Article 34:
"Contestability and fairness are intertwined."
Again a standard formulation. Mind you, I am just as far from a legal expert as most of you. For someone who has observed the market as a consumer and a seller, this argument in combination with unfairness as
"imbalance between rights and obligations"
...seems useful beyond my personal interest for the philosophy of law.
DMA critique is predominantly cheap shots of people that don't care much about the law beyond claiming that "We should decide for ourselves."
I have sympathy for truly liberal minds. But If "decide for yourself" was generally the best or only regulation, then seat belts should not be mandatory, the use of drugs and medication would be open to anyone's own judgement, including small children, hormone steered kids, addicts and very old people other people that tend to be higher than is healthy.
More importantly, if "we can decide for ourselves" is the only law we follow regulating technology, the option building itself would also be subject to companies selling and promoting them at will. You can't regulate that either if "we can decide for ourselves."
We see what is happening right in front of our eyes when people drink the Kool-Aid of Apple, Facebook and Google. It's not just wrong, it's utterly depressing, sad and repulsive. Ok, next paragraph...
Often overseen by the end user that only thinks about their direct personal risks when offering private data:
"The processing, for the purpose of providing online advertising services, of personal data from third parties using core platform services gives gatekeepers potential advantages in terms of accumulation of data, thereby raising barriers to entry."
Collecting personal data at big scale and using it competitively, will in itself result in an unfair (incontestable) advantage.
Paragraph 37:
"Not giving consent should not be more
difficult than giving consent."
Another (not un)reasonable regulation and knowing how sneaky our big friends are a very accurate description of how our big tech friends operate.
38 makes sense, too:
"Children merit specific protection with regard to their personal data, in particular as regards the use of their personal
data for the purposes of commercial communication or creating user profiles."
There is still a long way to go, but, so far, I haven't seen anything particularly uniformed or particularily stupid in this regulation.
39 addresses Apple:
"Where such restrictions relate to third-party online intermediation services, they limit inter-platform contestability, which in turn limits choice of alternative online intermediation services for end users. Where such restrictions relate to direct online sales channels, they unfairly limit the freedom of business users to use such channels."
Again, quite accurate. Critics of this paragraph like to repeat Apple's talking point about protecting customers from bad actors.
(42) DMA protects AppStore whistleblowers:
"Any practice that would in any way inhibit or hinder those users in raising their concerns or in seeking available redress for instance by means of confidentiality clauses in agreements or other written terms, should therefore be prohibited."
Meaning Apple (or Google) cannot prevent you or their employees from outing information that exposes malpractice via confidentiality agreements. Again, nothing wrong with that either.
43) Gatekeepers can't impose Webbrowsers. Well argued but my mind rushes to see if there are some edge cases where that would create issues. Can't find any—but I'm sure a smarter person could.
Avoiding edge cases is one reason why the law sounds so weird. At the same time—to a Western European—the law is not by the letter or how it can be bent, what counts is its intention (its"spirit"). Could be irritating for Americans with their case law and its quite common by the letter interpretations.
If you have trouble understanding the difference between the letter and the meaning of the law, I can't think of a better explanation than this episode of Kung-Fu. https://youtu.be/Gm_yMBUNY08
This episode is so good, I can't believe it's a TV show. (I remembered it differently, I thought they both ate the plums when the master was away for some days, much longer than expected, and they both ate them because they had nothing else to eat).
Paragraph 45 shows a very rare insight into how dodgy the business of online advertisement really is:
"The conditions under which gatekeepers provide online advertising services to business users, including both advertisers and publishers, are often non-transparent and opaque."
Yes. It's a black box. The money you pay to Google and Facebook and the value you get out of it are extremely intransparent. It boils down to "believe me."
This may trigger free market advocates.
A) If you don't like Google's or Facebook's results you can go elsewhere.
B) Millions of businesses use them, so they probably work.
C) You can see it in your sales performance. If it didn't work, no one would book them.
Did you choose them because you really have the choice? How good is the return? Why clicks really? Who measures?
I won't comment on every paragraph. But I'll mention good or bad things when I find them. No really bad things so far (12/66 pages). 49) Again completely reasonable:
"To enable end user choice, gatekeepers should not prevent end users from un-installing any software applications on their operating system. It should be possible for the gatekeeper to restrict such un-installation only when such software applications are essential to the functioning of the operating system or the device."
