Hacker News new | past | comments | ask | show | jobs | submit login
Don't Be a Free User (2011) (blog.pinboard.in)
202 points by oskarth on Jan 21, 2017 | hide | past | favorite | 99 comments



Unfortunately, it often seems that with too little money, a service disappears due to bankruptcy, with too much money a service disappears due to talent acquisition.

With software as an application, there was still the possibility to continue using software for years when a company shut down. Windows has terrific backwards compatibility and on UNIX you get far by keeping around old shared libraries and giving a hint to the dynamic loader.

With services, you are lucky to get a one-year heads up. Sometimes the service disappears in mere months. For this reason, I only give my money to services that are too big to fail (e.g. Dropbox & Office 365) or have an easy migration path (Fastmail). Other than that I refuse to buy anything but software where I actually own a copy.


> Windows has terrific backwards compatibility

Windows is most backward compatible software in existence.


> Windows is most backward compatible software in existence.

You were downvoted for saying this, but anybody who has read the Old New Thing blog (https://blogs.msdn.microsoft.com/oldnewthing/) or Raymond Chen's chapter on Windows backwards compatibility (http://ptgmedia.pearsoncmg.com/images/9780321440303/samplech..., PDF) will be both impressed and horrified at the lengths the Windows team went to in order to avoid breaking programs.


I think that's what danieldk meant. Terrific != terrible (english is not my primary language so I might be wrong)


I think user5994461 meant that calling windows backwards compatibility "terrific" doesn't do it justice. My dad wrote software more than 2 decades ago that, amazingly enough, still runs on Windows 10 and he still uses it.



Out of curiosity, what is that software?


You are correct


I think that title must be given to OS/360. Usable backwards compatibility for half a century, and counting.

Windows backwards compatibility took a dramatic turn for the worse with Windows 7 when they decided to run old software in a VM instead of keeping compatibility. People had to turn to DOSBox instead, which perhaps ironically was originally made for things like Linux to run DOS software.


I believe that the future will be for applications as SaaS or selfhosted on your own server and everyone will have an account with aws or any cloud provided or self hosted cloud. So all the data will be on your own cloud and the data will be stored in a standardized way so that we can switch to another service and data will always be with the user.


The problem is these days many desktop applications have mobile counterparts and some kind of service to sync the two. If that service shuts down, the application becomes a whole lot less useful. That, and iOS/Android have terrible backwards compatibility, so mobile apps that are not continuously updated often don't work correctly on new OS versions.


I think this made sense in 2011, but now the trend of acquihires has gotten so out of hand that it hardly makes a difference; products making plenty of revenue are still being instantly shut down after being bought out.


VSCO is a curious case.


This is why web services should be open source and available for self hosting:

- The geeks with the know-how to run a Web server benefit from your product for free, but in exchange you get bug reports, patches, etc.

- Everyone else pays, because you also offer a paid hosting option.

But because the product is fundamentally open source and federated, it can and hopefully will outlive its original developer.


My experience running a project/company like you describe:

1) Not the geeks, but the big multinational companies with disposable developer time end up using your open-source solution. Exactly the types of users I don't feel sorry for and would rather extract money from.

2) The "payback" in the form of bug reports and patches is negligible, because nobody likes to do work if they're not being compensated (either through money, or, more commonly in the open-source world, fame)

3) The users of the hosted version don't want to pay anything because "it's open-source, after all".

4) People that do not have the necessary technical capabilities to host your software, but also don't like to pay, harass you on GitHub and other support channels because "your software doesn't work", and it "throws error messages".

The key advantage is that some people (journalists, bloggers) tend to present you in a more positive light if you describe yourself as an "open-source solution" rather than yet another commercial app.


Open Core for the win.

You want to use the base functionalities? Be my guest.

You want Single Sign On through LDAP, AD, etc.? Commercial edition.

You want advanced user management and permissions, etc.? Commercial edition.

You want the product to work behind a NTLM proxy? Commercial edition.

You want to use your own domain name (if hosted)? Commercial edition.

There's a handful of requests that are sure-sign markers that someone is using the product in a corporate environment.

Of course, if the people asking for those features also provide a pull request that complies with the project's pull request guidelines, then in that case, things might change. Though it's quite rare that such a huge feature as AD integration is actually provided by a project outsider (especially one that also comes with good code coverage, respects project coding conventions, etc.).


