Slashdot Mirror
Long Live Closed-Source Software?
EvilRyry writes "In an article for Discover Magazine, Jaron Lanier writes about his belief that open source produces nothing interesting because of a hide-bound mentality. 'Open wisdom-of-crowds software movements have become influential, but they haven't promoted the kind of radical creativity I love most in computer science. If anything, they've been hindrances. Some of the youngest, brightest minds have been trapped in a 1970s intellectual framework because they are hypnotized into accepting old software designs as if they were facts of nature. Linux is a superbly polished copy of an antique, shinier than the original, perhaps, but still defined by it.'"
I'd like to say that the author of this article is completely clueless. Perhaps he should define his position more, and say something like "Open Source interfaces aren't creative" or "Gnome isn't creative," rather than paint a vast category of software, including quite a bit of highly creative non-Linux software, with a single brush.
Just look at Java opensource software. Eclipse, Spring and Hibernate are some of the most innovative opensource projects, massively used by the biggest corporate giants to boot.
Yes, I'm keenly aware I'm preaching to the choir. This article is the most flame-baiting piece I've seen on the front page in a long, long time. I have to admit, it'll be good for driving traffic, and unfortunately the author is probably going to make a bunch of money on it. He won't get my clicks, though... I flatly refuse to read TFA.
512 MB RAM, 20 GB disk, 200 GB transfer, five datacenters. $19.95/month.
Why did the adored iPhone come out of what many regard as the most closed, tyrannically managed software-development shop on Earth?
What, the same closed, tyrannically managed software-development shop that built a complete, adored operating system around BSD?
Tubal-Cain smokes the white owl.
Such remarks basically insult practitioners for a lack of imagination without giving any substantiation. "Who know how much better it could be" is an impotent whine [whinge]. The commentator reveals themselves.
Attaching open source to these statements clouds the issue. Serious innovation isn't being supported anywhere, except perhaps in Universities. Even there it's hard because the interesting stuff is at the fringes. Businesses aren't interested in it because that won't make them money any time soon.
OS creation isn't that interesting to most people, because once you know enough about it, you realize that while the Unix paradigm may not be perfect, getting to a current Unix's level of capability and stability would take decades.
All the technology in the world won't hide your lack of vision, talent, or understanding.
If you assume that Linux is the only open source stuff being written.
There is some very innovative open source stuff out there that has nothing to do with Linux. Including a few next-gen operating systems.
In fact, I think that the fact that open source programmers have gotten so much out of Linux that a 70s platform is *still viable and thriving* in 2007 says quite a bit about them - and the opposite of what the article was saying.
There are some legitimate criticisms of open source - this isn't one of them.
For linux tips: http://www.linuxtipsblog.com
Why I should pay any attention to Jaron Lanier.
./ or somewhere else, because somehow he got certified as a smart guy (TM), but for the life of me I can't think of anything interesting that he's done or contributed that would deserve that appelation. All I've ever seen of him is a bunch of tech punditry that's either obvious or empty speculation (which is supposed to be significant because he's a smart guy (TM)).
His name pops up every six months on Edge or
Anyone who loves or hates any language, platform, or manufacturer, doesn't know what they're talking about.
Better word
UNIX.
The original versions shipped with source code. It was only when AT&T tried to make money on it that the source code closed down, and then guess what happened? dozens of incompatible versions became the norm.
i thought once I was found, but it was only a dream.
Every piece of significant Internet technology designed, developed and deployed over the past 25-30 years has been open-source. Offhand, I could list everything related to Usenet and NNTP, Apache, perl, gopher, python, PGP, BIND, Firefox, archie, AFS, NFS, X, LDAP, MIME, majordomo and mailman, ruby, RCS, CVS, subversion, BSD Unix, Linux, sendmail, postfix, courier, exim, P2P and associated tools, IRC, a bunch of ASF projects, etc., etc., etc. These are the building blocks of what most people perceive as the contemporary Internet -- and I'd say that creating that has involved some serious innovation.
The biggest obstacle to innovation isn't open-source: it's software patents and the associated legal thicket that's being constructed to strangle innovation and thereby preserve the profits of the incumbents. I note with interest the the overwhelming majority of those engaging in this anti-innovation practice are vendors of closed-source software -- who are thereby admitting that they can't compete on merit, and so have to resort to unethical legal maneuvers to quash their competition. Oh, and the occasional open-source-is-bad propaganda piece.
