Roel Willems

Articles

    Obsidian with automatic GitHub syncing

    Some updates running Obsidian with automatic GitHub syncing.

    I ditched my homegrown solution1 for the Obsidian Git plugin. Both approaches are based on this article by Bryan Jenks.

    Using the plugin I got everything running on all my systems in a few minutes spend on each one. On Windows it took a little more time due to user error. Reminder for my future-self (and other Cmder users 2), don’t forget to also install regular Git for Windows.

    For backup purposes, I’ve cloned the repository to my NAS (Linux) and updating it using an hourly cronjob. A bit unnecessary as the files are already stored on GitHub, 2x OneDrive (work/private), and iCloud. So this exertion was purely in the name of science ;-)

    I ran into one issue when working on the same document on two systems at the same moment. Not a regular use-case for me, but will need to look into how this solution handles potential file conflicts. But other than that the process is running very smoothly. Besides purely syncing Git is also bringing version control. This is a huge plus to me for my every growing notes library.

    On iOS I have WorkingCopy and iA Writer working the Obsidian repository. iA Writer and WorkingCopy integrate well and this already is a solid solution. iA Writer can access the files in WorkingCopy via iOS’ Open in Place feature. No need to transfer files between both apps. And the built-in diff tool is great. One (minor) downside is that WorkingCopy doesn’t offer automatic syncing. You need to push/pull changes from within the app. GitJournal should offer automatic syncing but I’m running into problems. I’m only seeing one file from the repro, but this could well be due to my setup. I do have an exotic file naming structure to keep things organized inside of Obsidian. But until I solve this issue I’m more than happy using iA Writer and WorkingCopy.

    1. Albeit the plugin is running cronjobs it’s much easier to maintain as settings are also synced between devices (so changing the commit interval to more/fewer minutes will be synced to all devices running Obsidian) ↩︎

    2. If you’re like me and missing a proper console in Windows Cmder is your savior, just don’t forget that Git is not installed system-wide ↩︎

    Solid

    For some time I’ve been following the Solid project by Tim Berners-Lee. This November his startup launched the Solid privacy platform. Enabling organizations to build applications on the principle that the user owns and controls their data.

    The core idea is that the data isn’t owned by an organization but by users themselves. Users have their data stored in “Pods” (Personal Online Data Stores). Applications can request access to specific data within a Pod. This concept is a huge shift away from how most of the web has evolved. As most of the web has become one big data collection machine where organizations “own” data.

    The project is open source but since November there is an enterprise version available for use by large organizations.

    The whole project is super interesting to follow. Time will tell if it will succeed. But if it will it’s a major step for online data privacy. Regardless, I am intrigued by the concept.

    And if someone could pull off reinventing the Web, who better than the inventor himself?

    82.3 Kb

    Fiddled a bit more with optimization of the website. As the website is hosted by Micro.blog I a little limited on what is possible. But still the result is quite good:

    1. Scaled and compressed all images, resizing and reuploading isn’t the most inspiering to do, but still it counted for 90% of the bloat in size (and my website isn’t even very image heavy)
    2. Removed the Gravatar integration with a static favicon and header image. Gravatar isn’t privacy-friendly at all and the standard integration within Micro.blog is quite unneccesary. So win for privacy and reduced one request
    3. Stopped using the custom CSS function within Micro.blog and just added all to the the custom theme CSS, saving one request

    Overall some minor changes which resulted in reducing my homepage from 3,3Mb to 82.3 Kb uncompressed1.

    1. click on the Waterfall tab to check the uncompressed size of the site ↩︎

    How to batch export from Ulysses to plain Markdown

    This is a small guide to help moving files from Ulysses native file format (.ulysses files) to plain Markdown. Having your writing in plain Markdown is necessary when you want to move to a different writing app or just for backup purposes1.

    This guide works for files without images and Ulysses specific bells and whistles

    1. Create a temporary folder on your Mac for the exported Markdown files e.g. “Ulysses export”;
    2. Open Ulysses on your Mac and add the “Ulysses export” folder to the library (bottom of the left sidebar) → Add External Folder;
    3. Select the files you would like to export from Ulysses to standard Markdown;
    4. Move the files, within Ulysses, to the “Ulysses export” folder;
    5. Open the “Ulysses export” folder in the Finder and move/import to where ever you would like.

