going on. A computer where almost all configuration is done in universally readable text files.

I don't have to imagine. I used to use them. And this is one of the reasons I don't use Unix anymore. I don't give a rat's ass what format the config info is in as long as i can administer it easily. In the MS world, i push a button. In the Unix world, i open five text files and type like mad, hoping i don't screw something up in the process before i hit save. Now add 40 servers you need to do this simultaneously to and you see why this idea sucks.

> A system where people are invited to create and modify their own software using tools as basic as shell scripts and as advanced as low level languages

Support call: Um... I can't seem to print anything...
Support tech: ok, step #1...
<lapse 12 hour call>
Support tech: OK... WTF did you change in the code?!?! This is unsupported!
User: but... but... you said i could do this!

Bottom line: while most geeks think it's cool to modify the source of thier shells and OS's, the majority of the population thinks this is a BAD idea, and no company i know of would do it - this is why they pay hefty support premiums - they WANT out-of-the-box supported software.

Now, if you're talking about automation scripts and customizations, Windows has WSH, and you can write just about anything you want in .NET, which supports over 25 languages. So there isn't much imagination required for this in the MS world.

<flame>
All I can say is that I'm not too surprised by the high level of ignorance about Linux and Unix systems by people who have decided that the one-true-way is Microsoft's. From the passé argument that Linux users must choose between GNOME and KDE applications (simply not true for the majority of modern Linux distributions) to the belief that editing a commented, documented text file is somehow so much more difficult than navigating to a configuration panel with one word descriptions of toggle boxes, I don't see a serious understanding of modern Linux usage -- for goodness' sake most people can configure all they need to Linux these days in a fully graphical environment if they want to anyhow. A couple days with a Linux installation or four year old experience hardly makes you an expert in what working with Linux is like.

I know many, many Windows users who were plenty unhappy with Microsoft's efforts and remember no lack of such users back before Linux became a hot topic. They didn't have to claim that Mac/Amiga/OS2 was so much better just to come to the conclusion that Microsoft was far less than perfect. Does anyone remember the meme that Bill Gates was the devil? :) Frankly, Microsoft has been annoying users since the inception of Windows which, having since became catholic appears defendable to many as just the scapegoat du jour. Are you really so close-minded to not recognize that Windows has many flaws?

Where was the Microsoft fix before the half a dozen major e-mail virii to hit happened? The most recent fix for IE covered things that were reported more than 6 months ago. And, there are still more unfixed holes. Whose fine operating system is currently propagating the latest worm? Is it because they're the biggest and brightest? Nope -- it's because Microsoft has all of the security street creds of ratfink.

There is a huge difference in downloading the Microsoft .NET SDK, ensuring every computer that will use the script has the .NET runtime installed, writing a 20 line program in x language, compiling it and still figuring out how people are going to run it then doing a 2 line bash script.

Do people really think it is so terrible to go the command-line? I think I remember many pro-Microsoft people deriding Mac users for the same belief some years ago. Oh, I guess it has all changed now? Right, now we can configure everything in a binary blob that only one program can decode. How nice!

People can be so closed-minded and presumptuous. To think that all nature is some hierarchal power chart. lol...
Anthropomorphizing -- don't underestimate its power. For example, on what objective basis do we presume that the big mama any is being served by worker ants? Maybe she's a "slave". Perhaps they have all found "equality" already? What on earth are we talking about?? They are ants, not Europeans. It's like the way skunks used "rape" each other. Only recently has that description been revised in the scientific community as being a subjective, loaded term. Or how about the "war" that animal groups have on each other, nay, even plants and bacteria?? Of course if completely forget about the *other* chimp (pan paniscus), the work of Sue Savage-Rimbaugh, Frans de Waal, Jane Goodall and others we may very neatly accept that strict-hierachal-power relationships are the norm for chimps -- but what gives the right to take our own culture, our own human history, and presume that the oppression and slavery of our past and present is appropriate in describing another species? Come on, we've been doing it for hundreds of years. Its only been fairly recently that the mainstream scientific community has began to really rebel against such unscientific practices.

If a leaf is drifting down a stream is the stream the powerful master and the leaf the weak slave? Come on -- that's absurd! The normalization of hierarchal power is just a completely natural phenomenon of culture, like believing Christianity is normal, or homosexuality is abnormal, or that it is normal to wear head scarves or anything else. We are blinded by culture!

