I have literally never seen a sensationalist headline like this come true.
But seriously, Parse looks like it could help some people out, and seems to be doing so already, but let's not be ridiculous. When people say that it decreased dev time by 10-100 times, what they really mean is that they added features that otherwise would never have seen the light of day, and they have no idea how long it would have taken had they tried.
There will always be things like this to help build apps, and they will always come with trade-offs. Things may be easier, but slight changes to those things become impossible. Not everyone will think that the trade-off is worth it, and the app industry will chug along as always, if a bit shinier.
I don't think that's true: I've used Parse on several commercial projects where the cost and effort involved in developing internal back-end solutions was significantly more than using Parse. And it wasn't a case of adding in features we weren't going to do: this was implemented core feature sets.
Yes, there are trade-offs, and we still use internal systems and solutions for some of our projects where Parse isn't appropriate. However, I can categorically say it's saved us significant time and money - probably not a factor of ten, but large enough for us to continue looking to use it in the future.
This is why I'm calling it sensationalist. Think about what it would mean for someone to claim 100x faster - if they got it done in a week (and that's pretty damn fast for any decent piece of software, regardless of the environment or scope), they're claiming that they initially budgeted nearly two years. I'm saying that there is no way that ever happened.
Again, I said that there are benefits (and also trade-offs), but that this isn't game-changing. I'm not clear on where we disagree.
Yeah, I've taken that tack "the more sensational the title, the better I'm off reading the commentary to the submission to satisfy my curiosity rather than clicking through the link and rewarding a hypeman/bullshitter."
Will it change how the entire app industry works? Not really. Will it make life easier for some developers? Probably. Having used Parse in the past and ultimately dropped it for a custom backend, let's take a look at where Parse fits really well:
1) You're a one-off or small dev shop with no web experience, and making a basic data store with a JSON API as a backend is beyond your technical means (this is by far the best use).
2) Your data isn't that complex (stuff like high scores and the like), so the fact that this is just a big Key/Value store won't harm your performance much.
3) You also don't need any server-side intelligence, just a data store an an API.
4) Your growth and revenue model is in line with the pricing model of the service you selected.
Now here's the things I struggle with, and why where I work when we develop mobile apps, we stay away from Parse/Stackmob/Kinvey/Appcelerator ACS/Whoever. These apply to our customers / projects, and may not be the same for you.
1) We almost always want our server to be intelligent, not just an API-wrapped database. When I want to trigger push notifications, I want to do it with server-side logic, not everything running in the app. Our big beefy AWS servers are way more capable than someone's 2-year-old free Smartphone.
2) It doesn't take long for monthly fees to outpace the cost of just having your own server, or eat into your profitability. Most of these services work out to about 3-5 cents per user per month. Depending on the size and complexity of the app, and what the revenue generated from it looks like, that can be vastly more expensive than a small AWS or Heroku deployment.
3) It's rare that we need a data store as simple as "users + key/value". It's one thing to build a basic chat app or worldwide high score in a system like that. It's another to try and build an inventory tool or CRM.
4) Mobile data connectivity is high-latency and unreliable. Making a dozen API calls to populate a single screen is not only inefficient, it can be very frustrating to your users. An API call for a screen/view/fragment should return exactly the data needed, no more, no less. Let the database do what databases do best: sort and collect data. Let the phone just display it.
So really, for small apps or apps with relatively simple data requirements, Parse does a fine job, especially if you don't have server people on your team (in which case, BaaS is your only real option). If the app's data is going to be complex, require server-side logic or have any kind of complexity that would involve optimized API calls? You're probably better building your own.
" Using Parse can lower your development time by 10x to 100x. Developers go from taking weeks to build an app to being able to build one and push it out in a matter of hours.
"
So here is an interesting question. If you can build an app in a matter of minutes, how much is it worth? I ask because there is often a strong correlation between economic value and the time it takes to produce something. For example, movies produced in a couple of days often have a lower economic return than movies that take months to produce. Even including pornography in that mix, sans outliers like Paris Hilton, there seems to be a strong relationship between time and value.
Presumably Parse makes the implementation stage easier, but is that for all types of App? Or is it a class of App? It would need to be the former to 'change the industry' but what does that mean about the economic value of the industry to begin with?
Clearly this is part of a much bigger question of the economic value of 'Apps' in general. We are evolving models based on various metrics of course, and I'm curious does this 'destroy' value or does it 'grow' the market, or something else entirely.
You are grossly underestimating the effort it takes to find an interesting problem and then find an interesting solution.
Think of it in terms of Legos. Parse is an automatically sorted tray of Lego pieces for you to use. No more prickling your hand for 3 minutes while you search for that one piece -- no boilerplate.