    1. Just trying to backup the .ulysses files isn’t very useful as the .ulysses files have a hash for a name and contain two files (Content.xml and Text.txt). To get your writing out of the files you have to open all of them manually. If you use an iCloud-stored library the files are hardly reachable at all. ↩︎

    The beauty within

    I like words. I like plain and simple thoughts and stories written down. For as long as the internet existed it has been a great place to read and write.

    But while the web has been ever increasingly gaining advancements in design and complexity, I value… no, I crave, simplicity.

    Almost all of the web today is geared toward gaining attention and eyeballs. Every year, with every advancement, I feel we distance ourselves more and more from what is really valuable for these mere eyeballs (also known as humans).

    At most websites, you have to click through a jungle of stuff which only distracts from their main objective (just to name a few: Medium, Youtube and almost all news outlets on the web). And although I fully understand the business objectives driving these decisions, I feel there is a quality to simplicity and not being over-designed. In my opinion, it adds character.

    Most of the time your message will not be more clearly expressed when surrounded by fancy stuff let alone a bombardment of ads and messages.

    Plain words on a plain website are in my opinion just like a vinyl record. And just like for people enjoy listening to a vinyl record, it’s not about audio perfection or having all bells and whistles. It’s completely the opposite.

    To take the music analogy a step future. A super clean production of a mediocre song can become a hit (with the right number of infuencers and paid-airtime), but be all forgotten about in a year or so. “Perfection” gets boring most often. The song that an artist (possibly with less vocal quality or marketing budget) just had to put out there has the potential to become legendary. Because it has meaning and its roughness only adds to its quality.

    With words, like music, it’s also not about perfection but about meaning.

    When writing we should feel like an artist making a song. Making a song, not because it will show our vocal capabilities best, but because it’s a song that has to be made, has to be heard.

    The same goes for words, the beauty is in meaning. Just feel free to write.

    This post is as much reminding myself that my writing doesn’t have to be perfect as supporting everyone reading this to just write down what you feel you have to say!

    ∞ Sorry Mark Zuckerberg, Facebook isn’t a “positive force”

    In a Tuesday interview with CNN, Mark Zuckerberg defended Facebook’s handling of recent scandals and his own leadership of the company. He flatly rejected calls to give up his position as the chairman of Facebook’s board and said he had no plans to fire his top deputy, Sheryl Sandberg.

    People like to mock Google’s “don’t be evil” slogan, but the company really has made difficult calls in the past—like pulling out of China over pressure to build a censored search engine.

    By contrast, Facebook only seems to care about promoting the continued growth of Facebook.

    arstechnica.com

    The inability of Facebook to change after an increasing number of incedents and scandals is shocking. Although Google shouldn’t be presented as a poster boy, the differences does show. Talk the talk … walk the walk

    ∞ Computers have learned to make us jump through hoops

    The other day I had to log in to a service I hadn’t used before. Since I was a new user, the website decided that it needed to check that I wasn’t a robot and so set me a Captcha (Completely Automated Public Turing test to tell Computers and Humans Apart). This is a challenge-response test to enable a computer to determine whether the user is a person rather than a machine.

    But note also the delicious additional irony that the Captcha is described as an “automated Turing test”. The Turing test was conceived, you may recall, as a way of enabling humans to determine whether a machine could respond in such a way that one couldn’t tell whether it was a human or a robot. So we have wandered into a topsy-turvy world in which machines make us jump through hoops to prove that we are humans!

    Although it’s clearly human ingenuity transforming us to unpaid “click workers” training A.I, but regardless it’s still a very interesting case.

    It makes you wonder what could happen if an A.I. would learn and deploy this on its own. And although I don’t think this would happen in the very near future, we should think about unforeseen consequences (as almost all technology and legislations bring in there wake).

    P.S. Personally I find the types of CAPTCHAs presented to us very interesting. It gives insight in the current focus of developments in machine learning and A.I. Will be very interesting to see what will be next after road signs and store fronts.

    Computers have learned to make us jump through hoops - The Guardian

    ∞ To www or not to www

    For 20 years or so, there has been the debate over whether you should use www or not in your web site’s canonical hostname. So should you use www or not?

    If you’re concerned about the security of whatever you have on “example.com”, make sure you slap a “www”- in front of it. If that doesn’t help you choosing whether you should use www or not, I’m not sure what will.

    To www or not to www – Should you use www or not in your domain? by Bjørn Johansen

    Great read on the pro’s (prettier and easier) vs cons (security regarding cookies) of using www. For my own domain I don’t use cookies (besides one Cloudflare load balancing cookie), so using the origin domain without www. is perfectly safe. But still I wonder, should I switch just to support proper and safe web techniques?

