Slashdot Mirror
Bob Metcalfe On NPR
linux slacker writes "Bob Metcalfe, Inventor of ethernet, writer of a
weekly columnn on Infoworld, was on Boston U's NPR show
The Connection this morning. Among other topics, he predicted the eventual downfall Open Source and Linux, citing as one reason the old argument that "Open Source leads to security breaches". Now, this guy did support the Microsoft anti-trust case and opposes monopolies, but he just doesn't like the Open Source idea. " To be fair, he did talk about a lot of other stuff - it was a pretty good show.
The dudes been senile for at least 15 years. It
really is sad seeing the media displaying has-beens,
like geeks (original sence) in a side show. Though
to be fair, Bob is not so much as a has-been as a never-
was. His one claim to fame is not so much as inovative,
as appearing at the right time. Bob Metcalfe sorta
reminds me of Al Bundy, continualy reliving the
gloriouse day of scoring four touchdowns.
Over the years, the man has said so many assinine
things, that even the ignorant news media (including
oh hallowed NPR) probably realises he's an idiot.
Maybe they are just parading him about so we can all make fun of him.
If you're in Boston and want to hear the replay... 90.9 FM, WBUR
---------- Financial Crypto is the Only Crypto That Matters
The beauty of open source software is that it cannot die. While the current momentum of the ideological craze and the love affair with popular computing may taper off, the software will never die. Grow old, atrophy, stagnate, whatever--it won't die. It'll be there for me to pick up in 10, 20, 30 years and start working again, if for nothing other than the joy of it.
Free Software and Open Source Software has been around since before I was born. In fact, the idea that you DON'T get the source code is newer than completely closed source projects. Sure there were some NDAs and whatnot before (which I agree hamper progress) but the source was still there.
This kind of prediction is frivolous and basically a higher form of trolling--and the poster and Slashdot bit. We've heard about the coming death of USENET, IRC, UNIX(tm) and even the Internet--none of which have died. In fact, most have seen a massive resurgence in recent years.
Open Source will die when all of the hardware that runs Open Source software is gone. I'm not forseeing that soon.
I think, Mr. Metcalfe, you'd find your time better spent solving technical problems. Leave the punditry to persons more deserving of the ridicule that goes with that job.
-fp
Right......
Why hasn't there been any class-action suits against Microsoft right now?
I do agree though, that *right now* the open source 'movement' hasn't gained enough momentum to really compete directly with a proprietary-based corporations. Granted, there are exceptions like Apache for web-serving, but most all software (not most popular - most software in general) that is opensourced is very lousy. Things are improving though.
Pure hype must be running my DNS machines right now. And my Oracle Development environment.
Gee an I thought I actually had something. I didn't know I could get hype to function as if it was software.
A Dick and a Bush .. You know somebody's gonna get screwed.
War is necrophilia.
Do you think [E]thernet would have evolved to 1Gbps if that had been the case?
Hmm. Hardware engineering. Tell you what, let's repackage this in terms of more conventional engineering--civil engineering, to be exact:
"Now imagine a world in which 3com published the schematics for their [bridges] with instructions on how to build your own [bridge]..."
Okay, we already have this world. Whenever an engineer drafts a bridge, that blueprint is put on file somewhere and any engineer who wants to look at how it's made is free to do so. This engineer can then use anything they've learned from studying the design, provided it hasn't been patented, in their own designs.
How many times do bridges collapse? With the occasional glaring exception (Tacoma Narrows), not often at all. This model works extremely well for civil engineers.
"Now imagine the 3Com Public License which states that not only are they giving away the schematic [to their bridges], but you don't owe them anything... Make as many of these [bridges] as you want, give them to your friends and neighbors. No licensing fees, no royalties, etc."
Well, here's the real question: why would I want to? If I can study their bridge as much as I want, both by studying schematics and walking across it a few times, then I'll probably find things I'd do differently. I might not like the curve of this arch, or I might want to use a higher grade of steel, or any of a thousand other small changes. I wouldn't make a slavish copy, but an incremental improvement.
Again, this world is not hard to imagine. It's the world we live in today.
"Do you think [bridges] would have evolved to [the Golden Gate Bridge or other masterpieces of engineering] if that had been the case?"
Well, judging from the fact that we have a Golden Gate Bridge exactly because of this being the case? Yes, I do think the state of the art would have evolved to that point!
"Do you think 3Com would have survived 5 years in the [bridge-building] business?"
Sure. In fact, I think they'd have survived 17 years. They could patent the nonobvious and useful portions of their bridge design and deny other people the ability to duplicate it without permission. That gives them 17 years of profitability, provided their patent is a good and sound one. Hopefully, by the time 17 years are up, 3Com would have found a new and even better way to build bridges--and would have received another patent, and have another 17 years, etc..
Moral of the story: We already have all the intellectual-property protections we need, without needing to make things proprietary.
(Before the anti-IP crowd flames me to death here, let me say that patents need serious revision. 17 years for a nonobvious, useful invention that makes for safer bridges is one thing--17 years on XOR encryption is another!)
A Dick and a Bush .. You know somebody's gonna get screwed.
War is necrophilia.
I hope you know you have to be a big customer to get any clout. Otherwise, your complaints don't mean squat.
And it is getting to the point that some Microsoft licensees are finding out that they have to screw with the source themselves, Microsoft doesn't care about anything that doesn't affect at least 1 million people anymore -- seriously! And getting access to the source code of Windows is a tricky game. It's costly, and Microsoft could still withhold it from you at any time (e.g., Bristol Technology).
So, unless you are a Fortune 500 company, I really don't see the "somebody to sue" factor even being remotely applicable. The result is what we see in Internet servers ... big business chooses Microsoft, smaller businesses choose Open Source.
