Starlink has bufferbloat. Bad.
 help / color / mirror / Atom feed
From: Frantisek Borsik <frantisek.borsik@gmail.com>
To: sascha@thexlab.org
Cc: "Sebastian Moeller" <moeller0@gmx.de>,
	"Network Neutrality is back! Let´s make the technical aspects
	heard this time!" <nnagain@lists.bufferbloat.net>,
	"Michael Welzl" <michawe@ifi.uio.no>,
	"Dave Taht via Starlink" <starlink@lists.bufferbloat.net>,
	bloat <bloat@lists.bufferbloat.net>,
	"Cake List" <cake@lists.bufferbloat.net>,
	libreqos <libreqos@lists.bufferbloat.net>,
	"Jeremy Austin via Rpm" <rpm@lists.bufferbloat.net>
Subject: [Starlink] Re: [NNagain] [Bloat] The French have decided to destroy the Internet within their own country by taxing Gigabits (Dr. Peering)
Date: Tue, 17 Nov 2026 23:49:24 -0800	[thread overview]
Message-ID: <CAJUtOOiOAf1qVNvJ0s+TcNUaU5nHhzzeRBvfDSJ1pixFKBJ1Vw@mail.gmail.com> (raw)
In-Reply-To: <a361c388-ae33-4c9d-847e-f5aff66dd3a1@thexlab.org>

🙏

Frantisek Borsik reacted via Gmail
<https://www.google.com/gmail/about/?utm_source=gmail-in-product&utm_medium=et&utm_campaign=emojireactionemail#app>

On Wed, Nov 4, 2026 at 10:33 AM Sascha Meinrath <sascha@thexlab.org> wrote:

> On 11/4/26 12:03 PM, Sebastian Moeller wrote:> Hi Sascha,
> >
> > On November 4, 2026 5:20:01 PM GMT+01:00, Sascha Meinrath <
> sascha@thexlab.org> wrote:
> >> Hi Sebastian,
> >>
> >> My point is that if I VPNed to a point outside France, then a French
> ISP would have no way to know that I've accessed the type of media that
> should be taxed.
> >
> > A misunderstanding, the amendment prposes to tax every single bit
> independent of the source... it makes zero exemptions, not for FOSS, not
> for non commercially ial entities. Sure the motivation is to tax big tech
> primarily, but the current (half-baked) legal implementation is not
> differentiating...
> >
> >> In essence, VPNs (and proxy services) would create extraordinary
> problems for even instrumenting the proposed taxing regime.
> >
> > How? Like with a CDN the tax would hit the vpn provider... the ASN of
> the source IP....
>
> A VPN provider (at least any that I would recommend using) would not know
> the
> content being VPNed. In essence, someone in France could VPN to a
> different
> country; neither the ISP or the VPN provider would know the specific
> content
> being accessed (Note: while some VPN's do track their users, none of the
> privacy
> protecting ones would do so); meanwhile, the media company -- even
> assuming some
> sort of extra-territorial leverage of French law outside of France -- may
> have
> no idea that the end destination of the media in question is within France.
>
> Long story short, neither the VPN provider, the French ISP, nor the media
> company would necessarily know that a specific piece of media was destined
> for
> consumption within France. This is, of course, exactly the conundrum
> facing
> authoritarian and censorial regimes around the globe -- in this case, one
> is
> dodging tax surveillance, but the technical underpinnings are very much
> the same.
>
> All in all, a dumb & doomed (proposed) law.
>
> --Sascha
>
> >
> >
> >
> >>
> >> --Sascha
> >>
> >> On 11/4/26 10:40 AM, Sebastian Moeller wrote:
> >>> Hi Sascha,
> >>>
> >>>
> >>>> On Nov 2, 2026, at 15:30, Sascha Meinrath <sascha@thexlab.org> wrote:
> >>>>
> >>>> I can't wait until the French Parliament learns about VPNs... It
> would be absolutely *wonderful* if all the ISPs passed along the tax to
> their customers... and then taught all their customers that VPNed traffic
> would, for obvious reasons, be exempted.