And Gnome. And the media players on X. They're either superb copies of old tech, or they're just running behind whoever-sets-the-trend. It's also very untrue with regards to apache, perl, python, webbrowsers (who's running after whom in this game ?). But operating systems need an overhaul, that's for sure. Not that old micro/monolithic debate (that I couldn't care less about), but currently a whole lot of tech is ending up in userland where it doesn't belong: virtualization, network-distributed/scaled filesystems, network-distributed/scaled services. And APIs. I mean, by now, transactions on a filesystem should be part of your standard C-API; read, write, oh sorry, I didn't mean that: rollback. Why isn't it ? Standardized APIs with regards to shared memory, synchronization devices, events; the UNIX crowd seems to find it very acceptable to rely on backward compatibility here. Why ?
Religion is what happens when nature strikes and groupthink goes wrong.
But "UNIX" doesn't rebut TFA, it reinforces it! The article's whole point is that OSS has done little besides copy the work of closed-source innovators, with GNU/Linux copying Unix being the chief example!
It's because trying to lead open-source developers is like herding cats. Unless you're holding their can of food, they won't go where you want. And if you can't make all of them focus on the single project you want accomplished, you don't get anything done without a huge mass of so many people that everyone can do what they please and you'll still have enough people going your way. But the only way to get that size a mass of volunteers is to work on a "sure thing" project with an established design that moves towards a goal everyone can already see -- to copy an established product.
For example, wasn't the OpenMoko team supposed to have released a user-ready package of hardware and software by now?
Best word
wheel
There are certain ideas that are hard to improve upon beyond minor cosmetic and detail changes. There are a lot of things one can do to improve wheels -- materials, suspension, etc -- but changing the fundamental shape isn't one of them.
(And yes, one can invent radically new concepts for transportation -- e.g. wings -- but they don't solve the fundamental problems that wheels solve.)
Unix/linux, word processors, spreadsheets, etc solve certain fundamental problems. You want radically different software, look in radically different problem areas (as some other posters have noted).
There are certain shapes of non-round rollers that work fine, and even lumpy wheels work, but after continued use they'll both wear themselves to a circular wheel shape. Twenty years ago Henry Spencer's sig said "Those who do not understand Unix are condemned to reinvent it, poorly", and Microsoft (among others) has been proving him right.
-- Alastair
I would actually hand the prize for OSS development to Ubuntu Linux made by Canonical. They got around the "good, cheap, fast: choose two" dichotomy by using philanthropic funds, and the result is a system that manages to almost not betray its decades-old foundations. DISCLAIMER: I am an OS X user, though I can fully understand how Apple obviously takes the path of "good and fast" by throwing "cheap" out the window. Open source software greatly lowers the barrier to entry into the market which, imho, increases innovation in how business is done. Very probably, in fact, almost definitely. But TFA spoke of innovation in software and in computer science, and all the copying (and then the subsequent touting of a copy of a 30-year-old system as an "alternative" OS) hinders innovation in software engineering and computer science. Whether the trade-off is fair, I leave up to you. I think we should at least be fostering real innovation in academic CS and hobby programming, even if the market won't support it in business. I mean seriuosly, I don't see closed source software making amazing technical breakthroughs either. Plan 9 from Bell Labs. The iPhone's multi-touch interface. Bluetooth. The Nintendo and Sega games that actually make good use of the Wiimote. VMWare Fusion and Parallels. Portal. I believe what the original author is seeing is the waning of the industrial revolution in which we're seeing an extensive slow down in technological wonders. Sure computers are getting smaller and with slightly different/better interfaces, but going from steam engines to cell phones was a VERY rapid ramp up... we just don't see that kind of thing anymore. Since I have a huge hulking Dell desktop sitting a meter to my left that was, in 2001, only two steps down from the bleeding edge, and since I'm typing this post on a Macbook Pro (made in 2007) that outstrips that Dell in every possible way due to Moore's Law and the increasing capacity of hard drives while providing several capabilities, like the iSight camera and motion sensors, that weren't even available (at least not cheaply) in 2001, I'd have to say technological development is not slowing down. Maybe it has stopped accelerating, but everyone knows that exponential growth can't continue forever.
I'd say we've reached the point where people problems hold as back more than actual technological problems. If OpenMoko got their shit together (I really wanted one, so now I love to use them as an example of a failed project.), we could all be running mobile phones with multi-touch interfaces, cheap service plans, WiFi internet access, and best of all, software at least open enough to let us program the device. Such a device would, if sold cheaply enough, put established mobile phone and the less savvy mobile video gaming companies out of business, and we have the technology to produce it. It's just the people causing problems.
Closed source software is very important to how people use computers, even if they tend to use OSS. For example, if, say, Windows XP or Mac OS X were fully open source, would you really choose Linux over them?
In a nutshell, the point I'm trying to make is that closed source software can be very good. True, that can't be said of certain products, but Windows XP wasn't all that bad, Office 2007 (ignoring OOXML) is excellent, and since Mac OS X was introduced, Apple have always made a brilliant example of how to create good software; I'm typing this on Mac OS X Tiger now and it's excellent. True, its kernel is open-source, as are the GNU tools, and several of the APIs, but the rest of it is closed, and I truly don't mind using it.
