Meteor Dev's Thoughts on Laravel Spark?


#1

From what I can tell, the creator of laravel has released a framework for laravel called Spark. Spark is similar to meteor ( in that spark tries to save time and money by providing stuff out of the box like authentication, payment systems, user systems, etc.).

He is charging $99 for a perpetual license… so pay $99 once, and create as many projects as you want (I think).

What do meteor folks think of this?

The Open Source Religion

There are a lot of “open source zealots” who buy into the open source philosophy without every questioning it. Their reasoning seems to be that everything should be free and the more free that things are, the better they are. But are there others who are more open minded to a model like spark?

To me, this model makes much more sense then having companies like MDG give the framework away for free, and then try to make money on ancillary products (galaxy, consulting, etc.)…

The spark model directly aligns the incentives of developers who use the framework, with the team who is creating the framework.

The more time, effort and features you add to the framework, the more money your “framework company” will make. You also have a huge incentive to listen to the needs of developers who use your framework, and to act on that feedback ASAP…

Compare the above to the current “Open Source + Ancillary Revenue” model that MDG (and basically 99.9% of the industry use), which creates a situation where the framework creators are constantly scrambling to find a solid business model that is usually on vaguely tied to the framework. They end up spending the majority of their time and money building ancillary products and services, rather than focusing 100% on building out the framework, taking in pull-requests from contributors, and just generally trying to make the framework awesome.

Many frameworks end up asking for donations or patreon accounts anyway! Which is basically a half-ass backwards way of going about Sparks much more professional and organized strategy… with this “Spark Model”, there is no more praying that the creator of the framework you use is “raising enough beer money from his paypal button” or that he suddenly decides he needs to take on more consulting work and leaves the repo to wilter.

The Downside of Open Source that Never Gets Mentioned

For some reason, these very real, very common downsides of open source never seem to get attention. They are just taken as a given. Nobody ever asks if the overall cost may outweigh the benefits… and remember, Spark is $99 dollars. We’re talking a difference of Free vs $99… that price is so low it’s virtually open source, but DRASTICALLY reduces the risks and DRASTICALLY increases the quality/satisfaction for developers.

As a new developer coming to the industry with an open mind, and applying some economics 101, I think the spark model would result in much happier developers and a healthier “framework company” than current “open source” models.

What do you think?


Autoform is dying
What is Meteor missing?
#2

You’re comparing a package to a “framework”.

You could compare Spark to Meteor toys and Meteor to Laravel.


#3

I’m not comparing meteor to spark.

I’m comparing spark (the company) to mdg (the company), who are essentially operating in the “same space” with different business models.

MDG is saying “use our path” to make your next app. Spark is saying “use our path” to make your next app. One is “open source” the other is “paid”.

I am not asking which is better, or insinuating a javascript developer should switch to PHP…

My question is what business model is better for developers?

This is a high-level philosophical question, this isn’t a dev wars thread about the semantics of packages vs frameworks, or if you should jump ship to spark. I’m trying to get some conversation going about higher-level about the general direction that the software industry could go towards.


#4

TL;DR Personally I think payin for any code is really lame and limiting.

One of the major reasons I love computing is the essence of open source and really when I heard Linus Torvalds talking in a interview way back when I was only in primary school the whole concept he was talking about with GNU and Linux was really so more intelligent then what else was being presented; and I feel that essence has always been the predominant driving force in why Linux is the most powerful choice. Look at the history, it speaks for itself. Microsoft hasnt got anywhere. Mac is built on *nix and owes part of its success to the GNU world.

We did not reach these heights of web development through a focus on material monetary gains, in fact that was seen as a pathetic lesser intelligent option - a good developer should be able to benefit the world and know that a good life style / abundant security will be provided back by the wealth of knowledge they are offering. As a developer your cup should overflow as you are technically in the upper layers of societies intellectual endeavours.

Like a plant leaves its seeds for us to grow again, open source is required for all functionality of computing to benefit and thrive. If we close source or make it paid for it really just makes it restricted. If meteor was a paid for code base I doubt it would have many people using it.

Paying for hosting is another matter, as it is a service which involves resources, rack time and power and so forth.

I don’t want to sound rude but personally I feel the whole concept of galaxy was ultimately a pie in the sky silly stunt which ensured meteor got enough traction to survive against the plethora of other ‘me too’ frameworks. But it didn’t really need it, meteor is so good at what it does galaxy was just a sweet cream on the top. Completely unsustainable but really a very nice offering.

Well it worked and they have our attention, now meteor in my opinion would prosper with a simple publishing ppc / cpa model where as they sell media units on the forum here, on their own hosting and so forth and even provide a lower tier free hosting on shared units which forces advertisements on to the product. I made all my money so far in this method, so I cannot see how it would not work for meteor.

