Starlink has bufferbloat. Bad.
 help / color / mirror / Atom feed
From: Colin_Higbie <CHigbie1@Higbie.name>
To: "starlink@lists.bufferbloat.net" <starlink@lists.bufferbloat.net>,
	Juliusz Chroboczek <jch@irif.fr>
Subject: [Starlink] Re: The French have decided to destroy the Internet within their own country by taxing Gigabits (Dr.  Peering)
Date: Thu, 19 Nov 2026 16:11:31 +0000	[thread overview]
Message-ID: <BN7PPF7A8514BAD4A9831664A92746396C5F1792@BN7PPF7A8514BAD.namprd16.prod.outlook.com> (raw)
In-Reply-To: <BN7PPF7A8514BAD3E94DEE08B8478B5DCA7F1792@BN7PPF7A8514BAD.namprd16.prod.outlook.com>

Juliusz, sincere apologies for misspelling your name. I saw that just as I hit send.



Julius, 

This is not an engineering problem (except to the extent that implementation may or may not require parsing the sources of the data), but an economic policy one. When adding a tax, that punishes and diminishes the activity (if not familiar with this law of economics, the underlying cause and effect driving it is: any tax is added to the price paid by consumers, and at a higher price, there will be a lower quantity demanded by customers). That either affects everyone, or through some financial redistribution, the wealthier French take on all of that cost and poorer French are insulated from it, but either way, when price goes up nationally, quantity demanded drops, regardless who pays for it.

Now, if the French government can limit the tax to only American companies (not sure how that's possible with VPNs without setting up something like China's Great Firewall, which gives the government total control and ability to monitor all traffic in and out of the country), and if French ISPs want to deal with adding that complexity to their billing so that customers pay more for watching Netflix than for browsing French websites, then, yes, you could technically make French people pay more for American services and thereby incent them toward French alternatives. 

Also note that implementing this segregated model would add substantial cost, with the primary user-perceived effects being higher prices, more complicated billing, and loss of privacy. There would also be more tax revenue to the French government, which you could argue has a benefit to the French people (especially if they cut other taxes as a result), but that's not a direct benefit to the ISP customers paying for their Internet.

And if you don't segregate and just tax all traffic equally, which seems the more likely scenario given the technical challenges (but I'd yield to the many expert network engineers in this group for that part), then you're reducing access for everyone who uses the Internet to either send or receive data.

And back to the starting point: the effect of these taxes and increased prices is to reduce the quantity demanded, reducing overall revenue and profitability to suppliers. That in turn reduces the incentive to suppliers (ISPs) to invest in building out more infrastructure (the same problem Net Neutrality advocates ignore in the U.S.), which is the ONLY thing that can, over time, drive down costs of delivering Internet to end users (not the price paid, which could be offset with government subsidies, but the actual cost that somebody, whether the Internet users or the taxpayers, must pay). This means any such tax ultimately INCREASES the overall costs and REDUCES the quality of the service available over the long term (years to decades). It is possible that these effects are small, possibly even small enough to be considered de minimis, but directionally, that is absolutely what must happen, per the basic laws of economics.

The piece that most non-economists miss in reviewing how incentives affect supply, meaning pricing and availability, over the long term is the importance of growth and how incentives can affect it. Growth CREATES NEW SUPPLY, which ALWAYS results in effectively reduced cost and prices over time (due to economies of scale and because innovation only moves costs in one direction – down). Engineers tend to think of everything as a zero-sum game, which is typically fair in engineering problems. However, that is NEVER the case in economics in the long run (it is correct in the short run, where there is no time to innovate and improve, but presumably any tax policy should have a perspective longer than the current fiscal year). Given time, people and businesses innovate to cut costs and create new, better solutions to sell more because that enables them to earn more. That ALWAYS creates new wealth and capabilities that didn't exist before. Hence, NOT a zero-sum game. Devising those innovations and then implementing them takes time, so this is only a long-run effect, but in your policy considerations, please be careful not to neglect those benefits, and, conversely, the costs of reducing their efficacy through disincentivizing taxes.

Cheers,
Colin



  reply	other threads:[~2026-11-19 16:11 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <179507172508.1293.11534238119359794018@gauss>
2026-11-19 16:08 ` [Starlink] Re: The French have decided to destroy the Internet within their own country by taxing Gigabits (Dr. Peering) Colin_Higbie
2026-11-19 16:11   ` Colin_Higbie [this message]
2026-10-31 12:56 [Starlink] " Frantisek Borsik
2026-10-31 18:05 ` [Starlink] " Daniel AJ Sokolov

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://lists.bufferbloat.net/postorius/lists/starlink.lists.bufferbloat.net/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=BN7PPF7A8514BAD4A9831664A92746396C5F1792@BN7PPF7A8514BAD.namprd16.prod.outlook.com \
    --to=chigbie1@higbie.name \
    --cc=jch@irif.fr \
    --cc=starlink@lists.bufferbloat.net \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox