Audio Of The Week: What Can We Learn From The History Of The Internet For The Future Of Crypto?

In this podcast, a16z general partner Katie Haun interviews a16z co-founder Marc Andreessen. It’s a great conversation.


Comments (Archived):

  1. sigmaalgebra

    The sound quality is awful: I just bought DVDs with a dozen of the Rathbone-Bruce Sherlock Holmes movies, and all the sound is really easy to understand. For that “podcast”, somehow for nearly all “podcasts”, the sound is awful — I can understand less than half the words. Bummer. The podcast users need to get with some good audio engineers and improve their sound quality.Besides, I have no great respect for Marc’s background, information, or insight. E.g., his information and insight are essentially what is in the coffee shops along Sand Hill Road, and that is essentially all quite mediocre stuff, heavily dreams and floundering of self-taught hackers confined in an echo chamber, very visible, e.g., on Hacker News and on the Web sites of Sand Hill Road VCs. Occasionally someone gets lucky at the tables in Vegas, and Sand Hill Road is much the same except that people attribute real capability instead of just luck.Besides, I see no significant long term future from the blockchain technology since for essentially any candidate application there is older technology that is superior. E.g., recently I’ve been doing a lot of Internet shopping: For that, my bank, Amazon, Sam’s Club, Walmart, Hanes, FedEx, and several others, the Internet, Web sites, e-mail, my computer, and I all work GREAT, no use of blockchain anywhere. IMHO, same for essentially all candidate applications: Bluntly, blockchain is inferior technology. In time, as the speculation, dreams of a big new thing, fad, and hype fade, that inferiority will become of overwhelming importance. Yes, between now and then, blockchain assets will be volatile so that someone with a lot of information and insight might be able to make some money.

    1. hemang c subramanian

      Nevertheless, how many times have you bought off a relatively unknown large supplier in a foreign country who can get you specialized product – at a fraction of the cost you pay today. Such e-commerce is not facilitated by today’s payment networks and involve large transaction fees, often delay product deliveries. Also, have you sold anything on any of today’s platforms – globally? Have you transferred 50 cents online to anyone anywhere in the world? Have you transferred greater than 1 Million dollars online to a foreign bank and have the foreign bank issue it to your supplier in real time at the current exchange rate? Have you stored your photos, or anything securely in ways that you are 100% sure that only you have access to it? and not even the storage company that owns the storage? Have you really been paid for anything you have ever written or created- constantly, consistently and at a fair value —-The blockchain and cryptocurrencies can enable all of these use cases (and in many cases do it too). There are instances where these markets are being created at a much more rapid pace than the internet. This is not to say that blockchain will make today’s businesses irrelavant, but it will definitely enhance what each of these businesses do today.

      1. JLM

        .These opportunities have been staring crypto and the blockchain in the face for a decade. WTF?Many of these things are outliers or can be done by simple processes such as Zelle.Legacy businesses are not going to cede their franchises. A good example is Walmart’s crypto initiative. They will be their own best customer.JLMwww.themusingsofthebigredca…

        1. pointsnfigures

          There are assets that don’t trade very well. Digitizing them, and opening up markets might make them more transparent-and make capital more efficiently

          1. JLM

            .Call Air Traffic Control — the flying pigs will need to be controlled.JLMwww.themusingsofthebigredca…