> >>>
> >>> Not how this is supposed to work: the ISP aggregates its end-customer
> bound traffic volume by ASN and reports these aggregates quarterly, the tax
> office sends the ASNs the tax bill. VPN does not really matter, except now
> it is the VPN provider that is on the hook for the money. The amendment
> does not address VPNs but for CDNs that are in a similar picckle the idea
> is that CDN operators either pass on the cost or "eat" it themselves (both
> half-baked ideas). The only oddity VPNs add is that they typically receive
> no money from the actual content providers, but the end users.
> >>>
> >>>>
> >>>> I find proposals like this to be absolutely idiotic -- and, they are
> yet another step in the global data obfuscation arms race.
> >>>
> >>> Oh, that proposal is pretty half-baked, no arggument there, but I
> think we differ in which of its half-baked-ness we consider worst ;)
> >>>
> >>>>
> >>>> #encrypteverything
> >>>
> >>> Well, if you look what this is supposed to operate on encryption is
> not really an option, as this operates on source and destination IP
> addresses, that are somewhat prickly to encrypt if you want packets
> delivered...
> >>>
> >>>>
> >>>> --Sascha
> >>>>
> >>>> On 11/2/26 6:23 AM, Sebastian Moeller via Nnagain wrote:
> >>>>> Hi Michael,
> >>>>>> On Nov 2, 2026, at 08:44, Michael Welzl <michawe@ifi.uio.no> wrote:
> >>>>>>
> >>>>>> The tax is 5 cents per Gigabyte. This is insane.
> >>>>> Back of the envelop calculations for video streaming (assuming this
> to be the one category driving most of the sustained capacity usage):
> >>>>> common use seems to be HD streaming (say on average at 5 Mbps), with
> say 2 hours per day, 7 days per week:
> >>>>> (5 * 1000^2) * (60 * 60 * 2 * 30) / (8 * 1000^3) = 135 GB
> >>>>> -> 135 * 0.055 = 7.425 Euro/month
> >>>>> typical cost for midrange  (ad-free) video streaming in france: 8-15
> Euro
> >>>>> depending on the reference contract tier, the tax almost as large as
> the total revenue, no way this could be priced-in without significant price
> increases.
> >>>>> The proposed tax indeed seems quite steep, however it is capped at
> 15% of revenue in france or 2% of world-wide revenue
> >>>>> So that would get this more in like with a 15% tax... on revenue,
> still nothing to sneeze at, but a far cry from the numbers one gets when
> simply multiplying...
> >>>>> Regards
> >>>>> Sebastian
> >>>>> P.S.: My take this is half-baked and will not make it through the
> further legislative process in that form.
> >>>>>> It goes against how we agree to use the Internet.
> >>>>>>
> >>>>>> English version:
> >>>>>> "France has recently approved a highly controversial "Internet
> Traffic
> >>>>>> Tax" (or Gigabyte Tax) as part of the 2027 budget bill amendments.
> The
> >>>>>> new levy targets digital content providers based on the volume of
> data
> >>>>>> they transmit to French internet users, which is monitored by
> network
> >>>>>> operators and Internet Service Providers (ISPs) heavily utilizing
> >>>>>> Internet Exchange Points like France-IX."
> >>>>>>
> >>>>>> Google AI summary:
> >>>>>> "Key Metrics & Tax Rates
> >>>>>> • €0.055 per gigabyte charged to content providers.
> >>>>>> • Capped at 15% of a provider's French revenue.
> >>>>>> • Capped at 2% of a provider's global revenue.
> >>>>>> • Measured directly by French network operators and ISPs."
> >>>>>>
> >>>>>>
> >>>>>> On 2026-10-31 13:56, Frantisek Borsik wrote:
> >>>>>>> Crossposting, because this garbage idea might "inspire" more
> countries
> >>>>>>> and
> >>>>>>> have a horrible impact also on bufferbloat everywhere.
