Long Term Financing

[phime42]

With increasing popularity of the TextSecure-app, questions about financing the server costs arise. While developing is funded via BitHub, I would be interested in getting to know the long term financing concept.

Thank you very much!

[jedie]

Interesting question.

My 2cents:

financing concept can be to get money for the server via donations. Maybe a good idea is to create a p2p network to refrain from central server? Or it must be easy to create a grid of servers and everyone can add his own server without much work...

[phime42]

I have several ideas to finance it. One way could be the WhatsApp approach - one (very small) annual fee (or at least kindly asking for money once a year). Another way would be the the implementation of donations based on in-app-purchases and, as an alternative, Bitcoin-support. To encourage the user into donating, a badge-system (like a small indication on the profile picture that the user is a recognized donator with a defined colour-scheme which is depending on the amount the respective user has donated).

But, anyways, all of these approaches don't really fit into the Free and Open Source with maximum privacy context of TextSecure. Maybe its best to offer a central donation service on the web with an indicator how much money Open Whisper Systems has backed (maybe via relative values in form of a progress bar) currently has reserved for paying the server costs.

I think it's cruical to the success of this lovely application that the servers are extremely reliable and the service has the very best uptime possible. Only this way TextSecure can be the WhatsApp alternative we all are waiting for.

[benedetto]

For some people this also seems to be some credibility issue. They're like "Hey, those guys don't want any money for their stuff but they will have running costs for their servers so how will they get this paid. Surely they will sell our user data sooner or later like all the other seemingly free services have done before..."

So having some concept to present where the running costs should come from (or a concept that ensures that it is not necessary to care about the money) would take the wind out of their sails.

[grandeit]

Imho a much bigger concern are sms verification costs.

[generalmanager]

Just like @benedetto I also noticed several people who were critical how it's possible for you to pay for servers and sms gateways while there is no big "please donate" button on the website. Many also don't make the distinction between Whisper Systems, which was bought by twitter and Open Whisper Systems, which took the FOSS code WS released and improved on it. It also doesn't help that Open Whisper Systems uses the twitter handle "whispersystems". That results in quiet some people thinking that twitter is somehow behind TextSecure and RedPhone, which should also be made clear right on the front page. The best solution might be a new and prominently linked FAQ page which addresses these issues. That would probably also release pressure from your twitter account.

[jedie]

IMHO there should be a seperate webpage with more information... I find only the main page https://whispersystems.org/ with a small "section" about textSecure and only a few information...

There should be a page with information about textSecure and his feature, security aspects, donotation etc....

[generalmanager]

@jedie I agree, the current situation, where information is spread across the whisper systems page, the blog, the GitHub issues and two twitter accounts (whispersystems and moxie) adds to the confusion.

[moxie0]

Yep, we need a new website (@mkhandekar has done the bulk of the design for a new one). We also need to setup some kind of better community support situation. GitHub issues are great for bug reports, but I get a metric shit-ton of emails (2 per minute at the moment) from users just asking questions that should be covered in a FAQ or knowledge base (zendesk?).

As for money, the infrastructure costs really aren't that expensive. The SMS verification adds up, but I'm hoping we can talk with Twilio or someone in that space about getting a deal to support the project. Other than that, it's mostly developer time. Currently we've gotten some grants and stuff, but it'd be great if someone could setup a nicely done kickstarter or IAP donation thing or anything else to keep funding coming in.

[ghost]

@moxie0 A kickstarter sounds like a good idea for two reasons A) Money, and B) Getting the word out. TextSecure is only worth it if people use it. We'd need to have as few bugs as possible, and iOS support ready to go if we want to do Kickstarter. We wouldn't want to lose any hype because new users find a bug or something similar. I'd gladly help with a Kickstarter.

Also if you need any help moderating github I can do that too. Just go around and remove/close posts which are simple questions or general silly-ness. I'm reasonable.