There are 2 issues with that:

- People who only need basic functionalities don't bring any money

- People who need paid functionalities don't get to try them in the free version

So as much I as like free tiers as a user, for a business I think "free trial" is the way to go.


People who only need basic funcs provide you testing, i.e. product stability i.e. money.


Or open demo for the win.

Everything is paid.

Except there is a free edition for up to N users, or non-production use.

This lets people try the product, demo and experiment.


I can't see why any free software user would consider it unreasonable to charge for delivery of a service as long as the source code for the service is freely available. There's nothing in the FSF / OSI that says anyone is entitled to free operation of software!

There are consumer orientated services like Newsblur (http://newsblur.com) that do this. There are business orientated services such as SugarCRM (https://www.sugarcrm.com/) that do this.

SaaS is the likely saviour of "commercial" open source.

That said - yes users are unreasonable! All users, all the time. Whether it's open source or otherwise. Unfortunately, lots of open source users think the giving up of their time to report a problem means it should be solved, and they can't do it. The only thing I can offer is that it tells you how important what you created has become to them - think of it as a reverse compliment!


There's a common belief among regular people (not actual knowledgeable of the philosophies behind the concepts) that "open source" means everything is free. As a company which sells hosting and development based on an open source project, we see this all the time. That's why we generally avoid mentioning "open source" at all in our marketing (as does Newsblur and SugarCRM, from what I can see).


I think the Busines Source License has cleanly solved this dilemma:

http://monty-says.blogspot.com/2016/08/applying-business-sou...

Free as in Freedom but not free as in beer.

You get to use the software for free subect to limitations (can't have more than N users, can't resell hosted access, etc), until N years from now, when the license reverts to GPL. Point releases restart the clock (so you have to fork the N year old version if Oracle acquired the company).

I'd be interested in a variant that reverts to BSD, but maybe that can't work for some reason. I haven't thought too hard about it.


The Business Source License is not Free as in Freedom. The FSF states the first requirement for software to qualify as Free (https://www.gnu.org/philosophy/free-sw.html) is:

"The freedom to run the program as you wish, for any purpose."

Any limitations on use make the software non-Free.


The FSF doesn't say you can run it when you wish.

Seriously though, you download it now, and there is a contractual obligation that the software will be GPL later. Stallman used to charge people for improvements to emacs. This is the same, but with an extended delivery timeline that (hopefully) leads to better Free software.

Also, note that once you pay, you get all the benefits / protections of the GPL, including redistribution of modifications after the timer expires.

I'm not sure if you can redistribute modifications immediately or not. I would certainly hope so! (Any money for modified versions would go to the author of the original version, and there is no "we're missing feature X" like there would be with open core / demoware so why should they block such things?)


>The FSF doesn't say you can run it when you wish.

Yes it does. "As you wish" means no limitations. It's a superset of "when you wish". By your argument all software is Free software, because eventually the copyright will expire.

Being GPL later is better than being proprietary for as long as possible, but until it happens it's still non-Free.


Can #1 be mitigated with a provision in the software license? "Free for personal use only" kind of thing.


Yes, but that would not be an FSF or OSI compatible license. It would be more like Microsofts "shared source", or GitLabs enterprise edition https://gitlab.com/gitlab-org/gitlab-ee/blob/master/LICENSE where you're giving access to the source code under certain conditions/limitations.


> Not the geeks, but the big multinational companies with disposable developer time end up using your open-source solution. Exactly the types of users I don't feel sorry for and would rather extract money from.

These companies much prefer to pay for support, even for open-source projects.


Companies don't really want to pay for the software because every IT manager can go on long rants about idiotic and broken licensing but they will be quick to pony up for a support contract, white labeling, and hosted services.


Going open source does not save any project. It's just a different way of doing things. If the project is not profitable, it will still go under while being open source with a hosted business model. The question is does this sort of model work with _your_ project.

There have been open source projects that used it just as marketing tool and then shut down the open source version when they grew popular enough. There have been those which have severely limited the open source version to get people to buy the premium version which is hosted.

At the end you are dealing with locusts, they will consume anything of value to them for free without giving a damn about its implications to the future survival of the product. Altruism is nice, but it's not business savvy.


The comparison to locusts is quite beautiful.


