greattsclerouse

thanks. I've read different stuff pretty much every where I go so thought it might be worthwhile to ask this community about their thoughts. appreciate your input.

In defining and discussing important notions of social networks and platforms, I will borrow some ideas from the infamous Steve Yegge , and the one and only Martin Kleppmann for ideas related to “local first” data, as well as, to help the user to better understand some of the technical notions of a truly distrubuted data platform (Martin literally wrote the book on this!).

The discussion from here is going to get somewhat technical but I will do my best to cut through the nerk talk and get to the point. Based on our discussion in Part I, we know that there have been a long history of failed social platforms — many of which have tried to solve certain aspects of this problem. Core components that need to be understood well are notions of Identity Management, Anonymity, Security & some good old CAP theory!

Data nerds say things like, “In God We Trust, All Others Must Bring Data”. This is something I think most reasonable minded folks can get behind — especially when it comes to establishing a common truth! But, how would anyone know if I decided I was going to make everything up?

Use Matrix or KeyBase and self host. Im shocked how much people still trust these shady companies.

I think it pays a lot of dividends to understand some of the history of the past so that we can learn from them as we look to build something better for the future, so I’ve decided to publish some of my thoughts on this topic and also share some of my findings with the community so that nobody has to go down this massive rabbit hole again in the near term.

In 1000 years we will have ascended :) but if not, I hope to be chilling on Risa somewhere!

In all seriousness, in the future I suspect we will use a combination of these types of capabilities but they will all be linked together in a way that prevents totalitarian control of speach.

I think some of the best technologies today are keybase.io matrix.org briarproject.org and community.mattermost.com

My view is that the next emerging social network will be based on a secure and self hostable compute fabric that leverages zero trust compute and secure compute, data sharing, etc. I think Urbit had the right direction but went crazy and lost the plot.

I'm working on this problem personally and I hope to be in alpha / beta testing for certain aspects of the platform by end of year :) Feel free to reach out if you want some additional details.

I think it pays a lot of dividends to understand some of the history of the past so that we can learn from them as we look to build something better for the future, so I’ve decided to publish some of my thoughts on this topic and also share some of my findings with the community so that nobody has to go down this massive rabbit hole again in the near term.

History of Social Networking: https://october.substack.com/p/part-i-the-next-great-online-... https://october.substack.com/p/part-ii-usenet-a-genuinely-pu... https://october.substack.com/p/part-iii-slashdot-and-advogat... https://october.substack.com/p/yik-yak-secret-and-the-minima... https://october.substack.com/p/the-rise-and-demise-of-branch

One of the best readings on this topic is here: almost a library about the history of the net and Netizens http://www.columbia.edu/~rh120/

This is a pretty good overview of current state of alternatives: https://medium.com/hackernoon/so-you-want-to-leave-facebook-...

good write-up on layering complex systems and why this is a hard thing to solve for: https://jods.mitpress.mit.edu/pub/issue3-brand/release/2

MIT/Zitrain: https://dash.harvard.edu/bitstream/handle/1/4455262/Zittrain...

Activity-Pub/SOLID/SSB/Matrix/others: In my view, the links below offer us a real foundation for the future of the internet that can be used together to build a better social fabric

https://solidproject.org/about https://activitypub.rocks https://matrix.org https://briarproject.org https://scuttlebot.io https://joinmastodon.org https://keybase.io/blog/keybase-proofs-for-mastodon-and-ever... https://keybase.io

But even as we find our way with the technology, we are repeating the mistakes that we should have learned from in the past: https://blog.joinmastodon.org/2018/07/cage-the-mastodon/ https://blog.joinmastodon.org/2019/07/statement-on-gabs-fork... https://www.theverge.com/2019/7/12/20691957/mastodon-decentr...

"At the core of the open source ethos is the idea of liberty. Open source is about inverting power structures and creating access and opportunities for everyone." — GitHub employees' letter to GitHub

And Now, it’s about time to discuss the net bill of rights that our future social network should adhere to: http://www.columbia.edu/~hauben/Book_Anniversary/Appendix.pd...

Here’s the thing, once you read all of this stuff you begin to get the idea that this problem is actually bigger than just a social networking platform—there are lots of these and yet as every Netizen can see, the problems still exist. The fabric of our problems isn’t so much that we don’t have the technology to solve the problems of our times—the issue is that it almost requires that we burn down the existing system in order to make the world a better place! BUT WAIT!!!

Before we think about burning it all down, I think there are a few things that we should consider first and, of course, there will be lots more links and examples along the way! One thing that I want to emphasize as we go on this journey, is that form must fit function—up until now, I think social networking and it’s controlled/heavy-handed evolutionary path has left a rather sour taste in the mouths of many. But, this conclusion is both a result of the evolution of social networking, but also the result of a change in the way that we use technology and our relationship to our devices/machines.

I'll be publishing a series of blog posts on this very soon. Hope this helps!

I'd recommend starting here: https://ocw.mit.edu/courses/audio-video-courses/#physics

In my experience these are some of the best online courses you can watch to learn physics. Personally, I would look into the trying to watch the lectures from Walter Lewin--Walter is a fantastic orator and has a really great mad-scientist persona that is really captivating. Some additional archived lectures can be found here: http://dspace.mit.edu/handle/1721.1/34001 and here: https://ocw.mit.edu/courses/physics/archived-physics-courses...

I got my minor in physics from NYU many many moons ago (yes I'm getting old), but I found that the MIT lectures and OCW materials went way beyond the NYU coursework in both breadth and depth. I watched these lectures and worked through the lecture notes & assignments for Physics I, II, III, Quantum I, II, and several others in addition to digging into the Mathematics lectures / content. I found this material to be the most helpful out there. I'll also point out that I emailed the professors (Lewin, and others) and was pleased to receive a warm and helpful response on several occasions. I hope these are as helpful for your learning as they were for mine.

