Bob Metcalfe re-evaluates open source (Linux.com) [LWN.net]

Linux.com talks briefly with Bob Metcalfe. "It's the sustainability long-term of the open source model that I worry about. Who will take care of the software after the novelty wears off and the volunteers lose interest and get real jobs?" Mr. Metcalfe appears not to have noticed that an awful lot of those "volunteers" already have real jobs.

to post comments

Bob Metcalfe re-evaluates open source (Linux.com)

Posted Apr 25, 2007 20:29 UTC (Wed) by atai (subscriber, #10977) [Link] (3 responses)

The long term sustainability of Bob Metcalfe's relevance is what people should be worried about.

Bob Metcalfe re-evaluates open source (Linux.com)

Posted Apr 26, 2007 8:16 UTC (Thu) by job (guest, #670) [Link] (2 responses)

I think it's pretty safe to say that Ethernet is here to stay.

Bob Metcalfe re-evaluates open source (Linux.com)

Posted Apr 26, 2007 9:10 UTC (Thu) by tialaramex (subscriber, #21167) [Link]

Home Alone is going to be playing on deep cable channels and in the mid-afternoon dead slot on bank holiday weekends for years, but Macaulay Culkin is washed up. Ethernet is Metcalfe's Home Alone moment.

Bob Metcalfe hasn't been relevant since he was forced to eat his own words live on stage. He's predicted catastrophic Internet wide failure, and the death of Linux since before the dotcom boom and rather than correct himself, each year he simply makes the same predictions again. If sites like Linux.com and LWN stop publicising his rants maybe he'd have to "get a real job" ?

Bob Metcalfe re-evaluates open source (Linux.com)

Posted Apr 27, 2007 13:09 UTC (Fri) by clugstj (subscriber, #4020) [Link]

So, he invented Ethernet and started 3Com. Why does that make his opinion on "Open Source" at all relevant? It's like asking Rosie O'Donnell for advice on politics.

Bob Metcalfe re-evaluates open source (Linux.com)

Posted Apr 25, 2007 21:26 UTC (Wed) by dmarti (subscriber, #11625) [Link]

It's the sustainability long-term of the college football model that I worry about. Who will take care of the football playing after the novelty wears off and the college players lose interest and get NFL jobs?"

Bob Metcalfe re-evaluates open source (Linux.com)

Posted Apr 25, 2007 21:33 UTC (Wed) by jdahlin (subscriber, #14990) [Link] (2 responses)

It's the sustainability long-term of the proprietary software development model he should worry about. Who'll take care of proprietary software when the founding companies goes bankrupt or discontinues the products?

Bob Metcalfe re-evaluates open source (Linux.com)

Posted Apr 25, 2007 22:27 UTC (Wed) by pmilne (guest, #34533) [Link]

I worked for a power utility who used a Human Resources package from a local software firm. The firm ran into financial difficulties, the last straw was when the utility cut the power for arrears. The utility fairly promptly had to find new software (since such software needs regular updating to accommodate changing tax rules etc).

The HR manager was not amused when I suggested that he should have persuaded management to leave the firm's power on.

Bob Metcalfe re-evaluates open source (Linux.com)

Posted Apr 26, 2007 15:30 UTC (Thu) by ccchips (subscriber, #3222) [Link]

Grid help us, you're right.

This is especially painful in "nich" situations like music composition and notation software such as an old program called QuickScore for Windows. I loved this program a lot, and now it's orphaned.

At least the guy didn't copy-protect it. Still, it's sad. If a program like Rosegarden loses its volunteer support, eventually somebody who needs it will pick it up and fix it to work on new hardware/software.

Oh--here's something interesting. Which closed-source operationg system runs on as many hardware platforms as Linux? Never mind it's "long-term sustainability."

The alternative

Posted Apr 25, 2007 22:02 UTC (Wed) by bojan (subscriber, #14302) [Link] (6 responses)

> "Your movement's character assassins tend to forget I was as negative on Windows as on Linux -- both being back-to-the-'60s technology."

I always like reading statements like this. The question that follows is: "So, did you write the alternative and where can I get it?"

The alternative

Posted Apr 26, 2007 5:44 UTC (Thu) by drag (guest, #31333) [Link] (1 responses)

It's also probably telling that back-to-the-sixties technology is more succesfull then back-to-the-seventies, back-to-the-eighties, or any back-to-the-ninties operating systems.

Go figure.

The path to Windows and Linux is littered with the remains of dead or dying propriatory operating systems.. along with the skeletal remains of the corporations that attempted to make money by making, supporting, and selling them.

The alternative

Posted Apr 27, 2007 20:45 UTC (Fri) by man_ls (guest, #15091) [Link]

It might seem that licensing has been more important than technology in the end. In this area there have been basically two wildly successful alternatives: liberal licensing and GPL. Neither proprietary keep-your-hands-off nor free do-what-you-want have seen so much adoption (although they occupy some sizable niches, as Mac OS X and FreeBSD).

If it were true it would be funny.