    Micro guide: Getting your hosted Microblog IPv6 ready

    This is a small guide (hence the category “micro guide”) to enable IPv61 for your hosted microblog2.

    disclaimer: please make sure that you know what you are doing with your DNS settings. Although it isn’t really difficult or complex you can really mess up access to your website if the settings are incorrect. Please make note of the original settings before changing anything.

    Setting up IPv6

    To start, you should go to your domain registrar or system to manage DNS (personally I use Cloudflare and have written another micro guide with recommendations on how-to enable Cloudflare for your microblog).

    Now you can create an AAAA record (a record is an instruction for the browser to which server to connect and the AAAA version is used for IPv6 addresses).

    Make sure that the AAAA record is pointing to:

    2600:3c00:1::68c8:16d6

    This is the IPv6 address for hosted microblogs. Leave the A record (address for IPv4) and other records like CNAME unaltered. Based on you DNS provider it should look somewhat like:

    example of DNS records

    Save the settings and after a few minutes you can check if your website has support for IPv6.

    1. What IPv6 is and why it is beneficial will get technical very quickly, if your interested you could start reading Wikipedia or look in to this tutorial ↩︎

    2. hosted with micro.blog ↩︎

    Creative commons licence

    Redid the footer of my website and am quite happy with the result.

    Also moved to a creative commons licence for my content (the “Attribution-ShareAlike 4.0 International (CC BY-SA 4.0)” to be precise) which gives everyone the freedom to remix, reuse and build upon the content shared (with appropriate credit and indication if changes were made).1

    Although I don’t expect many people to use my content, consider it my contributions to a free and open internet.

    Image of the footer

    1. Although I believe it would be a good thing for the internet and it’s users if more people would move to more open licences, please do make sure to really understand the implications to determine if a licence is right for you. Read all about it  ↩︎

    Facebook accused of extremists to one another through ‘suggested friends’ feature (The Telegraph)

    Using sophisticated algorithms, Facebook is designed to connect people who share common interests.

    “Facebook, in their desire to connect as many people as possible have inadvertently created a system which helps connect extremists and terrorists.”

    The Telegraph article1.

    There is much to say about the role and responsibility of Facebook. But I couldn’t think of a comment as brilliant as the one by @ben ->.

    1. Picked a few specific quotes from the article to show the gist of the article storyline. ↩︎

    How very social

    Samsung Smart TV. During the first minute after power-on, the TV talks to Google Play, Double Click, Netflix, FandangoNOW, Spotify, CBS, MSNBC, NFL, Deezer, and Facebook­ even though we did not sign in or create accounts with any of them.

    Now you know why your TV keeps bugging you about GDPR consent

    ∞

    Who’s next?

    Step 1: get dating apps to build themselves on your platform’s data Step 2: cut them off from your platform’s data with no warning Step 3: build a competitor to dating apps with the data you are now keeping to yourself

    Seems kinda antitrust-ish…

    John Kneeland via Daring fireball

    Fool me once, shame on you(r news business); fool me twice, shame on me(re dating apps).

    So this begs the question; who’s business will be next?

    Disruptive vs Sustaining innovation and the Tesla case

    I always enjoy Ben en James talking business strategy in their Exponent podcast (Ben Thompson from Stratechery with co-host James Allworth). Their last podcast about Zillow (#149 - Zillow and Sustaining Aggregation) has a specific part really standing out. At 14:14 minutes in, James explains the difference between sustaining and disrupting innovation. And does so in very clear language.

    An excerpt from the podcast (emphasis is mine):

    James: It think what would be interesting here, just to drive home the point for people. Is to think about a technology right now and how it could be applied as a sustaining innovation and a disruptive innovation. It’s been a less popular topic in the more recent Exponents but obviously, if you go back far enough, something we would talk a whole lot about is self-driving cars and the impact that’s going to have on society.