And, it is the same with Microsoftism. Blinded by our own use of it and the use by others, we justify it, we presume its superiority -- this may be true or not true, but the defining point is when we stop objectively, rationally consider the whole picture and consider ourselves "A Windows User" or "A Linux User", etc. I know that Linux, Mac, and Windows users all do this and they all are just as wrong when they respond to new information about alternatives with flaccid truism and regurgitated propaganda. It has all the same intellectual foolishness of religion.

For me, Buddhism and Hinduism appeal more than the Judiac(+) tradition because of the open-ended nature of their philosophical discourse, because of the encouragement of rational consideration of doctrine, not blind following. As Guatama Buddha said, "If I say something that you find to be false, through it out; if I say something true, keep it." This is why I like GNU/Linux as well -- the responsibilities are transferred to the users and everyone can participate in the development. This is in sharp contrast to accepting the dictates of a company.

It is much more like anarchism where as the loyalty of Microsofties smacks of nationalism to me. :P ;)
</flame>

And if that riles you up, just wait till I tell you what I think about Krispy Kreme!

full of hot air,
chrootstrap

This is going much further off topic than I ever expected - I don't think it's a bad thing, but I also don't have time to argue about software in terms of the world's major religions and political systems.

So, I'll address several small things that caught my eye (and apologize in advance for not addressing the entire comment - interesting as it was, I seriously have too much to do this week to get that involved with this):

1. "A couple days with a Linux installation or four year old experience hardly makes you an expert in what working with Linux is like."

Aren't you just as guilty of this, but with Windows? I know you had it installed for a while, but given some of your previous comments (having to right-click on the Start menu to explore), I have to assume that you might not know Windows as well as you think you do.

2. "Frankly, Microsoft has been annoying users since the inception of Windows"

There are annoyed users of every OS ever invented. I've been annoyed by things in every OS I've ever used, including those from Microsoft.

In short, I don't understand the point of this comment.

3. "People can be so closed-minded and presumptuous. To think that all nature is some hierarchal power chart. lol..."

Thanks for attacking me instead of the argument. It's nice to know that because I have a different opinion of things than you do that I'm "closed-minded and presumptuous."

The mocking "lol" doesn't really help, either.

4. "Do people really think it is so terrible to go the command-line?"

Just users.

And Apple - try to get tech support to help you out with anything regarding the command line - they won't do it. There must be a reason for this, eh?

5. "For example, on what objective basis do we presume that the big mama any is being served by worker ants? Maybe she's a "slave"."

If "she's a 'slave'," then there's a power hierarchy. It doesn't matter who fits into what role - it just matters that the roles exist.

Animals kill other animals and use the environment to their own benefit - this is power, and there's a hierarchy involved. The bigger fish eats the little fish - that's a power hierarchy in which one creature has dominance over another.

6. "If a leaf is drifting down a stream is the stream the powerful master and the leaf the weak slave? Come on -- that's absurd!"

Absurd? It sure is, but *you* brought it up.

<shrug>

7. "It is much more like anarchism where as the loyalty of Microsofties smacks of nationalism to me."

Regardless of the "open-ended" nature of GNU/etc., I've encountered plenty of Linux/OSS zealots who are much more "nationalistic" than anybody I've ever met from the Microsoft camp, and that includes the people I know who *work for Microsoft*.

We've both had experiences with people who would defend ideas regardless of how bad they are - I'm not one of those people. If I think something sucks, I'll say it sucks. If you're talking about/to anybody who acts otherwise, then you'd probably waste less time by addressing them directly.

"Aren't you just as guilty of this, but with Windows? I know you had it installed for a while, but given some of your previous comments (having to right-click on the Start menu to explore), I have to assume that you might not know Windows as well as you think you do."

I've used (I think) every released version of Windows for PCs from version 2 on up except 2003 and Longhorn, including Windows 2, 3, 3.1, 3.11, Windows for Workgroups, 95, 98, ME, XP Home/Pro, NT 3 on up (admittedly not the earlier ones), 2000 Pro/Server/Advanced Server. I have Windows XP running at home which I happily use MinGW (http://www.mingw.org) on. I've done Windows development using VB, VC++, COM, Access VBA, OLE cruft, ADO DB junk, etc over many years and I still think it sucks. The example was of the round about way to open the Explorer. If you have a file in the desktop and show its location, it will open the Explorer too but that's not what I'm talking about.

