Musing on static blogs, caching, and ActivityPub

Every now and then I get an urge to convert this blog to a set of generated static HTML files. Could I also do that when publishing an ActivityPub feed?

Read the rest of this entry »


So, I’ve kind of lost the cycle of yearly summary blog posts. Things happened! Here’s a brief summary of happenings since my last post in December 2018:

  • We have a kid. It’s been mostly great. Challenging at times, but worth it. I’m still amazed thinking about how much the kid has grown.
  • There is a pandemic. It mostly sucked. I got COVID at least once (positive test in September 2023). We’ve been fortunate that COVID-19 hasn’t impacted us a whole lot – we ended up cancelling two Christmases, but no one in our immediate circle got seriously ill or died.
  • We bought a house. A house wasn’t really my first choice, but it is a better option than most apartments in Toronto. We contracted to have some renovations done in early 2023, and have more on the wishlist, but don’t really have the time or money yet.
  • I was working at a game company since mid-2019, it wasn’t bad, but I was laid off this month. It was initially in-office, but much like everyone else, we swung hard to work-from-home in March 2020 and continued this despite some tepid attempts at hybrid mode in 2022. I got to be a bit of a lead and to mentor a little bit. I was proud of what we achieved as a team, but towards the end it wasn’t working out business-wise.
  • Beyond a December 2019 trip to Cuba for a beach vacation (we went cheap, and it wasn’t all that), we didn’t travel – the farthest I’ve gone since 2019 was Niagara region, and Kitchener this month. Small child plus a pandemic will do that. Honestly, I haven’t missed travel too much yet. We’ve been wanting to go somewhere, but where, and how? A 3-4 year old isn’t an ideal travel companion. We’ve wanted to go to Newfoundland, or Iceland again, but the kid probably won’t care much for landscapes.
  • My music listening, already low, took a further drop starting in 2020, and hasn’t recovered. I’ve struggled to find time to concentrate on music, and a mental block prevents me from listening to music casually (except for Youtube music streams). I realized that I used to listen to music when I was by myself, and I haven’t been by myself in a long while, and I haven’t adapted well to listening when family or coworkers are around. Live music stopped after 2019 and I haven’t been back yet, partly due to COVID concerns, partly due to guilt of missing time with my kid, and partly due to laziness in planning.
  • Particularly in 2023, I’ve also struggled to dedicate time for reading. I ended up reading only kids’ books, and Brandon Sanderson’s Four Secret Novels. Notable reads in past years included Martha Wells’ Murderbot series, most of it during first lockdown in early 2020.
  • I also haven’t been doing much photography. In what’s sure to have outraged 2015 Jarek, I used my phone for vast majority of pictures. It’s a Pixel 3a and its computational photography does a lot (a particularly impressive example), but it’s laggy.
  • I didn’t really spend much time on personal projects, except for OpenStreetMap whenever time allows (which isn’t as much as I would like).
  • My diary has had several week- and month-long gaps, but I go back to it in the end. It’s nice to have a record.
  • I’ve been quite anxious about the state of the world – climate crisis, wars, fascism – doubly so since having a kid. It’s weird and scary to be looking at the real-world Gibsonian Jackpot likely up ahead. I try to come up with ways to help and contribute positively to the world, but it hasn’t been easy.

HTTP is sometimes OK

The 2021 post Plaintext HTTP in a Modern World by joshua stein elucidates some of the reasons I’m not entirely keen on forcing HTTPS for reading websites. Old computers and old browsers won’t be able to use encryption strong enough to still be useful today. But that’s no reason to stop them from anynomous access to websites where users mostly read or view content and aren’t able to submit credentials or personal information. Like, for example, blogs.

I have a 10-year-old browser still installed (Opera 12.16) which cannot read Wikipedia anymore. I don’t expect to be able to log in to Wikipedia, but reading text and viewing images should be possible with a browser released in 2013. That’s not ancient history. We had good CSS then!