You take a gymnasium filled with a 1000 7-year-olds and hold a contest to see who builds the coolest stuff. No amount of tools is going to help the poor kids who can't think of interesting things to build. On the other hand, one particularly ambitious venturist decides to build a 5-foot-tall T-Rex out of Legos. Parse rapidly accelerates the build process by providing the stuff he needs right when he needs it.
If anything, I foresee an increase in price in the future as we see more apps that provide more value in more sectors. App profit is not a zero sum game. The profit derived will be directly proportional to the value derived.
[TLDR] - The limiting reagent to successful apps is good problems and good solutions. Good tools merely speed up development once you have figured those two things out.
Parse is good for us because we can build iCloud-like applications without having to be solely iOS. We find storing small amounts of user-centric customizable data a sweet spot for Parse.
I think more powerful tools yield different apps, not just the same apps in less time. Because of the process of human design, I doubt we'll have solid, robust, lasting apps built "in a matter of hours", but potentially more and better version 1's.
Development aside, designing, testing, marketing, etc. seem to take a certain fixed amount of thought, discussion, and reflection. So it might still take a few weeks or months to build a great app, but in that time, you could experiment more. Going from, say, days to minutes to build certain features would be incredibly valuable. You could try things you never would've had time for otherwise, and end up with apps that wouldn't've even been possible with other tools.
This is an important question for software development in general. I find it interesting to see just how much simple tasks (say parsing data and combining it with other data) results in significant improvements in user experience.
I am not in the "everyone should learn to code" camp, but I sometimes think that everyone should know a few standard Unix command line tools.
Maybe the link between effort and economic gain will balance back out sometime. Maybe right now is just one of those boom times in history and useful tool creators are in position to reap the benefits.
You're talking about the supply side of the supply/demand equation. Essentially, the greater the supply of something, the lower the price one can charge for it. Reducing the cost of production effectively increases the supply, so it ultimately reduces the price as well.
Ultimately, what this means is that the competition will be harsher and that will drive the prices down. This won't drive down the value of the mobile market in general, just the prices of individual apps (though they are already fairly low.) In fact, generally when the supply goes up, the total income of the industry as a whole goes up as well because marginal consumers will now participate due to lower prices.
"You're talking about the supply side of the supply/demand equation. Essentially, the greater the supply of something, the lower the price one can charge for it. Reducing the cost of production effectively increases the supply, so it ultimately reduces the price as well."
No I am not. One of my areas of research has been the economics of information. Unlike classic 'real goods' economics which are driven by supply and demand, information economics are driven by other forces. One of the paradoxes this sorts out is how something can be 'free' to produce and still have economic value.
In the case of creative works, I theorize that the economic value derives from the creativity itself. One might pay to see a play rather than read it for free because the 'value' is watching the theater bring the work to life. Similarly with 'apps' in that you might pay for one that saves time, or automates a loathsome task, Etc. So the economic question becomes what does it cost to 'produce' (which is go from nothing to something) the work, vs what is the economic value of utilizing the work (aka the desire to have that problem solved). For some things, like Uber (which automates finding the nearest car available for hire) many people 'desire' to have that problem solved and it motivates them to give economic value to possess that capability. (or buy the app, subscribe to the service, etc).
So Parse lowers the cost of converting idea into App, which lowers the economic cost to develop an idea, does it have the same economic value to the recipient? Let me use an example that might illustrate what I'm trying to say.
Lets say you've got a neighbor with a rude dog which barks every time you walk out the front door. This is annoying. Now someone offers to fix it so that you don't have to deal with it, and their 'fix' is to put a giant tube down your front sidewalk so the dog can't see you leave. You might find no value at all in that solution because you could do the same thing and the giant tube looks stupid. Or the person might be some sort of dog whisperer type who could go over and train the dog to not respond like that when they see you. Well in the latter case there is a higher perceived economic value since you don't possess 'dog whisperer' skills and thus you could not implement the solution yourself. But the tube thing, well you could do that if you wanted too.
So back to Apps, what is the economic value of an app you really could create yourself in a 'couple of hours' ? Probably not a whole lot. But if it would take you 'weeks' to make that app, well you're willing to spend $4 to have someone else do the leg work.
Information economics are driven in part by a strange opportunity cost function which I can't say I've completely characterized, but I do know that if we released a tool tomorrow that let you speak to your phone and tell it the 'feature' you wanted it to have, and some sort of thing went out and created that and dropped it on your phone. It would destroy the app ecosystem as we know it today.
Perhaps Parse + Mechanical Turk + Siri can get to that point, if so it would really change things.