And it even goes deeper than that! 90% of those big businesses don't run their business on the Internet (and the ones that do choose UNIX/OSS), nor do they entrust Microsoft for most of their back-ends. It's still tried and true big iron and/or UNIX -- which is more of a testament to UNIX, and not so much Open Source (but wasn't UNIX developed akin?).
-- Bryan "TheBS" Smith
-- Bryan "TheBS" Smith
Independent Author, Consultant and Trainer
Yeah, Bob is kind of a genius flake. Unfortunately he gets paid to wander around and offer up his opinion which has been illuminated very little since CSMA/CD. At this point I think he is limited to being provocative (upsetting people). His value as a commmentator is based on how much call volume he generates. Riling people up rings the phones and keeps his value as a speaker up. I find it sad.
It's possible that when you were in diapers, he was tweaked out on drugs. Ethernet was a (and is a) revolutionary concept. How much of computing today relies on ethernet. Lets see, I use it daily, so do millions of other people. It is possible, like many ex-hippies I know and other older folks that had, substance altered mind frame experiences have messed up their lives.
I take issue with Metcalfe's assertion. I don't know about switches and such, although as far as I know there isn't much innovation going on in that area, but neither Intel nor Microsoft is leading the technology in its field. In fact, neither chips nor office suites nor any of the other areas those two companies are involved in are really doing anything exciting these days. Intel is consistently being bested by AMD, but continues to feel the benefits of its monopoly position.
One of the only areas of the hardware field that is experiencing qualitative, rather than quantitative, improvement, is that of 3D acceleration. By that distinction, I mean that CPUs are merely getting faster, not better in any radical way. However, video cards, products of a field with several major players, namely nVidia, 3dfx, Matrox and ATI, are getting better, with additions of things like better texturemapping, lighting, motion blurs, lens flares, and other improvements. I think that this competition is a result of the fact that the field enjoys fierce competition.
Of course, it's a little unfair to compare Intel to 3d card makers, because there isn't much you can really do to chips that wouldn't completely break backwards compatibility. That is, Intel couldn't start manufacturing DNA processors tomorrow, because besides the fact that they're not really extant, they would need massive engineering in order to be useful. The same goes for hard drives, which are getting bigger, faster and cheaper, but not more interesting, even though they are in a competetive field, because there, again, isn't really much you can do with a hard drive except those things. Microsoft, however, hasn't had an original thought in its life, except maybe in their mouse-making division, and that's an area where (ahem) they don't have a monopoly.
Switch the . and the @ to email me.
From the book "Switched, Fast, and Gigabit Ethernet":
"What made ethernet the ultimate victor was not technical superiority or speed but Metcalfe's vision to turn Ethernet into an industry standard and not keep it a vendor-specific proprietary technology."
That's why I always wonder what's up with Bob. From what I read, Bob got Xerox (Where he worked when it was invented.) to turn over the patents to the IEEE to make sure ethernet would be open. So apparently they gave up the IP for free. It's just seems like he would have more of an open mind...
Um, having the source known makes open source software open to security breeches? Kinda like how because I have the source code for TwoFish sitting here, I can break it with ease?
Certainly much older than the hat he promised to eat if the Internet didn't collapse (or was it some other non-comestible?) He "invented" Ethernet while at Xerox, though it's being debated how much "inventing" it really took, since--as has been stated elsewhere--much of the technology was based on established principles. Since then he seems to pretty much have rested on his laurels, as befits one after so much hard work. I'm not aware of much further ground-breaking work he did after that.
Anyway, he wields his thusly-won fame to create the impression of much wider-reaching expertise and authority. He loves sitting on all sorts of panels and boards (watch those splinters now!) and taking his mouth for a spin. I've read his column at InfoWorld for years, and on average he has less to say than most of our development team members on a good day. He certainly is very fond of name dropping and all sorts of allusions to the lofty company he circulates in.
My favorite was his vitriolic article a couple of years ago or so to "impeach Clinton." He pretty much exposed himself as a troll and a person of little emotional consequence. He is in bed with Big Business, loves everything Big Business--as any good Republican should--and from that perspective he certainly has no love lost on Open Source. As far as the Internet is concerned, he'd probably much rather have it converted to one LARGE Ethernet network and do away with this pesky network-agnostic protocol called TCP/IP.
Uwe Wolfgang Radu
The argument to stick with commercial software because of the culpability is the biggest pile of horse shit ever served up. EVERY shrink-wrapped piece of software denies any responsibility for failure whatsoever, and that also applies to OEM or site licensed software. I would like to hear of a SINGLE case where a commercial software vendor was successfully sued.
The only exception is custom written software. If you contract a software project out, the boundaries for success and failure are usually very clearly defined and can lead to successful legal recourse. But that's not what open source detractors talk about. They're talking about shrink-wrapped software, or its functional equivalents.
Uwe Wolfgang Radu
Strangely, the rest of the marketplace hasn't seen things quite the same way.
I think the lesson here is similar: no amount of controversial punditry is going to change the fundamental truth that people are going to make decisions about their operating environment based on suitability to purpose--zealotry (whether open- or closed-source) isn't going to be the force that shapes the market.
I had meant to post this when the last article on Metcalfe was on /. but couldn't find it.
From the September 16, 1996 issue of Newsweek, an article by Steven Levy on the topic of
Metcalfe's prediction that the Internet was soon to "melt down".
[excerpts]
(After talking about the slowness of the Internet) "Such annoyances are commonplace on the Net these days, but to Bob Metcalfe, networking pioneer turned cyberalarmist, they are evidence of impending disaster. In his thinking, the Net is a fish already hooked; those routine brownouts are but the first few twinges at its mouth. Soon the fish will find itself reeled in, and we will witness the pathetic spectacle of the once mighty Internet, the darling of our economy and the object of our millennial dreams, flopping aimlessly like snagged red snapper on a boat deck. Web sites will become cobweb sittes. E-mail will be dead-lettered. Stock prices will fall to Earth."