'Thanks for attacking me instead of the argument. It's nice to know that because I have a different opinion of things than you do that I'm "closed-minded and presumptuous."'

I'm sorry. I didn't mean that to be taken to heart. It was a flame after all! ;)

">Do people really think it is so terrible to go the command-line?
Just users."
But don't you remember all of the Microsoft users who used to argue out their ears about how silly it was that Mac users didn't have a command line? When did it become such a sticking point? Once an OS is predominately GUI oriented then apparently the CLI has the plague. Are words and text really that awful?

"And Apple - try to get tech support to help you out with anything regarding the command line - they won't do it. There must be a reason for this, eh?"
Because that's been the Apple way for more than a decade? Is that still true anyhow?

'If "she's a 'slave'," then there's a power hierarchy. It doesn't matter who fits into what role - it just matters that the roles exist.'
My point is that there is no such role their except what we project from our culture. The idea that she is a "slave" is completely absurd in as much as what being a "slave" is. Was she sold into the nest? Does she lack "freedom" and "rights" accorded to the other ants? We are confusing our culture with ants' when we presume those things.

"Absurd? [The leaf and stream example] sure is, but *you* brought it up."
But isn't that part of nature? How about the rain? Is there a power relationship in that? Is killing the definition of power? What is dominance? Is using the enironment for ones benefit hierarchal power? How? Just because a big fish might eat a little fish doesn't mean it is the little fish's master or oppressor. The complexity exceeds the treatment that pigeon-holing it into such human-specific terminonology yields.

'Regardless of the "open-ended" nature of GNU/etc., I've encountered plenty of Linux/OSS zealots who are much more "nationalistic" than anybody I've ever met from the Microsoft camp, and that includes the people I know who *work for Microsoft*.'

Sure, I believe it. It's just that you often seem to enjoy slamming free software and it seems (to me) like that's because you're somewhat zealous in trying to defend Microsoft. Some people like GNU/Linux just because it's not Microsoft, many people just like to feel like their tastes are better than others' irregardless of the facts. But, there are many, many people who geniunely like and appreciate free software and can see the many unique qualities that it has socially and technologically that go far beyond simply being free of cost. Sharing code can truly be a wonderful thing.

"I've used (I think) every released version of Windows for PCs from version 2 on up except 2003 and Longhorn..."

OK - you know your stuff.

"But don't you remember all of the Microsoft users who used to argue out their ears about how silly it was that Mac users didn't have a command line? When did it become such a sticking point? Once an OS is predominately GUI oriented then apparently the CLI has the plague. Are words and text really that awful?"

I was one of the MS users who argued that GUIs were stupid - but I also understood that GUIs seemed to be much easier for the people around me who wanted to use computers for everyday things like word processing.

I think so-called "Power Users" (and, please - let's not argue about the use of the word "power" here - we both know what is meant by the term) might get much more out of the command line than others, but that doesn't mean that *all* users would.

The thing about the GUI is that there's a *lead* - a way for you to get involved with the system without having to make the first move.

GUIs can be more interactive - there's more of a dialog going on, and that can help people who don't quite know how to make the first move.

With a command line, it's like writing a novel - you have to draw the knowledge out of yourself in order to make something happen.

GUIs work more like Mad Libs - they sort of ask users to "fill in the blanks."

I can get started with a GUI by just pointing and clicking - with the command line, you have to start by knowing some commands, and that takes more effort.

I personally wouldn't want to use an OS that didn't have a command line, but I wouldn't argue that the command line is the best general method of interacting with the OS for most users.

"My point is that there is no such role their except what we project from our culture. The idea that she is a "slave" is completely absurd in as much as what being a "slave" is. Was she sold into the nest? Does she lack "freedom" and "rights" accorded to the other ants? We are confusing our culture with ants' when we presume those things."

I didn't say anything about slavery - I'm talking about power. I think it can be pretty well argued that there are many places in nature in which one animal obviously has power over another.

I don't *want* to be eaten by a bear, but one could overpower me and that'd be the end of it. Sure, there isn't an org chart or anything, but there's a clear dominance.

If I have a gun, then I probably move up the chart so that I'm more powerful than the bear.

There's a relationship there, and power is involved.

