¿Tienes cuenta de YouTube?

Nuevo: habilita las traducciones y subtítulos creados por los usuarios en tu canal de YouTube

English subtítulos

← Reflection on NSFNet

Obtener código incrustado.
8 idiomas

Mostrar Revisión3 creada 07/29/2012 por juan.jose.zapico.

  1. So, in that video, the University of
    Michigan, the unexpected underdog in the
  2. competition for the National Science
    Foundation wins. And they win using, sort
  3. of an old fashion technique. Basically,
    many might say that was cheating, right?
  4. They didn't play exactly by the rules.
    This is becoming an increasingly common
  5. technique sometimes when, when an
    institution wants to win a grant. They'll,
  6. they'll you know, they'll, they'll bring
    so much more money to bear on but it's
  7. kind of an obvious, obvious thing that
    they can, that they're going to win the
  8. grant. So the, but, so there is actually a
    couple of schools that thought that was
  9. kind of unfair. But the other thing to
    think about in this is the lobbyists
  10. carefully crafted the15 million dollars. I
    would claim they carefully crafted the
  11. fifteen million dollars so this network
    would fail because they knew how much, if
  12. you leased the lines from AT&T, it
    would, it would cost so much that the only
  13. thing you could afford was 56 kilobit
    lines for the network that was being
  14. proposed. Now, when you think of 56
    kilobits, right? So here, take your, take
  15. your phone, right? You got your 3G and
    your 4G, and the thing that came before
  16. all those was EDGE. You remember EDGE?
    Now, of course, if you go to the wrong
  17. place, you end up in a basement or
    something, you go EDGE on your, on your
  18. phone. Edge, the thing that's your phone's
    doing when it's doing real bad connection.
  19. Edge is 128 kilobits, which is twice what
    the national backbone for all the
  20. scientists talking to all the computers in
    the country. 56 kilobits was the national
  21. backbone that the lobbyists carefully
    authorized funds for. And there will be
  22. only one conclusion at this point and that
    was that they wanted it to fail. And, as
  23. is said in the movie Jurassic Park, nature
    finds a way. And, and if, if you think
  24. about it the story that Doug van Holland
    just told is like a perfect storm. Mci
  25. just started existing. You know, they just
    started existing and so they wanted to do
  26. something cool , and so they were willing
    to take a risk. Doug had, just come from
  27. Carnegie Mellon University, where he had
    worked with IBM before. So, what's the
  28. likelihood that, here's a school that has
    a long network history, but didn't work on
  29. the Arpanet. Knows IBM intimately. Just
    it's just pretty amazing, you know, that,
  30. that this all happened. And so, but it
    did. Nature found a way just like in
  31. Jurassic Park and all the plans and all
    the attempts to, to, to box, to put and to
  32. draw a fine line around this failed. And
    so, the NSF net as Dough said, just took
  33. off, alright. And the, going back to what
    Larry said, the, the key was is that each
  34. school had to be, you know, first on it or
    you're going lose your physicist, right,
  35. it became a badge of honor. And so,
    schools basically panicked and they found
  36. money somewhere to run fiber, to run
    networks to people's offices. This is a
  37. whole bunch of infrastructure that, that
    needed to be installed. And I wasn't at
  38. University of Michigan at the time, I was
    at Michigan State University at the time,
  39. just up the road. And I saw the internet
    for the first time in the, in the building
  40. that's here in this upper left hand corner
    of the, of the slide. And that is the what
  41. we used to call SI North, the school that
    I'm part of. It's really a rather
  42. nondescript building but they used to have
    monitors and watching all these things.
  43. And so, the traffic grew, the performance
    grew there's all kinds of things that,
  44. that had to be solved through this thing.
    It started in 1988 and was supposed to go
  45. through 1993, for five years and it ended
    up going through 1995. Now, the key thing
  46. is if you think about that time frame, you
    know, by early 1990s, things were pretty
  47. universal. I mean we'd gone from will this
    happen to every, everyone had to be, you
  48. know, everyone that mattered was pretty
    much on the network. And the question was
  49. how much bandwidth, what with these
    servers, how do we work with all these
  50. things? And so the, the original NSF Net
    was aimed at research universities and
  51. they had made a bunch of rules about that.
    And, and there were some universities in
  52. particular Cleveland the Case Western
    Reserve at Cleveland it, well Cleveland
  53. area, Case Western, around Case Western
    Reserve University, had this thing called
  54. a Freenet. And these were bulletin board
    systems. And there was lots of bulletin
  55. board systems but they were all very local
    and some bulletin board systems started
  56. having sort of partnerships with
    universities and sneaking regular people
  57. onto the Internet. I remember personally,
    just in a working university, the internet
  58. was this cool thing. And, it was something
    that only we at universities could do.