"'Maybe the Internet has already collapsed,' Metcalfe says to me later that day, in a lecture that lasts the dinner hour. 'Everyone complains about brownouts every day. But it's going to get worse. Worse! Worse! Worse!' The inventer of the Ethernet networking system and founder of the 3Com corporation has risked his considerable reputation by publicly predicting a 'gigalapse' - in which a billion hours of access time are lost - by the end of the year."
The rest of the article is just as amusing.
So what's the application that will lend credible validation to Open Source?
Sendmail and Bind have been Open Source and running the internet for about 20 years...
Where have you been?
"Free your mind and your ass will follow"
> ...'good' closed software (ie on the space shuttle)
;)
...just kidding ;)
Hmm. I believe one of the latest missions admitted to running NT on a bunch of networked laptops for some sort of experiment... 802.11 of course
But I think you're talking about the actual operating system software - life support, propulsion, etc. etc. - in which case, how do you know it's 'good'? A moot point in your comment I know, but I'm curious... seems to me that the most obvious thing NASA could have done was whip up something along the lines of NASA-DOS that was programmed in a language even the Obfuscated C Contest winners couldn't decipher to make sure that all the 'talent' was kept in-house. Certainly, by OSS standards, this system would, well, suck. But it works, right?
Conversely, what if NASA used OSS systems? Wouldn't that be an interesting scenario...
Loading linux-2.2.17
updating gyroscope orientation...done
verifying CO2 filters...done
testing O2-N mixtures...OH MY GOD THE RING HAS FAILED WE'RE ALL GOING TO EXPLODE
challenger login:
Cheers,
levine
Hey Bob, had any "black-outs" lately?
Many companies are beginning to sell Linux support. With the exception of Red Hat, who has been around for a substantial amount of time, I don't expect any of them to survive the next three years. It's like the dotcoms - all built on hype to inflate a stock price. Everyone gets rich, then everyone gets out. Everyone who's left gets screwed.
I know Microsoft will survive the next three years, however, even if they get broken up (a stupid idea, if you ask me, but let the government have their fun). I can go to them to get support. I can find an MCSE (not necessarily a talented one, mind you, but I can find one). I can find training materials, software updates, tools and extensions, order replacement media, and do all the other things that keep my machines running.
Yes, Fuzzy Logic. I derived most of the theroms
and was playing at applications for it in 1977 and
1978, when I was a teenager. So what? Its an
obviouse idea like Ethernet, Metcalfe just published
befor someone, and someone published fuzzy theory
befor me ( actualy I did'nt think it was that big
a deal nor did the good folks who got the credit).
Isn't openness one reason ethernet is so successful?
Hmm. Maybe, but I'd have to say the main reason Ethernet was successful was because it was simple to implement, and worked. There were (and are) other LAN protocols standardized at roughly the same time (other IEEE 802's include Token Ring and Token Bus, among others) that haven't been nearly as popular, not because they have significant failings, but because they're more expensive to implement. The IEEE specifications for things like Token Ring were just as "open" as Ethernet.
You can see it for yourself in the relevant patent here. What is it that this guys keeps on telling HE invented it, without ever mentioning the (at least) 3 other inventors name ?
Now, as he always comes with the same provocative and dumb ideas (remember the "Linux sucks" article ?), one often speaks about him, which only encourages him to keep on writing the same kind of articles.
I don't read his column in Infoworld anymore for the same reason, and don't find I miss anything by doing so.
Did anyone say professional troll ?
"Naughty, naughty, naughty, you filthy old soomka !"
What is it with this guy? He's a good inventor; ethernet is great! Invent more stuff! Stop being the party pooper!
The trouble with this argument is that it hasn't shown itself to be true over the last 20 years of computing and it is in fact making great strides in the OPPOSITE direction.
I'm beginning to empathize with RMS (that scares me...really). I understand his frustration. People just don't pay attention. HP is contributing significant resources to the SAMBA project which is, in my opinion, one of the most significant pieces of open source software available. IBM has devoted manpower and hardware resources to various linux porting efforts (specifically RS/6000). Apache, though not predominantly on Linux, still runs most of the web. Sendmail routes enomous amounts of mail. BIND works...what else can I say? XFree86 is in heavy use among commercial entities. SGI, SUN, IBM, HP are all involved in OSS. Are there bigger names in computing? Anyone?
Open Source basically isn't catching on in the Wintel world. I'm sorry, I missed the part where Microsoft won. Last time I checked, the battle was raging and OSS is still on the assault.
No, open source software is in use every day by companies the world over. They may not know it, but they use it, they fund it, and they LIKE it.
-fp
Open Standard is not the same as Open Source.
Take your analogy... Now imagine a world in which 3com published the schematics for their devices with instructions on how to build your own.
Ok, that's not hard to believe. A lot of electrical devices have readily available schematics for use by repair technicians.
Now imagine the 3Com Public License which states that not only are they giving away the schematic, but you don't owe them anything... Make as many of these devices as you want, give them to friends and neighbors. No licensing fees, no royalties, etc.
Do you think ethernet would have evolved to 1Gbps if that had been the case?
Do you think 3Com would have survived 5 years in business?
That's all he's saying. Without income there is no incentive to improve the product.
He probably thinks blacks are still 2/3rd of a human being and women shouldn't be able to vote.
We should just ignore these relics and get on with it.
A Dick and a Bush .. You know somebody's gonna get screwed.
War is necrophilia.
Furthermore, Microsoft doesn't explicitly sell support, they sell a product. That gives them the big incentive to design the product in such a way to lower support costs.
Linux companies have the inverse model -- a free product and a per incident support model leads to products which (A) Are much more complex to support, often by being highly configurable and tunable, but (B) are also easier to support for the knowledgeable person.