Whoa, great news for camera, calendar or mail apps:
"To ensure contestability, the gatekeeper should furthermore allow the third-party software applications or software application stores to prompt the end user to decide whether that service should become the default and enable that change to be carried out easily"
Critiques of DMA suggested focus and modesty like "They should have focused on 'allow apps to communicate directly with their customers'" or some other minor details. LOfuckingL.
If you oppose the DMA as an indie dev you probably haven't read it. Look: No more private APIs:
"The gatekeepers should, therefore, be required to ensure, free of charge, effective interoperability with, and access for the purposes of interoperability to, the same operating system, hardware or software features that are available or used in the provision of its own complementary and supporting services and hardware." 🎉
What is short sighted, unrealistic or unreasonable about 63?
"Closing an account or un-subscribing should not be made be more complicated than opening an account or subscribing to the same service. Gatekeepers should not demand additional fees when terminating contracts with their end users or business users."
64 is one for Mastodon:
"gatekeepers should therefore ensure, free of charge and upon request, interoperability
with certain basic functionalities of their number-independent interpersonal communications services that they provide to their own end users, to third-party providers of such services"
Again, in a nitty gritty American interpretation you can spin flaws into it. If you read it as it was meant, you know it's the right thing to do.
Winner takes it all? Not in mama EU's house! Inform us of your planned acquisitions, give us all the relevant data and we'll use that data to understand market trends to prevent all your planned predominance tactics.
"gatekeepers should inform the Commission of all of their intended acquisitions, prior to their implementation, of other undertakings providing core platform services or any other services provided within the digital sector or other services that enable the collection of data."
72 dedicated to ensure privacy is one for the books:
"gatekeepers should at least provide an independently audited description of the basis upon which profiling is performed"
'At least'!🤣 I can only imagine the shock on US business people's faces when listening to their lawyers explaining what is in there and what it means for their business in Europe. "We'll just shut everything down in the EU!" Turns to BA: "Right?" BA silently shakes head.
The following paragraphs mostly regulate how the commission can ensure compliance. Okay, so what happens if they still don't comply? 86 gets more concrete:
"Compliance with the obligations imposed by this Regulation should be enforceable by means of fines and periodic penalty payments."
As we know penalties now speak big tech language and use the Billions, not the Millions scale.
Nothing spectacular or unusual in the following paragraphs, still focusing on execution of the strong rules specified before. 101 is somewhat interesting maybe: Who are the experts?
"…each Member State should be represented in the advisory committee and decide on the composition of its delegation. Such delegation can include, inter alia, experts from the competent authorities within the Member States, which hold the relevant expertise for a specific issue presented to the advisory committee."
Coming to a close for the regulations part. 104 explains that if you think that things go wrong you don't need to wait for the parliament to notice and act, but that as a consumer you can enforce your rights:
"Consumers should be entitled to enforce their rights in relation to the obligations imposed on gatekeepers under this Regulation through representative actions"
A good reminder in a time where the law seems to be different for the affluent and the common citizen. 😊
Okay, let's finish this up:
"III.5.7. The gatekeeper shall not require end users to use, or business users to use, to offer, or to interoperate with, an identification service, a web browser engine or a payment service, or technical services that support the provision of payment services, such as payment systems for in-app purchases, of that gatekeeper in the context of services provided by the business users using that gatekeeper’s core platform services"
Just this month, Spotify took action and informed its users that it won't accept payment for its Premium Service through AppStore anymore. They're in an ongoing battle with Apple for years.
News about this didn't make the rounds because this change only affected a small group of Spotify users, as Spotify AppStore subscriptions were only possible in a 2 year window.
Reminder: The discussed regulations only apply in the EU. In other territories... Apple's own rules about payment still apply.
So unless I misunderstand: If you advertise in Der Spiegel via Google, Google has to tell both you and Der Spiegel how much they charge for ads on spiegel.de.
This finally brings light into the ad business black box. These people know exactly what they are doing.
I'd like to go through this faster, but it's just too good to skip. Every paragraph is a little treasure trove of things I always wanted and never dared to hope for. Next person that casually critizizes the DMA gets a smack down.
It keeps getting better: this paragraph, translated into practical reality, tells Apple that they can't cut the 20 Billion Dollar deal with Google to set Google Search as a default, but they have to provide a lot of Search engines the first time a user opens Safari.
Yeah, "let the user decide" for real.