  59. And, we, we're the only ones that could
    use it and all you want to do is tell
  60. people about it. And then, things like
    these Freenets made it available to the
  61. average citizen and that just created so
    much more demand. And so, towards, you
  62. know, early 1990, late 1980s, this
    academic only rule sort of started to be
  63. relaxed and they made some rules about who
    could do what. And by, by this point in
  64. time, the cat was totally out of the bag,
    right. I mean the network was 45 megabits
  65. and all the schools on the planet. So, it
    was pretty tough at this point for the,
  66. for the lobbyists to shut it down [laugh],
    right? They had carefully laid plans, had
  67. failed and out we go. Okay, and so we have
    this connectivity. And one thing I
  68. remember about the connectivity in the
    early days is we used to have post-it
  69. notes of all the cool servers where you
    could download software or where email
  70. list were or, or newsgroups were. And, and
    you, you have to have post-it notes and
  71. keep track of all these host names
    etcetera, etcetera, etcetera. And so, once
  72. the connectivity was there, and this was
    happening worldwide and in additional states
  73. as well. And the question was is how would
    we organize all this information? How
  74. would we make sense of it? And so we
    started Illinois, University Illinois in
  75. the supercomput er centers and we moved to
    the University of Michigan, where the
  76. first NSF Net happened, and then grew and
    expanded. And the, the next place on our
  77. stop is CERN, CERN High Energy Physics
    Lab. And so, you probably know that the
  78. CERN is the birthplace of the web but
    that's not all it does. A matter of fact,
  79. it chose that the web was not it's purpose
    because it's purpose is Physics, high
  80. Energy Physics. And. One of the things
    about experimental Physics is that, kinda
  81. like super computers. The in order to make
    the next step in physics research, you
  82. need to you need to build a bigger
    experiment and a bigger experiment. So, it
  83. used to be that a physicist could learn
    things about electrons and neutrons by
  84. something about the size of this table
    like a cloud chamber. And something would
  85. go through and they could write a little
    paper. And then once, once you've seen
  86. everything you can see and things and the
    size of this table, you'll say, well, I'll
  87. do something ten times larger than the
    table. And then I get a new set of papers
  88. and a new set of results and at some
    point, you've, you've gained as much
  89. Physics as you can gain from that. And
    this repeats itself, and it gets larger
  90. and larger and larger. And it gets to the
    point where you literally can only afford
  91. to have one experimental facility in the
    entire world. And this is what CERN is
  92. dedicated to. It's dedicated to the notion
    of if you're going to build a physics
  93. experiment, let's build it here at CERN.
    And there's a whole structure around this
  94. where people come from all over the world.
    It physically straddles the border between
  95. France and Switzerland. It's got, you
    know, that Switzerland of course, has you
  96. know, has really good relationships and
    people can come and, you know, even
  97. Russian, you know, back when there was
    some tension between governments,
  98. scientists can still get together and
    work. And the lead time on these projects
  99. is fifteen to twenty years and the, the
    size of the things that they build is just
  100. gigantic like the, the Atl as the, the,
    the, the, most recent detector is like six
  101. stories tall and took years to build. So,
    these people have to work and build and
  102. think. And it turns out that they have a
    lot of fun. So I'm going to introduce you
  103. in this next slide to the Cernettes. So,
    the Cernettes, is a musical group from
  104. CERN. And, and so basically you [laugh],
    you, you can view some of the videos that
  105. I have here. You don't have to, but I
    suggest that you do. They're just pure
  106. fun. Their, their, their music is about
    the web, about high-energy physics, about
  107. colliders, and about supercomputers. And
    you see the women sitting around the
  108. supercomputer in one of their music
    videos. And so, it's just kind of fun. But
  109. remember that the reason that these people
    are together is to mix something that
  110. can't be made separately. So, take a look
    and then come back. Well, welcome back. So
  111. continuing. I have been to CERN many
    times. The first time that I went to CERN
  112. was to do a lecture recording. And I have
    been working for many years, since 1999,
  113. with a physicist named Steven Goldfarb.
    That's Steven Goldfribe right there. He is
  114. the lead singer of the, of the Canettes
    Blues Band. And everyone else in the
  115. Canettes Blues Band is also a physicist,
    [inaudible] physicist. She actually is I
  116. believe the secretary general something.
    So, these people are all physicist and,
  117. and, and so they, they play together,
    right? You saw the Cernettes, if you
  118. choose to see. And now, here we have the
    Canettes. This is a blues band. Part of
  119. the reason that they do blues is because,
    you know, these people come from America
