For some weird and unexplainable reason, people normally expect better services from private companies than from their own governments. This is not the case for our citizens in Estonia. https://qz.com/1535549/living-on-the-blockchain-is-a-game-changer-for-estonian-citizens/
@yogthos I met a lot of officials from Estonia when I worked on implementation of electronic signature in Poland. There are many factors to their success, but the main is -- they have 1.5m citizens. Doing a project for 1.5m citizens is hundreds of times easier than doing it for 40 or 80m citizens. The latter is possible, but requires much higher skills.
@kravietz sure, but a country of 80 million obviously has more resources than a country of 1.5 million as well. And the cost of scaling is not linearly correlated with the population increase for obvious reasons.
@yogthos did you? ;)
@kravietz I'm saying that cost of technology doesn't rise linearly with the number of users. This is not a controversial statement for anybody working in tech sector. If that was the case no tech companies could ever scale.
You're saying that something fundamental changes with tech being developed in public sector, and I don't see that happening myself.
I used to work in the private sector and now I'm working in the public one. The bureaucracy and efficiency are very much comparable.
Public sector has one distinctive feature: it's much more risk averse as compared to the private sector.
This is understandable since every failed project is met with a concerted hysteria in the media because it's "public money".
@kravietz yeah and I think this is where private sector can provide value, but it can also create problems in the long term when they grow.
One model that could work is to have a law that allows the government to buy out a company that's considered an essential service at market price.
@yogthos This I can only agree.
> cost of technology doesn't rise linearly with the number of users
It does not *have* to if you do it correctly. Often it's not done correctly and in public sector it's done incorrectly more often than in private simply because public sector can survive more failures.
Here's a whole article I wrote about electronic signature failures in EU that highlights these issues
@kravietz you're forgetting survivorship bias here though.
Majority of companies fail, so you have to factor those as well in terms of overall effort and resource expenditure. With the public sector failure is just more transparent.
@yogthos Well, that's part of the problem: public-owned companies rarely fail.
In Poland we have 15 state-owned coal mines that haven't been profitable for decades and state essentially pays billions each year so that they can dig some more coal. Because it's poor quality, state also buys coal for power plants from Russia. Any mention of reforms results in trade unions marches and nothing changes.
@kravietz in many cases this can be a positive though since you don't want essential services to fail due to market conditions. For example, broadband is not profitable in rural areas in Canada, so people can't get quality service from the private industry. And keeping employment up is often a benefit as well.
But you're right that in some cases you end up with vestigial services that aren't really providing value. So there needs to be mechanism for deprecation.
@yogthos That is an easy criterion here: if the service was *intended* to operate on non profit basis (like all the basic infrastructure services you mentioned) then, well, it should do just that.
That's one of the main advantages of public sector by the way: it does *not* have to generate profit. Private companies in most countries are required by law to generate profit and if they don't the directors may be accused of failing the company.
@kravietz yeah agreed, the interesting question is how to decide whether something is essential enough to become a public service. For example, and argument can be made that Amazon has become essential enough to warrant being public infrastructure.
But it's so large at this point that there isn't any way I can see that happening.
And here's a whole paragraph on Estonia from that article:
"As for the general architecture of the QES, Estonia is a notable exception from the negative trend. In Estonia, the government implemented a reasonably complete and consistent product (DigiDoc), which started from clearly defined objectives, an open technical specification, software packages and an internet portal"
And they indeed did that - a whole consistent and usable product.
@kravietz [citation needed]