Address that for me: Does the bear have power over me in the sense that it's able to kill and eat me? I think so. Does the big fish have power over the smaller fish in that it's able to kill the smaller fish? I think so. Does a hammerhead shark have power over about a billion other living things because it can kill and eat them? I think so.

I'm not talking about politics - nobody's saying "please" and "thank you" in these scenarios. It's just about raw power, and it exists in nature.

"But isn't that part of nature? How about the rain? Is there a power relationship in that?"

There might be, but I think it's probably reasonable to suggest that we just stick to your traditional creatures here. Bacteria/fish/humans - these makes sense for the argument.

The reason we brought inanimate objects into the discussion is that I mentioned a design doc serving as proxy for the author, which indicates power, and I still believe this - there was intent involved.

I might be wrong, but I don't feel that I'm nuts for arguing that the leaf didn't intend to land in the stream, or that the rain didn't intend to fall.

But, that's just me and my western background of not believing that rocks have souls.

"Sure, I believe it. It's just that you often seem to enjoy slamming free software"

When you say "slam," I think of trashing something just for kicks.

When I post these OSS things, it's because I've been stupid enough to believe that I've had a thought that adds value to some greater conversation.

I don't have anything against free software, nor do I have anything against Linux/etc.. My involvement with OSS, however, led me to find certain aspects of the culture really irritating. When I write about it, it's because I'm making an observation.

Just stand back from one of these arguments for once and really try to determine where the slamming is coming from: I say that I don't like the tendency to spend too much time on choosing a license, and you come in to accuse people of crimes against humanity. The conversation somehow gets into this "OSS vs. Capitalism vs. Christianity vs. God" thing, and that's something *you* bring to the table.

If you read my post without bringing any of that to the table, then you might see that I'm not slamming OSS - I'm talking about an aspect of it that I didn't like. If I wanted to slam it, then I'd head over to osnews and say something like "linux sux0rz d0odz."

What's funny is that it's all so very simple: I'm pro-MS because I like the products. I'm anti-OSS *culture* because I've had some negative experiences and seen things I really haven't liked.

I'm not against your ideas. I think the world would be lovely if all we ever did was roll around in the tall grass with pandas while coding hunger out of existence, but that isn't what the post was about.

That isn't to say that you shouldn't respond in whatever way you choose, but, for the love of god (just an expression), try to see that a lot of the "slamming" is being brought to the table from the outside and that I'm *not* trying to limit the freedom of people on this planet by saying that pompous license switchers suck. In a way, I'm saying "OSS would be better if we *made* it better," and that's not a bad message for "your camp."

Ah, thank goodness for your patience with my stodgy camp polarity. I'm sorry if I got hot headed; I guess I've been worried you were brainwashed into the Cult of Bob when you visited the MS campus. I know I can be overly political sometimes. It just all the capitalism, consumerism, conservatism, *-ism I observe day after day builds up and I feel very alone in my views and frustrated with what I think the status quo is. So when ever I'm frustrated about something specific, the more general frustrations spread into it.

I'm actually very frustrated with the majority of the "OSS" crowd for continually denouncing the political and social implications of free software, even the basic philosophy while meanwhile bashing commercial companies. There many who are just opportunists, with no more interest in the liberation that free software is about than which soda is on sale at Walmart. And the pompous power pushing of some of the people working on free software is just embarrassing.

I've found that the majority of software developers are quite conservative, though they tend think of themselves as being moderate-liberal. They are like young Republicans. There is a libertarian streak, generally a very pro-capitalist streak too. Well, maybe after they are brought down from the 'upper-class' through out-sourcing they'll rethink their market philosophies. I've met almost no one who believes in anarchism and few who believe in various degrees of socialism. I feel like a pariah even though I feel that I just advocate the simple reality (reality to me at least) that class, nations, and hierarchal relationships are fictions. Sometimes it just all bubbles forth like froth happily lost in a stream.

For me free software is about much more than just some tools. It is a functional social expression of ideologies that I cherish and believe in.

Thanks for one of the best descriptions of GUI-CLI interactions I've ever read.

"I guess I've been worried you were brainwashed into the Cult of Bob when you visited the MS campus."

If it makes you feel better, Bob doesn't get any respect in Redmond, either.

But, the visit was just pure fun. I was already a fan, and I got to meet some of the people behind some really cool work (MSDN).