That’s a reason why I have HTTPS enabled on this domain, but not automatic redirecting to HTTPS. The only thing I was unsure about was how to let visitors know that there is an HTTPS version of the site available, without force-redirecting everyone. I was thinking of having two sets of HTML pages generated, one to be served over HTTP with a warning banner and a link to HTTPS at the top of the page; and the other to be served over HTTPS without the banner. But the post by joshua shows a cleaner solution: the Upgrade-Insecure-Requests HTTP header. The header indicates the browser’s preference for HTTPS, and as of 2023 the fact that the browser knows about this header also implies that the browser supports a currently-secure encryption level – so browsers that send it can be upgraded to HTTPS, and browsers that don’t can get the response over HTTP. Presumably they’ll change Upgrade-Insecure-Requests: 1 to Upgrade-Insecure-Requests: 2 once more recent encryption algoriths are deprecated and browsers of 2020 or 2023 have to be cut off from HTTPS.

Do keep your login forms on HTTPS though!

There was also the 2018 post Securing Web Sites Made Them Less Accessible by Eric A. Meyer which featured discussion which pointed out a problem with leaving HTTP open: it allows tampering of the content between server and client, for example ads injected by ISPs. I’m not entirely sure where I fall on this. On one hand, I can see site authors not wanting their content to be changed in transit. On the other hand, if the choice is between an old browser getting possibly-tampered content or getting no content, is the possibly-tampered content still more useful? And, the conditional upgrading using Upgrade-Insecure-Requests seems to offer a kind of herd immunity: if 99.5% of traffic is using HTTPS, the effort of injecting ads or tracking into 0.5% of requests might not be worth it, at least from a commercial perspective. Targeted injection (like for websites useful for activists) might still happen, but a locked-down old browser being used to read general-purpose web content for hobby reasons seems reasonably safe.

Like all old protocols, HTTP will eventually cease to be useful. But there’s no reason to make it happen earlier than necessary.


Na Ars Technica jest dzi? artyku? o dial-upie, wi?c przypomnia?y mi si? i moje pierwsze kroki w sieci. Rok 1998, us?uga dial-up TPSA, 0202122, op?ata nie miesi?cznie, a co 3 minut – 29 groszy. Po 22 i przed 8 rano by?o pó? ceny, jako dzieciak po 22 nie podpada?o, wi?c pami?tam wstawanie rano i w??czanie komputera “na dole”.

Pr?dko?ci na dial-upie by?y na dzisiejsze standardy absurdalnie niskie, cho? do tekstu starcza?y, wi?c tylko multimedia i ?ci?ganie oprogramowania zajmowa?o. I p?acenie za czas, w którym tylko czyta?em, troch? nieszcz??liwie wychodzi?o. By?o oczywi?cie ?ci?ganie tylko nag?ówków w usenecie, pisanie emaili offline, ale zupe?nie wygodne to nie by?o.

Obecnie moje, do?? wolne, po??czenie internetowe jest oko?o 500 razy szybsze ni? dial-up 56k. Mo?e nie przypadkowo RAM w moim komputerze te? 500 razy wi?kszy ni? tamtejsze 32 MB, cho? nie wiem czy szybszy.

Wliczaj?c inflacj? z?otego od 1998 roku i obecny kurs dolara kanadyjskiego, za obecny miesi?czny koszt mojego (nietaniego) po??czenia móg?bym po??czy? si? (pomi?dzy 22 a 8) na oko?o 28 godzin w miesi?cu, czyli nieco mniej ni? godzin? dziennie. Przynajmniej mia?bym miej czasu na doomscrolling.

A poszukuj?c informacji ku potwierdzeniu moich wspomnie? znalaz?em stron? która nadal ma informacje i komentarze z roku 2001: GSM (jakby znikn??o, jest te? w Wayback Machine), jak równie? seri? artyku?ów z 2011 r. która opisuje te? moje pierwsze zetkni?cia z polskim internetem: Polbox, Frico, KKI.

(Teraz jest rok 2023, a mój system blogowy zepsu? polskie literki. Nie wiem jak mu si? to uda?o. Naprawi? pó?niej.)

Quick Heat and Carbon Numbers

I was thinking about carbon emissions and calculated some rough estimates for our house.

Between December 11, 2020 and December 12, 2021, we used 2031 cubic metres of “natural gas” (mostly methane) to heat our house and provide domestic hot water. (The heated area is about 120 square metres, the house is double brick and largely uninsulated.)