But in terms of charging for code it’s just asinine and futile, let code be free. Ownership of code is even a grey area I believe really, I mean sure you can trademark stuff and try patent pend it but it goes against the entire ethos if you are using open source packages to achieve that…

The knowledge required to run a good service and make it user friendly is a feature people don’t mind paying for, also considering the money available in advertising return it was a missed trick I feel personally currently they are leaving money lying on the floor.

Meteor galaxy could of been sustainable with just a pop under and some banner ads. Youtube and Facebook purely survive on this model of publishing.

Compare paid for frameworks like Zend to Meteor. It’s hardly infront is it? Or am I missing something :slight_smile:


#5

Interesting answer (the whole post).

One thing that stuck out for me was:

how would a one-time payment of $99 limit you?

To me, it’s just ensuring that somebody will be there to work on the underlying codebase. I don’t really see where it limits you. It doesn’t mean the “framework company” couldn’t take in pull-requests. I don’t see anything that an open source option could do that a “one-time payment of $99” option couldn’t.

It’s really just a formal way to ensure the underlying code base get maintained while you do exactly the same thing you would be doing with an open-source framework.

IMO it’s much more sustainable and better outcome for developers (almost to the point of it being blatantly obvious it’s the better model). I think the open-source dogma that permeates the software world is the only thing holding the model back… you just could never make it work because people take open source software as a given (and automatically assume it’s the ideal model).

Again, many frameworks/packages end up having to depend on donations, consulting, and things that do nothing to push the package or framework forward… all because people vehemently think open source is the only way.

How many times do you read a github issue where the framework maintainer is responding to (basically incredulous and entitled folks) for fixes and improvements? The response we’ve all seen is a sneering “PR requests welcome!”…

80% of the time that PR request never comes.

I guess one good example of open source “failing” is Blaze. Now there is talk about a third-party who takes donations that will maintain it… again… all could have been avoided with the spark model.


#6

Beyond Blaze, here is another good example:

These situations are literally EVERYWHERE (not just meteor, everywhere). But the open-source dogma is just so strong that nobody seems to notice or wonder why…

many folks would probably be happy to pay $19 for a perpetual license to use autoform… then aldeed and a team could work on it full-time. But instead, aldeed (like any human) has to leave the repo to die so he can pick up paying work to feed himself and his family…

Paying for something of value, or paying a programmer, all of this is totally okay with everyone. But the minute you upload code to github (e.g. meteor-easysearch) instead of making it a SaaS Api (e.g. algolia), and most believe paying for it would be just too “limiting” and “lame”

the result of this thinking?

another repo bites the dust. and you end up spending $100/hr on a developer to build whatever autoform functionality you needed (instead of that 1-time payment of $19).

Aldeed loses. You lose. Everyone loses.

I think Spark’s model is the future, and I hope that the open source zealots will approach the situation with an open mind (although I’m not hopeful of this…)


#7

Say a proprietary package is created to replace autoform.
You pay 19$, the maintainer can’t make a living out of it because X and Y reasons.

The package is abandoned, you lose and you cannot even fork the package.


#8
  1. This assumes the best practice for when somebody shuts down their “package business” is to abandon it completely and not allow people to fork the current version. I suspect more likely they would just release it to the wild. And for the small percentage that don’t, are you taking on that much more risk then the current situation? Plus spark is more like meteor (provides tons of stuff out of the box) than autoform which is one feature.

  2. One would assume it’s less likely to be abandoned if the maintainer is paid, no?

  3. $19 is just a random number.

  4. It assumes all the great parts of open source are mutually exclusive from this other model… e.g. the maintainer would not accept PRs… you could never fork it if the project “commercially” fails, etc.


#9

If all code was free, you would have no job… there seems to be little to no material difference between what the open source zealots deem okay to charge for and what they deem not okay to charge for… many times it revolves around whether they are doing the work or somebody else is doing the work.

And again, there is a massive graveyard of work that just fades away… blaze and autoform being JUST two examples that are close to home. If everyone that uses blaze or autoform “threw in” $20, both blaze and autoform would be in much better shape and it would save everyone who uses it many many man hours.

Advertising money may be a nice return on investment for an individual person, but MDG (and others) are taking VC money. They are expecting you to build a business that can do $10M a year and still be growing quickly. They aren’t going to get that with a few forum ads.


#10

Sure.

And then you have something like the Sublime text editor where you pay 60€ for a license.

Only for the guy not to do any kind of development for 2 years straight.


