5 ms·

I just want to second Andrew's list. I'm a developer with probably 1/100 as much money as he has, but this is the first list I've seen in a long time that ment

I just want to second Andrew's list. I'm a developer with probably 1/100 as much money as he has, but this is the first list I've seen in a long time that mentions specific low-hanging fruit that has the potential to actually improve lives (vs the mostly vacuous or profit-oriented areas I've seen in other lists). I look at problems in the world as a series of solutions built from first principles. So if you want a distributed mesh network, you need a supplier that can make millions of boxes for roughly 1/10 their retail price, or you need software that can run on existing phones/routers/computers. Then you need open software with tests for all the edge cases so you know it's free of bugs. Then you need an algorithm for things like web of trust or onion routing. Then you need a compelling use case (easy, no more internet and/or cell phone bill) and a way to prove it's safe so it goes viral. And so on. No one step of the process is insurmountable, or in many cases, especially difficult. Once we have that box, other ideas like content-addressable memory (things like IPFS) become feasible and we end up with an internet running thousands of times faster than what we have now, because media can be seeded once to each city and then cached by downloaders. So of course that's a barrier to entry, because few players in the existing paradigm want to disrupt the status quo. But that's no reason to settle for the wildly expensive and slow technology that we are stuck with now. Pretty much every item in his list works along these same lines.

lifeisstillgood 8y ago
Start with the routers. Build a better home router. Sort of like Ubiquiti kind of is doing. Build one that does - local caching and P2P sharing of media / photos etc this gives ISPs a genuine incentive to use it - if routers can build a mesh network (or at least not push back to ISP core) for 20/30 % traffic will that be savings? - Router acts as first responder to bot nets / malware from within the house Another good isp feature but great for the rest of us. IOT devices are only going to get worse and more vulnerable- but an intelligent router would be able to stop it sending ACK packets at cloudflare or whatever at source. And let you know that your fridge has been hacked and what to do about it. - Actually get updated with new firmware ... ever I think the household router / household hub is an important leverage that is under looked a lot.
- andrewflnr 8y ago
  What I've been told is that ISPs could easily stop most ddos attacks now, but simply don't give even the tiniest crap about it. Also, they tend not to like p2p, so I don't think you can sell this to ISPs.
  - Kovah 8y ago
    Absolutely, why should they bother? They make more money by routing more traffic - no matter if that's an YouTube video or a botnet request.
    - lifeisstillgood 8y ago
      I'm pretty sure (exISP) that routing costs. At a minimum you have to maintain free peering infrastructure at IXs perhaps, or if you are not a Tier 1 you basically pay for upstream access. Its why cacheing is all ISPs really care about. Its why Akamai makes ISPs happy by adding a edge router in their network. Presumably netflix too.
Sophistifunk 8y ago
There's a great list of products that should exist, but they're expensive and time consuming to do well, and nobody but JetBrains knows how to get people to pay for them. Even Microsoft is giving away most of their dev tools these days.
- Aeolun 8y ago
  Jetbrains can make people pay for them because their products are just insanely good value for money. I currently get their whole suite for $15 a month, and I get to use them at work and for freelance assignments. I mean, it’s a no-brainer.
  - irq11 8y ago
    You’re not rebutting the parent comment. I’ve built tools in this space. The bar for “insanely good” is very high, and the market size is quite small. It’s a crappy business from an investor perspective, and legions of nerds have washed up on this particular beachhead.
loceng 8y ago
Are all of the people, expert resources, available to build this if the monetary and time resources were available? The societal benefit, as a safeguard at minimum, is obvious, however of course execution and adoption is the more challenging part.
pjc50 8y ago
Mesh routing will necessarily be slower than backbone routing. And if you're pushing "no more internet and/or cell phone bill" you're saying that operating this network isn't going to be profitable, or will have to have a surveillance-advertising business model. You're also eliding the question of paying for content. > thousands of times faster than what we have now The last link is always the slowest. Moreover, much of the most interesting internet stuff really is global communications and dynamic websites, rather than easily replicated static content.

lifeisstillgood 8y ago

Start with the routers. Build a better home router. Sort of like Ubiquiti kind of is doing. Build one that does - local caching and P2P sharing of media / photos etc this gives ISPs a genuine incentive to use it - if routers can build a mesh network (or at least not push back to ISP core) for 20/30 % traffic will that be savings? - Router acts as first responder to bot nets / malware from within the house Another good isp feature but great for the rest of us. IOT devices are only going to get worse and more vulnerable- but an intelligent router would be able to stop it sending ACK packets at cloudflare or whatever at source. And let you know that your fridge has been hacked and what to do about it. - Actually get updated with new firmware ... ever I think the household router / household hub is an important leverage that is under looked a lot.

andrewflnr 8y ago
What I've been told is that ISPs could easily stop most ddos attacks now, but simply don't give even the tiniest crap about it. Also, they tend not to like p2p, so I don't think you can sell this to ISPs.
- Kovah 8y ago
  Absolutely, why should they bother? They make more money by routing more traffic - no matter if that's an YouTube video or a botnet request.
  - lifeisstillgood 8y ago
    I'm pretty sure (exISP) that routing costs. At a minimum you have to maintain free peering infrastructure at IXs perhaps, or if you are not a Tier 1 you basically pay for upstream access. Its why cacheing is all ISPs really care about. Its why Akamai makes ISPs happy by adding a edge router in their network. Presumably netflix too.

Sophistifunk 8y ago

There's a great list of products that should exist, but they're expensive and time consuming to do well, and nobody but JetBrains knows how to get people to pay for them. Even Microsoft is giving away most of their dev tools these days.

Aeolun 8y ago
Jetbrains can make people pay for them because their products are just insanely good value for money. I currently get their whole suite for $15 a month, and I get to use them at work and for freelance assignments. I mean, it’s a no-brainer.
- irq11 8y ago
  You’re not rebutting the parent comment. I’ve built tools in this space. The bar for “insanely good” is very high, and the market size is quite small. It’s a crappy business from an investor perspective, and legions of nerds have washed up on this particular beachhead.

loceng 8y ago

Are all of the people, expert resources, available to build this if the monetary and time resources were available? The societal benefit, as a safeguard at minimum, is obvious, however of course execution and adoption is the more challenging part.

pjc50 8y ago

Mesh routing will necessarily be slower than backbone routing. And if you're pushing "no more internet and/or cell phone bill" you're saying that operating this network isn't going to be profitable, or will have to have a surveillance-advertising business model. You're also eliding the question of paying for content. > thousands of times faster than what we have now The last link is always the slowest. Moreover, much of the most interesting internet stuff really is global communications and dynamic websites, rather than easily replicated static content.