Visited birthplace of the Web: CERN

I do so hate the references to TBL as father of the net. He really was just the guy who claimed paternity after the event…

The concepts of the web, predated TBL, by some forty place years. Hyper Information Systems were first theorized by Vannevar Bush, in his paper “As we may think”, published in 1945.

TBL didn’t even find out about this, by himself. As he admits in the section on influences in http://www.w3.org/People/Berners-Lee/FAQ.html he got the idea from Linkworks from Digital. More specifically, as I recall, something called “Memex Prototype 1″ developed by Bob Wyman and his team in Valbonne. Of course, Ted Nelson and Doug Engelbart were also working in this area.

Bob Wyman, who I was lucky enough to work with in the late 80′s, recently invented Pubsub.com, which sadly languishes on some tapes in an archive somewhere – one of the visionary organizations with respect to RsS, XML, JAbber and all that stuff…

He has just started with google, and I I look forward to more innovation from him.. He has also worked on Distributed Software Licensing, arguably the framework for Digital Rights Management of today.

Amusingly, Bob was also the force behind one of the first and major Office Automation Systems – Digital’s All-in-1 which started out as a Prototype called Cposs, as I recall. Bob tells the story at http://mamamusings.net/archives/2005/11/19/the_culture_of_the_deck.php (second to last comment) about how Apple’s Hypercard, would, it appeared, have grown from his work.

If it weren’t for people doing things smarter, and “stealing ideas”, we wouldn’t have advanced as we have, but we need to be cautious about attributing the original ideas to the right people.

MisterQ: the implementer usually gets the credit and Tim was the first to build a Web server, a Web browser, and show it to other people who adopted it.

Of course other people had other ideas earlier. And Tim’s work doesn’t take away from them, but it wasn’t Bush who showed the world the Web browser and server. It wasn’t Bob, either.

And, when you search Google for “Inventor of the Web” you get a pretty clear answer: http://www.google.com/search?hl=en&q=Inventor+of+the+Web&btnG=Google+Search

If you don’t like the answer, well, sorry. That’s not the way the world works. It might also have something to do with the fact that you won’t sign a real name to your post, which greatly reduces the credibility of your complaint.

I too loved CERN, and all its jumbled onsite evidence of collaborators all over the world–a battered wooden packing crate sent from India–a parking lot with license plates from all over Europe–the multiple varieties of accented English I heard.

Our guide told us that one detector’s brass fittings had been melted down from the casings of bullets once issued to USSR soldiers. I have to say, that moment brought tears to my eyes.

Robert, It was lovely to see you and Maryam at LIFT. Brian is very keen to get you back to CERN when you’re next over on this side of the pond. Say ‘hello’ to your wonderful wife for me!

Robert, MisterQ is undoubtedly Peter Quodling of Australia — and once of Digital Equipment Corporation… Knowing Peter, I doubt he was intentionally hiding when he wrote his note.

Robert, as you point out, TBL appears to have been one of the first if not the first to build a wide-area network web server. I know that Peter knows this, so I’d like to explain a bit of what he’s saying in his note: As I am usually very careful to say, I believe that I may have built the first “wide-area network web *browser*”… (not server) four years before TBL built his server. I didn’t need to build a “web server” since my work was done with DECnet on VAXen running VMS. DECnet and the RMS file system were integrated in VMS — thus a web server wasn’t necessary. Only a browser was needed. With DECnet, opening a file on a remote machine was identical to opening one locally. But, TBL did his work on Unix, which as an older and less modern operating system didn’t offer integration between the file system and the network. Thus, to overcome this shortcoming of Unix with TCP/IP, it was necessary for TBL to build a server — perhaps the first server. But, certainly not the first browser. (BTW: I called Memex a “web browser” and suspect that my usage of the term may have influenced folk we talked with at CERN. As Peter points out, the early name of “ALL-IN-1″ was CP/OSS — The “Charlotte Package of Office System Services” — since it was originally developed in DEC’s Charlotte, NC office. Given my connection to Charlotte and the fact that I had wanted “hyperinformation” in ALL-IN-1, I would joke that what we were building was a browser for “Charlotte’s Web.”…)

