Thicket data repository for the EEG
anil.recoil.org
1{
2 "id": "https://ryan.freumh.org/2023-10-30.html",
3 "title": "30 Oct 2023",
4 "link": "https://ryan.freumh.org/2023-10-30.html",
5 "updated": "2023-10-30T00:00:00",
6 "published": "2023-10-30T00:00:00",
7 "summary": "<div>\n <span> Previous: <a href=\"2023-10-23.html\">23 Oct 2023</a> </span>\n <span> Next: <a href=\"2023-11-06.html\"> 6 Nov 2023</a> </span>\n </div>\n \n \n\n <h2>HotNets</h2>\n<p><span>Organised travel dates, travel, and\naccommodation for HotNets.</span></p>\n<p><span>I was awarded a Queens’ College Perry\nTravel Award to the value of £740!</span></p>\n<h2>Met with Anil</h2>\n<p><span>Agreed to schedule dates to visit NYC\nand then find a time to meet Jane Street to discuss the polyglot package\nmanagement project proposal. I’m going to take the train from Boston to\nNYC! Anil said I’ve got to get a <a href=\"https://en.wikipedia.org/wiki/Junior_Ranger_Program\">Junior Ranger\nBadge</a> from a <a href=\"https://en.wikipedia.org/wiki/List_of_National_Park_System_areas_in_New_York\">National\nPark in NYC</a>.</span></p>\n<p><span>Anil recommended live blogging the\nconference on Mastodon.</span></p>\n<p><span>Talked about what a HotNets\npresentation could look like, and whether a live demo was feasible.\nMaybe I could pull AEON and Patrick’s OCaml R-Tree together into\nsomething workable. I think while very cool the APIs just aren’t there\nfor the AR demo on a hololens. A smartphone (e.g. iPhone LIDAR) could be\na lot more feasible though… I think considering the goals here is a\nuseful exercise. I’m going to say the goals of this presentation\nare:</span></p>\n<ul>\n<li>To get people excited enough to read the paper.</li>\n<li>To get people to come and talk to me about ideas for or issues they\nsee with the SNS.</li>\n</ul>\n<p><span>This should inform a SIGCOMM\npaper.</span></p>\n<p><span>Anil asked if AEON is\nproduction-ready, and I said I’ve been running it on my server for\n(according to my NixOS config git history) 9 months! There is a lot of\nstuff around tunnelling in there that complicates things, but that is\nuseful for the DNS over DNS Oblivously (DoDO) idea. We’ve also discussed\na signpost-like idea that AEON could be adapted to taking on similar\nfunctionalities to a TailScale control server. I think Jane Street has a\nNoise protocol WireGuard implementation lying around that could prove\nuseful.</span></p>\n<p><span>Anil cautioned me against spending\ntime on Hibernia. He said the idea of a service graph for things trying\nto run that could be dynamically pushed into machines and migrated is\nmore interesting. We’ve discussed a similar idea realting to Eilean\nbefore.</span></p>\n<p><span>Anil said when writing or presenting\nto consider the input mental state of the recpients and the desired\noutput state. Every paragraph is a process, control extraneous\nthoughts.</span></p>\n<p><span>Anil encouraged keeping multiple\nprojects in a ‘flywheel’.</span></p>\n<p><span>We discussed deploying Matrix servers\non eboards on physical spaces for the EEG.</span></p>\n<h2>Met with Anil and Magnus</h2>\n<p><span>Discussed the R-Tree based\nARP (a link-layer independant spatial resolution).</span></p>\n<p><span>Similar to ray-casting\ngame engines, responding with RRs.</span></p>\n<p><span>Eval - 2\nscenarios:</span></p>\n<ul>\n<li>low latency</li>\n<li>global & local</li>\n</ul>\n<p><span>Magnus mentioned frequency\nscaling.</span></p>\n<p><span>Anil encourage thinking\nabout the full connection pipeline lifecycle (in <20ms).</span></p>\n<p><span>Magnus is thinking a lot\nabout how to bootstrap this, e.g. how to get the DNS server you resolve\nto with 5g… or just your physcial presense without any network\nconnectvity.</span></p>\n<p><span>PUB/SUB came up a few\ntimes, but I think request/response is more approriate. You can return\nmultiple responses to a query. Maybe PUB/SUB would be useful for mobile\ndevices.