      2. sigmaalgebra

        ForThe blockchain and cryptocurrencies can enable all of these use cases (and in many cases do it too).Those “use cases” have long since been essentially enabled.For all those applications, need make only routine use of highly mature, polished, reliable technology in place long before blockchain, and blockchain is inferior technology. If there is a significant need for such an application that is legal, then there are companies around the world ready. E.g., US banking goes around the world. IIRC Citi is nearly everywhere.E.g., if some textile company in Pakistan wants to sell nice cotton bed sheets in the US for $2 each instead of $20, $50, or $100 (checked any of the prices lately?), then as I understand the legal issues, they just set up a little branch in the US, and US customers deal with that branch as they might with Hanes or through Amazon. Likely the company in Pakistan can ship by truck directly from the boat from Pakistan to Amazon warehouses, and then the Amazon Web site will say “Sold by South Asian Textiles, fulfilled by Amazon” or some such. Amazon may slap on “Amazon’s Choice”. Whatever, so far the Amazon customer happy guarantee is rock solid. In addition, if there were anything subtle and seriously wrong with the product, then soon Amazon would notice and no doubt quit selling it. Similarly for Walmart, Sam’s Club, Costco, J. C. Penny’s, and dozens more. Blockchain doesn’t help; it’s inferior technology and just gets in the way.Sure, there is the claim that blockchain enables cheaper data base operations. Hmm …. So far I’m no great expert in relational database, but the software for my startup uses SQL Server in a central and crucial way, and I did all the work, with Micrsoft’s ADO.NET (active data objects), .NET Framework, and Visual Basic .NET (don’t laugh; it’s different from C# essentially only in the flavor, I prefer, of syntactic sugar). So, I can imagine some of how the relational database operations work for all the on-line shopping I’ve done recently at Amazon, Walmart, Sam’s Club, J. C. Penny’s, Hanes, a restaurant supply house in CT, and maybe a dozen more. Gotta tell you, I shopped very carefully for the product, features, and price and may have executed 5000 SQL SELECT statements for each actual purchase. On today’s computers, relational database operations can be just DIRT CHEAP. And with the routine database caching in main memory, the fact that nearly all the database operations are read only, and what we can get today in solid state disks (14 TB in a 3.5″ form factor), we’re talking fast, FAST, FASTER beyond belief. E.g., Amazon sells about 1.5 million different items. “Big”? Right? Nope: tiny, teeny tiny — soon it will all be in the main memory cache. Heck, the first server I plugged together for my startup has 16 GB of ECC (error correcting coding) main memory: I don’t know just what database schema Amazon might use, but for only 1.5 million items 16 GB should go a long way, and the servers Amazon is using might have 16 TB or so of main memory, each. People have known for decades that relational database all in main memory is faster than Superman’s speeding bullet. BELIEVE ME, blockchain won’t beat that in ease of software development, speed, cost, reliability, recovery, security, system management, basic algorithms, etc. Ain’t gonna happen. There are possibilities for new, improved, the first good or much better technology, etc., but blockchain ain’t an example.

    2. jason wright

      “no use of blockchain anywhere.”Would it be possible to spot it if they were to introduce it at some point in the future?

      1. sigmaalgebra

        Initially for a small application, maybe not. Eventually for a large applications, I’d guess that the usage would leak out.Or, basically we’re talking encryption and database, e.g., with authentication, and we know a lot about how those work from small scale to large. For encryption and authentication, we’ve got public key crypto systems and RSA (Rivest, Shamir, Adleman). For database we’ve got DB/2, SQL Server, Oracle, and several more. No doubt these tools are running Amazon, Walmart, Citi, Chase, Morgan, Fidelity, etc. If there were a big change to blockchain, then we’d hear about it.

  2. awaldstein

    Yes it is.I think Sunal Chokshi is doing some of the best work in podcast series out there today.Thanks for surfacing this one.

  3. Tom Labus

    “enlightened policy” will have to wait for the next administration. Hopefully, things survive until that point.

    1. JLM

      .Factually unsupportable utterance.There has never been a more accommodating admin to “low reg” than the current admin.Crypto is having a hard time decoupling from its Dark Web ancestry.There is no killer app driving it.Fight fair. Use facts not emotional baloney.Crypto has done nothing to earn special consideration. World class whiners.JLMwww.themusingsofthebigredca…

      1. jason wright

        They want a free breakfast, lunch and dinner. Earn it.

        1. JLM

          .Drinks. You forgot they want drinks. Free drinks. Plus free artisan wines and small batch bourbon.Even if the US SEC let the crypto world write the regs, more than 80% of the deals would still be crap, fraudulent, scams, and unsuccessful.There is no real demand for any of this other than an artificially generated one.This has been going on for more than a decade and we are still waiting for the killer app to appear. Hard to keep the words “head fake” from crossing one’s brain.JLMwww.themusingsofthebigredca…

          1. jason wright

            Their collective state of mind might suggest that’s where they’ve already spent their hard earned money. They’re well stocked up on the ‘pop’.