The real problem with the OSS model is that companies will usually happily pay for a product, but hate to pay recurring charges for support, and only do it if they really have to. Not to mention the fact that many of them don't realize that you often need to spend money to save money in long run support costs.
When I hear the word 'innovation', I reach for my pistol.
MSCEs must be listening and they don't see much that they like so... the one they agree with, regardless of how lame, is moded up. Or, it could be the folks in Redmond have a new office in their public relations department. This has got to be cheap advertising next to all that ZDNet pulp. On the other hand..
If you do it with your left hand, it will feel like someone else.
Comments about open source security holes bring to mind the Fuzz tests of 1990 and 1995. (Yes, I know, somewhat dated, hopefully it will be updated). In the 1995 fuzz test, a Linux distribution was tested among other systems. (I don't believe a *BSD distro was tested). Interestingly enough, the Open source stuff tested performed far better against the Fuzz test than closed source did (with NeXT performing the worst by far.) If Open Source encouraged security holes, one would expect that at a minimum, fuzz tests on open source software would produce a comparable failure rate to non Open Source UN*X systems tested (including Solaris and HP-UX among others).
We can talk all we want about logical reasons why open source software does not produce greater security holes, but empirical comparison studies of a stable distribution are in many ways far more convincing.
For more information on the Fuzz test, see
http://www.cs.wisc.edu/~bart/fuzz/fuzz.html
(Note especially the conclusions of the 2000 NT Fuzz test)
"I may disagree with what you say, but I will defend unto the death your right to say it." -- Voltaire
Question: How many businesses have sucessfully sued Microsoft for products that don't deliver?
I don't remember any. But I'm not on Microsoft's side on this. What I'm really saying is that THIS WHOLE ARGUMENT IS A LOAD OF CRAP!!!!
I have yet to work in an organization that gets more than the technet CD's for support. Granted thay are pretty good, they're just one step away from being community support.
The two times I did call Microsoft support was for Visual Basic. I burned one of my free support calls to be told there was a bug in the installation builder. Another free support call was burned on a follow up call to get the update sent to me without charging me for it. Barring the three free support calls that came with VB each of these calls would have cost $100.
Most of the time Microsoft support is really just farmed out to the PC Vendors, and don't get me started on dealing with them.
I have had very good luck dealing with the Linux community to get support. I've had success with getting Voodoo Banshee to work, finding the approprite hardware to buy, troubleshooting netowrking problems, setting up networking configurations that would just be a fantasy on NT.
Basically I'll sum it up again. This support argument for Windows is a load of crap!!
Oh, and to finish up on that blame when things go wrong argument. Well you need to have that if you are using windows. When we split our Mail/Print/File server to be one Mail server and one Print/File server, it was a total nightmare we had to change the configuration of every client PC. You bet I'm pointing my finger at Microsoft on that one.
Old school ethernet, Bob Metcalfe style ethernet is highly insecure. You kids with your 100BaseT switches don't know what it was like back in the day of Big, Thick Coax (tm).
I'm not talking about this pussy 10base2 stuff. I mean the real deal. 10Base5 baby.
It was so secure that all traffic was readable by everyone on the network, including people who had clipped vampire taps into the wire.
You kids don't know how easy you've got it.
Well, other than the fact that he invented ethernet, who is he?
In a column released September 1, Bob Metcalfe announced that he is ending his column "after 51 columns for each of eight years." I'm sure he will find ample pulpits to do his chicken-little act regarding the Internet, but at least he is making room for someone at the rag I used to work at.
If I write closed source code for a company, a couple of people in the company might see it. It may last until the company decides to sell that product.
I'd much rather write open source code because I feel I can really make a difference. And that difference can be felt as long as the code is worthwhile.
Looking for a computer support specialist for your small business? Check out
Yes, he's wrong. he was wrong the last six times, too. So let's actually be polite and do him the favor of ignoring his statements, okay?
Remember, this guy's technical specialty in in networking, yet he predicited in 1995 that the Internet would cease to exist in 1996. Why in hell would anyone expect his predictions in non-networking areas to have any more validity?
Steven E. Ehrbar
Okay, so he figured out how to put the ALOHA radio network protocol on a wire called "Ethernet." He based a company on this and got wealthy doing so. That doesn't make him a visionary any more than ALTAIR BASIC makes Bill Gates a visionary or VI makes Bill Joy a visionary.
Ignore the visionaries. Write code and prove Bob wrong. Again.
Umm...
I agree with what you say, but what you say also applies to Solaris, and many other good proprietary (read: closed-source) Unixes. And Metcalfe was specifically talking about Free software-- and not just Linux.
I happen to disagree with him. Free software has typically had better staying power than closed source alternatives. That's why Sendmail is still around and going strong, as is tcl/Tk, Perl, Bind, Linux, and even Unix in general, while NT has gone through 2 major re-writes since its introduction-- the last amounting to over 75% of the codebase.
In any case, I believe Metcalfe is wrong. Very, very wrong.
Microsoft is to software what Budweiser is to beer.
A small example of it at work...
p e=snippet&id=35
http://www.phpbuilder.com/snippet/detail.php?ty
It all started with this itch...
PHP had no function for directly sending e-mail thru an outside server, instead assuming you had local mail services.
I hunted around for some means to get the prototype out the door, and managed to (quickly) hack together something that more or less did it.
I posted it, sloppy tho it may be, to help whoever else might use it.
And used it is! Somebody else came along and added to it, somebody else made a class out of it, somebody else extended the class, I fixed a few bugs when I needed it for something else, somebody else fixed something I missed...
It's a small example, but it communicates the idea...
This is open-source at work! Everybody contributes a bit, and everybody reaps the rewards!
Would I open-source my work? In many cases, YES! I can now produce better software for my boss because of open-sourcing this particular function!
It's really a no-brainer.
-Ben
I have no problem with your religion until you decide it's reason to deprive others of the truth.