Once, you are able to complete the video lectures here, OCW has a massive amount of content for some of the more advanced courses that aren't in video format. In my experience, going through these video lectures and some of the mathematics lectures should set you up well to be able to comprehend even the most advanced content across field theory and string theory.

Cheers!

I can't help but think that this isn't a very well thought out criticism. I'll go through each of them and spell out why I think this critique doesn't hold any weight. Let me start by saying that I use IBM Cloud and I've switched to IBM Cloud from AWS.

1. There are several IBM Cloud pages—and they’re all talking about different things --it's pretty easy to find the right IBM Cloud address and honestly, anyone that can't find https://cloud.ibm.com probably isn't qualified to provide a review.

2. During signup, the user experience is… lacking --If you can't figure out how to enter a password into a box then it's probably best to go back to elementary school or kindergarten to work on the basics. Pretty much any autogenerated password gets accepted--I have 3 separate IBM Cloud accounts and I've never had a single issue figuring out how to set my password when registering. if 20% of your criticism of IBM's cloud offering is based on this issue then you're already grasping at straws.

3. Privacy notices—who needs ‘em? I'm not sure exactly what the criticism is here other than this notice wasn't updated after CA adopted new legislation. You didn't state why they are non-compliant or what your actual issue is. I'd suggest you do a better job of spelling out what the actual issue is vs. making weak arguments. I've now addressed 60% of your argument and nothing actually has to do with the cloud offering--it seems you couldn't figure out how to actually set up a cloud account based on your own arguments--SAD!

4. The onboarding experience is. . .also lacking: "Whoever designed this needs to be put on a PIP. Services are sorted alphabetically, and there’s no guided onboarding. Language like this—Want to spin up a VM / storage bucket / toilet in which to drown yourself? Start here!—is always useful. Next, we learn that “Lite” and free are two different things. We also are given two compute options: Container Registry Cloud Foundry

Either IBM or I have no clue what a “Cloud Foundry” is; it could really go either way and at this point I’m too scared to ask."

--Based on your lack of cohesive arguments so far, I'm going to guess that you actually have no semblance of an idea about what you are talking about. If you're a user of Cloud Foundry PaaS then I suspect you don't know much about it. IBM has been a provider of Cloud Foundry PaaS for several years and it's actually quite simple to use and also fairly well documented. Not doing your homework and making meritless judgements about a cloud provider isn't winning you friends or helping you to influence people. I'd suggest you spend some time getting familiar with the site and actually trying to explore the offerings instead of making ridiculous observations about a platform you have obviously never used. By comparison, I find that AWS interface to be very cluttered and not very well laid out--but maybe that's just me. I quite like the IBM Carbon Design interface libraries... so much so that I use them in my own web/interface development projects--they are very clean and offer a well thought out look and feel. 4/5 of your complains have held no water and do not actually address the cloud offering--I can't help but think you are in some way biased or unqualified to offer a review at this point. Also, if you couldn't figure out that the "Lite" plan is free to use then I think you should probably revisit my critique from comment number 2.

5. As for using the actual services? Yikes. --at this point you've fully disgraced your self and rendered your own critique to be worthless. One of the key elements of the IBM Cloud is that it's extremely configurable. You can spin up any number of linux boxes with Ubuntu, Debian, CentOS, RedHat or bring your own image. Not only is IBM more configurable and more aligned to open source, but it's also more performant across number of key features--they dominate lower, general and high intensity workload performance benchmarks for IOPS, Throughput and HA/Durability across block and file storage offerings. In terms of performance for compute, if you use a standard unix bench script (https://github.com/kdlucas/byte-unixbench.git) you will find that the price to performance comparison is also favorable to IBM's VM's vs their peers. I've not run these tests for a year or so but the last I checked IBM had a better price to performance value vs both AWS and MSFT (I didn't check for google) for both bare metal and virtual servers when testing on CentOS. Nothing you've said really holds much weight and pretty much every observation you've made here was cursory. I think it would do you well to spend some additional time to actually understand how the offering differs and why it may be a better product or not for your particular use case.

I don't expect that everyone will fall in love with IBM Cloud but I think they have a pretty solid offering and it continues to improve at a rapid pace. I moved away from AWS a few years back and while I was happy with my AWS IaaS, I find that I like the IBM offering more. Just my $0.02.

I think you should. People are always attracted to new shiny objects and most folks don't understand the complexities of blockchain and AI. While I think it's worthwhile to educate your management team, I think the best way to do this is to align their blockchain / AI aspirations to core competencies for your company.

Deploying technology for technology's sake is a problem I see often--this comes down to people's attraction to new shiny objects; and it's extremely important to make sure that this tendency doesn't have a negative impact on the core business. I work for a leading AI and blockchain behemoth and I see this all the time--quite honestly it benefits my firm's sales but it's a massive problem for shareholders.

My advice for most leadership teams is to focus no your core business. Blockchain and AI are all the rage right now, but they are complicated to implement--especially in production! AI systems are very brittle and AI-Ops is not simple. There is a great paper from google titled "Machine Learning, the high interest credit card of technical debt", I suggest you have a read.

In terms of your company, if you believe these solutions will improve your core competencies and provide a competitive advantage then you should absolutely invest in these capabilities. However, it's really important to get the basics right before moving onto fancy new technologies that aren't well understood.

There is a great meme/cartoon that I like to show folks that depicts a teacher and a few students in a class where the teacher is demonstrating basic arithmetic on the blackboard and one of the students raises their hand and asks "when do we get to machine learning." I like to use this cartoon when I discuss these types of solutions because it helps me to put things in perspective.