If we do something like that, are we ripping people off?
No.
Anyone is always free to download the latest code from CVS and compile their own binary.
What we'd be doing is providing a "service" whereby the source code is precompiled and in a nice dimage installer for you. That should be all kosher w/GPL and open source, no?
Yes, and the more I read in this thread, the better it sounds. And if anyone wants to discuss the 'roll your own' scenario, I can definately open up on that side as a gcc-4.0.1 test builder. If you don't have anything better to do for about a day, go ahead. I will be willing to state that this is not very good usage of computer time, but I work in this world. However, for the ordinary user, a nice donation is always welcome.
As to the Yes above, take a look at what Red Hat is doing. Almost all of the code in a Red Hat distribution is GPL and they charge, not for the software, but for a well tuned installation package and support.
Uhm, 1.8 million downloads...
If someone charge 5 $ per download it would be a lot of money, even if they become 1/100.
But that's unfair, isn't it?
Let's walk through how much this really is. 1.8 million x 0.01 = 18,000 x $5 = $90,000 per year. After we pay web hosting costs, business taxes, and employer taxes (yes, I not only pay my income taxes, but I pay all those wonderful employer taxes that U.S. employees don't always know about) we would be left with about $70,000 in taxable income (which I would pay Social Security and income taxes on) with zero benefits. Since there are zero benefits, let's take out the health insurance premiums (assuming I can even buy them) at $900 per month to cover my family and now we are down to $60,000 per year.
Oh, I almost forgot to take out business insurance in case someone sues me because I live in a litigious country. All of this is theoretical because I don't make near that amount off of NeoOffice.
In other words, giving software away for free is really a crappy business model. So where exactly is the unfairness?
Since there are zero benefits, let's take out the health insurance premiums (assuming I can even buy them) at $900 per month to cover my family and now we are down to $60,000 per year.
? $900 per month for a single family? Bloody hell, the US healthcare system has seriously gotten out of hand (I am assuming you have no more than 2 children).
pluby wrote:
Oh, I almost forgot to take out business insurance in case someone sues me because I live in a litigious country. All of this is theoretical because I don't make near that amount off of NeoOffice.
In other words, giving software away for free is really a crappy business model. So where exactly is the unfairness?
Hmm. True.
But I have the impression that monthly contributions are not such a good idea. At least in NL, monthly contribution are to things people must do (power, insurance, etc), whereas all charity and amenities (TV license) are yearly bills. Also, if you go for annual contributions, they will not show up on someone's account statements, each of which is a chance for people to say they don't feel it is worth it anymore.
My second point is that I think it would be best if donations are asked for as frequently or less often than major releases. My mum pays annual maintenance contracts for some of her software and is really annoyed if she has to pay without having seen any significant update in that year (or two).
My third point is that most shareware costs between $20 to $30. I assume that this amount is not a coincidence, but rather an equilibirum of the price that people will want to pay. Even at $20, that is still four times as much money as you put in your $5 example. Unless you plan to put three kids through Stanford at the same time, that should be a workable income. $50 would scare me off, at least.
I don't mind the 'pay per delay' option, as you immediatly see what you will be paying for. OTOH, it would be good if you keep putting out patches that will update the app throughout a major release. Having to paying three times, for an alpha, beta and final, could be a bit discouraging and will not likely increase the testing member base.
I have one single legal hurdle to think about. In NL, if you pay for something, you are entitled to return it if it does not live up to expectations. But the schemes currently on the table do not have a trail period. I would argue that the previous release is the trial app, but IANAL.
good luck,
Oscar _________________ "What do you think of Western Civilization?"
"I think it would be a good idea!"
- Mohandas Karamchand Gandhi
Joined: Nov 28, 2005 Posts: 299 Location: Near Frankfurt/M. Germany
Posted: Sun Apr 09, 2006 3:38 am Post subject:
I have been reading here for a while, and was not sure whether or not to post. Patrick, as you can see from my engagement here, I am very much in favour of your project. I think I do understand what you mean in your various statements within this thread. I strongly agree with: Earning money IS a critical point, and I think the amount of money you earn must be raised.
