Slashdot Mirror
What Open Source Shares With Science
An anonymous reader sends in a philosophical piece at ZDNet about the similarities between open source development and the scientific method. Here's an excerpt:
"The speed of progress is greatly enhanced by virtue of the fact the practitioners of Science publish not only results, but methodology, and techniques. In programmatic terms, this is equivalent to both the binary and the source code. This not only helps 'bootstrap' others into the field, to learn from the examples set, but makes it possible for others to verify or refute the results (or techniques) under investigation. In an almost guided-Darwinian evolutionary fashion, this makes the scientific process a powerful tool for the highlighting, analysis and possible culling of ideas and concepts; less useful ideas and hypothesEs die, and likely contenders come sharply into focus. Newton made his famous comment about 'standing on the shoulders of giants,' in part, to indicate that his contributions to human knowledge could not have been achieved solely. He needed the 'firmament' beneath him hypothesized, tested and confirmed by generations of scientists, philosophers and thinkers before him, over centuries."
Sadly, education has yet to follow this trend. Computer Science and Computer Engineering classes have yet to implement significant group collaboration. And while the hack tenet of "something that has been done once shouldn't be done again" was a conceived by some bright students, educators still give identical tedious projects that have the students complete in isolated groups, many times of consisting by just. There has even been an instance of a student being threatened to fail a class because he posted the source code of his project. How can we expect future developers to collaborate when their education forces a way to work that is very alien to the open culture and resembles that of a proprietary company
Why hasn't the scientific community produced open textbooks, free to re-print, photocopy and distribute (a la Creative Commons license)
Why is it hard for pioneering ideas like that of the state of California trying to open their school textbooks to be implemented?
Are you sure? Because science is done by a handful of "qualified" people working in ivory towers. A cathedral staffed with priests, if you will.
Open Source, though, is more like a bazaar. Wild and eclectic, the bazaar atmosphere takes the best and worst of everything, stirs it together, and produces some of the finest things found anywhere. Everyone has a say and anyone can set up shop.
I'm no millionaire, but I'd say that Open Source is much more like a bazaar than a cathedral.
...are the controls.
Yay for factual errors!
Newtons comment with regards to 'standing on the shoulders of giants' was actually just a jab at Robert Hooke (the two eminent Physicists hated each other, with the phrase originating in a letter Newton sent to Hooke).
However, Hooke was of significantly smaller stature than Newton, so by 'standing on the shoulders of giants' Newton was telling Hooke that he had learning nothing from him.
Although a fantastic scientist Newton was a very poor example of a human being, he was rude, offensive and incredibly stuck up.
As long as scientific results and techniques are hidden in very expensive privately-run journals and conference proceedings,
it cannot in any sense be considered open in the same sense as open-source or "fsf-free" software.
I would like to pursue scientific research as an amateur, but am prevented from doing so.
And this problem doesn't apply only to me, but to countless fully qualified scientists whose institutions cannot
afford the knowledge.
Science badly needs a Bastille day.
Where are we going and why are we in a handbasket?
proprietary software the equivalent of intelligent design?
That doesn't quite sound right to me...
Open source is like a box of chocolates. I'm not sure why...I just wanted to say that. It's what I thought of when I read the story.
People would be surprised how much software is developed under grant from the gov and is NOT open source. Some institutions like Cal Tech refuse to release their source code or even license it under an open source license that lets them retain copyright.
crickets.
http://slashdot.org/comments.pl?sid=1259963&cid=28244145&art_pos=9
I think I posted it before but I can't find it off hand.
putting the 'B' in LGBTQ+
I wrote on that here: ...
"Society of Manufacturing Engineers (SME) Licensing issues"
http://groups.google.com/group/openmanufacturing/msg/da82d9fd52265dbc?hl=en
"So basically, SME is using post-scarcity charitable dollars and tax
exemptions to finance the creation and distribution of artificial scarcity
of manufacturing information, which otherwise they could put up for free for
all on their website.
To be blunt, I feel it is unethical for a tax-exempt non-profit to withhold
17000 papers on manufacturing from easy global distribution, just so some
current staff can make money for future activities by creating artificial
scarcity. That is 20th century ethical reasoning, now outdated by easy
distribution via the internet at no incremental cost (I'm sure there are
many places like Ibiblio or Archive.org who would host them for free if
bandwidth was an issue). And making things "members only" doesn't really
solve it, and I'd expect even members can't redistribute content or make
derivative works without special permission. "
From stuff I wrote around 2001:
http://www.pdfernhout.net/open-letter-to-grantmakers-and-donors-on-copyright-policy.html
"The alternative of allowing charitable dollars to result in proprietary
copyrights and proprietary patents is corrupting the non-profit sector as it
results in a conflict of interest between a non-profit's primary mission of
helping humanity through freely sharing knowledge (made possible at little
cost by the internet) and a desire to maximize short term revenues through
charging licensing fees for access to patents and copyrights. In essence,
with the change of publishing and communication economics made possible by
the wide spread use of the internet, tax-exempt non-profits have become,
perhaps unwittingly, caught up in a new form of "self-dealing", and it is up
to donors and grantmakers (and eventually lawmakers) to prevent this by
requiring free licensing of results as a condition of their grants and
donations."