The alternative

Posted Apr 26, 2007 18:26 UTC (Thu) by flewellyn (subscriber, #5047) [Link] (3 responses)

I'm uncertain what he means by "back-to-the-60s technology" in the first place.

It's true that the first UNIX was created in the late 60s and early 70s, based off ideas from MULTICS and CTSS in the early 60s, but things have moved and changed a great deal since then. A modern POSIX like Linux or the BSDs is a very different animal from Bell Labs Unix from the 70s, or even SysV in the 80s.

The kernels have changed and grown with the hardware, the userland system software is much different (especially GNU, given that many core GNU utilities were deliberately written in different ways from the old ones), and the range and type of applications are very different.

Sure, you can trace the evolution of the core concept back that far, but you can do that with anything in any science. The new builds upon the old, correcting mistakes, adding new knowledge and techniques, completing partial solutions, and adapting to meet new needs. Do we criticize modern physics as "back to the 19th Century science" because it uses Maxwell's equations?

And one final note: the Quicksort algorithm was invented in 1960, and is still the fastest sorting algorithm we have in the average case.

The alternative

Posted Apr 27, 2007 1:33 UTC (Fri) by drag (guest, #31333) [Link] (1 responses)

The guy probably considures Unix-design itself obsolete, not just specific implimentations of that design. (linux vs BSD vs sysV)

I'd realy like to get his opinion on something he thinks a 'modern' design is. I bet he doesn't have any.

The alternative

Posted Apr 27, 2007 15:14 UTC (Fri) by zlynx (guest, #2285) [Link]

I'm not the guy you're responding to but I have some ideas for improvements.

The Unix/Posix system call API is synchronous, which means that a program--kernel conversation looks like:

opendir() -- wait -- result
readdir() -- wait -- result
open(1) -- wait -- result
read(1) -- wait -- result
close(1) -- wait -- result
open(2) -- wait -- result
read(2) -- wait -- result
close(2) --
etc...

An improvement would be an asynchronous API, which would look like:
opendir()
readdir()
wait for kevent
open(1)
open(2)
open(3..255)
wait for kevent
read(1..255)
wait for kevent
close(1..255)
closedir()
wait...
done!

Async syscalls were discussed a bit on LKML, so we may be seeing it in the future.

The alternative

Posted Apr 27, 2007 1:37 UTC (Fri) by raf (guest, #35151) [Link]

He might mean that Linux/Windows are more similar to 1960s UNIX than to, e.g., anything based on capabilities instead of ACLs, or an OS that uses something like WinFS instead of the traditional hierarchical filesystem.

Bob Metcalfe listens to himself hold forth on open source (Linux.com)

Posted Apr 25, 2007 22:08 UTC (Wed) by beattie (guest, #22563) [Link]

I'm sorry I gave up half way through, his continued broad brush painting of the entire Linux community for the childish behaviour or a few, combined with his snide remarks, convinced me that he was more interested in dazzling us with his brilliance and wittiness than in actually saying anything useful. Thus lowering the value of his own discorse.

Bob Metcalfe re-evaluates open source (Linux.com)

Posted Apr 25, 2007 22:12 UTC (Wed) by Hanno (guest, #41730) [Link] (2 responses)

It's the long term sustainability of witty comments on LWN we should worry about. Who will take care of trolling this site after the novelty wears off, people lose interest and start professional trolling on /. et. al.?

Bob Metcalfe re-evaluates open source (Linux.com)

Posted Apr 25, 2007 22:21 UTC (Wed) by jlgreer1 (guest, #29948) [Link] (1 responses)

Who is Bob Metcalfe?

Bob Metcalfe re-evaluates open source (Linux.com)

Posted Apr 25, 2007 22:31 UTC (Wed) by JoeBuck (subscriber, #2330) [Link]

Inventor of Ethernet, founder of 3Com. Brilliant as a young man, went badly wrong in the 1990s.

Bob Metcalfe re-evaluates open source (Linux.com)

Posted Apr 26, 2007 1:25 UTC (Thu) by wilreichert (guest, #17680) [Link] (3 responses)

Do I actually have to read this article or can I just come to a simple conclusion based on the summary above and go on with my life?

Bob Metcalfe re-evaluates open source (Linux.com)

Posted Apr 26, 2007 3:13 UTC (Thu) by proski (guest, #104) [Link] (2 responses)

The article sounds like a summary itself. It's basically just a collection of quotes with little or no context. It's impossible to argue intelligently against something presented this way, which explains what we are seeing in the comments here.

Bob Metcalfe re-evaluates open source (Linux.com)

Posted Apr 26, 2007 7:37 UTC (Thu) by petegn (guest, #847) [Link] (1 responses)

But the thing is .. IS this supposed to be a super uber technical discussion group on the viability of anything OR just a place where like minded individuals can get together and voice there own INDIVIDUAL take on each article without geting lambasted by the small minoroty that THINK they are UBER technical but in reality are no better than the rest of US ..

Pete .

Re-evaluating

Posted Apr 26, 2007 14:52 UTC (Thu) by GreyWizard (guest, #1026) [Link]

Wow. Well said.