If you know what it is talking about, this paper is so *not* boring. They're walking into a Saloon in a corrupt village in the far West like Clint Eastwood and clean the place up.
"Let the users decide." Users can't decide anything if you let Google and Apple have their go, uncontrolled. User don't know anything. Few know why Google Search is the default search on iPhone and how much this is worth to Google. They pay 1/8th of their *revenue* to Apple.
With all the work they do to protect our privacy, 20B is the price for making them turn a blind eye and let nosy Google be your default search.
https://www.makeuseof.com/why-google-pays-apple-billions-of-dollars/
Thank you for all the compliments. And, really, "Thank *you*!" It's great that you care. I was prepared to die of suffocation reading this and then get NO response.
I am not done yet, but you can already guess that I haven't even begun telling you what my conclusion of this adventure is, it is slowly building in my mind as I read this. I don't know what is coming, but I'm already preparing the encore. Gonna take a break now, and continue tonight.
Read on through a series of compliance articles, there is not much to report that would be of interest for consumers or third parties. It mostly about how investigations are conducted, what obligations gatekeepers have at what stage of the investigation and what happens if they fail.
So what happens if they do not comply? Companies risk a fine of up to 10% of their *global* annual revenue. Worst case, they risk to be shut down partially or completely in the EU.
10% of global turnover is too low a fine? Maybe it's worth risking it and just pay and go on with the malpractice? Think again...
If "a gatekeeper has committed the same or a similar infringement of an obligation" within 8 years, the fine can go up to 20% of global operations. Still lower than Apple's 30%, 😊but at 20% it starts to hurt.
Okay, done, the rest is not *that* interesting for us. It mostly specifies process technicalities and then defines terms like 'active business users' or 'submission of information'... it gets very technical at the end.
Final note: Individual countries cannot oppose findings form the EU court. A fully logical and pragmatic rule but, politically, likely, not a popular declaration.
It's late, now. I'll sleep over it, then I'll try to draw some general conclusions from it.
I'm cleaning it up, turn it into a blog post where I add the conclusion. It'll take some time...
Putting it in the right chronological order would already have been quite a job. Luckily there's Python. (Well, I could have extracted it a bit more cleverly, too 🤣).
I'll try to finish it as soon as possible and add a link to the end of this thread, when it's ready.
I need to find out how come that the repackaging Social Media threads for the blog turns out in days of work. I copied the thread, rearranged, rewrote and rewrote the post at least 30 times until it was halfway publishable. I could have gone for another 30 revisions.
I cannot fully explain it. Part of it is probably that I am completely neurotic when it comes to publishing. Part of it is that the size of information bits changes the overall idea presented in a very essential way.
I never fully thought about what I presume when I consider the length of an online article. It seems too obvious to think about it. The longer the text, the more information was needed to fulfill its promise. Right?
Maybe, the longer the text the more dense it needs to be to not appear chatty. The longer the presumed scroll, the more likely I roll my eyes when getting to the article. I think "This content had better be exceptionally compelling otherwise I'll quit."
You may have read or heard that the ideal paragraph consists of one thought. Clearly, there are various ways to begin and end a thought. One way is to start with a claim or topic sentence, offer examples for your claim, explain how your examples support the claim, repeat the claim in the light of the examples, and build a bridge to the next thought. It is not clear if the ideal paragraph has five, six, seven, a maximum or a minimum number of sentences, as it is not clear what one thought is, where or how it begins, and where or how it ends. On second thought, one could as well argue that every sentence consists of one thought.
Version 6. Deleted version 5 because that was just a sleepy early morning whatever. You can se now that I've crossed the line where I need to change the story and structure all the time and I am starting to have fun adding color and spice to the core idea.
Unless this is a late evening sleepy whatever, it may be ready for publishing. Will sleep it over then decide. I really start liking it now. It makes me smile; it's fun to work on. The Radiohead reference is still a bit forced though.
Some afterthoughts. I might add an intro, footnotes to this thread and some background info on where I got this 5x5 stuff from.
Thinking about it, I realize that this "follow the grid, break the grid, follow the grid break it" type of writing mirrors the design process. I like that a lot, but, likely it's not for everyone. The older I get the less I can be bothered with popularity.
@reichenstein The idea is seductive, but I disagree with any suggestion that you can write to a formula. The idea of limiting the number of paragraphs, sentences, or words, ends up making writing feel sterile.
Some paragraphs should be short; some should be long. It is variety that makes writing interesting, not arbitrary rules.