A 21st century issue: the irony of technologies of abundance in the hands of those still thinking in terms of scarcity.
By argument, I mean in the logical sense, such as a "claim" bolstered by "proof, information, and example". (I'm trying to separate the /. process of somebody posting something and someone else immediately disagreeing. But, Hey, I'm not trying to start an argument here about slashdot postings...)
So, I wonder if this argument has been used in the patent "process vs. product" or "software patent" courts. It seems to me that patents are generally awarded to the "products of Science" rather than the science itself. If code processes, algorithms, and concepts are Science, then patents should only be awarded to the "products of Science" such as individual chips or other hardware that utilizes the software and not the software itself. This argument could help clarify the boundaries of the patentable domain.
"The mind works quicker than you think!"
Software, like science, produces a non-rival public good. (Nonrival means it is not consumed when somebody uses it.) But there are private research companies just like there are private software companies.
I used to work in a publicly-funded virology lab studying Hepatitis C Virus (HCV). My biggest result was finding this particular human gene that HCV required in order to infect a person. If you took liver tissue, knocked out that gene, and tried to infect it with HCV... no infection. Has anyone seen this before? Nothing in the scientific literature, but we found a dusty old patent from a company that had clearly found this connection years earlier, but never published it or followed it up. The company was likely hedging its bets in case it wanted to follow up later. HCV kills tens of thousands of people a year (liver cancer). Just makes me so frustrated.
Most people are already familiar with negative market externalities like pollution or overfishing. Science and software both exemplify positive externalities, which are just as problematic in free market capitalism. If only there were a clear way to internalize externalities!
Oh wow, who woulda thunk that computer science is science?
Code?
Nothinig, computer are not science
Programs are written in formal languages. Every program is an explicit mathemathical expression.
Hide the code, and your are invalidating the expression, just like showing a math result without stating the formula. That's not science, at all.
What's in a sig?
This article should be covered up, cast out, smothered, and wiped off the face of the intartubes. Somewhere in the Book of Gates, it must surely be written that Open Source is the work of the Devils, and that all who are contaminated by such heresy shall surely be cast into the lake of BSOD.
All who seek after the ways of wisdom are surely aware that both science and programming are best accomplished in secretive enclaves, and pursued by the holy clerics of Corporate America. Surely, in the pursuit of wisdom, lesser beings shall be confused and damned by their communistic, socialistic methods. No good can come of the curiosity of the little man.
Keep the science in the cathedrals, and keep the coding in Microsoft labs.
Thus speaketh the Gates.
All hail the great EULA!!
"Windows is like the faint smell of piss in a subway: it's there, and there's nothing you can do about it." - Charlie Br
Why not free up the educational system and put K1 right up through the Ph.D. level of education as free, open source, tools available in towns as well as on the net. Education is simply a form of information. Let's get the for money players out of the loop.
Without discussing the relative merits of Open/Closed source, are we so sure of the importance of computer code? I mean, code is important and all, but undocumented and poorly written code is worth very little. Explaining, documenting, even proving the idea (beyond having it) behind a certain computer program is the real science, while the code is more like the experiment, the "mere" execution of a theory. This doesn't change the fact that writing good and working code is often quite difficult...
My book: Friendly F#, fun with game development and XNA; my game: Galaxy Wars by VSTeam; my gamedev language: Casanova.
"the practitioners of Science publish not only results, but methodology, and techniques."
That's the theory. The fact is that:
- most results are twisted or faked because winner takes all and being faster is better than being rigourous
- methodologies are dubious for the same reasons
- techniques are incompletely described to keep competitors behind
Science done the open-source way would be amazingly productive and creative. But the reality nowadays is very different:
You have to know in science there is only one reward: the publication of an article in prestigious review. And each paper generally requires months of research work. But the process to get an article accepted generally takes even longer.
Moreover, for each article published, there are only 2 winners:
- the first author who did 95% of the work
- the last author who generaly does nothing in the work but is the boss so his name has to be on the paper
The other researchers names can be on the article but it doesn't matter much. When you are looking for a job people just look if you published as first or last author.
So the competition is so intense these days in science and there is so much fear of competitors stealing ideas, that no deep collaboration can happen.
Moreover, the reviewing process is complete black box: papers are generaly reviewed by 2 or 3 anonymous reviewers that can be either
- competitors trying to delay authors to publish before them
- friends pushing other friends even when science is buggy
So you end up in a world where most scientists are living in this schizophrenic state where they have to publish results to get recognition in the scientific community while at the same time keep as much secret as possible for fear someone will steal their knowledge and publish before they do.
Maybe in 20 years when Gen Y takes over there will be some openess and science will benefit amazingly for that. But the rules and the reward of scientific publishing will have to change before anything can happen.
