Hacker Public Radio

Overview I live on my own, but I cook for members of my family from time to time. Each week we all get together and cook dinner for Wednesday and Thursday. I usually do the cooking but we are starting to share these duties for certain meals. In 2019 I thought it would be useful if I had some sort of random chooser to decide what next week’s meal was going to be. I wrote a Bash script called choose_meal, using a simple CSV file of meal names and the date last eaten to avoid choosing the same one too often. The shortcomings of this approach soon became apparent! It wasn’t long before choose_meal was rewritten in Perl. This time I decided to use a database, and chose SQLite to create it. My database contained just two tables, one for the meals themselves (called slightly confusingly 'meal_history'), and another for a record of the choices made (called 'meal_log') – the ability to produce historical reports seemed like a desirable feature! In 2019 the design of this system was very specific to our needs: one choice per week on a Wednesday. It was not something that could be used by anyone else – which seemed like a bad idea. In late 2020 and early 2021 the system was redesigned, as will be discussed in the detailed notes. In May 2021 a more general design was added to the public GitLab repository and the preparation of this show was begun. I had never intended this system to hold recipes. This was partly because I have built a collection of recipes I have constructed from various sources and amended as I have made them. I print these and keep them in a ring-binder for reference as I cook. In some cases the meals described in the database are multi-component ones (such as the dishes that make up a curry for example), so it doesn’t seem appropriate to hold these here. I might rethink this in the future however. Long notes Follow this link to read the detailed notes associated with this episode. Links Perl Perl 5: (currently v5.35.0) Raku: originally called Perl 6, a totally different scripting language. Perl 7: updates to Perl 5, removal of some historical stuff SQLite: a C-language library that implements a small, fast, self-contained, high-reliability, full-featured, SQL database engine. GitLab repository for Weekly Menus

In this episode operat0r builds a cat feeder based on thingiverse.com

This episode is a response to hpr3367 by Andrew Conway and Dave Morriss. One of the topics they brought up was the thorny issue of page numbers in e-books. Most of the time you don't need to worry about page numbers in ebooks, if you're reading fiction for example. The whole point of an ebook is that the texts can reflow to fit the page no matter what size the screen is or what font-size you've chosen. This is a major accessibility feature of all e-book formats. One reason you might want to specify actual page numbers, though, is if you're dealing with a technical or academic book, and you need to be able to refer to specific passages in the book by page number, as you are expected to do in academic research. Or, as Andrew and Dave were discussing, you might need to create an index in your ebook that would send your readers back to specific pages like in a paper book. I've thought about this before but never really gotten into the weeds and figured out how to make it happen. In fact, when I was creating the new digital editions of the Counterpoint textbooks like I discussed in hpr1512, I actually took the trouble to put page number anchors through the entire thing, so that at a future date I would be able to enable real page numbers. This was a key part of the source file's infrastructure, which helped me quickly find the passages I was working on in my huge HTML file. Those anchors are not quite in the correct format for EPUB, but they are consistent and I will easily be able to write a script to fix them. I haven't done that yet, but now that I figured out how to do it on some smaller examples, this is on my to-do list. Anyway while I was listening to Dave and Andrew talk about this, I thought I remembered reading somewhere that in the newest ePub specification, EPUB 3, there was support for publisher's page numbers to deal with precisely this issue. Their discussion prompted me to see if I could make it work. I'm happy to report success, although with some qualifications, which I will get into. Converting to EPUB 3 The first thing to do is to upgrade your ebook from EPUB2 to EPUB3. There are a couple of ways to do this. The way I did it was to use the ebook editor in a recent version of Calibre. When you open up the EPUB for editing, go to the Tools menu and choose Upgrade book internals. This will create the new navigation file nav.xhtml to replace the old toc.ncx file. You'll need to edit this new file later to enable the page numbers. Insert page anchors Next you need to put your page anchors in there. This could be very tedious if you haven't done any preparatory work, such as putting visible page numbers in plain sight in square brackets [21] the way I did for a couple of ebooks. It wasn't very elegant, but at least it was easy to find where the page breaks were. I have a Blather voice command that triggers a python script to create these things. Here's an example of page number anchor, which goes in the main text of the book wherever you want to insert a page number. This will not be visible to the reader inline. This is for page 57: <span epub:type="pagebreak" id="page57" title="57"></span> Page List in Navigation File Finally you need to put a page list in the new navigation file. This is simply an ordered list with hyperlinks to every page anchor that you put in your ebook. This will not be visible to the reader, but it's critical to making everything work. Here's a minimal example from my first attempt. This only covers Pages 122 to 126. This is the kind of page numbering you might need if you created an ebook from a five-pa