> >>>>>>>
> >>>>>>> *Original post by Dr. Peering:*
> >>>>>>>
> >>>>>>> *"The French have decided to destroy the Internet within their own
> >>>>>>> country
> >>>>>>> by taxing Gigabits. Here’s Octave Klaba's excellent analysis [in
> >>>>>>> French,
> >>>>>>> link below] showing just how insane the finance bill actually is.
> The
> >>>>>>> IXP
> >>>>>>> taxation amount is, just for starters, ridiculous!"*
> >>>>>>>
> >>>>>>> https://x.com/drpeering/status/2116468659392807
> >>>>>>> <
> https://x.com/drpeering/status/2116468659392807381?a=E-elnbppn-MAi-LNIrU&s=46
> >
> >>>>>>>
> >>>>>>> *Octave's post:*
> >>>>>>>
> >>>>>>>
> https://fr.linkedin.com/posts/octave-klaba-3a0b3632_jai-lu-lamendement-i-1358-au-plf-2027-activity-7522219045576888320-uqkw
> >>>>>>>
> >>>>>>> *Copy+paste:*
> >>>>>>>
> >>>>>>> J'ai lu l'amendement I-1358 au PLF 2027, adopté ce vendredi à
> >>>>>>> l'Assemblée :
> >>>>>>> une taxe de 0,055 € par Go envoyé vers les abonnés des FAI
> français. Le
> >>>>>>> chiffre est abstrait, alors comparons-le. Corrigez-moi si vous
> voyez
> >>>>>>> des
> >>>>>>> erreurs.
> >>>>>>>
> >>>>>>> Deux comparaisons et un cas concret :
> >>>>>>>
> >>>>>>> 1. Le transit et le peering. Un port 100 Gbps coûte environ 6 300
> € par
> >>>>>>> mois chez un fournisseur de transit mondial. En interconnexion
> directe
> >>>>>>> avec
> >>>>>>> un FAI européen, compter 3 à 5 fois plus, soit 20 à 30 k€ par
> mois. Sur
> >>>>>>> ce
> >>>>>>> port (a), la taxe atteint 1,02 M€ par mois, soit environ 3400 % à
> 16300
> >>>>>>> %
> >>>>>>> du prix d'une interconnexion vers les FAI français. On change donc
> >>>>>>> profondément le modèle économique de la bande passante en France.
> >>>>>>>
> >>>>>>> 2. Les hyperscalers sont critiqués pour le prix de leur trafic
> sortant,
> >>>>>>> qui
> >>>>>>> va d'environ 0,08€ à 0,045€ par Go selon le volume. La taxe, 0,055
> €
> >>>>>>> par
> >>>>>>> Go, représente 68 % à 123 % de ce prix. On parle donc de doubler,
> pour
> >>>>>>> le
> >>>>>>> trafic vers la France, un prix déjà jugé trop élevé !
> >>>>>>>
> >>>>>>> 3. Le cas d'OVHcloud. Nous envoyons de l'ordre de 500 Gbps au 95e
> >>>>>>> percentile vers les FAI français, soit environ 324 Gbps en
> moyenne. Sur
> >>>>>>> un
> >>>>>>> an, la taxe brute atteindrait 70 M€. Le texte la plafonne à 15 %
> du CA
> >>>>>>> réalisé en France ou à 2 % du CA mondial, soit entre environ 23 et
> 70
> >>>>>>> M€
> >>>>>>> par an que nous devrions collecter pour l'État, et donc répercuter
> sur
> >>>>>>> nos
> >>>>>>> clients.
> >>>>>>>
> >>>>>>> Conclusions :
> >>>>>>>
> >>>>>>> 1. Une taxe porte en général sur une petite fraction de son
> assiette.
> >>>>>>> Celle-ci vaut plus de 3000 % du coût du transit, et environ 100 %
> du
> >>>>>>> prix
> >>>>>>> public des hyperscalers. Rappelez-vous la loi SREN, votée pour
> >>>>>>> supprimer
> >>>>>>> les frais de transfert qui empêchaient les clients de quitter un
> >>>>>>> fournisseur cloud. Ici, on parle de doubler le prix du trafic
> sortant
> >>>>>>> vers
> >>>>>>> les internautes français, un prix déjà très problématique.