Yet you had your suspicions reinforced, and others learned, how not to manage big DB projects. This, one could argue, is of some value. Science is like this too. There is much (wasted?) duplication of effort when failed* experiments go unpublished, just as there are many results in expensive journals that are hidden from many of us without the extra funds to subscribe; much less perform. Perhaps a good first step would be the open-sourcing of failed-experiment data.
* For various values of "fail."
You plainly don't know science, and apparently missed the educational boat as well, if you are seriously pitching that "ivory tower" metaphor. If I assume your post is serious and not troll bait, you've been paying to much attention to the media declarations about nullities like "scientific consensus." There is not one field of science where there is uniform consensus among practitioners about anything. In very field you will find that the practitioners are divided into cliques, some of whom may have the media's and the politicians' ears and some of whom do not. These divisions would not exist if there was a true acceptance among the practitioners that a given theory was _the_ way things work. Not only that, science is not a democracy and the fact that a view is widely accepted is no assurance that it is true. If this were true, we would still be using Ptolemaic cosmology and fires would burn because of phlogiston.
You will often find many individuals in a field who treat some particular theory as the right one. I'm inclined to think that about plate tectonics for example. That does not mean that the currently preferred explanation is the correct one or the last word. There are dissenters in every field (even some geologists who doubt plate tectonics), even in fields that look pretty well settled, and they are dissenters (justifiably so) because the current and preferred explanation is _not_ quite adequate to all data available in the field. This the case from archaeology to cosmology, and as long as a given theory is not adequate to all phenomena addressed in a study area, that area will continue to be a science.
------ The only greater hazard to your liberty than n politicians is n+1 politicians.
Invoking the philosophy of open source was used to encourage openness in one bit of research. The published research was ambiguous on the mathematics which were used, and making (some) of the analysis code public helped clarify what was done. That blog has all too many examples of research which is being kept secret, however.
Open Source may be taking hold in IT, but in science, openness is gradually being lost. As pointed out, the critical factor is disclosing methodology as well as results. The Methods section of scientific papers used to be a major component, prominently placed before the Results, and often of comparable size. But if you look at the newer elite journals like Cell (and more and more of the Cell-wannabe journals), Methods are a brief afterthought at the end, often less than a single page. The old rule of "sufficient information to reproduce the experiments" has gone out the window completely, replaced by "just the bare minimum needed to understand in principle what we did". The clear implication of this is that the current leaders in any field no longer have to tell you how they did what they did, or at least, you have to contact them and ask them nicely. Certainly, it helps the elite maintain their advantage and raises the barrier for new entrants. Add to that the fact that so many of the materials are now proprietary commercially produced kits, often with undisclosed solutions or other components, and it is becoming harder and harder to test the reproducibility of any results. As recently as 20 years ago, if you withheld the identity of even a single compound that you used, you wouldn't be allowed to publish in any respectable journal.
The theory of Intelligent Design has a lot of holes.
So does proprietary software.
http://www.cybersource.com.au/users/conz/shoulders.html
Abstract: This paper posits that the open source/free software development and distribution paradigm will eventually become dominant. It aims to show that this will occur as an inevitable process, slowly at first, then with almost critical-mass motion. A number of analogies to other areas of human endeavour, such as Science will be used to underline the power of the concepts behind open source freeware. Also, that the open source movement shouldn't be viewed as an attack against any single closed source vendor, but against the inadequacies of the closed source process. And finally, the hope is for this message to achieve some sense of resonance with enough readers, to add just a little more momentum to the accelerating adoption of the open source paradigm.
What the news snippet actually claims is that a certain branch of science operates according to the same mechanisms as science at large does (that is, if computer programming may be considered a branch of science). This is basically what R M Stallman has been saying for at least twenty odd years. Namely, that computer programming is about sharing, about getting to see the source code, about modifying and improving it. Not only are there philosophical analogies between science and programming, there are also very important sociological and social ones, especially wrt how the open research model benefits the research community. I could go on about the Cathedral and the Bazaar models here (proprietary/patented research versus open/cooperative science), but such analogies are certainly quite obvious to the savvy crowd of Slashdot readers.
Intellectual Property: an immaterial non-entity, most fiercely contended by those with no proper intellect to speak of.
I think it's important to note that open source is important for scientific ethics. As scientists, we are required to disclose everything about what we do - the methods, the errors, the methodologies - just as the article points out. If we don't, we are breaking a code of behavior that exists for the integrity and reliability of the field. How is it justifiable to go and hide that behind the shield of proprietary software in the DAQ or analysis just because somebody like SPSS wants to shut out competition? The answer is simple: it's not. Closed-source is actually scientifically unethical.
I like your analogy. Forcing people with no wheels to invent a car. Clearly, concisely, and accurately depicts the nature of the problem-- Thank You!
Censorship is obscene. Patriotism is bigotry. Faith is a vice. Slashdot 2.0 sucks.
The Open Source Model has been intentionally designed after the merits of the scientific method, I would say. Many people agree that one of the cental intentions of copyleft is to be able to review, improve and build upon the previous work of others. Since copyleft is probably the most prominent codification of the open source model, I think it could be seen as a central post the model relies on, thus describuing its design (and design intentions) adequately.