8 ms·

The Mobile Downturn

87 points by arnauddri 13y ago

arnauddri 13y ago
We surely reach a more mature phase in the 'mobile apps' market but I believe products will gain in quality in the future and that there are still some opportunities to be seized: - there was a tremendous amount of apps released over the last few years, it is no wonder that competition is getting fierce: products will have to adapt in terms of quality with better and more curated contents. - there is still a lot of room for innovation in re-engagement (http://tomtunguz.com/mobile-reengagement/ http://tomtunguz.com/mobile-reengagement/), deeplinks (www.urx.com) are a very good start
- scholia 13y ago
  Aside from the fact that most apps are rubbish, the real problem is the move from the open, free, creative web to heavily-censored walled gardens controlled by giant corporations. Also, apps completely invade your privacy.
Spearchucker 13y ago
He lists the following as big innovations over the last 20 years: TCP/IP, HTTP, the browser, search, social, mobile, and blockchains. I know that it's got something to do with Bitcoin, but have no idea what a blockchain is. Wikipedia only references it in a link to blockchain.info, which talks about blocks but not blockchains. Nobody I've asked can explain it to me either. If it's such a massive innovation (is it?), what is it, and why isn't it easy to discover what it is?
- czr80 13y ago
  From: https://en.wikipedia.org/wiki/Bitcoin#Block_chain https://en.wikipedia.org/wiki/Bitcoin#Block_chain "Integral to Bitcoin is a sequential record of all transactions known as the block chain. There is only one block chain, and it records bitcoin ownership both present and at all points in the past. By keeping a record of all transactions, the block chain mitigates double-spending, a core design challenge of digital currency. Approximately every ten minutes a new bundle of transactions, called a block, is added to the block chain." If you need a more in depth explanation, which also covers how this relates to a famous problem in distributed systems (the Byzantine Generals Problem), see: http://expectedpayoff.com/blog/2013/03/22/bitcoin-and-the-byzantine-generals-problem/ http://expectedpayoff.com/blog/2013/03/22/bitcoin-and-the-by...
  - rst 13y ago
    Some further context: Computers all over the world are competing to extend the block chain by solving a problem involving cryptographic hashes. The reward for solving the problem is that a wallet controlled by the machine that does it can claim a Bitcoin reward. The reward is currently, I believe, 25btc --- but it's scheduled to drop over time; eventually to zero, at which point, miners will have to be compensated by claiming transaction fees. The upshot is that Bitcoin has a system for maintaining a single, global log which is consistent across all connected Bitcoin clients. And while Bitcoin itself uses this as a transaction ledger, there are plenty of other things you might want to use this construct for once you have it. Namecoin, for instance, uses a Bitcoin-like block chain to manage a DNS-like naming scheme. (There are some weaknesses to the design: most notably, if a single party has more than half the mining compute capacity, they're able to pull off all sorts of chicanery against the community as a whole...)
- evgen 13y ago
  I realize Fred and USV is heavily invested in bitcoin startups, but the block chain as a big innovation of the past 20 years on par with TCP/IP and HTTP? Compared to DHTs, consensus protocols that enable orgs like Google and Facebook to reach operational scales that were inconceivable ten years ago, compared to deep learning and statistical ML? Seriously?
- msvan 13y ago
  Blockchains create scarcity among bits, enforced by a distributed network. The most well-known usage is in Bitcoin, but there are many different applications. Many systems that were previously very difficult to create without a central authority/point of failure can now be built in a distributed fashion. See Namecoin [1], Ethereum [2], or Protoshares [3] for examples. 1: https://www.namecoin.org/ https://www.namecoin.org/ 2: https://www.ethereum.org/ https://www.ethereum.org/ 3: http://protoshares.net/ http://protoshares.net/
