How Reliable Is Feedly?

Shortly after the Ides of March last year1, and for reasons I've never been able to work out, Feedly stopped updating the RSS feed for this site. The XML file that is generated for the site has been consistently deemed valid by the W3C Validator and other 10C-based sites continue to see their syndicated content properly updated in Feedly. I know this because I subscribed to a number of them. Mine, when pulling from the XML version of the syndication feed, will not show updates.

But why?

The XML version of the RSS feed will work with a number of clients just fine, as I've tested the site with standalone clients as well as alternative online services, such as FeedWrangler and Fever2. Has Feedly simply stopped visiting the site and, therefore, does not know that the RSS feeds have been operating just fine aside from the one little blip that took place on April 10th?

According to the service logs, this isn't the case.

Stats Show Two

The Feedly crawler is stopping by this site almost every hour for the XML feed and every three hours for the JSON version, which the service has no trouble at all reading. The XML file is being properly returned, too, as the number of bytes sent is the very same number of bytes as the cached XML file (compressed as a gzip). Those subscriber numbers look a little off, though ….

Feedly Shows Nineteen

Seems one or all of the reported numbers are false. 1 subscriber? Just me?3 Or is it 2 subscribers? My mum and I, perhaps? Or is it 19? Perhaps some other number?

Unfortunately, this problem does not seem to be limited just to me. A now-retired UserVoice.com forum had hundreds of people asking the same question for various feeds that just stopped updating one day. The service seems to be aware of the problem, but is unwilling to do anything about it.

Feedly Support Services

So what is a person to do when their RSS feeds stop updating? Readers who notice will hopefully continue to visit the site directly or change readers. Writers who want to be read, however, will suffer if people using Feedly just thought that the site had gone silent, like tens of millions of blogs have over the years.

Personally, I've lost interest in trying to make things work with the service. My syndication feeds are valid and have been for well over 8 years. Rather than mess around with someone else's service, I'll do the next best thing and make my own RSS service. As one would expect, 10Cv5 already has this functionality built in. All that needs to be done is to build a simple client, write the documentation for the /reader API, and share it with the world.

When paid cloud services drop the ball, I tend to pick it up and take it somewhere else.


  1. The Ides of March takes place on the 15th.

  2. Fever is a self-hosted application, but is still a valid means of testing.

  3. I subscribe to my own feeds primarily to ensure the thing is going out and being read properly. Sometimes I even spot typos this way, then I can go back and fix them.

Genuinely Missed

Rachel Kroll asked an interesting question the other day:

For those people who have worked at multiple companies and thus have left a couple of jobs over the years, do you ever think about stuff you miss? These are the internal systems, build environments, tools, features, or other neato things that you had which made your job easier, better, faster, or maybe even possible… but which didn't exist on the outside world.

Thinking through all the places I've worked as a developer over the years, from an appliance repair shop to a printing shop1 to language schools2, plus a myriad of organisations I've worked with on a freelance basis, I can honestly answer "no" to the question as it's presented. The vast majority of the things I've worked with have either been publicly available3, supplied by me, or written by me … which is probably not something to be proud of. What I do miss about many of the places that I've worked at (or with) are the people.

If we're lucky, every place we've worked has had at least one person we've looked up to, learned from, or just plain enjoyed being around. Even as a freelancer, there are certain clients who are an absolute joy to work with. For me it's the people who give me a reason to get to work on time and give it my all, learning new and interesting things from them along the way.

Often times when I think it's time to move on to another type of company to take on a completely different set of challenges, I think about the people that I would be leaving behind and reconsider. Sure, there's email and social places available to maintain relationships, but it's not the same as when you're working together towards a common set of goals.

Internal systems, build environments, tools, features, and other neato things that are unique to an organisation are certainly nice to have, but they're not something I think I'd ever miss. Mind you, I've not worked for organisations that are even half as well known or respected as the ones Ms. Kroll has. Perhaps with more exposure to the various tools that companies use internally, I'd find a reason to miss something.


  1. Not sure "shop" is the best way to describe the company, given they employed 400 people and had customers all over North America.