You've written a very elaborate reply that still fails to point out why astine's simpler explanation (i.e. standard supply-and-demand) doesn't apply. After all, other things being equal, at a fixed price there will more supply for service X as the cost of providing X decreases.
All digital goods have infinite supply and yet even for goods that can be acquired for free (say Ubuntu distributions), there are markets (the $10 starter DVD). Once a digital good is 'produced' replication costs and distribution costs approach zero, or at least to a marginal cost of zero given the sunk cost of the infrastructure (one pays of internet bandwidth but doesn't normally charge it against digital purchases).
So my claim is that in a digital marketplace the quantity supplied of a good can meet any quantity demanded. Thus classical Keynesian economic theory doesn't work well. Since by the math if supply is infinite even selling for a small cost can give good returns (cost to produce is now zero after all) there is no compelling reason to sell at a fixed cost, but demand isn't infinite.
So why do people pay for 'free' information? (and yes some don't, this isn't a piracy discussion). Given these sorts of questions I've been working on what it is that gives information value. What are the economics of information?
One answer to that explains Google's success. Information can become valuable if it is timely, which is to say it meets a need in a very time efficient way. So high frequency stock market traders pay through the nose to know a few milliseconds earlier what the trades are. Google sells the fact that someone has just looked for <query terms> to someone who thinks they have an answer related to <query terms>. Someone is looking, the advertiser can provide, so what makes a search advertisement so valuable is that the intent behind the search is 'known.' Previous venues tried to sell the 'kind of people who are looking at these pages' the demographics. A less precise form of targeting but similar.
But back to the Parse discussion, one of the factors that appears to influence information value is its rarity, or difficulty in obtaining. So a picture of some celebrity's new baby is 'rare' and 'hard to get' so a paparazzi gets a big payday when they get that picture. But a picture from the agent's press packet on the birth? Not so much.
Applications seem to be more valuable if they are perceived as being 'hard to write' (perhaps this is a variation on bike shedding) than ones that are perceived to be 'easy.' And that affects personal statements of value as well, how many developers refuse to buy an app they feel they could write in a weekend and yet non-developers flock to it?
My claim is that classic economic supply and demand models don't apply to digital goods because the supply of the goods can be infinite and the classic model suggests prices would go to zero in that scenario.
> So why do people pay for 'free' information? (and yes some don't, this isn't a piracy discussion). Given these sorts of questions I've been working on what it is that gives information value. What are the economics of information?
Of course it might be interesting to ask how digital goods are different from tangible objects in economics. But this particular question is made on a wrong assumption. Digital goods can be scarce and hence not "free" by definition.
> But back to the Parse discussion, one of the factors that appears to influence information value is its rarity, or difficulty in obtaining.
Yes, but that's no different from the classical concept of scarcity.
> My claim is that classic economic supply and demand models don't apply to digital goods because the supply of the goods can be infinite and the classic model suggests prices would go to zero in that scenario.
No, no, it doesn't suggest that. That's the problem with your argument. The theory might say that that happens in a world of "perfect competition", which is a very strong assumption. As long as someone has the publishing rights for a digital good, its low marginal cost doesn't imply higher supply and lower price.
This is not really limited to digital goods. Take a hardcover textbook or a cancer-treatment drug: its marginal cost might be a few dollars but given limited supply it can still sell for over $100.
[ disclaimer, I am a (barely) trained economist. ]
So goods have a higher perceived value based on the effort that went into them? That's probably true but I don't think that it really changes the fundamentals here. In economics, all value is perceived and supply and demand are part of the equation in every instance. If the reason dog whispering is so valuable is because it's a learned skill that I don't posses, then it's a lack of supply.
With regard to mobile apps, there is a supply of programmers with the skills to make mobile apps and this program, Parse, threatens to glut the supply of theses programmers by reducing the skills necessary to make mobile apps and this will devalue those skills in just the same way that if dog whispering were easy, it would also not be as valuable.
Now there are other effects which can lower and raise the value of something aside from supply and demand and you allude to perceived valuation in creative goods owing to perceived labor costs, and that definitely exists in certain social circles, but it's by no means universal. Otherwise, how would you explain the vast popularity of Thomas Kinkade reprints or Garfield? The people who consume this stuff generally stuff know that it's cheap to produce; they just don't care. Or why very lazy concept art often fetches a higher valuation than much more difficult classical artwork? Or why some people are willing to pirate a million dollar Hollywood movie but still pay for money for a much cheaper less labor intensive film to support and independent artist.
In other words, I might prefer labor intensive goods over cheaper equivalent goods, for the sake of novelty, moral values, economic signaling, or for many other reasons but I just as easily might not as fashions and priorities change. However, when perceived scarcity per-se makes something seem more valuable, that is literally a psychological effect of supply. Supply most definitely drives creative goods because otherwise copyright (whose whole purpose is to create artificial scarcity and reduce the supply to prop up the price of creative goods) would be a moot point.
