On being faceless
Update: Alright, I’m back in. (After trying the same recovery mechanisms that didn’t work before, but suddenly did work this afternoon.) Thanks also to the Facebook employee who emailed offering to help. Now I just need to decide the harder question of whether I want to be back in!
So I’ve been locked out of Facebook and Messenger, possibly forever. It started yesterday morning, when Facebook went down for the entire world. Now it’s back up for most people, but I can’t get in—neither with passwords (none of which work), nor with text messages to my phone (my phone doesn’t receive them for some reason). As a last-ditch measure, I submitted my driver’s license into a Facebook black hole from which I don’t expect to hear back.
Incidentally, this sort of thing is why, 25 years ago, I became a theoretical rather than applied computer scientist. Even before you get to any serious software engineering, the applied part of computing involves a neverending struggle to make machines do what you need them to do—get a document to print, a website to load, a software package to install—in ways that are harrowing and not the slightest bit intellectually interesting. You learn, not about the nature of reality, but only about the terrible design decisions of other people. I might as well be a 90-year-old grandpa with such things, and if I didn’t have the excuse of being a theorist, that fact would constantly humiliate me before my colleagues.
Anyway, maybe some Facebook employee will see this post and decide to let me back in. Otherwise, it feels like a large part of my life has been cut away forever—but maybe that’s good, like cutting away a malignant tumor. Maybe, even if I am let back in, I should refrain from returning, or at least severely limit the time I spend there.
The truth is that, over the past eight years or so, I let more and more of my online activity shift from this blog to Facebook. Partly that’s because (as many others have lamented) the Golden Age of Blogs came to an end, with intellectual exploration and good-faith debate replaced by trolling, sniping, impersonation, and constant attempts to dox opponents and ruin their lives. As a result, more and more ideas for new blog posts stayed in my drafts folder—they always needed just one more revision to fortify them against inevitable attack, and then that one more revision never happened. It was simply more comfortable to post my ideas on Facebook, where the feedback came from friends and colleagues using their real names, and where any mistakes I made would be contained. But, on the reflection that comes from being locked out, maybe Facebook was simply a trap. What I have neither the intellectual courage to say in public, nor the occasion to say over dinner with real-life friends and family and colleagues, maybe I should teach myself not to say at all.
Follow
Comment #1 March 6th, 2024 at 9:54 am
Related to your point about theory vs practice, reading about the C64 days always makes me nostalgic even though I wasn’t alive then. You could actually understand that machine pretty much down to the metal. When I grew up and was really interested in programming, I had to grapple with giant awful libraries and APIs from the beginning to do anything interesting, and understanding the whole stack was hopeless (I still did not become a theorist due to lack of smarts). Maybe we should revive the old school computers for education.
Comment #2 March 6th, 2024 at 10:29 am
Bad news for Scott (maybe), but good news for those of us who want to see more posts on this blog!
How about doing a few posts with comments turned off? That way you can get back into the groove of “real blogging” without dealing with trolls, etc.
Comment #3 March 6th, 2024 at 11:38 am
Hi Scott, I work at Meta. I just sent you an email and hopefully we’ll be able to get this resolved.
Comment #4 March 6th, 2024 at 12:21 pm
“What I have neither the intellectual courage to say in public, nor the occasion to say over dinner with real-life friends and family and colleagues, maybe I should teach myself not to say at all.”
I of course respect your choice to share whatever you want to wherever you wish.
But for myself, I’ve found that when I don’t have the opportunity to share what I think with other people, it limits my ability to think things through at all. Do you experience that?
Comment #5 March 6th, 2024 at 12:36 pm
I find it always amazing how we entrust and build such a significant part of our identity, our history, and our social network to private systems and platforms from which we can be irrevocably excluded due to a design flaw, a bug, by whim, or a combination of the aforementioned reasons.
One of the things that excites me about distributed systems like blockchains is the possibility to build social networks and identity representation mechanisms from which, by design, one cannot be excluded and where identity is safeguarded through cryptography and mechanisms like social recovery (like https://vitalik.eth.link/general/2021/01/11/recovery.html)
Comment #6 March 6th, 2024 at 1:46 pm
“Partly that’s because (as many others have lamented) the Golden Age of Blogs came to an end,…”
No Scott, this blog remains ever golden to aficionados of your prolific prose, astute insights, and the sheer joy of engagement with you. Here, the golden era persistently thrives. Thank you.
Comment #7 March 6th, 2024 at 2:12 pm
Francesco B #5
A centralized service with transparent workings and public control will be orders of magnitude more efficient than any blockchain system. Any democratic government is supposed to work like that.