There wasn't any brainwashing at all. In fact, I was really shocked when I brought up some Linux/OSS stuff during lunch (not in a negative light), and found nothing but interest in the reactions - they didn't try to burn me down or anything.

Absolutely no brainwashing at all (although they *did* pay for lunch).

"I feel very alone in my views and frustrated with what I think the status quo is."

You're not alone in feeling that way at all. You're rare in that you express it with so much gusto, but I've known plenty of people who have felt the way you do about life.

I feel the same way about some things, but not all, and I tend to approach it differently than you do - for me, it isn't a pervasive thing. When I'm talking about OSS, I'm talking about OSS - the fact that I think the medical system in this country is totally fucked is something I might address in another post.

"I'm actually very frustrated with the majority of the "OSS" crowd for continually denouncing the political and social implications of free software..."

Whoah - that's a new one to me. I didn't know there was such a strong presence of such people.

To be honest, though, I found it hard to see past the "M$ sux0rz!?! LOL" comments, so I might have missed a few things.

"There many who are just opportunists"

Three guys I know: I, B, and M.

"And the pompous power pushing of some of the people working on free software is just embarrassing."

An on-topic comment! Woo-hoo! (sorry - couldn't resist :)

"I've found that the majority of software developers are quite conservative"

Agreed - Regardless of the camp. I've never encountered more racism/sexism/gay bashing/this bashing/that bashing anywhere else.

"For me free software is about much more than just some tools. It is a functional social expression of ideologies that I cherish and believe in."

Believe me - after spending some time with you, I'm quite fully aware of this :)

And I wouldn't go attacking it. I was just nitpicking about one very small aspect of the OSS universe. I'm down with the philosophy - just not some of what I've seen in practice...

> Are you really so close-minded to not recognize that Windows has many flaws?

Did I say Windows was flawless? You make far too many assumptions. Stop it and your blood pressure might get back to something normal. I'm very aware of Windows' problems as i'm aware of the problems in many OS's, and when I find them, i pick them apart with equal vengence.

> it's because Microsoft has all of the security street creds of ratfink.

This reaks of hypocrasy. Here you are saying others have no clue when it comes to Linux because they have some old perception of it. While I would be the first to say that was the case four years ago, it's no longer true, your personal perceptions aside.

> Do people really think it is so terrible to go the command-line?

Yes. Doesn't sound like you've administered large server farms. And your average user doesn't like them either - remember, not everyone is a dyed-in-the-wool IT geek. Don't get me wrong, i like command lines. But your scenario was a BAD idea. Like all tools, each has its uses, but if all you have is a hammer, everything looks like a nail i suppose.

>People can be so closed-minded and presumptuous.

Says the pot to the kettle.

<snip a bunch of philosophical/political/religious statements that have little bearing on the technical merits of the conversation>

> And, it is the same with Microsoftism. Blinded by our own use of it and the use by others, we justify it, we presume its superiority

Again, you presume too much. I'm like Rory in the sense that I *LEFT* the MS camp in the early 90s when i got fed up with the shoddy nature of their stuff at the time. But unlike some religious zealot blind follower of the golden penguin, when they got their act together, i didn't brush aside the technical merits of their new products, which I now enjoy working with more than their competitors' products. I have ZERO loyalty to the company - or ANY software organization for that matter. If they suck tomorrow, i'll find someone else who has better tools. Your efforts to peg and stereotype me are so typical of the cultural behaviour i despise in OSS, and completely hypocritical on your part.

> I know that Linux, Mac, and Windows users all do this and they all are just as wrong when they respond to new information about alternatives with flaccid truism and regurgitated propaganda.

Good advice. Use it. For once, we seem to agree.

Honestly, there are a great number of good products coming from OSS, and they have some incredibly outstanding people working the various organizations. I've communicated with the Mono guys, and they are both cooperative and spectacularly intellectual. It's the bloody zealot idiots who ignorantly bash MS at the drop of a hat that tick me off. They are always quite guilty of the same crimes they accuse the MS crowd of. And more pathetically, they give the hard-working people in OSS a bad name. Frankly, i'm sick of it.

First off, those who call it Open Source (a la OSS) have missed the lecture on what Free Software is and why it is Free as in Freedom not Free as in Beer. The purpose of calling it Open Source had a lot to do with divorcing it from its philosophical grounds. The free software philosophy is very much a part of what I value about the products and they have a considerable amount to do with what I like about free software regardless of the technical merits, which I find generally better.