[generalmanager]

I'm with @Armestam on this one - kickstarter is great, and I think the publicity is even more imporant than the cash. But with these campaigns you only have one shot - and we can't afford to miss. Many of our possible contributors may have already contributed to heml.is (no comment) and may want to wait until it's out before they hand out cash for a similar app. So we are going to have to be ready with a joint campaign for Whisper - the combination of TS and RedPhone, with a really well made video and a slimlined, informative website as well as a TS that's really tweaked for usability. If that's done I'm sure the campaign will have a lot of success, considering how far the development has already come, where it's going (federation, GCM free alternative...) and who started the whole shebang. I mean this is the real deal - transparent FOSS development, all the features the Nerds love with the usability the masses need. And we have moxie ;-)

[mcginty]

Typically with Kickstarter (versus other crowdfunding sites like indiegogo), you need some tangible set of goals that will be completed with the money given, and it should probably be more compelling than "pay our server bills for a year." Any ideas on what that might be?

It could be Whisper (or whatever it ends up getting called), but I don't know if that's enough to get people pumped.

[generalmanager]

@mcginty:

1. More platforms sure wouldn't hurt. The iOS and Android devs are really busy, maybe we could pay someone to get off the ground with Windows Phone. Not sure if we should try to make RedPhone work there too. I think texting has priority.
2. Then there is of course the desktop client. I mean a browser plugin is certainly better than nothing, but plugins for pidgin/miranda/kopete would be a real killer feature I don't think any of the competitors has. The added compatibility an open protocol brings with it can really come into play here. If the campaign gets enough publicity, we may even find some people interested to work on that in their own time.
3. Making RedPhone work seamlessly in all kinds of network settings. I for one have a lot of problems with calls beeing dropped after a while, "server error" messages before the call can even start (probably because my router connects via VPN to my provider...UDP over TCP sucks) and lots of other "funny" behaviour. I haven't really bothered logging and reporting all of that but I'm rather positive RedPhone could use some serious work on that end.
4. And the RedPhone for iOS part, which isn't ready yet and seems to be quiet some time away from completion.
5. Pay our server and texting bills ;-)
6. Our own distribution channel out of Play and the GCM free websockets support for the client, as nobody is really working on that atm. Also according to moxie websockets have some serious drawbacks for battery usage. Mayyybe we could set up the first FOSS push network? I'm sure there are a hell of a lot FOSS Apps that would like to use that too.
7. People seem to dig "self-destructing messages". I'm not sure if it can even be cryptographically implemented. I guess the apps that implement this work kinda like that: Server: "Server here, please delete this message now. And don't you forget the burning message animation!" Client: "Sir yes sir, right away sir!" Modified client: "Who are you?"
8. I know this is really far out - but video support for RedPhone would be real sweet.

I think this should be enough for 2-3 kickstarters. My personal favorites are 2, 1 and 6, in that order.

[jedie]

About more clients. Anyone interests in a Python Client? I started a repro here: https://github.com/jedie/PyTextSecure (Currently emtpy)

[sagischwarz]

I see a desktop client as a very important thing which could give TS an edge over the other messengers. At least the people I know (including myself) hate to use the phone to chat while they sit at their PC, where they could use their keyboard. But I dont know how easy it would be to handle multiple clients for the same account conveniently. Isn't there a private key that the user would have to copy manually to every one of his clients?

[generalmanager]

Just wanted to share what I found: The Open Technology Fund is apparently funding Open Whisper Systems with $455 000: https://www.opentechfund.org/projects I don't know how this is split up between TextSecure and RedPhone, but it looks like a rather nice cushion for now.

[vorueber]

Here is another funding opportunity for Open Whisper Systems: https://pressfreedomfoundation.org/

[ccverg]

There's also Bountysource (full disclosure, I'm on the Bountysource team), and we not only have fundraisers but also bounties and project donations. Fundraisers are great for visibility and raising a large sum all at once, but a combination of bounties and project donations is great for longer term. Good example: Neovim. They started off with a fundraiser, but now direct their community toward project donations and/or bounties.

By the way, bounties are issue-specific and get paid out to the developer who resolves that issue (core team or not), so they are a great way of encouraging development as well.

[generalmanager]

@mcginty Maybe we could tag this as financing or funding? I'd still like to see a crowdfunding campaign for which we could use the above ideas for funding goals. When tagged it should be easy enough to find again.

[manuelfink]

@moxie0 wanna recap here. I join @jedie. For me its really intransparent how Signal and Whispersystems is being financed. Especially the powered by twilio on the app startup screen makes me a little bit suspect, since they provide commercial communication solutions and advertising integration services for phone-call tracking. Servers must be paid, not only developers. I like the ideas above. Also a sponsor is nothing bad, but I appreciate more transparency.