Slashdot discourse is low grade, guys, and only serves to confirm my Linux pessimism. You disagree with me about Linux by saying I'm old, clueless, corrupt, it's been a long time since I invented Ethernet, I didn't really invent Ethernet, Ethernet is no good anyway, 3Com was taken from me because I'm a bad businessman, 3Com is not Cisco, and I've been wrong about everything else (for example, Internet gigalapses in 1996), so I just must be wrong about Linux. RHAT and LNUX were both trading today below 19% of their highs, which has nothing to do with me. They'll lose at least another 50%, I'm sure. FYI: There are THREE business models at issue. There is the old IBM+Microsoft model. There is open source. And there is my favorite, the Ethernet model, or Internet model -- open standards with owned implementations competing fiercely. Now, if you really want to get back at me for thinking this, buy my book and burn it. It's Internet Collapses and Other InfoWorld Punditry, now in its third paperback printing from IDG Books. /Bob Metcalfe
Polaris Venture Partners
I'm not particularly sure that letting your competitors steal truly innovative work simply because the Management is unable to comprehend your inventions counts as "open source"... if that's the case at your current place of employment, I'd suggest finding a company that actually has a clue.
"Great men are not always wise: neither do the aged understand judgement." Job 32:9
Yes, and it has far outstripped that goal. Although the primary protocol used on top of ethernet is tcp these days it doesn't really matter. Let's look at these big fat ISP's who provide broadband internet access through nothing more than switched ethernet every other block.
Goal or no, it's still not secure.
I also think it's true about security too. If there is something open in Unix, it is easy to fix, or get answers on how to fix it. On NT, you either have to work around it or wait until a service pack comes out to fix it.
It's either on the beat or off the beat, it's that easy.
I moderate therefore I rule!
--
I'm glad that there is finally someone out there that agrees with me... Linux (and to some degree open-source) is pure hype right now... I'd like to go forward 3 years and see where everything stands. If you had to ask me, use a vendor that makes the OS and hardware (Sun Solaris, HP HP_UX, etc...)
I think that he is wrong that open source causes security breaches. I actually think that open source is a cause for less security breaches. If you look at the response time between a hole in an open source system getting fixed and a hole in a closed source system getting fixed, you will see that it takes far less time for the open source one to at least have a patch out for it. While the closed source one usually takes weeks or even months to even hear from the publisher that it is even a problem. Don't get me wrong though, open source isn't for everyone, but I do think that open source does seem to handle the problem of bad security faster than other types. Now if they are referring to possible back doors being inserted into open source software, how do you know the closed source software doesn't have back doors. In fact in closed source software, there isn't even a way to check if there is a hole without having to tear apart the product.
I agree with that. RedHat (which I'd like to nickname OldHat) has constant problems with GNOME crashing. If Solaris had so much problems, Sun would be hip-deep in complaints and lawsuits.
"Ancillary does not mean you get to rule the world." --U.S. Circuit Judge Harry Edwards, speaking to the FCC's lawyer
The person that invented technology for Ethernet which in turned lead to Internet would be down on Open Source. Correct me if I'm wrong, but wasn't it the open specification of protocols and critical pieces of software(like sendmail) that paved the way for the technology explosion on the Internet? Or does he believe that all of that was a fluke?
Hemos, are you trolling us? -Dave
We're on the road to Tycho.
Open source has its place. The hobby community will accomplish things that commercial development will never pay for using Open Source.
That said, it's never going to be the predominant model for software development, nor will Open Source programs ever be the prevailing supply of software for businesses. Why? Businesses like dealing with other businesses. When things fuck up (and they do), it's nice to have someone to blame, someone who has a responsibility to back up their product.
No IT staffer will want their server support to come solely from the community. And the Linux IPO's are all hype - they're not a serious alternative.
In short, it's not MS's OS that everyone's buying - it's the support. Granted, Win2000 is an excellent OS, but that's not why it gets bought.
The impression I get from somebody who has to have his 20-year-old accomplishments tacked onto every page he writes is that of somebody who really doesn't have any good ideas left.
I haven't read the whole article yet, but I was thinking anyway...
Since he is the creator of ethernet, then obviously he has been in the industry for awhile (back to the beginning even). And way back when, most developing (like ethernet) was done in a pretty much open-source enviornment.
I was just wondering if his objections to open-source is only in relation to how it is going now (linux users fighting with *BSD users etc.) or if he doubts all of the open-source ideals (including the time when he used to work under them). And if the latter is true, if he thinks this because he has been at 3COM too long...
Devil Ducky
Devil Ducky
MY peers would get out of jury duty.
Igor Stravinsky once said that there are only two types of music - good music and bad music. Perhaps its time that such a philosophy was applied to software. Software that is robustly designed and tested will always be preferable to software that isn't (obviously); but being 'open' or 'closed' doesn't necessarily have anything to do with it. There are examples of 'good' closed software (ie on the space shuttle) and 'bad' closed software (You Know What).
It can be argued that open source increases the chances of a program being robust, but is not a panacea; at the end of the day it boils down to whether correct software engineering and testing principles have been applied. In many ways the open source movement is a backlash against appalling software quality in over-the-counter commercial applications - maybe it mightn't have been so popular if there had been more 'engineering' in software development from the outset?
-- briggers
-- briggers Remove blinkers to email me.
Oh wait, you said "Ethernet..."
Well, he probably invented that, too! ;>
im not really sure how much bob knows about linux. if the stocks go down, then that's that. linux won't need the stock market to survive. it just needs debian (pity i can't use that on my ppc g4).
--- Hey, Jesus is coming! Everyone look busy
Metcalf on Open Source:
"The idea is that the modern, high tech modern capitolist corporation, is an out-moded model for the development of software and that there's this new model, this sort of co-operative comunity and that will work better, and I don't think it will. I think that ummm co-operatiave..ummm.. uh you have a day job and go home to your programing..has it's limits.
uhhh I have yet a third paradym which I call the ethernet pardaym where the standards are open but the implemtations are owned, so that a corporation has property ..and economics can work.