Just a simple example of how it could be done. A service like Facebook could have a public certificate with which it can sign interactions with its users. So, it can’t just erase user profile and his posts like nothing happened – because a user still possesses signed proofs of interactions and authority over an account (like a receipt), which would be a proof in a court. In theory, Facebook could sign and share with the user the local copy of all of its data and posts, so in the case of total destruction of Facebook servers everything could be recovered from local user caches – in the untampered form.
Of course, in today’s world capitalistic megalomaniacs would never implement any such system. But it’s not the problem of centralization, it’s the problem with people who control it.
Comment #8 March 6th, 2024 at 6:23 pm
As someone who hasn’t gone on facebook in years, I didn’t know that you were active on it, Scott. It’s funny that this is how I found out about it, given the small but real chance that you won’t be back.
Comment #9 March 6th, 2024 at 6:44 pm
> You learn, not about the nature of reality, but only about the terrible design decisions of other people.
As opposed to being a theoretician, where you learn about the terrible design decisions in the creation of our universe.
Comment #10 March 7th, 2024 at 6:37 am
Greg Baker #9
Where are all these terrible “design desicions”?
I don’t see many…
QM & QFTs work very well in their domain.
Ok, perhaps there’s a problem with decoherence in quantum computing etc., but that’s unavoidable! Could have been otherwise in a world so complicated with myriads of interactions everywhere all the time?
As for Special and General Relativity, these are working almost perfectly well ( and moreover GR is a Work of Art aesthetically!)
Some people worry about singularities (especially timelike “naked” Singularities that violate CC), but personally I’m Ok with them, I find them kinda cool…
There are many things unknown yet (invisible matter, quantum gravity etc), but these are due to our own ignorance, not a fault of Nature.
We don’t know many things about emergence, the origins of life etc, but, well, only a few centuries of science behind us…
Perhaps the “Theory of Everything” when it’ll be found will be the ultimate masterpiece!
Comment #11 March 7th, 2024 at 6:49 am
“ What I have neither the intellectual courage to say in public, nor the occasion to say over dinner with real-life friends and family and colleagues, maybe I should teach myself not to say at all.”
This makes me sad.
Comment #12 March 7th, 2024 at 8:38 am
“The truth is that, over the past eight years or so, I let more and more of my online activity shift from this blog to Facebook. Partly that’s because (as many others have lamented) the Golden Age of Blogs came to an end, with intellectual exploration and good-faith debate replaced by trolling, sniping, impersonation, and constant attempts to dox opponents and ruin their lives.”
A tangent related to “sniping”: I feel that pointing out problems with peoples’ projects, ideas and reasoning comes much more naturally to me than, well, accomplishing things myself. I am torn on what to do about it, and how much space I should give to these doubts both in my own head and in discussions: On the one hand, 90 % of everything is crap, so if someone presents a project that takes up the attention and funding of people, I feel that it should be criticized. (For example, consider all the software complexity that introduces more problems than it solves, and can’t realistically be opted out of). On the other hand, I would like to be the person who accomplishes stuff more than being the naysayer. And ultimately progress comes from those who do and not from those who criticize, and those who do are not even those with the most realistic view of their projects’ feasibility and merits.
Comment #13 March 7th, 2024 at 10:22 am
Greg #9 depending how theoretical you want to be you can always “step outside” our universe; switch off gravity, allow closed-timelike curves, or whatever …
Scott – the number of us who like you and value your usually original and interesting take on the world surely outnumber the trolls etc … at least on this blog. But I get it.
Comment #14 March 7th, 2024 at 10:40 am
@ Danylo #7,
While I agree with what you wrote, I wonder what your opinion is of e.g. the disapora* project. I think it strikes a healthy balance between “totally centralized by a closed-source private for-profit entity that has total editorial control” and “completely decentralized Blockchain-based data storage”. Its pods are essentially the same model as Mastodon, but it was released already in 2010 — well before these ideas of decentralization were in vogue.
I think the model and software is good. I also recognize that diaspora* has never grown, essentially coming before its time, and with no event like Musk’s Twitter takeover to precipitate an actual exodus and (pardon the pun) diaspora.
Perhaps for the day when Jeff Bezos buys Meta?
Comment #15 March 7th, 2024 at 12:20 pm
haha! I just requested FB friends. FB should be like seeing Scott through a wormhole.
Comment #16 March 7th, 2024 at 1:51 pm
I understand your problem, Scott, and empathize. Some months ago I accidentally got locked out of Facebook and was unable to get back in. None of the recovery methods worked. I wasn’t able to get back in until I upgraded my internet service to a higher speed. Then I was able to transmit a photo-ID that got me back in.