"Did I say Windows was flawless?"
....
"Your efforts to peg and stereotype me are so typical of the cultural behaviour i despise in OSS, and completely hypocritical on your part. "
Every comment in your previous post was negative towards free software and positive towards Microsoft. You clearly indicated the supposed superiority of Windows software in your points. For example:
"In the MS world, i push a button. In the Unix world, i open five text files and type like mad, hoping i don't screw something up in the process before i hit save. Now add 40 servers you need to do this simultaneously to and you see why this idea sucks."
Based on this, I'm not sure how I'm being hypocritical and (incorrectly) steorotyping in suggesting that you do "not recognize that Windows has many flaws" or that you "presume its superiority".

"While I would be the first to say that [Window's poor security] was the case four years ago, it's no longer true, your personal perceptions aside."
Hmm... Four years ago was before the Slammer, Blaster, Sobig, Mydoom, etc not to mention the I Love You virus and its friends. I would generally agree that Windows security was worse 4 years ago, primarily because more people were running 95, 98 and ME. NT and 2000 were not especially bad and XP is almost as good as 2000, but there are several serious design flaws in the Windows system that contribute to its low level of security. To this day Windows has retained its heritage of being designed as a single-user system that defaulted to 'admin-level' access in most (previously all) things. Part of the problem is security through obscurity, part of it is the aforementioned heritage, and part of it is Microsoft putting marketable features before secure design. There are literally hundreds of exploits that allow people to crack into Windows boxes and almost any badly written Windows network application can allow an intruder to take over the machine.

"In the MS world, i push a button. In the Unix world, i open five text files and type like mad, hoping i don't screw something up in the process before i hit save. Now add 40 servers you need to do this simultaneously to and you see why this idea sucks."
....
"Doesn't sound like you've administered large server farms."

What are you talking about? So to administer a Windows server farm I just press one button?? I think you mean the Mac OS X server farm ;). Seriously, administering Windows servers is NOT that simple. I have in fact done Windows server adminstration for an ISP before and there were lots of things a person needed to know and typically the online help (and especially the on-window help) was insufficient. This is especially compounded if you want to ensure your servers are secure.

The more important point I want to make, though, is that it can be very easy to administer Unix networks. I mean, come on, Unix has been used in networks and multiuser environments from the very beginning more than thirty years ago! There are many, many ways to update many boxes at once in Unix ranging from using a NFS server to automatically pushing the information with scripts to regularly polling for updates to many other solutions. Please go explain to Google how they should replace their Linux GWS boxes with Windows so that they can 'push a button' instead of clumsily editing 'five' text files. One of the nice things about text config btw is the ease with which a GUI configuration tool can be wrapper around it.

I'm sorry if I'm coming off as a dick in this. But I think these are legitimate counter arguments and are not just some rehashed anti-Microsoft spleen venting. Degrading Unix because it isn't dependent on a GUI seems to be really missing the big picture to me.

Cheerfully yours,

chrootstrap

> The purpose of calling it Open Source had a lot to do with divorcing it from its philosophical grounds.

I have no idea how true that is. The fact that there are so many different licenses can only reflect the fact that people have different goals and philosophies. If so, I hardly think there's anything wrong with that. You live in an extremist world, and most of the others are trying to exist in this one.

> Every comment in your previous post was negative towards free software and positive towards Microsoft.

I countered three of your comments, and you automatically assumed i found MS *faultless*. That's a very large leap of reasoning.

> Four years ago was before the Slammer, Blaster, Sobig, Mydoom, etc not to mention the I Love You virus and its friends.

