<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    <p>During the FOSS4G Oceania community day, Bruce Anger and John
      Hildebrandt made great progress toward running OSGeoLive in AWS.</p>
    <p>Bruce's notes are here:
<a class="moz-txt-link-freetext" href="https://docs.google.com/document/d/12Hix3gAlOkbpyBj9EjOza-JLMEvS6l135Z-Hpuw7ZlU/edit?ts=5bf77367">https://docs.google.com/document/d/12Hix3gAlOkbpyBj9EjOza-JLMEvS6l135Z-Hpuw7ZlU/edit?ts=5bf77367</a></p>
    <p>James, your points are valid when considering setting up a
      production server. However, if we initially limit the target use
      case to a workshop/demo type setting, with nothing on the VM to be
      considered of value, then I assume security should be able to be
      ignored.</p>
    <p>(A future iteration could address security). <br>
    </p>
    <p>Brian, I probably should have been more specific about the use
      case I was suggesting.</p>
    <p>Cheers, Cameron<br>
    </p>
    <div class="moz-cite-prefix">On 15/1/19 4:09 am, James Klassen
      wrote:<br>
    </div>
    <blockquote type="cite"
cite="mid:CALDi5PQGO2gbxHT6DscdTRWNzZOK3=phc0p8jhNU66rk89-HwA@mail.gmail.com">
      <meta http-equiv="content-type" content="text/html; charset=UTF-8">
      <div dir="auto">
        <div dir="auto">Technically, I suspect it would be relatively
          easy to convert the OSGeoLive VM image into the formats
          accepted by various cloud providers.</div>
        <div dir="auto"><br>
        </div>
        <div dir="auto">My main concern is that historically OSGeoLive
          has been setup prioritizing ease of use on a single user
          machine to let a new user explore and learn the software with
          as few hurdles as possible.  This seems in fundamental
          conflict with best practices for putting a server on the open
          internet (in the cloud or otherwise). 
          <div dir="auto"><br>
          </div>
          <div dir="auto">For example, we have easily guessable and well
            documented passwords, generally permissive permissions in
            applications and on the filesystem, unrestrictive firewall,
            way more than the minimum software installed than is needed
            for any given task (increasing attack surface area), ...
            <div dir="auto">
              <div dir="auto">
                <div dir="auto"><br>
                </div>
                <div dir="auto">I think we would need to put some
                  serious thought into how to secure an OSGeoLive cloud
                  image and what that might mean for usability.</div>
              </div>
            </div>
          </div>
        </div>
      </div>
      <br>
      <div class="gmail_quote">
        <div dir="ltr">On Sun, Jan 13, 2019, 20:04 Brian M Hamlin <<a
            href="mailto:maplabs@light42.com" moz-do-not-send="true">maplabs@light42.com</a>
          wrote:<br>
        </div>
        <blockquote class="gmail_quote" style="margin:0 0 0
          .8ex;border-left:1px #ccc solid;padding-left:1ex">
          <div>
            <p style="margin:0;padding:0">Hi All --</p>
            <p style="margin:0;padding:0"> </p>
            <p style="margin:0;padding:0">> <span
                style="background-color:#ffffff">creating a "OSGeoLive
                in the Cloud" instance</span></p>
            <p style="margin:0;padding:0"> </p>
            <p style="margin:0;padding:0">  It might be said that
              OSGeoLive is ...</p>
            <p style="margin:0;padding:0"> </p>
            <p style="margin:0;padding:0">- a flagship for OSGeo dot org</p>
            <p style="margin:0;padding:0">- a proof-of-work for
              UbuntuGIS
              integration</p>
            <p style="margin:0;padding:0">- a service to all of the
              member software
              projects, incubated or others</p>
            <p style="margin:0;padding:0">    literally multiplying the
              leverage of
              interoperable data toolchains when combined</p>
            <p style="margin:0;padding:0">- a service to all of the
              science communities
              worldwide, in all major human languages</p>
            <p style="margin:0;padding:0">- a service to students of all
              kinds, in all
              places touched by technology</p>
            <p style="margin:0;padding:0">  ... others not mentioned</p>
            <p style="margin:0;padding:0"> </p>
            <p style="margin:0;padding:0">So, what might a "cloud"
              OSGeoLive be ? 
               news flash, many long-standing projects now</p>
            <p style="margin:0;padding:0">included on OSGeoLive are
              quite active in the
              cloud and continue to be..</p>
            <p style="margin:0;padding:0"> </p>
            <p style="margin:0;padding:0">Major cloud players say -
              Ubuntu OS is the
              most popular cloud service OS by number of customers</p>
            <p style="margin:0;padding:0">OSGeoLive is already an Ubuntu
              platform
              project. so ....</p>
            <p style="margin:0;padding:0"> </p>
            <p style="margin:0;padding:0">companies handling geospatial
              data chains are
              already using directly, UbuntuGIS PPA to build...</p>
            <p style="margin:0;padding:0">Some.. but another fact ..