@mcelhearn That's what the articles says, as well.
"Use numbers to keep things tidy—while you keep writing music. Find your rhythm by letting it find you. Drum up shorter and longer sentences just as you feel."
Was more explicit in previous versions. "Use the grid then break it." I took it out in the end because the word "grid" always felt alien. I even had use it break it use it break it again.... Using music and rhythm as the corrector felt closer to how it actually works.
@garychou yeah, f.i.
5 characters per word: https://www.wyliecomm.com/2021/11/whats-the-best-length-of-a-word-online/
5 sentences per paragraph: https://wordcounter.net/blog/2015/11/30/10988_how-many-sentences-in-a-paragraph.html
5 words per sentence is very ambitious. Average is more like 15 words. 5 words is Hemingway style, but obviously you need to vary the length to make good music.
@reichenstein I quite like it.
What I like even more is the thread of drafts that led to it. Not as pretty, but far more educational. At least for me. 😄🙏
@b3n I started trusting the process. If the idea is strong and survives rewriting it will be fun at some point.
I've been writing a longer piece for quite some time now. It is still nowhere near the fun phase, but the idea holds and holds. So eventually the pain will stop.
@reichenstein I guess part of my pain comes from all the many other ideas I cannot write about at the same time ;)
But thank you! I agree that just going with the flow and trusting the process is a big part of the solution. The other is taking more time for writing 🤭
@reichenstein Yesterday, I wondered why you suddenly focused about the outer form (5x5x5) instead of the inner, the focus on a thought.
But I do guess that this helps: looking at long sentences, and at their compounds (paragraphs), a writer could try to cut and restructure everything. To do that, he needs to *also* look at the outer form, the length.
@reichenstein Before you mused about the outer 5x5x5x5 form of things, I read and processed your links, and that gave me fodder for a very substantial 'layer' in my work -- that ideally 1 thought correspond to each sentence *and* each paragraph and, I'd extrapolate, also each article and book.
I'm currently exploring more practical tips derived from this in the open in our small community[1] -- this really was the theme of the week. Thank you for this thread!
@ctietze I came up with a hierarchy that I found somewhat arbitrary at first:
Idea > Argument > Thought
Corresponding to
Text > Paragraph > Sentence
But it persisted through a series of revisions and helped me getting a clearer understanding of what was a vague intuition at first. I think I'll stick to it.
@reichenstein I can see how larger/smaller compounds of mind-things would correspond to larger/smaller parts of a text.
The label "argument" bugs me, though, because that's a rather specific structure -- With my Uni background I do think of arguments first when I think of structured sentence clusters, but there's also non-argumentative model-building, for example 🤔
@ctietze Not sure I understand your hesitance. Argument (paragraph: group of sentences, meaning: thoughts) is part of an argumentation (the whole text).
Your description of arguments (paragraphs) as a sentence cluster sounds fine to me and my Uni degree. (What is yours?)
You can't pitch one useage of a word against another, expecting perfection, because you'll always find conflicting use cases.
@ctietze my own hesitation was: "In a way, everything, text, paragraph and sentence is thought or idea, depending how you use those words, so creating a hierarchy between them would be arbitrary and hard to understand and accept. And there is nothing in between idea and thought. Wait... What about an argument? Am I making up too much stuff now?"
I still cautiously like and understand it, well knowing I made it up.
@reichenstein I would think the same rule applies: one thought per article.
It's just that on each level of abstraction, you can produce more complex thoughts.
Which each require smaller, supportive thoughts to be intelligible.
Thought about this some more. The patternI see:
1) Limit a unit to one thought.
2) Complex units represent complex thoughts.
3) Thoughts decompose into sub-thoughts. Simple thoughts compose into new abstractions.
4) Understanding requires decomposition (to follow along with something new), insight requires composition and abstraction (to grasp a complex idea as one idea)
That escalated into a couple of cross-connection and #atomicity in the #zettelkasten
It appears almost complete, but I'll be putting this out into the open for feedback:
Oliver Reichenstein
Kirk McElhearn
Gary Chou
Cole Peters
Yoshi Carroll
Ben
Christian Tietze
Daniel 🏳️🌈 Jilg 
Erik Mogensen
Moritz Z.
Thomas Depierre
Bastian
zıəs uɐɟəʇs ✪
Rafael “Akai” Masoni
Dale Nash
Jürgen Siebert 