But I was not a happy camper. On the one had, I do a lot on Facebook that I enjoy and value, though I keep most of my intellectual work on my blog. But the hassles it puts me through from time to time are really annoying. If the metaverse of the future turns out to be at all worthwhile, I do not want it to be run by those jokers. Never.
Comment #17 March 7th, 2024 at 6:53 pm
It sadden me TREMENDOUSLY to read you saying
> The truth is that, over the past eight years or so, I let more and more of my
> online activity shift from this blog to Facebook. Partly that’s because (as
> many others have lamented) the Golden Age of Blogs came to an end,
> with intellectual exploration and good-faith debate replaced by trolling,
> sniping, impersonation, and constant attempts to dox opponents and ruin their lives.
The very thing you lament as the *cause* of you needing to use FB, is instead the *effect* of people using FB (and other social networks).
I am old enough to have a “physical social network” and I remember when we could just agree to disagree. Now the FB and friends have one goal: keeping you hooked on their apps and websites. What do they do for that purpose? They hire psychologists to know how to make it hard for people to stop using their crapps and they have realized that if you like something, feeding you “more-of-the-same” (among many other tricks) keep you “engaged”. So where have we ended? Each one is fed more and more of the same, become extremely insular first, and very separate and often extremist. In the end, as a population at large we can’t even agree to disagree anymore. And we end with the congress we have, and with all other malaise we have including the ones you lament about your commenters on your blog.
For this very reason I have quit the few social networks I had accounts on, and I would not create any one more, even at gunpoint.
I saw this coming, but I don’t say it as something to be proud of, I say it only to share that the 😭 started a while ago and it deepens every day, especially when I see a post like this.
Comment #18 March 7th, 2024 at 7:08 pm
I have a genuine question for any “professional commercial software product debuggers” out there (I assume that there’s a better term for that concept within industry).
Is it in fact true that fixing these kind of crashes is indeed “not the slightest bit intellectually interesting”? Is it always just a question of finding the one “terrible design decision of other people” out of 10 million lines of code, but the problem itself usually turns out to be pretty trivial and uninteresting once located? (Incidentally, if true, then this would be a nice illustration of the distinction between P and NP!) Or is it sometimes a genuinely interesting intellectual puzzle to track down the problem, which requires understanding the surprising behavior of a complex emergent system at a deep level?
In my experience, debugging my own code usually lies somewhere in the middle of that spectrum. But the situation may be different for professional commercial code created by many people.
Comment #19 March 7th, 2024 at 7:58 pm
Scott,
“the applied part of computing involves a neverending struggle to make machines do what you need them to do”
This saga continues with Chatbots…for the same reason of divergence between theory and empirical. Ironically the most intelligent species cannot do any better about intelligence than empirical post-facto analysis. Maybe its a fundamental limit of the universe, like speed of light ?
Comment #20 March 7th, 2024 at 9:11 pm
I had to remove the Facebook app. Was spending way too much time getting bogged down in pointless arguments trying to persuade people by using facts. That platform is ground-zero for flat Earth, anti-vax, anti-science, anti-nuclear, anti-intellect, anti-police, anti-capitalism and just general nastiness.
Comment #21 March 7th, 2024 at 10:39 pm
@Ted #18 What’s interesting to one person can be boring to another. And what’s trivial for Scott might be intellectually stimulating for mere mortals 🙂
So, mere mortal here. Salt, mileage, etc.
FWIW I’d say both types occur. Sometimes it truly is just a pointless nightmare – matching GPU drivers, cudnn, tensorflow, numpy and opencv versions so that they’re all compatible comes to mind. But sometimes, even if it’s all about “finding the poor design choice”, it’s not always trivial to fix it while not introducing further complications. Hey, we love problems with constraints! Also, a “poor design choice” is not always a static concept. Perhaps there was a reason for that design. A decent one. It could’ve been faulty, or circumstances could change, but understanding what could have led to that design is at times useful for further development.
And all that aside, sometimes stuff is just funny! Very early on in my life, I was parsing some cumbersomely named images with formats like timestamp_idx1_…_image_…_…_. And one couldn’t say the names were crazy or useless. Due to a combination of youth, a lack of firm “contracts” regarding the names, and time constraints, my parsing was fairly hard-coded and crude.
And one time… mid-parsing… an image of a clown kept appearing. Out of nowhere. These were _not_ images of clowns.