Glad you brought them up, because that's precisely what i'm talking about. MS had patches for the vulnerabilities that each of these worms took advantage of months before the worms were introduced. In fact, it was MS who posted the vulnerability that Blaster took advantage of, not some 3rd party CERT organization. However, MS suffers from the worst breach of security all OS's hava - the user. People didn't want Windows Update to be automatic because then it would "be too intrusive", but when it's not automatic by default, nobody pathces the system. Let's not forget that OSX, Debian, and Gentoo all had vital kernel security flaws during the same time period you mention here that could lead to remote control with root access. Mac is feeling the same sting as MS because people aren't upgrading (and they are charging $129 for the upgrade, whereas the MS patches were free). In fact Debian had 2 seperate incidents of similar critical security flaws, and it's still unknown how many other linux copies that aren't using the newest kernel are still affected by it. Let's say that for a week, Linux magically traded positions with MS and had 96% of the desktop market. I'm still waiting for someone to show me the uber-cool linux security feature that is going to stop 150 million technically-ignorant users from opening a trojan they just got from their nifty free P2P file sharing program they downloaded. That's the position MS is in right now. Now four years ago, there was certainly a reason to complain. Unix, probably the most-hacked OS in history, set the stage for nearly every Windows attack to date (particularly the buffer-overrun worms). It took Unix decades to get to where it is now in terms of security. It ticked me off that MS hadn't learned a thing from Unix until they got their heads beat bloody. But it would be exceedingly naive to think that Linux wasn't there either. I still have the T-Shirt that says "My other computer is YOUR linux box" from back in the day. So, despite you softening your stance (which I can appreciate), this is in fact, a rehashed anti-MS spleen venting.

> almost any badly written Windows network application can allow an intruder to take over the machine

It's very naive to think this statement doesn't apply to any other operating system. And poorly-written applications are hardly the fault of the OS maker. I wouldn't blame HP or IBM if what screwed the HP-U or AIX server was a 3rd party network app - i'd be jumping down the appropriate throat.

> What are you talking about? So to administer a Windows server farm I just press one button??

Your missing the entire point. You were talking about how great it would be to have all config in text. The point is that it really doesn't matter to admins. Admins want tools that ensure administration is *easy* and *not prone to errors*. And yes, these tools certainly exist for other OSs - in fact, we use 3rd party tools to administer Windows clusters because the MS applet is rather poor at this. The format of the data is rather meaningless if you can support standards and protocols like WBEM or DS. On the other hand, *forcing* admins to use nothing but text editors and command prompts is the bad idea i was highlighting.

> This is especially compounded if you want to ensure your servers are secure.

Windows hardening kits are available.

> Seriously, administering Windows servers is NOT that simple.
> Degrading Unix because it isn't dependent on a GUI seems to be really missing the big picture to me.

Oracle and IBM seem to differ with your opinion. Both companies have launched new initiatives to create more and better GUI admin tools. In fact, one of the top-5 features of the next version of Oracle and IBM's Stinger are a host of admin tools that make them more "like MS SQL Server". That speaks volumes, IMO. Furthermore, I'm not degrading Unix because it doesn't *depend* on a GUI, but rather that in many cases, there is no GUI *option* - and that holds true more for Unix software than for Unix itself, where this problem has been alleviated to some extent by various other software vendors and Unix distributors with value-add software.

"The fact that there are so many different licenses can only reflect the fact that people have different goals and philosophies. If so, I hardly think there's anything wrong with that. You live in an extremist world, and most of the others are trying to exist in this one."

I'm not sure you understood what I was talking about. It is not a matter of licensing, it is a matter of the deliberate effort to 'sanitize' the free software concept by removing the element of liberty. Taking the efforts of the FSF and GNU and removing all of the intentions of freedom to make it just another collection of software packages isn't exactly a neutral thing to do. It is a stance against expressing the views and intentions encapsulated in free software because it might not be 'corporate-friendly'.

"I countered three of your comments, and you automatically assumed i found MS *faultless*"
When did I say that you found Microsoft faultless? I believe what I said was that you
did "not recognize that Windows has many flaws".

"MS had patches for the vulnerabilities that each of these worms took advantage of months before the worms were introduced."
Oh really? The SoBig virus (the most infectious in history) was not addressed prior to going wild IIRC and even then Microsoft only told users to download one patch when really they needed several -- dozens in some days. Similarly the current MyDoom virus wasn't addressed previously. ILOVEYOU, Melissa and Core Red were all propagated by the poor default attachment handling setting in Outlook, and I don't recall the patches to change the default behaviour being out before the first worm.

"Let's not forget that OSX, Debian, and Gentoo all had vital kernel security flaws during the same time period you mention here that could lead to remote control with root access."
Bzzt! ;) While there was a fairly serious exploit for OpenSSH, I don't recall OS X having a kernel security problem (if so, please correct me). Debian has two very distinct branches, development and stable and I wouldn't be terribly surprised if there was an exploit in the development branch or at a certain point for Gentoo; indeed I remember a kernel level exploit not that long ago, but it was hard to do, no actual exploit code had been found, and there was considerable debate by the kernel developers about whether it could actually be done. On the other hand, security risks are regularly discover and discussed in the LKML (Linux Kernel Mailing List) and it would be disingenious to imply that Linux was immune to security problems.