The work we did in Valbonne, France on building Memex Prototype 1 was unfortunately too early to be really useful as anything other than an inspiration to others — and we were fortunate to have inspired many… For instance, in addition to our presentations to folk at CERN we were also the first to introduce Daniel Dardailler to hypertext. He later became, and remains, the W3C’s current Chair for Europe. (Daniel was the husband of Pascale Dardailler, one of the developers in my group in Valbonne, and thus he was one of the first people to ever see wide-area-network hyperinformation.) Also, the now late Gilles Kahn, a leader of INRIA and someone with close ties to W3C over the years worked closely with my group in Valbonne (his office was in the building next to ours.) Quite a few other “Memex” graduates have had strong roles in the development of the web. For instance, Ward Clark, one of the Valbonne developers, was an early employee at Open Market in the early 90′s. Open Market was one of the first “eCommerce startups” and the folk that first created FastCGI… I’ve done a few things myself since then…

The primary reason that Memex wasn’t successful when developed was that in the mid to late 80′s, there simply weren’t enough people connected to networks to get much use from a wide-area network anything. That’s actually the reason Dave Stone and Mike Horner of DEC in Geneva had us focus so much “marketing” on the people at CERN (a little north of us on the French/Swiss border and close to our Geneva HQ). CERN had one of the largest DECnet networks in Europe until they began to switch to TCP/IP. The world would have been different if they had stuck to DECnet… (So, Ben Segal’s work was influential for reasons you probably didn’t realize! Web servers were only needed *because* of the switch from DECnet to TCP/IP!)

Anyway, I would like to object strongly to your suggestion that Vannevar Bush shouldn’t get credit for having inspired the web. I know as a fact that many of the people who worked on what you now know as hypertext were inspired by Bush’s article. I first read the article in 1974 and it was *the* reason I got into computing — to build Memex… Throughout my career I have come across many, many others who have said similar things. Certainly, prior to 1990 or so, Bush’s work was well know to everyone working on the hypertext problem. Even Ted Nelson admits that he was inspired by Bush back in the 60′s. (and TBL read Nelson’s work before building his server…) Robert, you are certainly quite knowledgeable about many areas. However, you are new to this business and, I would suggest, should not really be making strong statements about things that happened long before you joined us…

Bush *did* inspire the web — no matter what any young folk may say today. I know, I was there when it happened… It may have taken me, and others on my team, to teach TBL about Bush’s writings, however, the fact that TBL was unfamiliar with Bush until he met us doesn’t mean he wasn’t already strongly influenced by Bush’s ideas — which by the time TBL became interested in Hypertext were well embedded in the thinking (and dreams) of those in our business.

bob wyman

Robert,

So tell us, what OS are these grid computers running? Linux or BSD, I’ll bet. Something that big either has to run Linux or one of the BSDs. Nothing else has the horsepower.
If I was building a huge grid computer, it would be running FreeBSD most likely, since it’s the most robust and stable OS platform on the planet bar none. Fastest TCP/IP stack, best metadata soft uopdate file system. The bennies go on and on…

Talk about straw-man arguments–who’s claiming Bush didn’t inspire the web or deserves no praise?

The real issue here is whether Scoble should be able to praise (even in passing) the work of TBL. So, someone who implemented big ideas in a form that got widely adopted and had a big impact–when that person gets praised, you have a big issue with that?

I see the exact same kinds of arguments made against Dave Winer, by Wikipedia vandals insisting that all credit for RSS should go to the people who built RSS .90 on top of RDF, or maybe to Netscape which employed those people, or maybe to the people who created RDF, or maybe to–well, basically all the credit for RSS keeps getting assigned to just about anybody except Dave Winer.

The same arguments are used to insist that Dave Winer deserves zero credit for implementing the “enclosure” tag in RSS (and getting his company’s blogware and aggregator to support it what was then called “audioblogging”) because somebody else once suggested tags for “sound” and “video.”

Hypertext is a smart baby with many smart ancestors, many of whom deserve praise. Scoble’s praise of TBL is also deserved.

My wife and I did a tour round CERN on a short holiday to Geneva a few years back. Great, except that I had accidentally booked us on a French-language tour. Fine for me, as a language student of 11 years and claims of bilingual ability in French. Not so great for her, with the more familiar British approach of ‘English or else’. Oops.

Only catch: I didn’t know what the French word for ‘magnet’ was. Kinda important in context. And my wife even worked it out before I did.