In the internet world, there is an open standard .. like the ethernet and internet world both have open standards but fierce comitition between companies, corporations who own the implementations of the standards and I think that's a superior model to open source movement which suspect will sorta fritter away.
(Interviewer interupts) If a hacker can improve a product and introduce to the rest of the world, umm why shouldn't the world be like that...in other words, why does ford have to make every car the same . . we understand that well, because I, tinkering with my V8 engine can improve yours . .but in software you can't.
Well, you could tinker with my engine too . .there are chips, I mean . people don't do this!, I mean. . btw Ford is not the only car manufactuer there are many others you could choose from. You're allowed to let your neibor tinker with your car, but no one does , or very few people do because it's dangrous.
There are very few people qualified to muck with the inards of an operating system like Linux, so the notion that there will be thousands of them, or hundreds of thousands of them ...
(Interviewer interupts)
It only takes one to improve it.
yea, but it's very hard to tell what's an improvement..the software development process is very very complicated and it's all about the control of complexity and uhh testing and uhh btw, who's to say one of these hackers isn't putting software in the system that say a security breach that they may want to exploit later, I mean there are many issues . umm
(Interviewer interupts)
It leaves a lot of the choice in the world to Bill Gates=, tho, doesn't it ?
Bill Gates isn't the only software developer. There's also such likeable fellows as Larry Ellison, I mean . anyone can develop software, start a company and try to vy for it uhhh look at Linus Torvalds, the champion of all this, a really great guy, but he works at a company called Transmeta. There products are not open source for a very good reason they need to have a product, they need to control it's content, they need to make money by selling it. "
a downfall of Internet, now he moves to Open Source and Linux. Everything will fall -- eventually ...
In 4 billions years our sun will become a supernova.
We use them because even we can evaluate 20 of them without having to get spending authorisation every time.
even if it's small potatoes the bean counters tend to go red if i ask for another cheque every day without deploying anything.
photoshop may or may not be better than the GIMP. but its easier to write plug-ins for the GIMP, its easier to experiment with it before committing to it (because its free) and the GIMP revs more often (its getting better faster).
That and the uncomfortable truth that RMS is a prophet. And like all prophets he says things we don't want to hear, smells like something we don't want to smell, and look slike something we don't want to hear.
And much to our chagrin he's probably right. At least the things he says are not self serving.
So i've got more reason to trust him than the rest of the liars out there.
If you've got a limitless budget and want to be lazy then pay for something off the shelf from a bozo trapping you into an upgrade path.
or you can buy a book, learn something new, and make something better.
'There is a Light that never goes out.'
And to be honest, this comes as quite a surprise to me. Not because of his apparent mental aptitude as we have all encountered, at one time or another, a well respected, intelligent person whom is a non-believer in the "open source" model. Rather, I am perplexed by his position on this subject in light of his past accomplishments with Ethernet.
After all of the adversity and disbelief in his ideas that he faced during his innovative work with Ethernet, one would imagine that he would be very supportive and optimistic with the potential for advancement that the "open source" approach brings to the software world. I, for one, would have thought this man to be a vocal, public proponent of the "sharing" of source code and standards to better a product or technology. Especially when considering that it was this type of openness and willingness to cooperate that not only made the advent of Ethernet possible, but eventually aided in its acceptance as a widely used standard.
All that said, however, I still have a great deal of respect and admiration for this great man and his achievements. As does anyone, he has a right to his own opinion.... even if it is wrong!!! :-)
- J
It's MY-way or the HIGH-way!!! - my father
I'll say it again: Bob Metcalfe is a genius.
And he's a perfect illustration of the fact, abundantly illustrated throughout history by geniuses greater and smaller than Mr. Metcalfe, that being a genius doesn't prevent you from being a crank, from holding firm to ideas that the rest of us consider, well, odd, to put the most charitable face on it.
I respect a lot of what Mr. Metcalfe has to say. And I consider a lot of what he says to be pure bunk. He'd probably think the same of me if he knew me. Such is life.
I count "the most secure" OS not as which one theoretically should be more secure, but which one I am more likely to be rooted using. I LOVE linux and I am not into microsoft, but I think it is MORE than fair to say that in the year 1999, you were far, far more likely to be compromised if you were using a Unix-like OS than using NT. I am not saying you could not secure any OS to be virtually intrusion-proof, but it really does seem now that there are many more exploits being discovered for *nix.
Now, don't get me wrong, I know that Microsoft probably has lots of un-found vulnerabilities, but that is my point: they are un-found, and unexploited. I know what you'll say next: "But Happystink, when they ARE found, they take months to be patched"... Is this really true? Microsoft respond very quickly now, just as fast as Redhat, Mandrake or whoever else.
So enough theory about which source model is more secure in a perfect world, let's re-evaluate and see which model works in the real world before we start posting comments like "I don't know who this Metcalfe guy is but he must be new to computers cause I've been using Linux 4 months and never ain't gotten rooted yet!" ok?
sig:
sig:
See the "..for smart people" banners Wired runs here? Look elsewhere guys.
Umm....if you have a sound card and can play RealAuido .rm files, you can hear it any time you like at: the link from the originally referenced page -- which goes to The Connection's archive of it (at Boston University).
"The girl makes Godot look punctual." -- Buffy
3COM should be where CISCO systems is now...Why aren't they? Bad leadership. (this is paraphrasing an article by Bob Metcalfe, talking about when he was in charge of 3COM).
Sure Metcalfe invented ethernet when he was at PARC and started the spin-off (3com) that was created to commercialize it. But, he also lost control of 3com long ago in a rather ugly battle. Seems to me that he's not the most astute businessman out there.