While it's good to have something for free, it will take something enormous to get open-source on almost every machine in the way, say, Windows is. For example, a real innovation that makes open-source software dead simple to set up, and different to anything before it. Because - let's face it - Linux is a jargon minefield for the inexperienced user, and while Vista is no better, XP and Mac OS X are dead simple - two editions, that's it.
That said, I do have a problem with fierce monopolisation of software using closed-source, which makes Vista my case in point. So my case briefly is that I don't mind using closed-source software if it's good enough and reasonably priced. If it's open-source, that's the icing on the cake.
Those using pirated Tinysoft signatures(TM) are a real threat to society and should all be thrown in jail.
Lanier invented gloves-and-goggles virtual reality. I tried his original VR system back in the 1980s (novel concept, terrible lag), and met him back then. Lanier tries too hard to be cool, but he has done real work.
He does have a point about the Unix/Linux/open source ecosystem. Face it, Linux is pretty much like Unix, which dates from the 1970s. The Berkeley stuff from the 1980s (notably BIND and Sendmail) is still in use, buried under layers of cruft and still breaking. C programs are still crashing all the time. C++ didn't help much. X-Windows, which was never very good, has survived all its successors.
I never dreamed when I started using UNIX in 1978 that thirty years later it would still be a major system. I thought the future of operating systems would be more like Multics, with rings of protection, on cheaper hardware. Or like Tandem, a transaction processing system where the mean time between system failures was measured in decades. Or like UCLA Locus, where distributed processing really worked. But no. It's just minor variations on UNIX, forever.
That's what Lanier is pointing out. We have roughly the same problems at the bottom we had thirty years ago.
You just made the authors point for him, bravo.
To say that UNIX is a "wheel" is garbage. UNIX (and Windows, which is based on similar concepts) is a moth eaten dirty piece of cloth. It's got giant problems. Look at malware for one. Look at how many jokes revolve around software crashes of some sort, for another.
Before claiming that UNIX is like a wheel, go read up on modern operating system research. Seeing as you have a low opinion of Microsoft, might as well start there - try reading Singularity: Rethinking the software stack from Microsoft Research. They describe an operating system that, amongst other things, operates in a single address space without using hardware memory protection. There are no traditional processes, or syscalls. Instead the basic unit of software is a "Software Isolated Process" or SIP that is statically verified and compiled to machine code at install time. SIPs cannot be arbitrarily modified after installation. The whole thing is a single address space microkernel, except without the performance problems that scuppered previous microkernel attempts (because there are no context switches). A new security model based on verifiable type systems, state-machine based messaging and pre-declared intents allow for the construction of systems that are far more resistant to malware and unstable 3rd party extensions than today.
And they only just got started.
Static linking isn't always bad. It's a balance. It would be an easier choice if a large percentage of library authors would learn how to version and manage a stable ABI. See, that's where the "Oh but you only need to replace one library if there's a security hole" argument starts to fall apart: most library authors couldn't keep the ABI stable if you held a gun to their head, so what actually happens is that existing software doesn't use the new library because the version has changed. You may as well be static linking in these cases.
It would also be easier if software authors got a grip and stopped creating huge and unwieldy dependency trees for every little feature they add. Do we really need to ship a shared library for libquxblortsnort when it's all of 10k and used by exactly three projects in the known universe? Perhaps the author should just include the library with their source and link it statically.
DISCLAIMER: I am an OS X user, though I can fully understand how Apple obviously takes the path of "good and fast" by throwing "cheap" out the window.
... wait for it ... they're too innovative and different. Look at SymphonyOS. Heck, look at bittorrent. You think that didn't innovate?
I understand now. Throughout this discussion, I just thought you were an idiot. Now I see that you're a fanboy.
There are so many truly innovative open-source projects I couldn't name them all, but most just don't get much support because
It's just a lot easier to build a new OS that can leverage 30 years worth of programs, so people try to make sure the GNU set will run on them. Even Apple took a bunch of open source tools as the base of their operating system.
By the way, Apple didn't innovate with the iPhone. That kind of shit has been in Japan and Taiwan for a couple of years now. Just because you haven't seen it, doesn't make it new.
Put identity in the browser.
So basically, you're saying that the community doesn't support truly innovative open-source projects, causing them to die for lack of volunteers? That's pretty much my point right there.
... shock ... it's even true for OSS. Welcome to society.
No, I'm saying that radical ideas almost always fail in the marketplace. There's a business rule which says something like "Never be the first to market with a new concept. Let that company fail and use what you learn from them to succeed in the market." The market generally accepts what is tried and true. That's true for software. It's true for movies. It's true for music. And
Put identity in the browser.