Betsy (#7), Dave Winer should be given credit for what he did and not for what others did. Dave didn’t invent RSS and should get no credit for that. What Dave *should* get credit for is years of hard and dedicated work without which it is unlikely that RSS, Atom, podcasting, etc. would be what they are today. Anyone who diminishes Dave’s contribution is making fools of themselves — but, anyone who claims that Dave did more than he did is simply wrong. A similar situation exists with the Web. TBL had an exceptionally important role in the process of making Hypertext a reality. However, it is wrong to give him sole credit or to refer to him as the “inventor” of the web. The reality is that hundreds, if not thousands, of people laid the foundation on which the success of Tim’s work was laid. Those people deserve to have their contributions recognized as well — even if less glorious than TBL’s. When people like Scoble say that TBL built the first server, the first browser, etc. they are simply wrong. Heck, if you want to see a hypertext server *before* TBL’s, just look at the HAM (Hypertext Abstract Machine) which was part of the Neptune system that Norm Delisle and Mayer Schwartz wrote about in 1986.) TBL’s ideas weren’t new but that should not diminish in any way our respect for what he did. We should be careful that in praising TBL we don’t neglect the work of others. The same goes for Dave Winer. I believe that is what Peter Quodling was trying to say in his note…

bob wyman

Robert:

Damn, I wish I’d known you were here, i would have come along. I know Ben well and he really is as you describe him, a great guy (whatever one’s opinion of the birth of the web). He would certainly be the last to make grand claims, as you say he’s too humble. Sadly I couldn’t take the time off work to come to Lift, i should have tried to get work to pay for it. If you come back drop me a mai.

Peter: The Grid CERN is a (major) part of (under the aegis of the EGEE project – eu-egee.org) runs on Scientific Linux, a redhat derivative made by CERN and Fermilab, though we hope to get it onto other platforms in the near future. CERN also works with the US Open Science Grid, and Grids ite sin the nordic countries. I think he data for the LHC will be processed at some 200 sites worldwide. If you want a pretty visualization of the scale of the Grid, see http://gridportal.hep.ph.ic.ac.uk/rtm/ ).

I am sooooo jealous that you got to tour CERN. The Hadron Collider is going to provide us with a wealth o data that has the potential to set the world of physics n its head and hopefully confirm some of the components of string theory.
I hope you got a bunch of video fro the tour to share!

A general for comment for the people here, if you want to see CERN, or more specifically the LHC, you have to come pretty much now. By the sumemr the tunnels will all be closed for commissioning, and once the machine turns on at the end of the year it will be a little ‘warm’ down there. Eeven if you can make it before it shuts you probably need to book now as the palces in the main tours are filling up.

For those that do miss it, there will be a very nice virtual tour of at least the ATLAS detector, which I’ve seen a very early version of. Look out for it on the CERN website.

I’d like to put into a bit more context some of MisterQ’s and Bob Wyman’s concerns about TBL’s respect for his antecedents from my own memories and notes from that time.

As Bob admits, Tim was in fact unaware of the work of Bush(Memex), Englebart(NLS) and Nelson(Xanadu) until after he had already conceived his essential Web ideas by 1989, based on his own system Enquire. He heard about Bush from the Linkworks documentation but MisterQ is wrong to say “he got the idea” for WWW from Bush, DEC Linkworks – or other hypertext systems like Guide, Dynatext, Apple’s Hypercard etc. that TBL was aware of. So where does the blame come in?

Tim’s idea was to combine hypertext and hyperlinks with Internet, which he was watching as it struggled for recognition at CERN. But even more importantly, he designed WWW to be UNIVERSAL and OPEN, i.e. cross-platform, cross-OS, and with its code open-source (then only just emerging as a movement). Tim solved the problem of “open naming” with the URI (“the most fundamental innovation of the Web” as he put it) and the problem of “open data-presentation” with HTML and HTTP. These were major problems that people like myself had despaired of ever seeing solved in the proprietary jungle of that time. And he implemented the whole lot, client, server and support protocols, single handed in 3 months.

As Bob points out, working in the real messy world is much nastier than doing it inside a proprietary OS where e.g. file opening locally or across (the proprietary) network was identical. Going from this to implying that we were wasting our time promoting Unix and TCP/IP instead of just arranging to run (and pay for) DECnet and VMS everywhere is a bit strong. “The world would have been different if they had stuck to DECnet…”. Maybe this was tongue in cheek, right Bob?