Blog post: https://claudiomiranda.wordpress.com/2020/01/04/my-geeeky-experiment-part-1/ Asus Eee PC 901 info via Wikipedia (first paragraph in link): https://en.wikipedia.org/wiki/Asus_Eee_PC#Other_Eee_90x_models OpenBSD Web Site: https://www.openbsd.org/ Duplicating installed packages on another machine, via the OpenBSD FAQ: https://www.openbsd.org/faq/faq15.html#PkgDup Video of Eee PC 901 running OpenBSD 6.6: https://imgur.com/KHJj8lK Mastodon post from @solene@bsd.network about the Firefox port on OpenBSD: https://bsd.network/@solene/106481939809930365 CPU on my Dell Latitude E6410 via sysctl: hw.model=Intel(R) Core(TM) i5 CPU M 520 @ 2.40GHz

I had a spare moment and found this show on my hard drive. Cheers MrX YouTube video I found that explains how to fix a rattling heat shield for free Picture of the fix!

Apologies for the heavy breathing. I chose (unwisely, in terms of audio) to walk up a steep hill.

Federated social media can open up some wonderful possibilities to reimagine some of the social apps we already use and find ways to do them better. In this episode I want to highlight a talk that aimed at such a reimagining involving an app I already use, Goodreads. My main use of it is to manage my book library, but it also involves a social aspect where you can be friends with people and share book reviews, recommendations, and so on. So seeing how this can be done differently with ActivityPub was very interesting to me. Links: https://conf.tube/videos/watch/c42604a8-d71d-4bd0-8081-d2c77210f206 https://www.goodreads.com/ https://www.palain.com/im-learning-spanish/ https://learnawesome.org/ https://github.com/learn-awesome https://www.zwilnik.com/better-social-media/activitypub-conference-2020/activitypub-conference-2020-nilesh-trivedi/

In this fourth part of our three-part miniseries on Deep and Machine Learning our two heroes shed some light on a DL architecture called Generative Pre-Trained Transformer (GPT), a pretty sophistic piece of software that fools most humans when it comes to authoring text (ideal for budding writers with a block in place). Other topics of discussion includes OpenAI (the company behind this framework), Elon Musk, Bitcoin, Microsoft and if the GPT can actually pass the Turing test. All will be revealed - don't miss this episode! Links: OpenAI: https://openai.com GPT: https://openai.com/projects The Turing Test: https://en.wikipedia.org/wiki/Turing_test GPT-2 source code: https://github.com/openai/gpt-2 GPT meta-progamming: https://www.lesswrong.com/posts/zZLe74DvypRAf7DEQ/meta-programming-gpt-a-route-to-superintelligence GPT-3 interview: https://www.youtube.com/watch?v=PqbB07n_uQ4 DSDS: https://en.wikipedia.org/wiki/Deutschland_sucht_den_Superstar GPT-3 sample 1: https://linuxinlaws.eu/files/padawans.txt GPT-3 sample 2: https://linuxinlaws.eu/files/HGttG.txt

Starlink website: https://www.starlink.com/ Starlink from Wikipedia: https://en.wikipedia.org/wiki/Starlink Starlink is a satellite internet constellation being constructed by SpaceX providing satellite Internet access. The constellation will consist of thousands of mass-produced small satellites in low Earth orbit (LEO), which communicate with designated ground transceivers. The SpaceX satellite development facility in Redmond, Washington houses the Starlink research, development, manufacturing, and orbit control teams. The cost of the decade-long project to design, build, and deploy the constellation was estimated by SpaceX in May 2018 to be at least US$10 billion. Product development began in 2015. Two prototype test-flight satellites were launched in February 2018. Additional test satellites and 60 operational satellites were deployed in May 2019. SpaceX launches up to 60 satellites at a time, aiming to deploy 1,584 of the 260 kg (570 lb) spacecraft to provide near-global service by late 2021 or 2022. SpaceX started a private beta service in the Northern United States in August 2020 and a public beta in October 2020, service beginning at high latitudes between 44° and 52° North. On 15 October 2019, the United States Federal Communications Commission (FCC) submitted filings to the International Telecommunication Union (ITU) on SpaceX's behalf to arrange spectrum for 30,000 additional Starlink satellites to supplement the 12,000 Starlink satellites already approved by the FCC. Astronomers have raised concerns about the constellations’ effect on ground-based astronomy and how the satellites will add to an already jammed orbital environment. In response, SpaceX has implemented several upgrades to Starlink satellites aimed at reducing their brightness during operation. The satellites are equipped with krypton-fueled Hall thrusters which allow them to de-orbit at the end of their life. Additionally, the satellites are designed to autonomously avoid collisions based on uplinked tracking data.