  2. There are other places I've worked at, but these are the ones at each major geographic location I've lived where I've worked the longest.

  3. Available either commercially or free.

Nine Years

Nine years have passed since the Great Tōhoku Earthquake shook half of Japan and triggered a tsunami that wiped costal cities off the map while claiming the lives of thousands. In this time a great deal of rebuilding has taken place and people have moved on with their lives; some with far more success than others. Reiko, Nozomi, and I were quite fortunate that we were largely unaffected, though Nozomi did have some pretty serious eating problems until mid-way through the following year due to stress and anxiety. Every large truck driving by would result in our apartment shaking, which would freeze the puppy1 in her tracks. Getting her to eat was a trial of patience, though she did pull through. Moving back to the Tokai region of the country went a long way to helping all three of us calm down and relax.

Moving back is something that I am quite grateful for. Reiko was not happy very often while we lived in Kashiwa, as I was gone most of the day to work in central Tokyo while she was at home in a largely unfamiliar city. We moved to the Tokyo area because I had accepted a job at a tech startup. It paid quite well and there was so much that I learned while solving the problems of the day, but what value is this when your wife is unhappy? Sometimes I joke that the only reason we moved to the Tokyo area was because fate demanded we bring Nozomi into the family. Other times it doesn't seem like a joke at all.

The quake did leave some lasting marks, though. My mind continues to conjure up tremors that don't exist. Both Reiko and I are careful to keep our emergency preparedness packs ready and properly stocked with dry food, clothing, and various supplies. We also have several external batteries for our devices in the event we cannot charge them for extended periods. The car never has less than a quarter tank of gas, either. If something happens, we can leave the house and drive at least 100km to safety with just a moment's notice. This is something that I never would have done in Canada despite living through the Great Ice Storm of '98.

Yet, all in all, life has drastically improved in the nine years since. Reiko and Nozomi are both healthier. The boy has come along to provide a great deal of entropy to our lives. We no longer live with just a couple of months' worth of savings in the bank. We have a house in a safe neighbourhood that sits 120m above sea level and 30km from the nearest beach. The 3/11 Earthquake was a time of crisis, but we're in a better position as a result of getting through it. Again, I am grateful.

The gratitude I feel is not for the quake, of course, but for how Reiko and I responded, overcoming the challenges and accepting opportunities that were different from what we might have desired in order to regain some semblance of stability. When we had that platform by the middle of the following year, we worked towards the goals that have led us here to where we are today. It wasn't easy. It wasn't always positive. It was educational, though.


  1. She was actually a puppy back then.

Fewer Subscriptions

For a little over two years I've been using Setapp as a "developer's App Store", picking up applications that allow me to solve specific problems encountered both at the day job and in my personal projects. With it I have been able to try out a number of different tools to find the best one for the particular tasks at hand and, for $10 a month, it proved to be worth the subscription rate. That said, as my work has become more specific, there has been much less need to quickly find an application to solve the occasional "Problem X" as there was just a couple of years ago. For this reason, I pulled the plug on the subscription and instead bought the two applications that I use most often: TablePlus, which I am now using daily on account of all the PostgreSQL work, and ForkLift, which is used a couple of times a month to upload large sums of data to Amazon's S3 storage buckets.

This move is the continuation of a process that I seem to have subconsciously embarked upon, where subscriptions that no longer offer the value perceived in the past are cancelled. This month it's Setapp. Last month it was Netflix. The month before was Amazon Prime. None of these subscriptions are particularly expensive, but they do add up and it has me thinking about the business model of collecting relatively small fees for access to a walled garden of digital goods and services.

Over the last couple of years a lot of us have slowly accumulated a large number of subscriptions to all kinds of services. Sometimes this is to support the developer(s) of a project. Sometimes this is to access commercial materials. Sometimes this is for sheer convenience. These fees add up pretty quickly. Is every service worth the amount provided? It seems the only time I ask this question is after an email hits the inbox letting me know that I've been charged for another month or year of service. I'm all for supporting people in the quest to do good, though I sometimes wonder if there's a better way to use the money.

This reduction of subscriptions should be a good opportunity to examine the lot of them. By being a little more careful with what services remain active, it will be possible to invest a little more intelligently in independent projects as well as my own.

Still a Pretty Penny

Every day the morning news programs interrupt their soft journalism with semi-serious updates on anything and everything related to the coronavirus situation. There's always news of more patients, more quarantines, more market problems, and more challenges as people scramble to deal with an invisible enemy as best they can. One of the more common stories that these news programs like to focus on is how much trouble the tourism industry is in, as people choose to stay home rather than risk contracting the Chinese virus. Hotels are offering discount rates. Restaurants are offering free (non-alcoholic) drinks. Smaller companies are also offering discounts or special deals in an attempt to drum up enough revenue to float for a month or two while the global contagion runs its course. One group that is heavily dependent on people travelling long distances doesn't seem to have changed their prices, though: airlines.

Two years ago today I was flying from Tokyo to New York to attend a week of meetings at corporate HQ. Non-essential travel has been cancelled at the day job but, if I did have to buy a ticket to the other side of the planet, how much would it cost?

Tokyo (NRT) to New York (JFK)

Seems the price hasn't changed at all since the last time I went.

Five Things for March

This past week has shed a little bit of light on what it might feel like to be a pet, locked inside and unable to go outside or enjoy the freedoms so many of us take for granted. In a stunning turn of events, time has actually moved slower this week than at any point that I can recall in the last 40 months. This is most likely due to a drastic reduction in physical activity and the daily arguments with the boy around simple things like picking socks off the floor. That said, there have been some things of note over the past seven days that are worth a paragraph or two of time.

Star Trek: Picard Begins

Episodes six and seven have renewed my hope in the series after four and five went down poorly. Heck, I didn’t even finish episode five, choosing to end the show with ten minutes remaining because it was just … terribly predictable. Mind you, Star Trek has had ridiculously predictable stories since the late 80s, so it’s no surprise that every modern episode of the franchise follows similar patterns. By the end of episode 7, I pieced together what I see as being the outcome of the show. Only time will tell if the various hunches are correct.

Depression Returns

With the lack of exercise and fresh air, I’ve found myself sliding into the familiar cycle of self-loathing and equivocacy towards just about everything. A good amount of personal time has been dedicated to getting more of the 10C platform built out in an effort to get back into a productive cycle, but its not really helping dislodge the unyielding lack of motivation I feel nor the dark thoughts that crowd the mind’s eye. Hopefully there is a resolution to the Covid-19 situation in the near future so that I can get back into almost daily treks around the neighbourhood without the stresses and hassles that come with every exit from the house.

Intel (and AMD) Are to Blame

Over the last couple of weeks I’ve been working my way through a commercial Learning Management System in an effort to get it ready for use by the day job. The software was purchased through a vendor alongside a bunch of customizations and now the system has been handed over to me to coordinate and manage the future direction of the tool. One of the things that has become incredibly apparent from the source code of this very common learning system is that the developers have not had to think about resource consumption at any point in the last decade. The number of blatant inefficiencies boggles the mind and gives me more incentive every day to toss the expensive software into the garbage and write a couple of dozen python scripts that read from flat files to deliver the same experience with 1/1000th the processing power and 1/10000th the network usage. These numbers are probably grossly inaccurate but, given what I see in the source of the LMS, there is no way any developer contributed a line of code from a machine with less than 16GB RAM and a quad-core i7 (or equivalent) CPU. Schools running this system likely pay many times more for servers and bandwidth than they should, all because of inefficient and unnecessarily bloated code.

Then again, maybe I try too hard and focus on the wrong things. The number of people who say “Who cares? So long as it works!” is … disappointing.

Coffee Without Milk Isn’t Coffee

Thanks to China’s uncanny ability to export highly contagious viruses, Reiko has “forbidden” anyone in the house from going into crowded buildings, such as grocery and convenience stores. As a result, we’re ordering our food online and having the store deliver it to the house1. This means no quick runs for milk, which means using a powdered substitute or drinking the caffeinated beverage black. Of the two, a powdered substitute is preferable, but just barely. Coffee was meant to be had with a bit of milk. Anything else is just spoiling the beverage.