> >>>>>>>
> >>>>>>> 2. Le garde-fou, un plafond calculé sur le chiffre d'affaires, est
> >>>>>>> incohérent. Ce chiffre d'affaires couvre des serveurs, du
> stockage, des
> >>>>>>> datacenters et des équipes sans rapport avec la bande passante.
> C'est
> >>>>>>> comme
> >>>>>>> créer une taxe calculée sur la consommation électrique des usines
> >>>>>>> automobiles en France, plafonnée à 15 % du chiffre d'affaires des
> >>>>>>> constructeurs. Quel rapport entre les deux ?
> >>>>>>>
> >>>>>>> 3. Une taxe nationale de ce type frappe à coup sûr les entreprises
> >>>>>>> établies
> >>>>>>> en France, alors que leurs concurrents mondiaux sans présence en
> France
> >>>>>>> livrent le même trafic aux mêmes FAI français sans payer cette
> taxe en
> >>>>>>> pratique. Cela crée une distorsion de concurrence au détriment des
> >>>>>>> acteurs
> >>>>>>> nationaux et déclasse les entreprises qui embauchent en France. Une
> >>>>>>> taxe
> >>>>>>> qui produit l'effet inverse serait une meilleure approche.
> >>>>>>>
> >>>>>>> (a) Facturé au 95e percentile, ce port n'est pas plein en
> permanence :
> >>>>>>> avec
> >>>>>>> un profil journalier type (22 % vers 4 h, 90 % vers 21 h), il
> >>>>>>> transporte en
> >>>>>>> moyenne 57 % de sa capacité, soit 18,6 millions de Go par mois.
> >>>>>>>
> >>>>>>>
> >>>>>>> All the best,
> >>>>>>>
> >>>>>>> Frank
> >>>>>>> Frantisek (Frank) Borsik
> >>>>>>>
> >>>>>>> In loving memory of Dave Täht: 1965-2025
> >>>>>>> https://libreqos.io/2025/04/01/in-loving-memory-of-dave/
> >>>>>>>
> >>>>>>>
> >>>>>>> https://www.linkedin.com/in/frantisekborsik
> >>>>>>> Signal, Telegram, WhatsApp: +421919416714 <+421%20919%20416%20714>
> <+421%20919%20416%20714>
> >>>>>>> iMessage, mobile: +420775230885 <775%20230%20885>
> <775%20230%20885>
> >>>>>>> Skype: casioa5302ca
> >>>>>>> frantisek.borsik@gmail.com
> >>>>>>> _______________________________________________
> >>>>>>> Bloat mailing list -- bloat@lists.bufferbloat.net
> >>>>>>> To unsubscribe send an email to bloat-leave@lists.bufferbloat.net
> >>>>>> _______________________________________________
> >>>>>> Bloat mailing list -- bloat@lists.bufferbloat.net
> >>>>>> To unsubscribe send an email to bloat-leave@lists.bufferbloat.net
> >>>>>> _______________________________________________
> >>>>>> Bloat mailing list -- bloat@lists.bufferbloat.net
> >>>>>> To unsubscribe send an email to bloat-leave@lists.bufferbloat.net
> >>>>> _______________________________________________
> >>>>> Nnagain mailing list -- nnagain@lists.bufferbloat.net
> >>>>> To unsubscribe send an email to nnagain-leave@lists.bufferbloat.net
> >>>>
> >>>> --
> >>>> Sascha Meinrath
> >>>> Director, X-Lab
> >>>> Palmer Chair in Telecommunications
> >>>> Penn State University
> >>>
> >>>
> >>
> >
>
> --
> Sascha Meinrath
> Director, X-Lab
> Palmer Chair in Telecommunications
> Penn State University
>
>

  parent reply	other threads:[~2026-11-18  7:49 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-31 12:56 [Starlink] The French have decided to destroy the Internet within their own country by taxing Gigabits (Dr. Peering) Frantisek Borsik