Meanwhile, he keeps predicting the "megalapse" where the internet will be crushed under the weight of its own traffic. He thinks traffic will grow faster than bandwidth and one day it will all come to a screeching halt, kind of like when TCP's re-xmit algorithm gets out of hand.
The thing is he has been predicting this event every year for something like that last four years. So far he hasn't been anywhere near close to being right, but that hasn't stopped him from blindly continuing with the predictions.
Meanwhile he's using this "megalapse" theory as one reason to argue for metered access. Geezus! Metcalfe may have been at the centre of a fundamental technology in networking, but he's clearly an old-school thinker when it comes to the Net at large.
When information is power, privacy is freedom.
He thinks eventually we will have a world where the The SPECS are open, but the implementations are not.
Open standards, NOT open source. I suspect he's right, but it's really too early to be sure.
tj
I'd rather bind up my defense into one post.
.DWG files.
Yes, big names are playing the Linux game now. Sun HP, and IBM would be among them.
Who really trusts Sun anyway? They're as bad as MS sometimes.
IBM - yeah, they're going Linux now. But they've still got tons of other stuff. It's a new fad to them - can you guarantee they'll support it? What happens when they decide Linux development is no longer financially viable? Then you've got an expenditure in Linux support and hardware that won't run most of IBM's proprietary software.
HP - do they really have any software of their own?
Now, some people have stated that they can buy support for their Linux software. What about when supporting it no longer is a profitable business? Then it's back to the web. And if your web browser doesn't work... now yer screwed.
But what about LUG's? Well, you can't count on your friends to solve your problems. A friend who went MCSE recently found out that those eight hours I spend at my desk, I'm actually doing work for a company, and I can't spend time to fix his problems at work. Unless they're living on unemployment, all the people in your LUG have jobs and have to be at them. And if you're down for the entire day, well, you're going to be explaining yourself to some very irate people the next day.
Now, this is NOT a slam on Linux, or a troll, as some have mislabeled me. Linux is a very good OS, and will do things that Windows will never do. But the fact remains that it's not ready for prime time. You can't depend on the community the way you can depend on a major software vendor.
And who said this is pro-MS? I'm just pointing them out as the oft-maligned example. Since you're all so infinitely familiar with them, they're the best example I have. You'd rather discuss another company, that's fine - we'll talk about Macromedia or Adobe or Autodesk or something.
In fact, let's do that. The firm I work for is a major manufacturer of flow control products. In english, we make pipe. Lots of it. This requires very complex machines, with many moving parts, which need to be fabricated at a moment's notice for installation during the next shift. To do this, we have a drawing database we're developing using a combination of AutoCAD 2000 and some proprietary software written for archiving
Those drawings are our business. If they go down, it's someone's job. Given that, can you trust the parts supply for a mill that can make a thousand feet of tube in one minute to a community project like Linux? One that doesn't have credentials attached to it; professional reputations that live or die on the support of their products? I can't go to my boss and say, "I'd like to use this program I got online. It won't cost us much, but the only support we can get comes from either:
A. Startups cashing in on a fad.
B. People on the Internet."
He'd laugh me out of his office. Now, not to slight the community - it's a great resource to me. But I don't rely on it professionally. I won't stake my reputation on software that isn't sold to me by a reputable business. And I'm relatively liberal. Imagine explaining to your stockholderw why you can't get your product distributed because the system is down and there's no one to call.
When the Metcalf article came on slashdot, I looked up some of his older articles. He actually said that the entire Internet would die. His reason was well, weird.
He had written this during the time of the CDA, when people were really frightened about censorship on the Internet, naturally people were using the internet to discuss this. It was Metcalf's hypothesis that since people were using so much bandwidth discussing pornography, and censorship actual porn would be drowned out. Without porn, he figured, the Internet was pointless.
Yup, if there's thing wrong with the internet today, its not enough porn.
He also predicted (incorrectly, he admitted) that the internet stock bubble would collapse in 1997, and that Yahoo stock would be worthless. While stock went down a bit in 1999/2000, and new IPOs didn't go so crazy, he was way of with that to.
The Internet, pornography, E-commerce, dot-coms, and Linux. Clearly Metcalf knows a loser when he sees one....
ReadThe ReflectionEngine, a cyberpunk style n
The internet will fail in 1997(?) from too much growth.
The stock market will crash last fall (and 11k dow and 3k Nasdaq don't count).
He's not often right. But yeah, hear him as representing OSS backlash. He makes good points. He's just wrong a bunch.
Open Source as "everyone will work for free for the good of society and make money based on charitable feelings of corporations" will fail.
Open Source as "groups of people can do good project work and will do it for personal gain and better products" DOES work (apache comes to mind - people wanted a web server that worked and worked on bits that made their own lives better - personal gain - PLUS they got other people's hacks on top of that.)
Other models about (I want the freaking source for the product I just paid a lot for because I need to support, debug, improve it).
I've been paid by my bosses to do work which has made it back into OSS projects. Not out of charity but out of "I need to fix this; I'll also share those changes" and there's little problem recognizing that.
Instant Karma.
open source has already gained critical mass. this ignoramus is acting as if this is something new. some gnu utilites are over 25 years old! where is this even heard of in the commercial/wintel world??
--
And Justice for None
--
And Justice for None
Bob has a Security Breach in his head, and the alien underground mafia is using it to fill him up with anti-linux propoganda.
why?
because if the source code is available, the alien underground mafia cannot sneak thier backdoors into our software, and that will ruin thier plan to TAKE OVER THE WORLD.
Oh wait, that was pinky and the brain.
s/alien underground mafia/pinky and the brain/g
I think the whole Open Source thing has a few more good decades in it. It's just now gotten back on it's feet. The 60's and early 70's had some great free software. Then we had the Unix wars, late 70's and 80's were somewhat mediocre. But the mid to late 90's saw a huge surge in open source popularity.