Polyphasic Breakdown

Back in December I started a polyphasic sleeping pattern that would see me get two (or three) power-naps during the day with a 4-hour block of rest at night. This allowed for 18 and 19-hour day lengths, making it possible to work on personal things, spend time with the family, and tend to the day job. This worked out relatively well until around the end of February when the old habit of falling asleep in the middle of the day when napping was not an option started up again. I’ve since has to re-adjust the body to seven hours of sleep rather than five, and this seems to be working a bit better. There’s still another year or so of the “special project” at the day job, which means a proper sleeping pattern is at least twelve months away. Hopefully the new schedule will survive that long.

There are a few other items I could have written here, but then this post would have become a “Nine Things” article, which would probably be a bit too long. One thing I can say, though, is that it’s good to hear my kid laugh at British comedy. This is something we both enjoy greatly.


  1. I know. I know. Believe me, I’ve argued the faults in the reasoning for hours to no avail.

Phantom Rumbles

Almost 9 years have passed since the Great Tohoku Earthquake that shook half the country's land and triggered a devastating tsunami that laid waste to entire coastal towns. Aftershocks were frequent and concerning, hitting the improbable number of 1,000 just thirty-three days after the event. People were wary of the ground beneath their feet and for good reason. Since 3/11 quake I have often tensed whenever the ground shook on account of a large truck going by or someone turning up a decent sound system nearby. Every time the floor moves, either in my mind or otherwise, I'll look for some indication that the ground really is in motion. This often means looking at a glass with liquid inside or a hanging ceiling light. If there is a real earthquake, these objects would show signs of its existence. Over the nine years since 2011, I've been right fewer than 5 times.

Last year I wrote about these phantom quakes and how I generally deal with them. Lately there's been a new twist to the imaginary seismic activity: phantom rumbles.

Before each earthquake is a low rumbling from the ground as the incredible energy contained in the earthquake's shockwaves ripple across the land. I've been hearing this sound once or twice every day for the last couple of years and, just like with the phantom quakes, I look for a glass or a ceiling light to verify the concern. Lately the sound has become a lot more frequent, though.

Every truck that drives by and some of the vans driven by neighbours all seem to carry the telltale sign of an impending quake. Try as I might to rationalise it, there's no getting around the fact that I'm clearly worried that there will be another 3/11-type event in this area. It's irrational and I know it to be such, but the deep-rooted concern that we'll have such an event cannot be ignored.

How long will this fear of large quakes last? The family and I were physically unscathed by the giant quakes. There's no logical reason for the concern to exist, yet it's very hard to get rid of.

Better Information

Since the news of COVID-191 started to circulate around Japan, people have wanted to know where specifically cases have been found. When there is a highly contagious strain of a foreign virus running rampant, the last thing a lot of people want to do is put themselves or their families at risk. Unfortunately, the municipal, prefectural, and federal governments in the country have tried to keep locations under wraps in an attempt to minimise the risk of a temporary economic disaster in any given area2, but the data is available if you know where to look. Fortunately a university student from Kyushu University invested a little bit of time, learned how to code up a quick website, and put the data online in a visual manner that is both easy to understand and appreciate.

001 - nationwide

The map is interactive and contains data from the ministry of health and welfare, breaking down the cases numerically and by municipality. Red dots are people who are currently infected, identified by order of the infection3. Green represents the number of cases where the infected patient recovered.

As one would expect, everybody is looking forward to seeing more green.

002 - nagoya area

Based on this chart, the nearest case to my home is about 13km away, not too far from the dealership where I bought my car. Given how many people in the neighbourhood work in and around the area with known cases, there is an obvious need to pay attention, refrain from licking sidewalks, and observe proper hygiene when using and returning from public spaces. That said, it's great that a university student with no background in programming was able to learn enough to build this website and share the data with people across the country. My wife and I are both indebted to him as it saves a great deal of time when sifting through the news for recent updates.

Hopefully some level of government takes the initiative to copy this website and provide people with easy-to-understand, reliable, and recent information. Sadly, governments are generally not known for being able to communicate effectively unless they're demanding something from citizens.


  1. Colloquially known as the coronavirus, though there are many coronaviruses. This XKCD explains the term's usage quite well.

  2. People are often too cautious about things in Japan. It can become very tiring very quickly. There are some advantages to the conservative approach, though.

  3. If the number is 350, that means the person identified is the 350th known case, not that there are 350 infected people in the area.

Everybody Should Have 32GB

A colleague of mine will have the good fortune of a RAM upgrade in her computer today, bringing the device up to a respectable 16GB in total. However, given the number of phones that are starting to ship with as much as 12GB right from the factory, it seems bizarre that the digital devices that we use at the day job continue to be hindered by a lack of relatively inexpensive upgrades. Given the complexity of the tasks we ask from our computers, everybody should have at least 32GB in their notebook or desktop machines1.

The reasoning behind this is pretty simple, but I'm going to over-simplify it: the more RAM a device has, the faster it is.

Sticks of RAM

Before using my current notebook, I was using one of two devices with 8GB of memory. While these machines had enough processing power to do a good bit of what I was asking, the lack of RAM made working with large sets of data rather time consuming. Often times I would find myself sending several gigabytes of data to the upstairs server, with its 32GB of RAM and really fast SSDs, in order to process the results without negatively impacting the performance of the main work notebook. The new machine with its 32GB of RAM has never run out of capacity regardless of how much I've thrown at it. More than this, memory is rarely flushed back to disk for later retrieval, making the system performant in every metric that counts. Is there a large Excel file to work with? Does an 80GB database need to be restored and processed? Will there be a series of videos to convert for use as a web-embedded resource? It really shouldn't matter. People deserve to have a machine that can keep up with them.

Generally the argument that comes up when I make statements like this is along the lines of "Well, not everyone is working with a lot of data. Some people just like to use Facebook and watch YouTube.". Fair enough. But why should anyone want to keep their machines limited if they're going to be used for any sort of work? Yes, there will be "wasted" memory sitting in machines if everyone has 32GB (or more), but it's better than the alternative.


  1. I say this fully understanding that a lot of modern motherboards simply cannot handle more than 16GB for various reasons. The fact remains that people should have ample RAM.

Thinking About "Smarter" Crawlers

This past week has seen a number of much-needed updates to 10C get released, one of which was a direct result of seeing an excessive number of requests coming from unabashed slurping bots. There is no denying that organisations that exist for the sole purpose of earning money from the efforts of others tend to follow each and every URL that is found on our websites, but is this really the best use of everyone's resources? Any crawler that wants to index each and every bit of public content that is hosted on 10C will need to read several million pages worth of text. Even accessing 5 URLs per second would mean that a crawling engine would need about 291 days to access each an every post. This is terribly inefficient for both the content-scraping scum and my infrastructure.

This got me thinking ….

As of this moment, every post on 10C works out to about 261 megabytes of text. Compressed it would be closer to 64MB. Updates are sequentially tracked and can be rolled back to any point at the drop of a hat. Would it not make more sense for crawling engines to have a URL that could generate a complete package of content on demand for them to download then, when the machines stop by in the future to look for updates, they can use the same URL but with a query string saying something akin to "give me all updates since last Tuesday at 3 o'clock" to download a much smaller package than the first. Doing this would be a win-win for everyone in that web servers could save their resources for actual visitors and companies that crawl every page online will not need to wait darn near a year to completely archive a platform as sprawling as this one.

Bandwidth consumption would go down. Page load times would drop. Server hosting requirements would decrease ever so slightly. All for the sake of an open mechanism that allows marketing companies to get what they want.

Then again, why would we want to make this easy for marketing organisations? No … I'm happy to have the recent updates that identify the bots and block them outright with a nice and clear message. The bandwidth usage has dropped. SQL operations per second have been halved. The server can actually relax a little bit now.