2026-10-31 14:51 ` [Starlink] Re: [Bloat] " Sophie
2026-10-31 14:58 ` [Starlink] " Inemesit Affia
2026-10-31 18:05 ` [Starlink] " Daniel AJ Sokolov
2026-11-02  7:44 ` [Starlink] [Bloat] " Michael Welzl
2026-11-02 11:23   ` [Starlink] Re: [Bloat] " Sebastian Moeller
2026-11-02 14:30     ` [Starlink] Re: [NNagain] " Sascha Meinrath
2026-11-04 15:40       ` [Starlink] Re: [NNagain] " Sebastian Moeller
2026-11-04 16:20         ` Sascha Meinrath
2026-11-04 17:03           ` Sebastian Moeller
2026-11-04 17:57             ` [Starlink] Re: [Cake] " David P. Reed
2026-11-05  9:05               ` [Starlink] Re: [Cake] " Sebastian Moeller
2026-11-16 23:04                 ` [Starlink] Re: [Bloat] Re: [Cake] [NNagain] " Juliusz Chroboczek
2026-11-17  1:43                   ` David P. Reed
2026-11-17 12:33                     ` Juliusz Chroboczek
2026-11-18  7:53                     ` Frantisek Borsik
2026-11-18  9:25                     ` David Lang
2026-11-18  7:48               ` [Starlink] Re: [Cake] Re: [NNagain] [Bloat] " Frantisek Borsik
2026-11-18 12:12                 ` Sascha Meinrath
2026-11-04 18:33             ` [Starlink] " Sascha Meinrath
2026-11-04 19:09               ` Sebastian Moeller
2026-11-04 19:14                 ` [Starlink] Re: [Bloat] Re: [NNagain] " David Lang
2026-11-04 19:20                   ` [Starlink] Re: [Bloat] " Sebastian Moeller
2026-11-04 19:26                 ` [Starlink] Re: [NNagain] Re: [Bloat] " Olivier MJ Crépin-Leblond
2026-11-18  7:49               ` Frantisek Borsik [this message]
2026-11-18  1:48       ` Raquel Renno Nunes
2026-11-18  7:46       ` Frantisek Borsik
     [not found]   ` <bf2f7d692a5e42738403850d419c4a94@OS5P279MB0502.NORP279.PROD.OUTLOOK.COM>
2026-11-02 11:33     ` [Starlink] " Michael Welzl
2026-11-16 16:11 ` Juliusz Chroboczek
2026-11-16 16:31   ` [Starlink] Re: [LibreQoS] " Frantisek (Frank) Borsik
2026-11-16 20:41     ` [Starlink] Re: [Bloat] Re: [LibreQoS] " Linus Lüssing
2026-11-16 22:53       ` Juliusz Chroboczek
2026-11-17 15:59         ` Michael Richardson
2026-11-18  0:13           ` David Lang
2026-11-18  0:36             ` [Starlink] Re: [NNagain] " Richard Roy
2026-11-18  1:16             ` [Starlink] " Michael Richardson
2026-11-18  9:17               ` David Lang
2026-11-18 13:37                 ` Juliusz Chroboczek
2026-11-18 23:22                   ` David Lang
2026-11-17  6:36     ` [Starlink] Re: [LibreQoS] Re: [Bloat] " Sebastian Moeller

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=CAJUtOOiOAf1qVNvJ0s+TcNUaU5nHhzzeRBvfDSJ1pixFKBJ1Vw@mail.gmail.com \
    --to=frantisek.borsik@gmail.com \
    --cc=bloat@lists.bufferbloat.net \
    --cc=cake@lists.bufferbloat.net \
    --cc=libreqos@lists.bufferbloat.net \
    --cc=michawe@ifi.uio.no \
    --cc=moeller0@gmx.de \
    --cc=nnagain@lists.bufferbloat.net \
    --cc=rpm@lists.bufferbloat.net \
    --cc=sascha@thexlab.org \
    --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