#11

As I see it, if you want to make money from software, you have basically 3 options:

  1. Sell once: You charge an amount ( 9,-, 99,- or 999,- usd) for a permanent right to use it. This is what you describe, and has always been the traditional model. Think Microsoft Office. However, after 20 years of developing MS Office, people are happy with version 2010. They don’t need version 2013. Good for them, but you stop making money. Mmhh, problem. Secondly, why keep pouring money into a product that you have already sold. People won’t buy it a second time, just to support you.

  2. Subscriptions: Take Adobe: every designer already has Photoshop, and it’s not the small improvements you keep making that will convince new customers. So, they’ve started asking an annual price for everyone( even the old customers eventually). This keeps their revenues steady. Same with Microsoft Office 365: the money keeps pouring in without having to invent stupid ‘improvements’ ( MS Office anyone?) to push for an upgrade.

  3. Give away your software for free, so hopefully you get the largest userbase and become the de facto standard( keep the barrier as low as possible for people to start using it). As not everybody is an expert or has that much time, people( and certainly companies) want professional support. And who better to get ( paid) support from, than from the people who’ve designed it.

A lot of software-companies are turning away from the ‘buy-once’ principle. It works fine in the beginning, but the better it becomes, the less you earn.

Subscriptions are nice( from a seller’s perspective) for stand-alone software like Photoshop.

The ‘Service-model’ works best for stuff that requires a lot of support. And Software Development( frameworks, apps, db’s, …) fits best in this category.

Another point pro Open Source as a business-model: You give it away for free, but you also get a lot of stuff for free in return. Many people contribute for free to your product ( in PR’s, plugin’s, writing docs, …). With a closed model, you have to do all that yourself. Imagine how many extra paid Dev’s MDG would need to build and maintain the Meteor ecosystem all by themselves.

eLeventy.


#12

These are mostly (or all) moot points.

If you use spark (which is just built on top of laravel), you can do exactly what you would with laravel-- for instance you can download any package you want off of composer/npm. Spark doesn’t have to get “extra paid Dev’s to build and maintain the ecosystem”. Also, I don’t see any reason somebody couldn’t send Spark a PR, and Spark accept that PR. Additionally, I’m sure many people will make Spark-specific packages that will be free too.

Again, spark’s model is basically EVERYTHING that is good about open source, plus you get a little more “convenience” by paying $99 bucks per project (production projects). You get a a 6-12 month head start for $99 dollars and you increase the chances that Spark can actively improve the framework.

Once again, all you’re doing is letting a person (or team of persons) focus on that framework or package you’re using by throwing them $99 bucks. You get a 6-12 month head start and a team there to help maintain it… for $99 bucks.

All I’ve heard so far are extreme edge cases and moot points as to why its not good. Which translates to “Spark’s model is better 99.9% of the time”.


#13

Dude if you want to sell spark go do it on some newbie webmaster forum not here at meteor seriously :smile:

From reading over this thread it seems you’re hearing what you want to hear. You seem to be missing the point entirely, saying if all code was free I wouldn’t have a job? It’s just insane bro. Do you know there are people who are drivers who are not mechanics and just drive cars for their whole living right? And furniture sales men that never touch a carpentry set. Or chefs which serve meals without ever planting a seed once in their life. That is the nature of economics we delegate and move with our lifes.

Not every CEO is a CTO or even close. Infact technical knowledge of these type of things is a minute sliver of the I.T spectrum of skill sets. Many workers in a office will work on a system but only a handful will overstand the core structure entirely. A few will understand from documentation but asking to modify the system is another point. And here is why we need open source. Here lies the pulse to why we have such a plethora of development tools, it is not due to a proprietary closed based closed mind system at all.

It has already been stated by many people here how obvious the reasons for GNU/GPL Open source is more beneficial to productivity but as you seem to be stuck on this spark thing let me try another approach.

Let’s look at React. It’s a popular view engine from Facebook’s team which is open source. Strange how facebook choose to do this. If you are so correct, but hey - that’s another point for now. I digress.

React works with any web language literally php, asp.net, js, python, ruby and React Native works on the iOS and Android devices.

Looking inside React we see it works with Babel as a transpiler. Also we see people using it alongside Meteor, another project. Or it’s working with Rails. Maybe people are writing some code with Coffeescript, using ImageMagick to adjust thumbnails and serving their pages with Nginx using a Linux distrobution and editing their code in Vim, atom or emacs. Maybe for deployment they use docker or vagrant, or capistrano or mupx… who knows.

The search may be performed by elastic search on their site, or maybe sphinx and their database model could be MySQL or Mongo, maybe with Redis and memcache bolted on for some key based caching.

All of the above are open source.

There is no way you can do what we can do without open source.