The term "free software" is used rather ambiguously in the article. The free software that is referred to in the article is free as in beer. The more commonly used meaning of the word (among hackers) implies free as in freedom. If you want to learn more refer here https://www.gnu.org/philosophy/free-sw.en.html


It seems like the author is indeed referring to Free Software, but then makes a comparison to free-as-in-beer web services.


I disagree: it seems like the author is completely unaware of the "free-as-in-freedom" definition, and all usages of "free software" refer to free-as-in-beer web services.


I'd not take any bet that maciej is unaware of any major tech movement.

Look in the third paragraph from the end:

> I love free software and could not have built my site without it. But free web services are not like free software. If your free software project suddenly gets popular, you gain resources: testers, developers and people willing to pitch in. If your free website takes off, you lose resources. Your time is spent firefighting and your money all goes to the nice people at Linode.


Can we live in a world where we get paid for writing free software?


Yes we do. You can double-license, you can offer consultancy services, there are options.


The article doesn't even talk about software at all, services is what it means.


These arguments fall apart when you're living in a country with no access to online, international payment methods.

And I realize that you don't want to bother with people that are not currently potential customers, but I'm just saying that you're locking entire countries out.


One approach is to get some small percentage of users to pay, and use the rest to lock the minority in (like craigslist, linkedin recruiters).

Figuring out the right scheme (and one that crosses borders) isn't easy though.


Could Bitcoin solve this?


To some extent, yes; You could buy Bitcoin p2p locally.

But it is still a gray area, as it could be classified as foreign currency.


So it needs ads then? Or it needs to be self hosted? What's the alternative for someone to build a business on the internet like pinboard does?


What would be the point in advertising in a country without payment infrastructure? You would have to find local businesses who want to advertise online and who would also have barriers to paying you.


In Tunisia for example, there is a booming payment infrastructure. It is just limited to local currency and local stores. Local Google and Facebook ads are flourishing.


I don't know if there is a generic answer to this problem.


Average user can't afford pay for all the free service they receive. If each user had to pay $20 a month for even only the top twenty websites they use and this would be beyond the reach of the average user; Google, Facebook, Free Apps, LinkedIn, Wikipedia, Yahoo, Twitter, etc.


Is there a way to actually estimate how much money I would have to give per month to compensate for the ads I see on the web? Perhaps some browser plugin that counts ad impressions? I doubt that any one web site makes $20 per month from my visits.


http://www.forbes.com/sites/tristanlouis/2013/08/31/how-much...

Relevant data repeated here. Per-user revenue for some sites:

Google: $10.09

Facebook: $1.63 (although has since doubled)

LinkedIn: $1.53

Yahoo: $1.81


Interesting. Note that these numbers are per quarter, so according to this, average revenue even for Google is only around $3 per user per month.


Appears AVG used to have an estimator called PrivacyFix, but site is now dead. Here's a story covering it:

http://adage.com/article/digital/worth-facebook-google/29304...

Author claims the tool estimated that their value to Google alone would have been $200+ and explains why too.


I don't know about that. I'm sure $20 is too much for some people, but in the United States, at least, $20 isn't much compared to the other costs of accessing the web. It's very easy for a DSL, cable, or mobile contract to cost $50-150 a month already.


$20/month for each of the top 20 sites you use is $400/month.


He's saying $20 per service, though.

The way to make the economics work is the same way the economics of cable TV work - bundling.


Oh, sorry, I misread.

$20 per month per service seems like a ridiculously huge jump for services that are currently "free with advertising," though. Surely nobody is making $20 per user per month with advertising.

At $20 a month per service the bigger problem would be people like me who can afford it just fine but don't think most services provide nearly that much value.


I pay for pinboard, for exactly this reason. I like it enough that I want it to be there tomorrow. Same with LastPass, Fastmail (don't know if they still have free accounts), The Guardian and The NYT.

There is much, much more of the web that I use for free, either because they have their own reasons for for not charging at all, or they appear to have enough of a paying base that I'm merely small-guilty but not worried, or because I don't care enough about them to think about the issue.


I agree with what you're saying, but I stopped paying for the NYT for similar reasons. My subscription isn't enough to keep them in business, so either I'm not paying enough or they're wasting my money.

If I pay for something that I could have easily and legally gotten for free and then it's plastered in ads, either charge me more or revisit your business model. I want to see them stay around, yes, but I wasn't getting the value for my money that I expected to get.