Given that I do not speak against this, I may say the following:
First, I am not convinced of the idea of "early access for money" as it has been discussed until now. The critical point of this is: The first users of any milestone are those, who do much of the testing. Especially the Alpha releases can have various bugs (although 1.2 alpha worked good for me). And this could be (could! not will!) true for 2.0 Alpha and for MacTel Alpha more than for 1.2, because in both cases afaics the changes of the codebasis and for MacTel even hardware-environment are bigger than they were between 1.1 and 1.2. Thus at least for the Alphas I would suggest: let them be tested by volunteers, who know the risks, not by paying people. But you could limit the Alpha-tester-circle by having testers subscribe, so that "the latest and greatest" will not spread all over the wold before the Beta is out. Then, beta and final could have an "early access"-solution. Another question is: what, if another serious bug like that installer-thing that made 1.2.2 necessary repeats? In such a case, I think, for reasons of liability, access to the patches MUST be given to everyone immediately. How would this be solved within an early access solution?
My second point is: Ben brought up the thought of giving a possibility to pay a bill to larger organizations. This has not been discussed very deeply until now. I read, what you said about user support versus admin and techpeople support, Patrick. Discussion of the "organizations" thing ceized somewhat after that. But I think, support is not necessarily the only possibility to make a paid version differ from other ones. One could think of perhaps a "technical documented" version for admins (e.g. $50 per milestone and per 10 users, what makes $500 in a 100-user organization), or one could think of "enhanced install", e.g. brought to audience with an easy to use multiple-user installer or thelike. If the problem, Ben named (I can only donate, not pay a regular bill) is a problem for more than one organization - there could be ways to earn money for certain services without having to provide 24/7/365 support. Other possibility would be "enhanced installation" with packages of languages or templates preinstalled (users could give a choice out of wish-list of what they think they need and for getting that preinstalled, they pay).
Now, I do not know, whether any of my thoughts is realistical. But perhaps they help thinking further - at least that would be my goal with that. (If all is nonsense, just throw it in the dustbin)
T.
Joined: Nov 28, 2005 Posts: 299 Location: Near Frankfurt/M. Germany
Posted: Sun Apr 09, 2006 4:31 am Post subject:
@advertizing: afaics there is an ad from google at the patches page. I do not know how much this does bring in, but if it does some good: I would not mind having some advertisement on planamesa-home or other pages involved as well.
T.
Edit: On trinity, if one wished to integrate some ads, there would be space to the right of the trinity-banner.jfif. I am not sure, but, would it be possible to ad a cell there for an ad?
Joined: Dec 08, 2005 Posts: 291 Location: Berlin, Germany
Posted: Sun Apr 09, 2006 6:30 am Post subject:
As an alternative to the "Donate & Get Early Acess"-model I would like to suggest another: "If You Want The Binary, You Have to Pay". The project is at an important milestone, the upcoming release of NeoOffice 2.0 and an Intel version of it. Why not take money of all users for the new version? It would mean, pay once & register for the right to download all binary versions (alpha, beta, official, updates & patches). If the developer has to live from his work, and the project is in danger of being discontinued, I would suggest, that everybody should give some money. That way, the amount could be kept small, say 10-15$ maybe. Additionally, I would offer institutions to pay more for the right to redistribute.
As to the question: People will want to try out the software, I'd say: They can try the (still unpaid) version NeoOffice 1.2. If somebody doesn't want to pay, there's always the X11-version (Well, and then there is the freely available source code - ).
The amount of downloads will certainly decrease, but there is no harm in losing those, who download the software, because it's "free" & then they never use it.
Whatever model is being planned & used, I would suggest, that it is thoroughly & extensively communicated and explained to the public. I am sure that most non-geeks have absolutely no idea about how open source software projetcs are structured, that big things like Firefox, Apache and OpenOffice are financed by corporations & NeoOffice simply not & that doing NeoOffice is a 70 hours a week job without the guarantee of a fixed income.