  - ForHackernews 13y ago
    Isn't it horrendously inefficient to be copying the entire blockchain around all the time? What happens if any of those projects achieve mainstream adoption? Are we looking at potentially petabyte-sized blockchains being copied around everywhere?
    - wmf 13y ago
      You don't need to store the whole blockchain as long as some CDN is willing to serve the pieces you need. In a well-designed blockchain you can verify that pieces are legitimate without downloading the whole thing.
- kenjackson 13y ago
  TCP/IP is a lot older than 20 years old. It's around 40 years old. I'd say that JavaScript should be on that list -- not that I like JavaScript, but a scripting language in support of DOM was an important piece of work.
  - oscargrouch 13y ago
    Then C is much more important.. it gave birth to Unix(another big contender on the list), Linux, and pretty much every piece of infrastructure in the world.. javascript will be pretty small compared to those giants.. Other trends are big data, deep learning.. i bet deep learning will me much more revolutionary.. But you know SV have a love affair with hypes
    - kenjackson 13y ago
      You do know what C is also a lot older than 20 years old too, right? The statement was about stuff invented in the past 20 years.
bambax 13y ago
> the smartphone emerging as the platform of choice vs the desktop browser This is repeated from blog to blog but is it really true? Smartphones outsell desktops because you change phones every 18 months while a desktop will last maybe a decade or more; but do people really prefer to use their phone if they have a choice? It would seem to me people who use a phone most of the time are people who don't have easy access to a desktop: young people, or people who don't have a job or whose job keeps them away from a desk (lucky them!) When I'm around people who use a smartphone, I never see them engaged in anything more complex than listening to music, texting, or playing some super-simple game such as CandyCrush; they're never "browsing" anything, searching, writing, composing, etc. So if you're doing social or communications, sure, mobile first makes a lot of sense, but how about other domains? Do people use Github from mobile platforms? The Github app for Android apparently has "100,000+ downloads", which doesn't seem like a lot? It would be interesting to know the numbers for Github (for example), of mobile vs. desktop usage.
- smacktoward 13y ago
  > When I'm around people who use a smartphone, I never see them engaged in anything more complex than listening to music, texting, or playing some super-simple game such as CandyCrush; they're never "browsing" anything, searching, writing, composing, etc. The thing is, for the vast majority of people that's all they were doing with their PCs anyway. They weren't composing librettos, they were just checking email and playing Farmville. So to them it doesn't feel like they've lost anything.
  - jarrett 13y ago
    That rings true, yet casual users are not the only profitable market for software. Not by a long shot. Look at companies like Microsoft, Autodesk, and Adobe. They do very well selling professional software--software that's never going to translate well to a phone-sized screen.
    - trhway 13y ago
      >software that's never going to translate well to a phone-sized screen. an here is the next multi-billion of dollars :)
fredgrott 13y ago
He missed something.. Let me explain, what are big networks buying? Think Yahoo, Facebook, Google, MS, etc.. The mobile app now represents user data to buy...and that is what we are seeing in some of the acquirehires buying sprees is that the mobile app is kept alive after the buyout because the firm doing the buying is buying user data. Thus we still will see some $1B or more home runs if we know and understand what they are buying
- paulbaumgart 13y ago
  User data is useful for showing ads that convert well, but the fundamental unit of value in consumer apps isn't data, it's attention.
programminggeek 13y ago
I really don't think the block chain is the next platform or even that interesting. What exactly am I missing?