Turns out, the word “image” in Matlab is reserved for a specific “test-case” image of a clown. And under some conditions, I was extracting the string “image” rather than some index.
Were there multiple bad design choices involved? Sure – mine, the images namers’, Matlab’s. I still laughed for an hour after discovering the bug.
Comment #22 March 8th, 2024 at 1:21 am
There are the last romantics of the Internet: the “IndieWEB” movement. Here is the piece from their manifest article:
—-
…the web and using it for easy communication has become fraught with fragmentation and walls that often make communicating online far more difficult than it should be. Wouldn’t it be better if you had a single website that represented you online and through which you could easily communicate with everyone?
By analogy consider the telephone system which, just like the internet, consists of wires and hardware to access the network. Every user on the network has their own phone and phone number. What would it be like if AT&T users could only speak to other AT&T users and needed another separate phone, account, and phone number to speak to friends and family on Verizon and yet another to talk to friends on Sprint? To a great extent, this is what the internet has evolved to become with monopolistic, for-profit, corporate services like Facebook/Instagram, Twitter, LinkedIn, and all the rest.
Is there a better and more robust solution than these multitudes of social media sites which all come with their own onerous terms of service, limitations on your creativity, reach, ownership, and control of your online identity?
A growing number of people on the web are sure there is and they’re working together in an open yet coordinated way to improve the democratized nature of the decentralized internet. This movement is known as the IndieWeb.
…
Comment #23 March 8th, 2024 at 1:56 am
Del # 17
I think you’re right that popular social media are mostly responsible for the growing hostility and fanaticism on the internet (not only there, also they have visible negative impact on many people’s social lives, their extreme political views and choices, the inclination of some of them towards conspiracy theories nonsense , perhaps even worsening mental/ psychiatric issues…)
Unfortunately, they should have known better back then when the whole thing started growing ( without any serious preparation or consideration of the possible dangers, as it’s usually happens with such things…
… the same situation happens nowadays with the rapid growing of the AI / chatbots uncontrollable production and use without much consideration of the potential dangers for people’s everyday lives and their mental health, the impact on economy etc…
Given the already horrible impact that some of the social media had on people’s attitudes on the internet and their acts and choices in the real world, the situation only worse will be in the near future…
…and it is too late for tears and worries…
Comment #24 March 8th, 2024 at 2:21 am
Danylo Yakymenko #7 (and Alex Meiburg #14) Yes, IF there were a fully trustable, unhackable, uncorruptible single entity, decentralization might be less of a thing. It can also be argued that democratic government has its fair bit of decentralization (checks and balances, separation of powers etc) with respect to more “efficient” forms of governance such as dictatorship.
And as far as I understand, making Facebook sign all the interactions wouldn’t solve Scott’s problem: if Facebook wants to freeze your account, they can always simply ignore your requests to post, and you wouldn’t be able to prove they received them. And also, appealing to a court may be less efficient than even blockchains 🙂
Anyways, there are already platforms that implement your interesting proposal at scale: in Farcaster for instance users have “decentralized”, self-sovereign identities stored on a blockchain, posts are kept on a distributed storage that anyone can keep a copy of, and only the client and UI are centralized, but anyone can implement an alternative. The UX is on par with classical social networks, and the vibes are far less toxic (at least for now…)
Comment #25 March 8th, 2024 at 3:26 am
Alex Meiburg #14
As for diaspora, Mastodon and other fediverse projects I support their enthusiasm, of course. But I think people still need incentives (and liabilities) to run server nodes and care about them. Without that it’s impossible to compete with for-mega-profit companies. There must be a balance. We’re already witnessing how OpenAI excuses why they can’t be really non-profit and “open”.
Comment #26 March 8th, 2024 at 4:15 am
Ted #18
Indeed, the complexity of our computer systems is approaching a point where new people will find no difference between studying them and the laws of universe. More and more it becomes unfeasible for one man to understand every aspect of how e.g. an operating system works. Here [1] is a comment on how the development of ZFS and BTRFS, two major modern file systems, is going on. “Everybody is guessing” is what happens when there is a subtle bug.
I’m scared about what we can expect in the future when we lock up ourselves and our critical infrastructure in systems we don’t fully understand anymore. Putting this burden on an AI is even more worrying, in my opinion.