I dont think anyone would object to paying $10-15 as DocType suggested to download NeoOffice 2 with v1.2 remaining as a free download.. Maybe the alpha Neo2 could be released publicly on the site when it's usable in order to tempt people into upgrading to the final version when released?
Joined: Mar 15, 2006 Posts: 2 Location: Nice, France
Posted: Sun Apr 09, 2006 8:20 am Post subject:
I agree with Doctype and Longshot : if you want it absolutely free, you can : use OOo (but with X11), compile it yourself, the sources are freely available, or use the old 1.2 series ; if you want the brand new and full-featured Neo 2 beta or final as an easy to install binary, we ask you to pay 10-15 $, in order to support Patrick, and help him going on with his great work.
If you're still using 1.2.x, then don't you dare file bugs
A) they might already be fixed and B) they won't be getting fixed in 1.2.x
Do we support 1.2.x until 2.0 final is out? Traditionally, when any new milestone was out, the previous version or milestone became unsupported, and all users were told to upgrade (the one exception being that we waited until Neo/J 1.1 Alpha 2 to declare it more stable than 0.8.4)....
ovvldc wrote:
OTOH, it would be good if you keep putting out patches that will update the app throughout a major release.
That's only viable as long as Neo 2.0 remains on the same OOo 2.0.2 codebase forever...if Alpha 2 or Beta has to jump to 2.0.3 to get fixes (or, more likely, to stay in sync with Neo-Intel which will need OOo fixes as OOo-Intel gets more usage/testing), patches that seemlessly upgrade don't work.
Smokey _________________ "[...] whether the duck drinks hot chocolate or coffee is irrelevant." -- ovvldc and sardisson in the NeoWiki
Posted: Sun Apr 09, 2006 10:04 am Post subject: More on financing NeoOffice development
I've read through this thread and am pleased that my comment generated more discussion on the matter. If people are interested, I have a few more comments after reading the follow-ups:
NeoOffice Extras
I quite liked the idea of an optional pay-for download that had additional languages, clip art, fonts etc. You'd still get the core program for free, but interested users could get a few extra features. I know I'd be interested in a version that included the LaTeX integration that I know can be set up with ooolatex. It's been on my list of things to look into for a while now. If a replacement icon set is ever finished, this might also be a nice optional extra. I know I would like to contribute, but really have little interest in digging into OOo's code. Perhaps this is something that the rest of the NeoOffice community could work on without directly involving Patrick. We could put together the addons, and hand the finished product back to him for sale on the site.
Withholding binaries
It has been suggested that easy-to-install binaries be withheld. I think this is a mistake. Something like this was tried with Libranet Linux with rather poor results. It also increases the chance that someone will get fed up with this, compile it themselves and release a free binary a la CentOS. If you stick with a fixed, limited delay, those eager or in need will happily pay, others will happily wait. In my opinion, happy users are long-term users. While people here aren't doing this solely for charity, I'd be surprised if all involved weren't touched by chance9's post. This is a great aspect of OSS development that wouldn't happen without an easily installable .dmg. I'd bet that Patrick would miss it.
If I was to just discover NeoOffice now, I'd be rather annoyed if I could only download the source or pay for the binary. This actually just happened to me with an X11 server for OSX. I just gave up and stuck with the less than optimal X11.app that Apple provides. Perhaps I would have just walked away from NeoOffice if it had been pay-only instead of donating to it after having used it for a while.
I much prefer the idea of providing paid, early access to development and final versions, with slightly later finals for free. This accomplishes a few things: It limits testing of the buggier versions, thus relieving Patrick of some of the support burden. At the same time, it gives tangible benefits to subscribers by allowing early access to finals, and still allows a quality release to the general, non-paying public in a timely fashion. This also means that only well-tested bug-free versions would end out on cover CDs.