              Docker technology
              easily runs Ubuntu PPA systems on a RedHat base, today.</p>
            <p style="margin:0;padding:0">The Docker container shares
              the kernel only
              with the host, while the rest of the OS layers are
              Debian/Ubuntu.</p>
            <p style="margin:0;padding:0"> </p>
            <p style="margin:0;padding:0">Is there an economic arguement
              to be made,
              with respect to a "cloud" OSGeoLive ?</p>
            <p style="margin:0;padding:0">I find many economic
              arguements to be badly
              self-contradicting, when viewed broadly.. </p>
            <p style="margin:0;padding:0"> </p>
            <p style="margin:0;padding:0">The OSGeoLive project has
              continuity and
              serious utility.. beyond that, opinions vary widely..</p>
            <p style="margin:0;padding:0">Is a "cloud" OSGeoLive the
              right move ?  with
              what resources ?  open questions</p>
            <p style="margin:0;padding:0"> </p>
            <p style="margin:0;padding:0">  best regards from Berkeley,
              California</p>
            <p style="margin:0;padding:0"> -Brian M Hamlin   
               darkblue_b  dbb</p>
            <p style="margin:0;padding:0"> </p>
            <p style="margin:0;padding:0"> </p>
            <p style="margin:0;padding:0"><br>
              On Sun, 13 Jan 2019 20:23:28 +1100,
              Cameron Shorter <<a
                href="mailto:cameron.shorter@gmail.com" target="_blank"
                rel="noreferrer" moz-do-not-send="true">cameron.shorter@gmail.com</a>>
              wrote:</p>
            <blockquote style="border-left:2px solid
#000000;padding-right:0px;padding-left:5px;margin-left:5px;margin-right:0px">I'm
              really hopeful that
              we'll see some people follow through with <br>
              creating a "OSGeoLive in the
              Cloud" instance which could be spun up and <br>
              used during training sessions. (This could be a great
              Google Season of <br>
              Code topic). I think this could be
              done with free developer AWS <br>
              instances, but it would be good to have access
              to budget if required to <br>
              support a first deployment.<br>
              <br>
              Also, it would
              be good to have discretionary funding to pay for <br>
              OSGeoLive USBs for
              conferences that ask for it during the year. Maybe do <br>
              a print run of 500 to
              1000 USBs (That would be 50 to 100 USBs at an <br>
              OSGeo table at a
              conference/workshop/code sprint, for 10 to 20 events).<br>
              <br>
              On 13/1/19 3:11
              am, Astrid Emde (OSGeo) wrote:<br>
              > Hello folks,<br>
              ><br>
              > yes - it
              is time to think about the budget for 2019.<br>
              ><br>
              <br>
              -- <br>
              Cameron
              Shorter<br>
              ...<br>
              <br>
            </blockquote>
            <p><br>
              --<br>
              Brian M Hamlin<br>
              OSGeo California<br>
              <a href="http://blog.light42.com" target="_blank"
                rel="noreferrer" moz-do-not-send="true">blog.light42.com</a><br>
            </p>
            <p style="margin:0;padding:0"> </p>
          </div>
          _______________________________________________<br>
          osgeolive mailing list<br>
          <a href="mailto:osgeolive@lists.osgeo.org" target="_blank"
            rel="noreferrer" moz-do-not-send="true">osgeolive@lists.osgeo.org</a><br>
          <a href="https://lists.osgeo.org/mailman/listinfo/osgeolive"
            rel="noreferrer noreferrer" target="_blank"
            moz-do-not-send="true">https://lists.osgeo.org/mailman/listinfo/osgeolive</a><br>
        </blockquote>
      </div>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <pre class="moz-quote-pre" wrap="">_______________________________________________
osgeolive mailing list
<a class="moz-txt-link-abbreviated" href="mailto:osgeolive@lists.osgeo.org">osgeolive@lists.osgeo.org</a>
<a class="moz-txt-link-freetext" href="https://lists.osgeo.org/mailman/listinfo/osgeolive">https://lists.osgeo.org/mailman/listinfo/osgeolive</a>
</pre>
    </blockquote>
    <pre class="moz-signature" cols="72">-- 
Cameron Shorter
Technology Demystifier
Open Technologies and Geospatial Consultant

M +61 (0) 419 142 254</pre>
  </body>
</html>