TBL’s WWW succeeded in the end because, due to his foresight and determination, it was and remained an open, universal system, unlicensed, non-proprietary and consensual.

Ben, You wrote: “concerns about TBL’s respect for his antecedents”… The subject here has not been what TBL says about his work but rather what *others* say about it. I took MisterQ’s comments as a reaction to Scoble’s comments — not a suggestion of any failing of TBL himself. I believe I was careful in my own comments to make sure that they didn’t speak against TBL — but rather addressed what others say about him.

The *world* would have been different if CERN had stuck with DECnet. Yes, that was intended to be an ironic comment… If you hadn’t been successful in getting CERN to adopt TCP/IP, any hypertext system used by CERN (TBL’s or Memex, or whatever) would have ended up with the same fate mine did — i.e. failure. That would have happened since CERN would have been stuck working with a network protocol that the world soon would pass by. TBL wouldn’t have had any reason to build the server that was vital to supporting his system on Unix, Windows, and dozens of other OS’s that didn’t have networking integrated into the file system. Eventually, someone else would have been the one to finally get a useful “worldwide” web running. Yes, the world would have been very different if you hadn’t talked CERN into adopting TCP/IP. But, I’m happy with the world we got.

bob wyman

First of all, good job! But I want to ask a question. Why can’t the SEARCH ENGINE be ‘harnessed’ to the OS in a progessive manner for schools? Wikis are big. And
blogs are unmanageble in the main. But a school-curriculum, which is your OUTSIDE ‘intranet’…? Huh? Hey! That is a COOL IDEA!
Thanks-

Kulafwrnei

Photos from the visit here: http://www.23hq.com/mygdal/album/1652579

Robert and Bob, Ben and others.

Yes, was I, the Pseudonym is a simple one, to reduce the spam (and it’s what most peers know me as)

My original issue, was mostly around the fact that, with all due respect to TBL, sorry Sir TBL, the development of technology is one of development from many contributing factors and people. I myself have several “ideas”, that build upon the work of others, that I hope to unleash on the world at some future point.

Having had a long involvement in the “industry” and having had the privilege to count amongst my friends and mentors, the like of Bob W and others, I am perturbed by the current day lack of understanding as to how (and why) we got where we are.

I never cease to be amazed when talking to a current day “Security Expert” and I make mention of the work of Bell and Lapadula at Mitre Corp, in the early 70′s. (and subsequent work by Biba and others).

Or talking with a SQL DBA, and start talking about E.F. Codd, and his “laws” from the late 60′s. Or the work by Licklider et al, on human computer interaction, and so on.

Or listening to XML devotees, and asking them what they understand about Lisp and s-expressions in Lisp (which are arguably where XML should have gone)
The blank stares in return are a sad reflection on the lack of understanding by many of the true “science” behind technology, or moreso, the ability to work “as a team”.

I think that one of the biggest failings of the present day, is that people do not understand how and why we have gotten to where we are.

re comment 2. and the reference to google. This raises an interesting point. The internet is now the largest authoritative source of knowledge. sadly, it is also the largest non-authoritative source of knowledge.

The anonymity that it inherently presents, and the lack of accountability for delivery or information, or validation of expertise, is such that the overall “signal to noise ratio” is probably getting worse as the knowledge base expands at the exponential rate that it is.

[Aside: Bob W, will be amused by the fact, that a few years ago, I was taken to task by a project manager that interpreted software licensing her own way. It was only when I quoted line and verse from the design document and the business policy document that drove it, that she deferred to a much more authoritative source - otherwise she would have continued to spread mis-information. ]

Blogging while a wonderful concept, also has it’s flaws, as I am sure that Robert has experienced.It gives the power of the journalist to the the individual, yet not the training and or governance imposed by publishing houses held accountable for journalistic works.

What would be more interesting would be mechanisms for a) validation of the authoritative “quality” of the data, and b) better mechanisms to challenge, and or expand on others ideas with due attribution. Things like CC Licensing and tagging go part of the way, but not far enough yet.

Of course, who knows, in 20 years time, history may be trying to tell us that Scoble invented blogging, or even the internet (in conjunction, of course, with Al Gore) ….