Business purchases
I would also like to reiterate that, for my workplace anyway, it's important to get some kind of a receipt for products or services rendered. It's orders of magnitude easier for me to justify paying for something if I can say “we get this in exchange†and then I get a real invoice to follow this up.
As for the “Enhanced Support†version that I suggested and that Patrick responded to, it's too bad that someone doesn't set up a business that could handle this. (Not that I'm volunteering...) I wonder, if you set the price appropriately, if it would be much of a support headache for Patrick et al. if what you were paying for was just a direct email to Patrick, Ed and other major NeoOffice contributors for fast bug questions and a chance to bend their ears. On the other hand, if you did set the price high enough for that, you might not get any takers.
Last edited by ben on Sun Apr 09, 2006 10:37 am; edited 1 time in total
/me goes back to his cave _________________ "[...] whether the duck drinks hot chocolate or coffee is irrelevant." -- ovvldc and sardisson in the NeoWiki
Interesting. I though about the "charge everybody $X" approach. Initially, it sounded good, but there were were things that bothered me about it:
1. Bandwidth and labor costs - Once I start charging everybody, I will most likely lose all of my mirrors as my mirrors volunteer to distibute free stuff, not commercial stuff, so if more than 10% of the existing downloads purchase actually pay under this approach, new bandwidth costs will eat up some of the new payments. In addition, when people have to buy a product without trying it will likely raise expectations that it should not crash. Since I do not have the time to extensively test NeoOffice's underlying OpenOffice.org code, performing this task would cost a huge amount of money that I do not have.
2. Liability - Under this approach, I am selling software. As such, I'd have to accept refunds and be subject to general product liability. Right now, I do not sell NeoOffice and there are no buyers, there are only donors. In constrast, under the "early access" approach, you are buying a service (i.e.early delivery of the binary before everyone else) so there is no refund to claim. Some might claim that the early access fee is a product sale, but the fact that I will openly advertise that "if you wait two more weeks, you can download the binary for free" makes it pretty self-evident that you are not paying for the product.
3. Not really open source - Under this approach, I would be really stepping or the line from an open source, community-driven product to a commercial product. While the code would be accessible, requiring payment to use it is really a commercial product. Ed and I created NeoOffice as an open source project because we wanted it to be open souce. I am not ready to throw that away.
While the "charge everybody $X" approach sounds OK on the surface, it is really a very radical change and so I am going to stick with the "early access" approach for now.
To clarify a bit further, the "early access" is really simple and, in a few posts, I think some of you are splitting hairs. Basically, the idea is as follows:
1. Instead of announcing a new release after the fact like I do now, I will pre-announce the next release on the download page. The download page will still have the last free release. In other words, the top of the download page would have text like:
Quote:
The current release is NeoOffice 1.2.2 and can be downloaded for free by clicking here. The next release, NeoOffice 2.0 Alpha, will be downloadable for free after XX/XX/XXXX. For those of you who want NeoOffice 2.0 Alpha now, click here to find out how you can participate in our "early access" program.
2. Since users only need to wait a few weeks for the binary to become a free download, I have no fairness issue as you will get it for free. The marketing is those who really want the latest and greatest before everyone else will click on the link, see the text about how much required to get an early access download and a PayPal button.
There are roughly 200,000 hits on the NeoOffice download page per month so I really wouldn't be surprised if somewhere between 1% - 5% of the people decide that it is worth paying a small amount of money to get the binary a little early. You may think this is silly (I do) but apparently this is a very common phenomena: there are a certain small percentage of people that want stuff before everybody else and will pay to "be the first" to have it.
So if you don't want to pay, what changes? The simple answer is that you just have to wait a few weeks longer than you did during the NeoOffice 1.2.x release cycle. Some of you may not like that, but that is too bad. The existing system has proven to be unsupportable and has, in essence, become like a government entitlement program: very costly with no revenues.
If the early access approach doesn't work, I can try something else without having crossed the line to being a commercial product.
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum You cannot attach files in this forum You cannot download files in this forum
).
