740 Days Later

It’s time for another COVID update–and possibly the last I’ll be doing in this format.

This series began 50 days after the start of the pandemic and has had irregular updates 120, 200-ish, 250-ish, 300-ish, 320-ish, 333, one year, 420, 500, 600, 640, 670, and 700 days later.

Last time we were still reaching insanely new heights of new infections, but, thanks to vaccination, we got past that with much less impact in hospitals (or deaths).

But even as cases decreased and restrictions were (again, probably prematurely) loosened, we’re currently experiencing another bump after the big surge:

This new bump, like the peak before it, seems to be sustained by the sub-20 demographics, and, again, matches what I’ve seen regarding vaccination (I suspect many have only first doses, if anything):

Hospitals are coping, but admittances always lag behind new case figures and things are still evolving:

And this is why I’m wary of looking at the data without a long-term view, and taking in all angles–COVID-19 might be becoming endemic, but it is far from being gone, and I suspect that with the usual rainy months we have around this time (which were belated but seem to have finelly arrived), it may well rise again, and the Rt figures certainly hint at it doing that now:

But, like I wrote above, this may be the last post I do with this amount of detail.

The reason for that came in another display of premature incompetence a few days ago, when the government stated they would stop issuing daily reports on the COVID situation and move to a “simpler” weekly bulletin.

This move puts at jeopardy community initiatives like dssg-pt/covid19pt-data, which have relied on scraping the data (or entering it manually) from the DGS website, and continues the two-year-old tradition of utter inability to do something as simple as providing a daily updated consolidated .csv file (let alone a decent API).

Also, I’m very much out of steam as far as this is concerned.

Months of typing in figures from PDFs, weeks spent trying to scrape data and the progressive de-escalation towards the the (now very minimal) wrapper I use to import the data from dssg-pt folk and the minor adjustments I do to the dashboard still mean too more time and energy than I can spare.

I’ve also long given up on pointing out that if someone can build a fully working dashboard in 5 minutes with nothing but Excel and a free Power BI account, the Portuguese government could surely do better–and yet they haven’t, for the past two years, and are now certain to make it worse.

I get there’s a war on, fuel prices are through the roof and the economy is tanking, but to assume the pandemic has gone away and lowering visibility on data is just… stupid.

Which is why I’d give this government (which, incidentally, got reelected recently) a D- in technology savvy, because I still can’t believe they’re still issuing the same half-assed PDF reports in 2022.

If you’re tuning in just now, yes, our government used PDFs, with charts without axis indicators or supporting tables, as their official reporting tool.

That said, things are improving overall, even if we do keep hearing about co-workers, friends and family getting sick every week.

Remote work is no longer mandatory, so most people are back at work (although most sane companies have moved to hybrid regimes, since it’s become plain over the past two years that remote work works for many people and roles) and wearing masks only in enclosed spaces (like classrooms) and public transport.

I know it’s just a matter of time until we catch COVID, and am still worried about long-term consequences even if vaccines will protect us from most ill effects (and even that protection is sure to wane), but having lunch with other people is fast becoming a concept we can entertain again, and something I can actually look forward to–but I plan to do so warily and with appropriate restraint for many months to come still.