</span></p>\n<p><span>We talked about access\ncontrol, getting quite fine-grained (guests in your house not being able\nto turn your themostat on). I think some of this should be in the pervue\nof the naming system (e.g. can you address this device), but some should\nbe left to the other policies and mechanisms otherwise it gets way too\ncomplicated too quickly (expressing access control over functionality of\na device). DNS update keys could be one way to authenticate access\ncontrol.</span></p>\n<p><span>Talked about iPhone\nLIDAR’s. How you could scan to create zonefiles (automatically\n‘statically survey’), translating 3d space into a 1d network\ntopology.</span></p>\n<p><span>Magnus is thinking about\nthe load placed on these services especially for geodetic\nresolution.</span></p>\n<p><span>We think the local\nlow-latency resolution might be the best use case for the SIGCOMM\npaper.</span></p>\n<p><span>On EIO Anil said it’s not\nworth making this a unikernel, just to use the Mirage\nlibraries.</span></p>\n<p><span>Anil said Magnus might\nwant to try deploying AEON as a next step.</span></p>\n<h2>Shark</h2>\n<p><span>I believe the plan is to get a prototype\ntogether in time for SOSP.</span></p>\n<h2>SRG Seminars</h2>\n<p><span>Organised SRG seminars for Noa\nZilberman’s students to give practice talks. Advertised the Xen SRG\nseminar.</span></p>\n<h2>Website</h2>\n<p><span>Added disseration(s) to papers\nsection.</span></p>\n<p><span>My patchwork of makefiles, bash scripts,\npandoc invocations, and more is starting to get a bit unwieldy. A good\nproject for a rainy day would be to adapt Patick’s OCaml Static Site\nGenerator.</span></p>\n<h2>Matrix\nSynapse authentication with Raven Shibboleth</h2>\n<p><span>Got a\ngood resource at <a href=\"https://nickx.hu/posts/2020-07-13-matrix-synapse-shibboleth-saml.html\">https://nickx.hu/posts/2020-07-13-matrix-synapse-shibboleth-saml.html</a>\nSpun up an experiment on a Matrix server I had lying around on an Eilean\ntest instance.</span></p>\n<h2>Teaching</h2>\n<ul>\n<li>TA’d the last L50 lab.</li>\n<li>Marked and supervised Distributed & Current Systems for\nPembroke.</li>\n</ul>",
8 "content": "<div>\n <span> Previous: <a href=\"2023-10-23.html\">23 Oct 2023</a> </span>\n <span> Next: <a href=\"2023-11-06.html\"> 6 Nov 2023</a> </span>\n </div>\n \n \n\n <h2>HotNets</h2>\n<p><span>Organised travel dates, travel, and\naccommodation for HotNets.</span></p>\n<p><span>I was awarded a Queens’ College Perry\nTravel Award to the value of £740!</span></p>\n<h2>Met with Anil</h2>\n<p><span>Agreed to schedule dates to visit NYC\nand then find a time to meet Jane Street to discuss the polyglot package\nmanagement project proposal. I’m going to take the train from Boston to\nNYC! Anil said I’ve got to get a <a href=\"https://en.wikipedia.org/wiki/Junior_Ranger_Program\">Junior Ranger\nBadge</a> from a <a href=\"https://en.wikipedia.org/wiki/List_of_National_Park_System_areas_in_New_York\">National\nPark in NYC</a>.</span></p>\n<p><span>Anil recommended live blogging the\nconference on Mastodon.</span></p>\n<p><span>Talked about what a HotNets\npresentation could look like, and whether a live demo was feasible.\nMaybe I could pull AEON and Patrick’s OCaml R-Tree together into\nsomething workable. I think while very cool the APIs just aren’t there\nfor the AR demo on a hololens. A smartphone (e.g. iPhone LIDAR) could be\na lot more feasible though… I think considering the goals here is a\nuseful exercise. I’m going to say the goals of this presentation\nare:</span></p>\n<ul>\n<li>To get people excited enough to read the paper.</li>\n<li>To get people to come and talk to me about ideas for or issues they\nsee with the SNS.</li>\n</ul>\n<p><span>This should inform a SIGCOMM\npaper.</span></p>\n<p><span>Anil asked if AEON is\nproduction-ready, and I said I’ve been running it on my server for\n(according to my NixOS config git history) 9 months! There is a lot of\nstuff around tunnelling in there that complicates things, but that is\nuseful for the DNS over DNS Oblivously (DoDO) idea. We’ve also discussed\na signpost-like idea that AEON could be adapted to taking on similar\nfunctionalities to a TailScale control server. I think Jane Street has a\nNoise protocol WireGuard implementation lying around that could prove\nuseful.