Agreed, my sub fee doesn't pay for the NYT's existence. And worse, I block ads. I'm sure I wouldn't pay the actual fee required to support their current operations.

I think, or at least hope, that we're in a transition period for large operations that are currently ad supported. If ads didn't suck up so much of my resources, and were benign and didn't track me, then I'd relent. But as it is, ads don't just buy space, they buy people and any information the ad networks can get away from collecting, gleaning or stealing. "Homey don't play dat."


Cool, you can pay and be screwed a second time when the company is sold anyways.


It's some years later and pinboard shows no signs of slowing down, at least as far as I can tell. The guy who runs it makes a good living and he plans on keeping it going indefinitely.


I believe what you say about pinboard (and I pay), but Joel makes a factual observation. I can't blame anyone for using a service for free, when free is an official offering.

Merely paying for a service does not ensure that the service will continue, and if your data or operations are integrated into a service then when it dies it can be anywhere between inconvenient to impossible to extract yourself, paid or not.

But that's the nature of the web especially, with its relative low barrier to entry. I shuddered when Fastmail was acquired by Opera, and was relieved when they split. Even old established IRL businesses come and go for various reasons.

Nothing in life is permanent, and so logically some of what you use and know is going to disappear in your lifetime.


Meh. I don't like the subscription model. I have to pay for too many things monthly. Every silly web service is probably not worth the headache of trusting them my credit card details. So I will subscribe to as few services as possible. If there's a lifetime subscription option, I will consider it.


I think many services become popular and see wide adoption, because they are free. When the free element goes away there will not be nearly as many users willing to try it or to use it and love it so much that they will recommend it.


While I don't disagree with the sentiment, I do feel like $5 here and $5 there in monthly charges quickly adds up.

mail, web site/blog (some blog services want a minimum of $10-15!), maybe git hosting, another 5 for a bookmark service (pocket is $5, though pinboard is only $1) a back-up solution, and so on and so on. Should I also have a paid subscription to my internet search and social network services?


So what about HackerNews?


I don't think of HN like a product meant to make money though, more like a big ad for YC


A big ad and recruiting board for the right demography.


> more like a big ad for YC

That is dumbing HN down a lot. This community is much more than that: hackers share cool articles, have vibrant discussions and the comments section is probably the least toxic of all comments sections. YC related posts are very popular and YC itself is promoted heavily, but the community is much more than just YC.


He doesn’t mean Hacker News promotes YC in the literal sense. He means Hacker News is a popular hub for developers which are a target audience of YC. By association, Hacker News as a property of YC helps YC stay in the minds of developers.


I think the point he's making is that its business purpose is advertising for YC. Come for the discussion, and if you happen to have an idea, YC is the first thing in your mind that pops into place when you think, "How can I get this off the ground?"

Hell, as a casual user and a corporate employee, YC is the only startup incubator that I know of.


Its actually very simple: do not use web services that do not provide data liberation into some standard format.

In other words, you take a free ride wile its there then you switch to some other solution. You might be irritated at first because you liked the tool and its workflow but that is the only downside which is not that serious as typically something similar will appear soon enough.


I thought people who "clone them" was considered unethical? a copycat? And then if you start charging money from it, well, be read for the storm...

I totally agree with the logic of the article. And I do hope this will become a cultural thing. I just don't believe it will be widely accepted because of ingrained "free and open spirit".

I think, in practice, you'll be competing with fund-raising start-ups who provides free and better services with "free and open spirit" and lots of money ready to burn, after they took the large share of the market, small competitors will not be the only game in town, instead, they'll be out of the game.

That said, it's amazing that pinboard.in is still alive after so many bookmarking services closed. I guess the philosophy of Don't Be a Free User could work well on specific areas.


Polly Lingual https://pollylingu.al is a little mom-and-pop software creation. We're probably in the lightest shade of green on that chart. Someone help us!


HTTPS error when trying to view. Chrome isn't happy with the ciphers supported.


Http works though. I pinged them on Twitter, but maybe mods should change the URL to http.


Nothing changes. Except, perhaps Facebook might be around a while. What are your thresholds for relying on a service? I still backup my google and fb data.


Reminds me of trello. They gave everything free and made it so hard to pay them money.. And now they are done. Slack is next.