This seems to correspond to about 21 MWh or 75 GJ of energy, and as a rough approximation 4400 kg of direct carbon dioxide-equivalent (CO2eq) emissions. (Numbers rounded due to uncertainty in conversion rates, emissions values, and exact properties of gas being delivered to our house. Extracting and transporting the gas also causes emissions, but this seems to be at least within order of magnitude.) Our gas bills for this period were about $1000 (all dollar figures CAD).

Electricity Map / Tomorrow gives CO2eq emissions goal to meet Paris Agreement objectives of limiting global warming to 2°C by 2100 as below 5 tons per person per year by 2030, and below 2 tons by 2050. Atmosfair gives 1.5 tons CO2 per person per year until 2050 for a 1.5°C warming goal.

With three people in the house, our per-capita emissions from burning gas are around 1.5 tons, using up 30% of our 2030 yearly carbon budget and most to all of our 2050 yearly carbon budget.

At a very simplified glance, replacing gas with purely resistive electric heating would drop emissions by at least half, and possibly up to five times: Ontario’s electricity carbon intensity usually varies between 30 and 100 kg CO2eq per MWh, depending on grid load and how windy it is. (Currently highest intensity is during summer due to cooling load, but with a wider shift to electric heating the winter peak might also approach these levels.) 21 MWh of electricity would then cause about 600 to 2000 kg of emissions for the house per year, or about 2.5 to 3.5 tons CO2eq less than gas.

With our effective electricity price of about 12 cents per kWh (excluding flat fees, but including per-kWh delivery and taxes), and effective gas price of about 43 cents per m³ or 4 cents per kWh, swapping to purely resistive electric heat would cost us a premium of about 8 cents per kWh, and higher use would push us into a higher price tier adding another 2 cents per kWh. 10 cents per kWh would then add up to $2100 extra per year, or average of $175 per month. Roughly calculated, this is about $0.70 per kg of CO2eq avoided.

Of course, there are much better options than purely resistive electric heat. Heat pumps aren’t a perfect fit for our house, because our old cast-iron radiators work best with high water temperatures, and heat pumps work best with lower water temperatures. Ontario’s winter low temperatures can also push air-source heat pumps to limits of their efficiency. Ground-source heat pumps are an option, but at higher capital costs and substantially more installation work. But on the whole, it should probably be feasible to do all domestic hot water heating (if using a tank heater) and probably about half to two-thirds of space heating using electric heat pumps, supplementing with resistive electric when it’s particularly cold.

Outside of heating season, our use — which was then entirely domestic hot water — seems to average out to about 1.5 m³ per day, which is about 15 kWh or 54 MJ per day. Since our hot water use doesn’t change much over seasons, our yearly hot water use seems to be about 5.5 MWh or 20 GJ, meaning about 15.5 MWh or 55 GJ is used for heating.

A basic heat pump used for space heating would likely be about 3 times more efficient than resistive heating. (Assuming HSPF 9.8 for a system based on a MXZ-3C24NA2, which is primarily aimed at cooling, but this should be a decent initial reference. Resistive electric has HSPF 3.4.)

Using a heat pump like this to provide 5.5 MWh for domestic hot water and two-thirds of 15.5 MWh for space heating would then use about 5.3 MWh of electricity, and the remaining one-third of space heating would add about 5 MWh, for a total of roughly 10 MWh of electricity per year. This would bring our house electricity emissions down to about 300 to 1000 kg CO2eq per year (3.5 to 4 tons less than gas), and our cost premium compared to natural gas to about $1000 per year, or average of $83 per month. The operating cost per kg of CO2eq avoided is about $0.27. (There would also be a capital cost of the new devices and installation.)

For further comparison, carbon offsets from a respectable offset provider like Atmosfair cost about 23 euro per 1000 kg of CO2, which is about $31, which is about 3 cents per kg offset, or about nine times cheaper. But we’ll have to lower our own emissions at some point. A realistic solution would improve our house insulation, sealing, and heating efficiency before diving into heat pumps, but this post offers some baseline numbers for more comparison later.