And now there are a whole generation of students who have always had free OS's, dev tools, and all sorts of software. When they get out of school and in to the business world they will bring some of that with them. Much like students past brought Unix from the campus to the work place.
And it's more likely to happen now because there are so many young execs (usally coder turned suit) that aren't afraid to use open source porducts at the enterprise level (Yahoo! is a great example)
I think we are just getting started.
Ethernet's goal is carrying electrons down a wire. That's it. It is as secure or insecure as the high level protocols you are running on it. Perhaps you are confusing Ethernet with the Internet (in the form of TCP/IP)?
After hearing his predictions, I'm convinced that while you can come up with some great things in the '70s, it doesn't mean you're going to grok a lot of what's happening now...
Maybe Metcalf should spend less time on that sailboat of his and more time on line.
-------------------------------------------------
I bent my wookie
On the other hand, I don't think he had any idea what's new about Application Service Providers. He called amazon.com an ASP because you submit information to them and they run a program to handle your order. By that definition, just about every site with a form and a CGI program would be an ASP.
Original article is at http://www.in foworld.com/cgi-bin/displayNew.pl?/metcalfe/bm1204 95.htm
Here's a good story about the result: http://computer.muni.cz/internet /v1n3/eats9702.htm
I wouldn't worry about Metcalfe. He Has apparently has lost a few clues since inventing Ethernet.
Just remember that he was the one who predicted the fall of the internet, citing that its popularity was just a fad.
Being a good engineer doesn't make you good at seeing what uses people will put products created by engineers to. An engineer only understands the motivational structure of other engineers, unless he or she is uncommonly perceptive. It is possible that Metcalfe could be so perceptive. Of course his idea that the internet was doomed buries that idea six feet under.
I expect to see him on the Psychic Friends Network sitting beside Dionne Warwick any day now.
Lee
Muslim community leaders warn of backlash from tomorrow morning's terrorist attack.
with the effect of weakening the political power of the slave states.
Actualy, the northern states said that the southern ones shouldn't get to count slaves in their population whatsoever, since, of course, they couldn't vote, wern't citizens, were not protected by the constitution, and were certanly not going to be represented fairly by the people in the house. Southern states wanted one black person to count as a whole person for purposes of assigning house seats.
ReadThe ReflectionEngine, a cyberpunk style n
Ethernet isn't completely open. You have to pay some kind of licence fee to get the hardware MAC. One fee for the manufacturer code and it used to be one per device.
In the old days, when we had field circus in, it was normal practice to try to conserve MAC addresses by moving the PROMs between replacement hardware, and it wasn't because anything needed the MACs.
... that's why it's stable (according to Metcalfe). Just one of the many brilliant things said on the show. I also liked the bit where Chris brought up the $0 marginal cost issue -- how one person can improve free software and everyone benefits, unlike the brick and mortar world, where improving your own car doesn't improve everyone else's. Metcalfe was either too dense to understand, or purposely avoided the question. It's hard to tell which, as he babbled about "you can have your neighbor work on your car". Doh!
This is perhaps the most fantastically ignorant commentary I've heard re: free software.
Who gives a big rat's ass what this schlunk thinks? If he thinks. Sounds to me he just mouths the line is all. blah blah blah. Like the pinnacle of human endeavour is to make a buck. Spare me.
What about OpenBSD? It is Open Source, but in my humble opinion, it won't probably ever lead to security breaches. Certainly not more than commercial counterparts (if there are any now or in the future).
It always makes me wonder what Bob's smoking when he writes about Open Source (or "Open Sores" as he sometimes puts it).
Isn't openness one reason ethernet is so successful? Most competing technologies have lost. We used to have ATM and FDDI in our LAN at work, and since last year (actually ~18 months) we have had a pure ethernet LAN. Switching out 622 Mbps ATM and 100 Mbps FDDI for 1 Gbps ethernet, to connect up the buildings. If I remember correctly when ethernet was first released, it was just an open spec. Then 3Com (which he helped start) made up some hardware to sell, built to that spec. Then when everybody else made there hardware, it all worked together.
I know ethernet is software vs. hardware, so maybe I'm just seeing it wrong.
It's real hard (almost impossible) to predict what happen in the future. Mr. Metcalfe may be predicting that OSS will fail on one radio station, while another talk show guest is predicting that it will be the next stage of evololutionary development.
Is he basing his prediction on more than "people won't like the security risk it presents"? Regular US Mail poses a security risk as workers can easily steal your letters (especially around Christmas time, when letters can be easily lost). Even credit cards at the store are handled by someone who can very easily write your number down and use it to make phone purchases. But these methods continue to grow.
Certainly there has to be better reasons for predicting the downfall of a piece of technology or a methodology. Even using statistical analysis is risky at best, as statistics are almost always skewed one way or the other.
--
http://slashdot.org/articles/99/06/21/126233.sht ml
Hasnt Bob learnt anything from the previous episodes? How about these :
http://slashdot.org/articles/99/06/27/1353200.sh tml
http://slashdot.org/articles/99/07/06/213259.sht ml
There is no such thing as luck. Luck is nothing but an absence of bad luck.
Whenever you hear about Bob Metcalfe you hear him referred about as the "inventor of ethernet". He talks about how insecure and terrible open source is and how awful it can be. The one thing that gets me is that Ethernet was developed in a very tight corporate world -- and gee guess what? It's about as insecure as they get.
Chalking another one up to windbagism...
Bob's view is cool by me, it may be flawed by our own interpretation of where OSS is going, but hey, the world of full of people who eventually lose their way. Remember the world is what we make it. If we want OSS we can have it. If we want companies like M$ to decide whats good for us we can have that too. It's our call, make your choice. Personally, I like my freedom of choice.
PS Thanks TT, KDE is a fine piece of work and I look forward to KDE2.0
There are none as blind as those who will not see.. (unknown)