87 points by arnauddri 13y ago

We surely reach a more mature phase in the 'mobile apps' market but I believe products will gain in quality in the future and that there are still some opportunities to be seized: - there was a tremendous amount of apps released over the last few years, it is no wonder that competition is getting fierce: products will have to adapt in terms of quality with better and more curated contents. - there is still a lot of room for innovation in re-engagement (http://tomtunguz.com/mobile-reengagement/ http://tomtunguz.com/mobile-reengagement/), deeplinks (www.urx.com) are a very good start

scholia 13y ago
Aside from the fact that most apps are rubbish, the real problem is the move from the open, free, creative web to heavily-censored walled gardens controlled by giant corporations. Also, apps completely invade your privacy.

Spearchucker 13y ago

He lists the following as big innovations over the last 20 years: TCP/IP, HTTP, the browser, search, social, mobile, and blockchains. I know that it's got something to do with Bitcoin, but have no idea what a blockchain is. Wikipedia only references it in a link to blockchain.info, which talks about blocks but not blockchains. Nobody I've asked can explain it to me either. If it's such a massive innovation (is it?), what is it, and why isn't it easy to discover what it is?

czr80 13y ago
From: https://en.wikipedia.org/wiki/Bitcoin#Block_chain https://en.wikipedia.org/wiki/Bitcoin#Block_chain "Integral to Bitcoin is a sequential record of all transactions known as the block chain. There is only one block chain, and it records bitcoin ownership both present and at all points in the past. By keeping a record of all transactions, the block chain mitigates double-spending, a core design challenge of digital currency. Approximately every ten minutes a new bundle of transactions, called a block, is added to the block chain." If you need a more in depth explanation, which also covers how this relates to a famous problem in distributed systems (the Byzantine Generals Problem), see: http://expectedpayoff.com/blog/2013/03/22/bitcoin-and-the-byzantine-generals-problem/ http://expectedpayoff.com/blog/2013/03/22/bitcoin-and-the-by...
- rst 13y ago
  Some further context: Computers all over the world are competing to extend the block chain by solving a problem involving cryptographic hashes. The reward for solving the problem is that a wallet controlled by the machine that does it can claim a Bitcoin reward. The reward is currently, I believe, 25btc --- but it's scheduled to drop over time; eventually to zero, at which point, miners will have to be compensated by claiming transaction fees. The upshot is that Bitcoin has a system for maintaining a single, global log which is consistent across all connected Bitcoin clients. And while Bitcoin itself uses this as a transaction ledger, there are plenty of other things you might want to use this construct for once you have it. Namecoin, for instance, uses a Bitcoin-like block chain to manage a DNS-like naming scheme. (There are some weaknesses to the design: most notably, if a single party has more than half the mining compute capacity, they're able to pull off all sorts of chicanery against the community as a whole...)
evgen 13y ago
I realize Fred and USV is heavily invested in bitcoin startups, but the block chain as a big innovation of the past 20 years on par with TCP/IP and HTTP? Compared to DHTs, consensus protocols that enable orgs like Google and Facebook to reach operational scales that were inconceivable ten years ago, compared to deep learning and statistical ML? Seriously?
msvan 13y ago
Blockchains create scarcity among bits, enforced by a distributed network. The most well-known usage is in Bitcoin, but there are many different applications. Many systems that were previously very difficult to create without a central authority/point of failure can now be built in a distributed fashion. See Namecoin [1], Ethereum [2], or Protoshares [3] for examples. 1: https://www.namecoin.org/ https://www.namecoin.org/ 2: https://www.ethereum.org/ https://www.ethereum.org/ 3: http://protoshares.net/ http://protoshares.net/
- ForHackernews 13y ago
  Isn't it horrendously inefficient to be copying the entire blockchain around all the time? What happens if any of those projects achieve mainstream adoption? Are we looking at potentially petabyte-sized blockchains being copied around everywhere?
  - wmf 13y ago
    You don't need to store the whole blockchain as long as some CDN is willing to serve the pieces you need. In a well-designed blockchain you can verify that pieces are legitimate without downloading the whole thing.
kenjackson 13y ago
TCP/IP is a lot older than 20 years old. It's around 40 years old. I'd say that JavaScript should be on that list -- not that I like JavaScript, but a scripting language in support of DOM was an important piece of work.
- oscargrouch 13y ago
  Then C is much more important.. it gave birth to Unix(another big contender on the list), Linux, and pretty much every piece of infrastructure in the world.. javascript will be pretty small compared to those giants.. Other trends are big data, deep learning.. i bet deep learning will me much more revolutionary.. But you know SV have a love affair with hypes
  - kenjackson 13y ago
    You do know what C is also a lot older than 20 years old too, right? The statement was about stuff invented in the past 20 years.

bambax 13y ago

> the smartphone emerging as the platform of choice vs the desktop browser This is repeated from blog to blog but is it really true? Smartphones outsell desktops because you change phones every 18 months while a desktop will last maybe a decade or more; but do people really prefer to use their phone if they have a choice? It would seem to me people who use a phone most of the time are people who don't have easy access to a desktop: young people, or people who don't have a job or whose job keeps them away from a desk (lucky them!) When I'm around people who use a smartphone, I never see them engaged in anything more complex than listening to music, texting, or playing some super-simple game such as CandyCrush; they're never "browsing" anything, searching, writing, composing, etc. So if you're doing social or communications, sure, mobile first makes a lot of sense, but how about other domains? Do people use Github from mobile platforms? The Github app for Android apparently has "100,000+ downloads", which doesn't seem like a lot? It would be interesting to know the numbers for Github (for example), of mobile vs. desktop usage.

smacktoward 13y ago
> When I'm around people who use a smartphone, I never see them engaged in anything more complex than listening to music, texting, or playing some super-simple game such as CandyCrush; they're never "browsing" anything, searching, writing, composing, etc. The thing is, for the vast majority of people that's all they were doing with their PCs anyway. They weren't composing librettos, they were just checking email and playing Farmville. So to them it doesn't feel like they've lost anything.
- jarrett 13y ago
  That rings true, yet casual users are not the only profitable market for software. Not by a long shot. Look at companies like Microsoft, Autodesk, and Adobe. They do very well selling professional software--software that's never going to translate well to a phone-sized screen.
  - trhway 13y ago
    >software that's never going to translate well to a phone-sized screen. an here is the next multi-billion of dollars :)

fredgrott 13y ago

He missed something.. Let me explain, what are big networks buying? Think Yahoo, Facebook, Google, MS, etc.. The mobile app now represents user data to buy...and that is what we are seeing in some of the acquirehires buying sprees is that the mobile app is kept alive after the buyout because the firm doing the buying is buying user data. Thus we still will see some $1B or more home runs if we know and understand what they are buying

paulbaumgart 13y ago
User data is useful for showing ads that convert well, but the fundamental unit of value in consumer apps isn't data, it's attention.

programminggeek 13y ago

I really don't think the block chain is the next platform or even that interesting. What exactly am I missing?