However nothing on the scale of any of the previously mentioned worms and viruses has happened with Linux or Mac OS X. Also, for what its worth, all updates to the "major" version of OS X are free which include all the security patches that I'm aware of. I remember a whole slew of them for Internet Explorer ;).

"...it's still unknown how many other linux copies that aren't using the newest kernel are still affected by it"
The kernel is a fast moving target. The new stable release series is 2.6.x. I think you might be refering to security problems with the 2.4.18 version, I'm not sure. By the way, Debian has an excellent upgrade system, apt, that helps a lot with these kinds of patches.

"I'm still waiting for someone to show me the uber-cool linux security feature that is going to stop 150 million technically-ignorant users from opening a trojan they just got from their nifty free P2P file sharing program they downloaded."
There is, in fact, just such an uber-cool feature but I'm not sure if it'll help the 'technically-ignorant users'. Namely: USE THE SOURCE, LUKE. It is SOP to compile your own software packages which is usually just a matter of './configure; make install' in these days of autoconf and friends. Unless its from a big company, most Linux software packages are recieved in two forms: (1) as a precompiled package (a binary RPM for example) from a known source such as a MD5 checksummed ISO from a Linux distributor (e.g. a Fedora install CDROM) and (2) as a source tarball which is compiled by the user.

"> almost any badly written Windows network application can allow an intruder to take over the machine
It's very naive to think this statement doesn't apply to any other operating system. And poorly-written applications are hardly the fault of the OS maker. I wouldn't blame HP or IBM if what screwed the HP-U or AIX server was a 3rd party network app - i'd be jumping down the appropriate throat."

Oh, but there are some very distinct differences. Let's say you crack through Apache into a Unix box. What have you got? Not root access. Many servers work on Unices without root permissions, in fact with very limited permissions. This includes databases, FTP and HTTP servers, and remote shells. The fundamental multiuser security design of Unix is well tested and quite good; this was something inherited by Linux and not really something they deserve specific credit for. However, the security personality stuff that NSA has been doing and is now part of the 2.6 kernel is very cool security-wise.

Here's the whole truth: while I'm an ardent supporter of the FSF and GNU, I actually much prefer NetBSD to GNU/Linux in terms of the kernel and the overall philosophy. In particular, I don't like the Linux kernel that much. Having seen and worked with the innards of quite a few free kernels, I think Linux, while having loads of drivers (whose importance shouldn't be underestimated), is not at all the best.

I also think Microsoft has improved quite a lot. On the other hand, they're still Microsoft, trying to co-op the features of Java and pass them off as something original for example, paying a lot of lip service to security while taking a wait-and-see attitude with many known holes, etc. But, all in all, they are not that bad anymore -- certainly nothing like the 16-bit days of Windows..ME.

"You were talking about how great it would be to have all config in text. The point is that it really doesn't matter to admins. Admins want tools that ensure administration is *easy* and *not prone to errors*."
This may or may not be true depending on the admin. An experienced Unix sysadmin may very will prefer to be able to see the comments and whole picture of a config file rather than have to learn and wade through a GUI program. I know I prefer it. Still, many people do prefer GUI config tools and that's part of why they have proliferated on Linux. I think the format does matter because text gives you the option to do either. Text is also easily mungable with non-GUI tools for that matter.

"Oracle and IBM seem to differ with your opinion. Both companies have launched new initiatives to create more and better GUI admin tools."
Of course they are. It is an absolutely typical thing that they can sell for the 'next version', etc, particularly aiming right at the Windows-only admins who want/need a GUI tool. And a GUI can be a real help, particularly with a very complex database like Oracle. Still, just like navigating through menus, it can be might inconvenient compared to simply knowing the short-cut/editing the config file directly. It is not a bad thing and I recognize that.

Of course, a very simple answer to all this would be, "Yeah, well look at Mac OS X -- that's both worlds" and it's true to an extent. But, some of us codgers do sincerely enjoy the CLI and have a deep appreciation for the way text files, source code and commonly available compilers have given a system that has not just been suited towards one particular time's trends, but has been able to grow and evolve over the past thirty years.

Cheers,

chrootstrap