With regard to apps, the value of an app depends both demand-side factors such as the actual utility, social factors, etc, including for some people the perceived labor cost in its production, as well as supply-side factors such as capital costs and labor costs. If I could make the app I needed in a matter of hours, having someone else make it for me might be worth nothing, or thousands depending on how I valued my time and what the app can accomplish for me.
"I do know that if we released a tool tomorrow that let you speak to your phone and tell it the 'feature' you wanted it to have, and some sort of thing went out and created that and dropped it on your phone. It would destroy the app ecosystem as we know it today."
Of course, remove scarcity and an economy ceases to be necessary. That's a much simpler explanation than the one you posit.
On the other hand, last I heard, porn tends to make money no matter how badly/quickly done. But perhaps that has changed in the years since I took economic geography.
I think the economic value of the app industry period has been basically kneecapped by apple anchoring $1 as the proper price of an app. There's just so much work you can put into something that earns $.70 and still make a living. And yes, nobody has a right to a living, but software gets complex fast and it's shocking how quickly even relatively simple software starts to demand a lot of time from a developer.
I'm also not privy to any of Parse's plans, but I'd bet they're going to allow server side logic. A small dev without anybody particularly experienced at being a sysadmin seems like one of their sweet spots. Learning the ins and outs of ssl / nginx / caching / security / the many unexpected ways server software will fuck you may be better outsourced.
Anyone knows how Parse's server side is done ? By looking at their API i would say something like AWS + MongoDB, but they don't seem to have known the same kind of problems as Heroku when Amazon went down the other day. So maybe they're using else. Anyone ?
As a sometime developer myself and late-blooming CS student I see things like this and start to feel insecure about my place in the world... if the barrier to entry keeps getting lower so that anybody can build an app, what's the point?
Obviously, I need not feel this way. Why? b/c this trend where eventually 'civilians' w/ no training in info tech or knowledge of conventional programming languages will be in a position to build stuff on the fly implies that the state of the art of software engineering is going to be miles from where we are today, and to me this is a good thing: hardcore software engineers have always been focused on making tools.
How about lock-in? What's your (the users/owners) stance on that? It seems my app will be completely hanging on Parse.com; they can change the pricing, go bankrupt, get hacked, get bought, go down(AWS) or whatever scenario making me wish I didn't use them? This functionality seems a bit too crucial/low-level to be not in-house or open source.
Yeah, but exporting is not really an option... When you are running, have massive traffic, 100+ million api requests/month etc, it's not very interesting that you can 'export all data'. I need the services otherwise I need to rebuild them right? I would consider this a massive risk for my business. Which happens with most PAAS/SAAS thing ofcourse, but this seems to be very fundamental to my apps; I mean; apps won't work without this service.
I like the Ning proposition; if you want you just download the source + data and run on your own servers. Or there should be a buy-out for the software, maybe depending on the number of average monthly API requests you did until that time with some kind of minimum. For stuff this low-level I think a (restricted) open source model is the only way to go to prevent lock-in to be honest.
just an article of "One Ring to rule them all" phenomenon. Entire app industry contains lots of different usage scenarios. It is literally impossible for a platform/library to fit all type of requirements.
Hey, Tikhon, we know you're proud of Parse and you really should be because it's awesome but taking liberties with a headline like you did there might not be so cool. It can be taken one of two ways:
1. The founder of Parse is a total egomaniac.
2. The founder of Parse is really proud of his company and has a great sense of humor.
Parse is awesome though. I signed up back when you guys were first opening it up to people and I remember you sent out a lot of emails to get people to come back and keep using it and you have great documentation and tutorials. I see it "changing the way the app industry works" in the same way AWS, Heroku, and others changed how people thought about web app deployment. The big deal is in giving more people access to the tools they need to run a mobile app that they either couldn't build or afford before. That's awesome because lots of people will build some cool things with Parse... And amateurs like me will be able to churn out crap much quicker and cheaper. Awesome!
How did he take liberties with the headline? "This Startup Could Literally Change The Way The Entire App Industry Works" is the actual headline from Business Insider.
But seriously, Parse looks like it could help some people out, and seems to be doing so already, but let's not be ridiculous. When people say that it decreased dev time by 10-100 times, what they really mean is that they added features that otherwise would never have seen the light of day, and they have no idea how long it would have taken had they tried.
There will always be things like this to help build apps, and they will always come with trade-offs. Things may be easier, but slight changes to those things become impossible. Not everyone will think that the trade-off is worth it, and the app industry will chug along as always, if a bit shinier.