[1] https://www.phoronix.com/forums/forum/phoronix/general-discussion/1424461-openzfs-is-still-battling-a-data-corruption-issue/page2#post1424496
Comment #27 March 8th, 2024 at 6:24 am
Ted #18
I usually write software for proof of concept or prototype applications, where production-level resilience isn’t a concern, but achieving an end-to-end functioning system is crucial. Debugging, when approached with the right attitude, can be a very satisfying endeavor for me. It’s surprising how many ways something that seems obvious and simple can go wrong. Often, you encounter a problem that seems to completely elude your understanding, yet you know it’s possible to resolve it (meaning, you can’t blame incompleteness or some deep unknown unknown). This can make it especially frustrating (knowing there’s a solution), but it also prevents you from giving up. And when you finally solve it, it feels awesome! You’re aware that you haven’t proven $P \neq NP$, but you feel pretty good nonetheless, and you also got the thing working! Sometimes, you have to delve deeper into the stack, which requires time and effort, but this rewards you with a deeper understanding of how things work. For an engineer like me, it might just hit the sweet spot. Paul Graham also has a stance on that: https://paulgraham.com/work.html
Comment #28 March 8th, 2024 at 11:42 am
Glad you got your account back.
Re posting, though, I would suggest the option of posting your stuff here anyway without fortifying it against every possible attack, and then simply ignoring most of the attacks, and deleting the rude or bad-faith ones.
It is OK to get attacked online. It doesn’t actually hurt you unless you let it. And any great work that brings notice will be attacked. If you’re not catching flak, you’re not over the target.
Comment #29 March 8th, 2024 at 3:54 pm
When I was younger, my only connection to the wider intellectual world was through those that could easily have had all the stimulating discussions they wanted with their peers and friends, but inexplicably chose to post essays on the internet instead, exposing themselves to bad faith arguments masquerading as censure for no good reason. There are others who have figured out how to profit from the negative psychology on the internet, but they’re to a one terrible – if eventually the only essayists left on the open web are those with an explicable reason to be there… a great thing will be lost.
Comment #30 March 8th, 2024 at 6:30 pm
Where does bad code come from?
Comment #31 March 8th, 2024 at 8:18 pm
@ Ted # 18 – I’m a professional programmer of long experience. I haven’t worked on Facebook-scale operations, but I have debugged some decently large problems in business systems. Yes, in my experience, the fix itself indeed typically isn’t “intellectually interesting”. The problem is very often something along the lines of “There’s a memory leak in this subroutine down deep in the innards – make sure it frees the memory used under all failures”. Or “Sometimes a database connection is getting lost under this set of circumstances, and that’s why over time there’s no database connections available, then the server stops serving”. The “intellectual puzzle” isn’t at all anything of “surprising behavior of a complex emergent system”. At best, it’s more like “Which one of all many the assumptions about how this system works, is subtly wrong, because the assumed behavior isn’t the observed behavior” (more parts in the system means more things which can go wrong!). In a way, it’s a very scaled-up version of technical support – “Why doesn’t my computer turn on when I press the power button?”. Helping someone with their computer support problem may be very personally gratifying in a sense. But “because it’s not plugged in” isn’t intellectually interesting in itself.
Comment #32 March 9th, 2024 at 9:01 pm
My first job in the long ago was as an “Applied Computer Programmer” for a large engineering organization. Writing and debugging code seems idyllic in memory–thinking and solving logical problems all the work-day. They were not huge systems though. I recall someone who did program a large system announcing, “All known bugs of this system have been fixed. All future bugs will be new ones.”
After seeing that movie about Zuckerberg, I resolved never to use Facebook. Yes, I take a lot of quixotic stands.
Comment #33 March 11th, 2024 at 9:20 am
Does being on Facebook help your research? does it put you in better contact with fellow researchers? If so then here is a dichotomy.
1) being on facebook is GOOD for your research since you are in contact with others.
2) being on facebook is BAD for your research since you are also getting into discussions on other topics, some of which are stupid. Or at least not intellectually stimulating (on my blog there is a debate about whether or not `combinatorist’ is a word).
Comment #34 March 12th, 2024 at 9:36 am
Many people go to industry because it pays better than academia, but also because they feel it is easier to have an impact on people’s lives.
Academia requires a different mindset.
No one likes debugging I think. It is just part of creating real things, rather than proving theorems or creating prototypes.
Most of the problems are not other people’s bad choices but are the result of work that needs to be completed under resource and time constraints.
Having a safe place to share and discuss and be ourselves is important. It is just that Facebook might not be that place.
Comment #35 March 12th, 2024 at 12:38 pm
No, Scott, no!
*Maths* is not general enough.
It even simply can’t *be*.
It always is *too* specific for the uses for which the Platonists *used* to advocate it.
Which isn’t even the case any more with the maths `advocastists’, to coin a new word.
–Ajit