    Now the technology is going to be the same if you are an existing manufacturer or something like Google, and maybe Google drops it into an existing manufacturer. This is a technology in and of itself business model agnostic, but you think about the incentives of different players and how they want to apply that technology. You give this to someone like BMW, an existing car manufacturer. And their instinct is to think; how can I take this technology and sell the vehicle that I have for more money as a result. And you get cool features like self-driving on Tesla or adaptive cruise control on BMW. ……… But it’s fundamentally a sustaining innovation. This is exactly why self-driving vehicle; people love to talk about Tesla’s disruptive, and maybe there is something in terms of the way they are thinking about building out a network and so on, this is true. But in terms of the actual vehicles, this is why Tesla vehicles are fundamentally not a disruptive product. They are using a technology to sell a better product to the existing people for more money. It is fundamentally the same business model. But take that to a very different context which is; something like Uber or Lift, doing ride sharing. And their primary cost of business is paying for the drivers in the vehicles. And that is disruptive to the existing car manufacturers. Uber and lift want to take the self-driving technology and they are not going to take that and trying to sell the vehicles for more money as a result of it. They are going to take the technology and use it to drive down the cost basis of offering ride sharing. The same technology applied in fundamentally different ways, one is sustaining, one is disruptive.

    Highly recommend listing if you care about online/tech business strategy. James (and Ben) make a very compelling case that most (new) tech companies still running sustaining business models and most of them are in fact are no disrupters at all. Even if popular media will consistently describe them as “disruptive”.

    Tesla is also a great explanatory case for disruption vs sustaining business models. The technology deployed by Tesla is very new (and Eldon’s vision maybe futuristic) but in its basic form, Tesla tries to build (better) cars and sell them for profit.

    My personal take out is: technology in and on itself isn’t disruptive, the way technology is applied and to be more precise; the business model around it, make it disruptive for the status quo. Based on the business model behind technology/innovation can make it sustaining or disruptive.

    I also like the Tesla example. It shows the difficulty faced when entering a market with new technology/innovation deployed in a sustainable manner. Tesla isn’t a tech company or disrupting the market. They are a car manufacturer in competition with other car manufacturers.

    And as much as I like Tesla for what they did for electric driving1, I do see many challenges in keeping ahead of the competition while solving the scalability issues they are facing.

    Entering a market with a sustainable innovation is a rat race. And this is no different for Tesla. So how much time does it take Tesla to get their production efficiency on par with the traditional car manufactures? And how long will it take those manufactures to get to the same technology level?

    Keeping in mind that traditional manufacturers very likely will be able to buy their way into newer and better technology at an increasing pace. Taking into account that there are a growing number of potential suppliers for self-driving tech (who haven’t shown interest in mass producing cars).

    With every minute passing it will be easier for those traditional manufacturers to add more advanced technology into the already established line up of cars (which already are built at a massive scale).

    So the biggest problem faced by Tesla isn’t how to build cars at scale. Given enough time I’m sure they will figure this out, other companies already have done so.

    But the challenge for Tesla is keeping their edge over the competition. Keep outpacing other car manufacturers on a technology front long enough. Making the most innovative cars and reaching the hearts&minds (and share of wallets) from their potential customer. Which in turn will buy them (much needed) time to sort out car production at scale.

    And if Tesla succeeds I still wonder what will differentiate Tesla when the rest of the market obtains the same level of technology.

    You are basically betting on your ability to innovate at a faster pace then the incumbents can play catch up when competing on technology in a sustaining business model.

    Which is all well and good, but clearly very different from disruption. Self-driving technology will make ride-sharing services cheap, practical and maybe even irresistible to consumers that they (potentially) will just stop buying cars altogether. Which would be very disruptive to the current market.

    If this will happen at all, no one can tell. But I know that although innovative we shouldn’t be calling Tesla disruptive anymore.

    1. how, in my opinion, Tesla altered the course of electric cars will be a topic for a different post ↩︎

    On turning off unnecessary notifications

    • turn off all lock screen notifications (this is better for privacy anyway)
    • turn off vibrate and sound notifications, I leave badges for important things like messaging apps
    • turn off notifications for apps that really don’t need it

    Nathan Toups makes a case for a dumber phone with great recommendations overall.

    One thing I would add from personal experience1 is to make very conscious use of the silence/mute switch on phone.

    No sound/viberation 99% of the time, but if I need and enable sound I also have the super annoying keyboard sounds enabled by default (so I don’t forget to revert to silent ASAP).

    Via Ben Brooks

    1. Personally I turned off sounds and vibration on my phone when I started wearing my Apple Watch. Now I’m back to wearing a classic watch as wrist based notifications over time became an unwanted distraction. ↩︎

    This will be interesting..

    WhatsApp will require users in Europe to be at least 16 years old

    I can imagine many reasons why the age limit would need to be increased under more strict European privacy law, none of them any good for something so fundamental as a messaging app.

    just a recommendation: Signal or Keybase

    ∞ Collected Goodness Good material. Good art. Good living.