$99? My company made $300,000 per quarter in its first year, all from open source tools.

Huge companies like Adobe, Avid, and Mac at the heart of the creative world. A world which pumps a much stronger pulse then web development. Hollywood movie cinemas, television broadcast, government propaganda and such is created using these ‘industry tools’ which are closed source. It does work for these endeavours yes, yet these companies are now moving to a SaS style hosted in a cloud more and more and I believe in the future that will be the defacto method for creative software to have a cloud you pay a monthly fee and have access to various tools for media / recording and broadcast.

So you see the difference here right? If you are selling a software which produces media or in some way helps in a creative industry role, go ahead license it. But for web development and making apps it’s really a totally different environment which has solely progressed due to GNU/GPL and I am as much a dedicated enthusiast then any next Linux system admin.

For me - If they wish to sell hosted laravel for $5 a month I would have no problem, but the code must be open source and able to be hosted from your own server additionally.


#14

I second the motion to pay. Open-source does not mean that it’s free, it means you can access the source. If you are cheap, it’s likely because your project is not making money, so you should be allowed to keep using Meteor for free until you become profitable.

Linus makes a decent living because companies and people pay indirectly to the Linux foundation. React is given for free by Facebook to get more developers they can hire. And so on (Webkit, Chrome, Atom, CUPS) … you need a sponsor who will make enough money to keep developing.

I don’t see how Galaxy can compete with the myriads of hosting solutions out there. If I am right, Meteor can only continue to grow thanks to the community or MDG will have to switch to another model. It’s a matter of time …


#15

I’m not implying no technology should be open source.

I’m trying to start a discussion about if there is a place for something like Spark, and if it could be be a model that carries all the benefits of open source, plus more stability for the given “tool”.

I am not sure if you looked at Spark, but it’s built on top of npm/composer packages… 1000s of open source projects. There is no reason you can’t use it with other open source tools… again it’s just $99 that saves you 6 months in development time and gives the maintainers some money to improve and maintain it…

Most people responding in this thread seem to not understand what Spark is.


#16

Laravel is great and Spark looks also. Monetizing the framework is a more typical business model than MDG’s.
tbh i much much much prefer MDGs way - through Galaxy and supporting real-world use cases
Perhaps naively, I consider this also lowers the barriers to entry for adopters

NB: We should be careful of equating open source with ‘free’.


#17

in an efficient market price is asymptotic to the cost.

If C is the cost of making software, and U is the number of customers, dC/dU = 0, which means that the cost doesn’t change with the number of user, thus lim(C/U) U-> +inf = 0, thus the price per unit of software tends to zero.

This is economics 101.

Closed ecosystems can’t compete with open ecosystems, especially on the interoperability side. If what you say would be true, then the linux kernel wouldn’t be everywhere, we would have Windows NT instead.


#18

I think this is just bad mathematics and made out of false assumptions.

If you have luck a few people would buy a license… but most probably a lot of people would simply take a free alternative or spin up their own. So how much $ value is in a package like this? What would you pay for it per year/month? Most probably not much more than the stated 19$/month. I don’t know aldeed, but the package is nice so I think he’s at least an “advanced developer” with ~40$/h or more on salary. So you’d need two subscriptions to let him work for one hour… In one hour you can’t do much - especially in a codebase like this. So this business model simply wont work out for a single replaceable package with a bunch of alternatives available.

So this isn’t really great and build up on assumptions, but you didn’t even consider if aldeed wanted to code on a package he perhaps doesn’t even use anymore.
I’m coming from cakephp and maintained a few packages with a moderate userbase - when I went away from php I just added a note to the repo and never committed anything again. I did this not only because it wasn’t commercially worth it (not even 100$/year via donations) - I did it because I just didn’t care any more. I simply had better usage for my time and no motivation to work on code I’d probably never use again.

So to answer your initial question: I don’t like the pricing model and I don’t think it would help in general. Technic evolves and packages die. You can try to slow down the second part with money, but in the end it will be wasted time and money.


#19

For the 10th time, spark is not a closed system. It is built on laravel, npm packages, and composer. There is literally nothing that you can’t do without spark that you can’t do with spark. The only difference is it gives you a (very material) head start (similar to how using meteor gives you a head start). And it does it for $99 dollars per project.

The closed system vs open system is irrelevant to the conversation.


#20

Well, I guess we’re all lucky because we don’t have to theorize about this. Spark exists and we’ll be able to see how they fair, from the number of people using it, to the viability of their business model, to their ability to compete with “free alternatives”.

The conversation in this thread has been interesting (to me at least).

I’ll try to keep an eye on Spark and come back here in 6 months to provide an update of how this all played out in real life.

all the best!

  • me