There's a glut of information on the internet. Great information, even. Free great information. The scarcity of free, great information is over.

If some of these sources dry up because of lack of money, that's too bad, but no great loss, because many other sources of free, great information will arise to take their place.


I am a proud free user. When a service closes because of bankruptcy, a new one always appears. Or maybe two or three. So who cares?

There's always a free alternative to every service, especially web ones.


Thank you Mr. Locust, that will do.


A workable analogy only if farmers were actually clamoring to market their fields to locusts.



Ah! Thanks.


Its not predatory but rather a symbiotic relationship. Customer/Users, Organizations (Management/Investors), Employees all are inter dependent on each other a view of the world that devotes attention to one of the three groups is misguided.


@devoply, I think the best term to describe akjainaj is a parasite. A parasite is an organism that lives in or on another organism (its host) and benefits by deriving nutrients at the host's expense.

Based from his response here, I think he is just a troll.


He has a point, consumer psychology is set up this way. As entrepreneurs we have to understand it and deal with it.


Most businesses of all sectors use all kinds of techniques to maximise consumption from customers, some of them very shady. Billions are poured into behavioural studies of customers and voodoo tricks so people spend money on stuff sometimes they don't even need.

As a customer, I feel entitled to take advantage of everything I can. I don't understand the negativity around it. Us consumers have a salary each month, and we are supposed to get everything we can out of it. It's that simple.


Absolutely this. Businesses have honed to great effect their ability to extract money from people. It's just the yang of capitalism that people extract as much as possible from businesses at the lowest price possible (or free).


I don't see how I'm trolling here.

For example, I run Spotify with an ad blocker, so I don't have to pay for it. And years ago, when they limited the amount of minutes of music you could listen to every month, I made tons of accounts to bypass that limitation. Why? Because, in the unlikely case they go out of business, by then some other service would've appeared, run by someone who thinks he won't commit the same mistakes the owners of the previous service made.

Common sense says my system of abandoning/taking services is not sustainable in the long term, but by now it hasn't failed me. And I'm also sure a sizable percent of people do the same I do. Am I a troll just because I'm open about it?


You do have a point, shipping 50lbs of pet food for free was not sustainable for pets.com in 2000 but one could hardly fault customers for taking advantage of such foolishness.

Let's set aside any moral/ethical implications of ab/using free services for a moment.

The problem for you as a customer is that once the service closes it might take some time to find a new service.

Second, your time is valuable too. Making multiple free accounts to save $10 a month is an activity which might make perfect sense for a poor college student but would draw comparisons with Mr. Burns if you are a middle-aged high wage earner.

It took a while for amazon.com to get into pet food business.

When del.icio.us became slow and uncared for it took some time for me to find pinboard .


At what point is it not acceptable to use free accounts or discounts?


Ok let me correct that. I thought you were just trolling. But you're really a parasite.

As defined: a person who habitually relies on or exploits others and gives nothing in return.

It's ok to use free services, to abuse them, especially if you are not capable of paying. It's understandable. A lot of people do that. But to be honest in saying that you don't really care about the company failing or if they go bankrupt for whatever reason even if you are an avid user because a new service will pop up that is free. Seems to be morally wrong.

Your reasoning also shows that you don't have your own company or you're not a developer creating or selling softwares. Because if you are, you would be afraid of people like you who will abuse your company the way you are abusing it now. This is just an assumption. I might be wrong.


Still gonna use Linux, sorry. There should be a disclaimer, "Does not apply to Open Source projects."


He addresses this in the article.


Oh wow. Read it but somehow missed that. Well I'll leave the foot in mouth comment for posterity.


I missed it as well. For future readers:

> free web services are not like free software. If your free software project suddenly gets popular, you gain resources: testers, developers and people willing to pitch in. If your free website takes off, you lose resources. Your time is spent firefighting and your money all goes to the nice people at Linode.


It depends on the service, but I think the vast majority of free products are over engineered in terms of using cloud services.

It costs something like $60 a month in maintenance fees for 1 VPS. Double that and you are safe from hardware outages.

As a dating site for example, you don't need S3, cloud formation, etc.


Oh good apparently owning up to a mistake and leaving it up should also get downvoted. Gotta love this community.


> Please resist commenting about being downvoted. It never does any good, and it makes boring reading.

https://news.ycombinator.com/newsguidelines.html




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: