Ask HN: How much is HTTP worth?
If HTTP was for sale or at auction today, how much do you think it could go for? And which companies or people do you think would be interested in buying it?
12 comments
HTTP was for sale
For instance, "selling" exclusive rights to serve HTTP would likely devalue the protocol to zero as quickly as systems could switch over to something else.
> What would this mean, exactly? What would "buying HTTP" mean for the buyer?
Ah, it's an NFT hard problem.
Ah, it's an NFT hard problem.
Oh you know, like NFT. A link to a .jpg of a HTML. Im putting it at a couple $mil.
Make an NFT for it and put it up for auction. Sure, it doesn't make sense, but neither does the question.
I guess you could also ask "What is language worth?" or "What is democracy worth?", "What is habeas corpus worth?". Some things really don't have an economic value attributed to them, because they are part of the fabric of the market from which economic value is defined.
You mean the protocol? Well it can't be sold as the standard is open and there is (as far as I know) no reference implementation which include special/exclusive features which might have some intrinsic IP value.
I think the extensibility of the protocol together with some proprietary features layered on top might have value though. I can't think of an example at the moment.
I think the extensibility of the protocol together with some proprietary features layered on top might have value though. I can't think of an example at the moment.
I guess what you're trying to do is ascribe some value to HTTP? Difficult thought experiment imo. Never the less: Let's say this ubiquitous protocol was mystically developed and owned by a private entity. I suppose the closest parallels in terms of direct commercialization would be Oracle or RedHat. If you wanted to play out that hypothetical you'd either have people buy into a license to your protocol somehow, adoption might be difficult, you could let everyone adopt it without enforcing some type of license, then you'd retroactively try to strongarm the project, you'd likely end up with a Joyent Node / IO js situation where someone forked it and that became the thing.
All of that is to say, HTTP is effectively infinitely valuable to society at this point, and it's so valuable precisely because it was never directly commercialized.
All of that is to say, HTTP is effectively infinitely valuable to society at this point, and it's so valuable precisely because it was never directly commercialized.
HyperText Transfer Protocol. Designed to transfer - mostly - hypertext, in form of HTML, now is used for anything. The form is messages over a reliable channel, messages have header and body (it's more complicated if it's used for continuous transfer). So the defining features of HTTP are that it's a deliberately small addition of structure over telnet.
Are those additions unique? Tim Berners-Lee certainly hit a good set of features, since it led to such a development. Having telnet as the base simplifies tinkering; having key-value set as the header can perhaps be improved, but still rather good. Suppose there is a comprehensive enforceable patent for HTTP - how much it would cost?
There is a history of Wright's brothers patent - https://wrightstories.com/articles/patent-wars/ - "The patent covered the control system of the 1902 glider involving the simultaneous use of wingwarping and rudder to perform a controlled turn and the ability to maintain roll stability." We know how it went - basically, this feature is used everywhere, even though alternatives could perhaps be devised (some more direct air pushers at the end of wings for control?) Does it illustrate possible fate of HTTP patent? If this is what's in cards for HTTP, for how much it could sell?
We need to find something which would hit a sweet spot with companies willing to buy rather than copy and trivially change, and where genuine re-development is too costly. I don't see something like this right now; something like royalties for RAMBUS memory was perhaps paid because there is a small window of time for the opportunity?
That is, if we consider the question as a matter of intellectual property. If there is different scenario - like, a scientist comes to a company and announces the magical process which would bring profits, but needs initial roll-out and payments to the inventor, the company's CEO would probably be hesitant.
I guess the answer is - not too much, both in immediate and subsequent payments. Small millions at most, perhaps, and that is assuming it's not here already and there is no corresponding environment around it - much less if an alternative already exist? One should look for people and companies with some special needs - military? unique markets? fundamental research? government-related economical think tanks? - where buyers would be easier to find?..
Are those additions unique? Tim Berners-Lee certainly hit a good set of features, since it led to such a development. Having telnet as the base simplifies tinkering; having key-value set as the header can perhaps be improved, but still rather good. Suppose there is a comprehensive enforceable patent for HTTP - how much it would cost?
There is a history of Wright's brothers patent - https://wrightstories.com/articles/patent-wars/ - "The patent covered the control system of the 1902 glider involving the simultaneous use of wingwarping and rudder to perform a controlled turn and the ability to maintain roll stability." We know how it went - basically, this feature is used everywhere, even though alternatives could perhaps be devised (some more direct air pushers at the end of wings for control?) Does it illustrate possible fate of HTTP patent? If this is what's in cards for HTTP, for how much it could sell?
We need to find something which would hit a sweet spot with companies willing to buy rather than copy and trivially change, and where genuine re-development is too costly. I don't see something like this right now; something like royalties for RAMBUS memory was perhaps paid because there is a small window of time for the opportunity?
That is, if we consider the question as a matter of intellectual property. If there is different scenario - like, a scientist comes to a company and announces the magical process which would bring profits, but needs initial roll-out and payments to the inventor, the company's CEO would probably be hesitant.
I guess the answer is - not too much, both in immediate and subsequent payments. Small millions at most, perhaps, and that is assuming it's not here already and there is no corresponding environment around it - much less if an alternative already exist? One should look for people and companies with some special needs - military? unique markets? fundamental research? government-related economical think tanks? - where buyers would be easier to find?..
so HTTP is roughly (2 billion devices)^2, with an average of $500 per device is (2b*$500)^2 =
$1,000,000,000,000,000,000,000,000
it's official. Jeff Bezos can't buy HTTP.
it's official. Jeff Bezos can't buy HTTP.
Now I want to start an MLM that buys Android devices for $50, sells them to recruits for $100, and convinces them the $500/device HTTP revolution is just around the corner.
HTTP has value because it cannot be "sold". If you tried to charge for it its value would go to zero.