I've made a special show covering the problem with chromebook support. I cover the latest AntiX release. I cover the latest NomadBSD release and I cover GUIX 1.3 and what I ended up using GUIX for.

Hosts Andrew Conway (mcnalu) Dave Morriss Handling incoming HPR shows Dave, with his Janitor hat on Processes incoming notes, to generate HTML Plain text is turned to Markdown. Otherwise various Markdown flavours are acceptable, or an HTML5 fragment If images are included the Markdown can refer to them with URLs such as: http://hackerpublicradio.org/eps/hprNNNN/name.jpg The hprNNNN element is a directory using the number you chose for your show. In the directory will be the images you sent. Other assets will go there too so you can refer to these in your notes as well. If you’re sending plain text, then markers such as >> Picture name.jpg here << will let me make the appropriate Markdown. The directory also needs an index.html file, but my scripts will generate this if needed. I intend to document this soon. How Dave makes HPR shownotes Uses scripts to manage show note generation Uses Markdown when writing the notes Scripts generate note templates and a Makefile Automation using make Uses Template Toolkit features inside notes to allow extra features Notes go through a pre-processor (a script with access to the show metadata) The end product is Markdown which is processed with Pandoc Template Toolkit (TT2) A Perl (and Python) tool kit for making templates Good for generating HTML, but can generate any text TT2 macros can be used to display scripts in shows relating to the subject, and to run them and capture the output. This makes it certain that the script on display really generated the output shown! The 'Falkon' browser (originally called QupZilla) is great for monitoring notes since it updates when the HTML changes. Dave confused it with Pale Moon (a fork of Firefox/Mozilla) when talking about tab grouping features Should this (personalised) bundle of software be released to the world? Probably yes, since the thoughts in it may be useful even if the code is not. Consolidating show notes into a book We had discussed the following topic a little in other contexts, and offline, but didn’t really look at it in this show. In brief, and for the record, the plans are: There was a series on 'sed', the stream editor from the GNU project. It was called “Introduction to sed”, and consisted of 5 parts, which ran through 2016. Each episode had short and long notes as well as several examples. Work has begun on consolidating all of the long notes into a single document which will be released on the HPR site, in HTML and PDF formats. Perhaps ePub will be included if feasible. It’s seen as critical that an index be provided so that topics can be found easily. At the moment this is simplest to achieve with the PDF version, using Andrew’s index generator as discussed in part 1 of this pair of shows. Digression about experiences in UK Higher Education Change of funding through the 1980’s to the 2000’s, particularly in IT Movin

On-board diagnostics (OBD) From Wikipedia, the free encyclopedia On-board diagnostics (OBD) is an automotive term referring to a vehicle's self-diagnostic and reporting capability. OBD systems give the vehicle owner or repair technician access to the status of the various vehicle sub-systems. The amount of diagnostic information available via OBD has varied widely since its introduction in the early 1980s versions of on-board vehicle computers. Early versions of OBD would simply illuminate a malfunction indicator light or "idiot light" if a problem was detected but would not provide any information as to the nature of the problem. Modern OBD implementations use a standardized digital communications port to provide real-time data in addition to a standardized series of diagnostic trouble codes, or DTCs, which allow a person to rapidly identify and remedy malfunctions within the vehicle. Torque Pro (OBD 2 and Car) See what your car is doing in realtime, get OBD fault codes, car performance, sensor data and more! Torque is a vehicle / car performance / diagnostics tool and scanner that uses an OBD II Bluetooth adapter to connect to your OBD2 engine management / ECU Forum Thread Search on NicoClub.com for Infinity Manuals

The upgrades made to the machine were a 500 GB ssd, 16 GB of ram, and went from El Capitan to Catalina.

You can purchase Starfinder directly from Paizo, and you can usually find Starfinder Society play at your friendly local game store. The alternate starship combat rules mentioned in this episode are available from drivethrurpg.com (warning: this is an affiliate link, but any purchases made through it provide Klaatu with store credit so he can buy more RPG books for future HPR Club reviews).

Welcome to the 9th Annual Hacker Public Radio New Years Eve Show Hunting food cpap machines Music Audio books podcasts Archive.org is great (support if possible please) Moss has a fun 2020 Politics Health care grey hat