  4. Richard

    Music Radio Photography/Cameras Phones Automobiles Airplanes Movies Television Computers …CryptocurrencyWhich one of these things is not like the others?Take away all these real things and the internet is still sitting between a 6th grade science class and a mainframe.To date, Cryptocurrency’s only market is greed and greater fools. If USV really thinks global warming will consume the planet in 10 years, where are the investments that might do something about it?

  5. William Mougayar

    “Enlightened Regulators”, yeah . What was common sense in 1993 is now an oxymoron.

    1. pointsnfigures

      Meh, regulators have never really been enlightened. “We are from the government and we are here to help”. I think the level of lobbying has scaled since 1993, but the size/scope of government have also gotten a lot larger. I’d love a government with a 1993 footprint.

  6. Richard

    The principal competitive advantage of the US is it’s currency and it’s military.Take away its the $ as the worlds currency – the carrot – and guess what we have left to keep world order ?

    1. sigmaalgebra

      > competitive advantage of the USAlso products of Microsoft, Intel, Oracle, Adobe, AutoCad, Google, Facebook, beef, wheat, soy beans, oil, natural gas, coal, medical products, the stock markets, banking, airplanes, electronics and engines for airplanes, academic text books, movies, weapons, and no doubt more.

      1. jason wright

        …and not forgetting the chlorinated chickens 😮

      2. Richard

        Of course we dominate in those areas but they have little to do with national security – which is ultimately the most important function of a society. The $ allows the US to exert pressure on nation states without the use of our armed forces.To all the scoundrels – who have made $ off the backs of the poor from via bitcoin, add that to your peace pipe.

  7. jason wright

    That trying to imitate and mimic the ‘old’ internet with blockchain tech is a mistake because by doing so we inhibit the native potential of the tech, and that to innovate a new market with the tech it is essential to move as far away from the crypto herd as possible.

  8. Mike

    Very interesting. A few quick thoughts:To say the Internet as currently constructed is limited to mostly advertising revenues I think discounts significant revenues from existing SaaS, subscription and ecommerce platforms that do not rely on advertising. The Internet is a Media platform for content distribution but also a Distribution Channel for all sorts of products and services. The entire Enterprise Software industry is being transformed. It will be interesting to see what new opportunities Crypto and Blockchain might introduce. NFTs look very interesting. Other ideas posted on this board.The “Internet” is not just software. There is a very robust hardware and infrastructure component, both wired and wireless, that has enabled the network over the decades, and will continue to do so into the future, 5G,optical networks, routing technologies, a host of advanced communications and data processors that enable the transmission, processing and analysis of increasing volumes of data, AI, mobile broadband etc. Not to mention entire new industries of edge devices, smart phones, tablets etc. Should we expect a similar impact from crypto or blockchain networks over the coming decades?If you look at the original Bitcoin blockchain concept published in 2009, as clever and groundbreaking as it was, what is the fundamental technical innovation? Why could this not have been implemented in 1999? Or what specifically about the state of Internet, or other technology in 2009 enabled it at that time? Mobile, processor power/cost, network BW? From what I have read the NSA has been implementing crypto algorithms for decades. Was it the proximity of the worst global recession in generations?This is not a treatise for or against crypto or blockchain. Mantra from this blog – business model innovation can be more economically impactful than technology innovation (or at least quicker returns). Just some thoughts on the comparison of these technologies to the Internet of the 1990s, and the potential of what followed.

    1. Michael Elling

      Agreed, one has to look at all the layers of the stack and boundaries across which the marginal supply and demand of information clear.

  9. Michael Elling

    It’s not what Marc says, but what he doesn’t say. Crypto is one of many potential means to an end; it is not the end. Nor has he defined the end well.What about the layer 1-2 price arbitrage brought about by the WAN competition model of the 1980-90s and only found in the US. What about the fact that the WWW rides over the TCP/IP stack? What about all the other competing trends in packet networks and service providers (public and private). The original sin is not about lack of a payment model. It is about lack of settlement models east-west AND north-south in the informational stack which take into account marginal differences in supply and demand. Only through properly designed (via the market yet monitored by an intelligent regulator) settlement systems can geometrical value captured at the core and top be redistributed to linear costs borne at the bottom and edge.The original sin arguably dates back to the 1913 Kingsbury Commitment and next 20 years when regulators and industry could not figure out the mechanisms for universal phone service and so the government took on that mantra with the Telecom Act of 1934.