  120. and they like the blues, and there might
    not be a blues band so they just make one.
  121. They actually have a club. The CERN
    provides a place for them to play music
  122. and they have lots of fun. And so, so, the
    other thing I like about CERN, if you ever
  123. get a chance to visit go in the back, this
    is the cafeteria right by the building 40.
  124. And they have steak. I mean, the food is
    just magnificent. I mean, it's Europe.
  125. It's, it's France and Switzerland, after
    all. And, and there's, I just have so many
  126. fond memories of hanging out at this, at
    this cafe. And so, then we have over here
  127. not a lot of people probably have a family
    photo, deep inside the detector pit, of
  128. the Atlas. And so and so, let's see,
    where's my, there's my wife, Teresa, right
  129. there, my son, Brent there's me and
    there's my daughter, Mandy. And so we took
  130. a family trip to CERN and now, you can't
    go in, let me clear this bit here, let me
  131. clear this. You can't go in the detector
    anymore, right. So, this is the five
  132. stories, this is where the beam comes in.
    So, down there that`s the 2.5 stories and
  133. then up is another 2.5 stories. We happen
    to be in the middle when we are taking
  134. this picture. If you are to go into, into
    this area right now, this is full of
  135. equipment, okay. Just absolutely, just
    full. Go look for the Atlas detector. And,
  136. and basically you actually can't go in cuz
    it's all full of radiation now, too. And
  137. so we got in one of the, this was not,
    this was early on. I came back and got
  138. another tour when it was almost done. And
    so, this just is fun, but really smart
  139. people. And so, that's sort of the, the
    key message here. So I'm going to show you
  140. another video and this video is optional
    and this is a video of blues, our blues
  141. band, Steven Goldfarb, and other
    physicists and then I, I showed up and
  142. sang with them. And so, this again is
    optional. So again, what I say is as you
  143. see the crowd shots, most of the crowd are
    also physicists. They work really hard and
  144. they play really hard. Okay. Welcome back.
    So, why did I waste all that time showing
  145. you music videos and other silly things?
    Well, I mentioned that innovation springs
  146. from a culture. It doesn't spring from
    sort of like someone saying you innovate
  147. right now. It springs from a culture of
    fun and accepting new ideas and trying new
  148. things and then attempting to do
    something. And maybe, maybe you don't even
  149. accomplish w hat you tried to do but you,
    you run into something really cool along
  150. the way. And that really is the story of
    the web. You know the internet was there.
  151. The packets were moving. We could move
    files. We could, move images. We could
  152. move video. But we couldn't find it and it
    looked ugly. But frankly, to those of us
  153. using it, who cared if it was ugly. Once
    you figured it out, it was just, just
  154. totally awesome. And so, this group,
    Robert Cailliau and Tim Berners-Lee were
  155. working at CERN. And they had a problem to
    solve. And the problem they wanted to
  156. solve was the distributed documentation
    for physicists. Now, these people have to
  157. build things that take twenty years with
    thousands of people spread all over the
  158. earth. So, they have to like come up with
    designs. They have to share the designs.
  159. They have to write reports. They have to
    tell, you know, they have to do some kind
  160. of an audit. So, they needed to sort of
    write documents. Except they wanted to
  161. share them. And they wanted to use this
    new internet thing to allow the collective
  162. editing of documents. Now, that starts to
    sound a little bit like Wikipedia, but
  163. it's not. What they wanted to, what they
    wanted was the documents to be on
  164. different computers and then link the
    documents together and make it so you
  165. could edit them. So, certain documents
    might be in Poland, and certain documents
  166. might be in Switzerland, and certain
    documents might be in the United States of
  167. America, other documents might be in
    Japan. And they wanted to be able to edit
  168. them all and then link them all together
    and edit them both locally and remotely as
  169. well. And so, they created effectively a
    hypertext text editor with hyper links in
  170. it. Alright, these links from one web, one
    online document to another. They had to
  171. figure out the format to write these
    documents in. Then they had to figure out
  172. a way to represent links. And then they
    had to figure out a network protocol to
  173. move the data back and forth, to store and
    retrieve the documents. And all that
  174. becomes HTTP, the hypertext transf er
    protocol, HTML, the hypertext markup
  175. language. Web servers, which is where the
    HTPD web servers, which is where the
  176. documents are stored. And web browsers.
    And so, they had to build a complete
  177. infrastructure to create the document,
    distributed documentation environment that
  178. they had imagined. So, lets go meet Robert
    Cailliau at CERN, in his office just
  179. across the street from where the coffee
    shop and the steaks are at in the CERN