</span></p>\n<p><span>Anil cautioned me against spending\ntime on Hibernia. He said the idea of a service graph for things trying\nto run that could be dynamically pushed into machines and migrated is\nmore interesting. We’ve discussed a similar idea realting to Eilean\nbefore.</span></p>\n<p><span>Anil said when writing or presenting\nto consider the input mental state of the recpients and the desired\noutput state. Every paragraph is a process, control extraneous\nthoughts.</span></p>\n<p><span>Anil encouraged keeping multiple\nprojects in a ‘flywheel’.</span></p>\n<p><span>We discussed deploying Matrix servers\non eboards on physical spaces for the EEG.</span></p>\n<h2>Met with Anil and Magnus</h2>\n<p><span>Discussed the R-Tree based\nARP (a link-layer independant spatial resolution).</span></p>\n<p><span>Similar to ray-casting\ngame engines, responding with RRs.</span></p>\n<p><span>Eval - 2\nscenarios:</span></p>\n<ul>\n<li>low latency</li>\n<li>global & local</li>\n</ul>\n<p><span>Magnus mentioned frequency\nscaling.</span></p>\n<p><span>Anil encourage thinking\nabout the full connection pipeline lifecycle (in <20ms).</span></p>\n<p><span>Magnus is thinking a lot\nabout how to bootstrap this, e.g. how to get the DNS server you resolve\nto with 5g… or just your physcial presense without any network\nconnectvity.</span></p>\n<p><span>PUB/SUB came up a few\ntimes, but I think request/response is more approriate. You can return\nmultiple responses to a query. Maybe PUB/SUB would be useful for mobile\ndevices.</span></p>\n<p><span>We talked about access\ncontrol, getting quite fine-grained (guests in your house not being able\nto turn your themostat on). I think some of this should be in the pervue\nof the naming system (e.g. can you address this device), but some should\nbe left to the other policies and mechanisms otherwise it gets way too\ncomplicated too quickly (expressing access control over functionality of\na device). DNS update keys could be one way to authenticate access\ncontrol.</span></p>\n<p><span>Talked about iPhone\nLIDAR’s. How you could scan to create zonefiles (automatically\n‘statically survey’), translating 3d space into a 1d network\ntopology.</span></p>\n<p><span>Magnus is thinking about\nthe load placed on these services especially for geodetic\nresolution.</span></p>\n<p><span>We think the local\nlow-latency resolution might be the best use case for the SIGCOMM\npaper.</span></p>\n<p><span>On EIO Anil said it’s not\nworth making this a unikernel, just to use the Mirage\nlibraries.</span></p>\n<p><span>Anil said Magnus might\nwant to try deploying AEON as a next step.</span></p>\n<h2>Shark</h2>\n<p><span>I believe the plan is to get a prototype\ntogether in time for SOSP.</span></p>\n<h2>SRG Seminars</h2>\n<p><span>Organised SRG seminars for Noa\nZilberman’s students to give practice talks. Advertised the Xen SRG\nseminar.</span></p>\n<h2>Website</h2>\n<p><span>Added disseration(s) to papers\nsection.</span></p>\n<p><span>My patchwork of makefiles, bash scripts,\npandoc invocations, and more is starting to get a bit unwieldy. A good\nproject for a rainy day would be to adapt Patick’s OCaml Static Site\nGenerator.</span></p>\n<h2>Matrix\nSynapse authentication with Raven Shibboleth</h2>\n<p><span>Got a\ngood resource at <a href=\"https://nickx.hu/posts/2020-07-13-matrix-synapse-shibboleth-saml.html\">https://nickx.hu/posts/2020-07-13-matrix-synapse-shibboleth-saml.html</a>\nSpun up an experiment on a Matrix server I had lying around on an Eilean\ntest instance.</span></p>\n<h2>Teaching</h2>\n<ul>\n<li>TA’d the last L50 lab.</li>\n<li>Marked and supervised Distributed & Current Systems for\nPembroke.</li>\n</ul>",
9 "content_type": "html",
10 "categories": [],
11 "source": "https://ryan.freumh.org/atom.xml"
12}