    Collected Goodness.com by Drew Coffman

    Very interesting and inspiring stuff, will need to look in to it more but the first recommended reads/listens seem very worthwhile.

    I also like the format a lot, it strikes a balance between linking to interesting content supplemented with quotes supported by the voice-over by Drew. Very well done altogether.

    Found via: @patrickrhone (thanks!)

    Scaling is hard

    Most of the design tolerances of the Model 3 are already better than any other car in the world. Soon, they will all be better. This is not enough. We will keep going until the Model 3 build precision is a factor of ten better than any other car in the world. I am not kidding.

    Our car needs to be designed and built with such accuracy and precision that, if an owner measures dimensions, panel gaps and flushness, and their measurements don’t match the Model 3 specs, it just means that their measuring tape is wrong.

    I’m quoting a quote of Eldon Musk by John Gruber.

    But after reading the article and memo I can’t suppress the feeling that people waiting in line (after forking over cash to enter the cue) would be happy with clear delivery times1 and quite satisfied with tolerances on par with current electric cars that are already build at scale2.

    In my opinion the real challenge for Tesla (or any car manufacture for that matter) isn’t reaching near perfect build precision but getting to a good enough level at scale.

    Tesla taking on the challenge to build 350.000+ model 3’s currently in backorder3 is a marvel on it’s own.

    This will play out to be the greatest achievement in automotive history or Tesla being overtaken by the traditional manufactures moving to full electric.

    1. Tesla halts Model 3 production second time this year -telegraph.co.uk ↩︎

    2. In Europe you could have a full electric VW e-Golf on your driveway, two months after you place an order. ↩︎

    3. Based on quick calculation on numbers provided by Bloomberg ↩︎

    ∞ It’s all about perspective

    Mobile apps are still often lacking in efficiency, Apple’s Smart Keyboard is mediocre and reaching up to touch the screen is far less comfortable and efficient than using a mouse.

    engadget.com

    The whole article feels a bit shortsighted and only focussing on comparing the iPad to a regular PC. All of the statements could be made in favour of the iPad as a PC replacement. It’s all about perspective I guess.

    Micro guide: Additional steps and recommendations when using Cloudflare for your DNS

    Cloudflare users wanting to use a custom domain name for their hosted Micro.blog will be greeted by a “custom domain isn't pointing to the micro.blog" error message when trying to save the custom domain.

    Luckily there is a work around:

    How to get your custom domain working with Cloudflare

    1. First follow the steps as outlined above and add the proper “A” and/or “CNAME” records to the Cloudflare DNS panel;
    2. Before adding the custom domain to your Micro.blog account make sure that the “DNS and HTTP proxy (CDN)” (orange cloud-icon) is disabled and set to bypass (grey cloud icon with the hover text “DNS only”);
    3. After updating your DNS records and Cloudflare set to bypass for the “A” record and/or “CNAME” go to “Account” on Micro.blog and fill hostname (e.g. yourdomain.com) that you’d like to use for your Micro.blog;
    4. Micro.blog should save and enable to custom domain without issues. If you get a message “custom domain isn’t pointing to the micro.blog" make sure that you have set Cloudflare on bypass and wait for a few more minutes (the DNS change can take some time to get in effect);
    5. After successful saving the custom domain you can re-enable Cloudflare functionality on your domain/Micro.blog by remove the bypass, make sure that the cloud-icon is set to orange (hover text should state: “DNS and HTTP proxy (CDN)”).

    So there you have it, you are able to use Cloudflare (DNS/CDN/security/caching/DDOS protection) with hosted Micro.blogs, with some additional steps to get the initial setup working.

    Some additional recommendations:

    Using Cloudflare to redirect www. to root domain

    If you would like to enable www. and root version of your custom domain it’s advisable to redirect one to the other (to make sure that there isn’t any duplicated content). Cloudflare can redirect the www. version to the root domain:

    1. Go to “Page rules” within your Cloudflare account;
    2. Create a page rule;
    3. Add <www.yourdomain.com>/* within the first field (“If the URL matches”);
    4. Under “Then the settings are:” select “Forwarding URL”, “ 301 - Permanent redirect”;
    5. Add: <yourdomain.com>/$1 to the destination URL field;
    6. Save and deploy the page rule for the redirect to take effect.

    Cloudflare, SSL and your Micro.blog

    You should not use the SSL functionality provided by Cloudflare (email help@micro.blog to enable SSL).

Older Posts →