Monday, February 21, 2011

Another one bytes the dust

I really didn't see this coming: www.itworld.com/open-source/136873/qts-future-bigger-nokia
Qt was the only professional free-software cross-platform C++ development kit out there, supported by a big name with a big promise. So now what?

Thursday, February 17, 2011

Some business ideas

As i stated in the first post of this blog, one of the main objectives of P2P OS is to have it released as an open source product, and, furthermore, the underpinning P2P protocol should also be released as an open specification. While keeping within the constraints of these self-imposed bounds, this post gathers some business ideas that i thought of, and which could add a business case perspective to the P2P OS project.
  • note: by its very nature this page is a perpetual work in progress as it has to reflect the up-to-date technological and business environment, and i'll keep adding/editing/removing ideas as they come and go; feel free to jump right in with any suggestions/criticism/etc that you may have

For starters, several important things that need to be mentioned at this point:
  • first, "open source" does not mean free, neither as in "free beer" nor as in "free speech". Although a common perception is that open source software must necessarily be free software all the way (i.e. both as in "free beer" and as in "free speech"), the truth of the matter is that "open source" literally only means that the source code of the application is open for inspection to anyone who'd like to peek into the code. In other words, releasing a piece of software as open source simply means that anyone is allowed to inspect the inner working of said software, with the main benefit being to allow the potential users to check for themselves the claims that the software makes for itself.
  • second, keeping in mind that P2P OS is, at its core, an application server (specifically, it provides P2P connectivity services to third-party applications), an important technical detail for the business perspective is that the P2P OS API is based on sockets communications, with P2P OS always running as a background service in parallel with any P2P OS-based application (i.e. as a separate process or thread).
  • third, it is important to keep in mind that P2P OS is intended as the foundation for an application ecosystem which creates its own self-supporting network(s), such that P2P OS-based applications (e.g. collaboration platforms, social networks, etc) will barely require any centralized infrastructure, may that me technical, financial, or of some other form, for their operation.
  • last but not least, there's a new kid on the block starting to flex its muscles, namely the WebRTC project. WebRTC was little more than a mere statement of intent two years ago when the P2P OS project has been started, but it has made some significant progress during the past two years and it currently looks like a serious potential competitor to P2P OS should it evolve into a truly free and open generic P2P technology (which is not the case in its current incarnation).

Now a few words about what i have in mind w/r to how P2P OS can be introduced on the market in a three-stage licensing approach:
  1. in the first stage, start by releasing P2P OS as a multi-platform executable only (i.e. closed source), while also making a promise that it will be released as open source after a certain period of time (about 1 year); this first stage will essentially allow P2P OS to penetrate the market, mainly with the help of early adopters
  2. in a second stage, release P2P OS as open source such that the code can be freely inspected, but do not release it as free software yet; instead, specify in the license itself that the code will [automatically] transition to free software status at a well-defined date (specified in the license); this stage should span several years, and it will be used to accomplish two main objectives:
    • provide the opportunity to increase market awareness and, based on the availability of the code as open source, establish P2P OS as a trusted brand
    • evaluate new business ideas/opportunities, and possibly initiate projects around them
  3. finally, release P2P OS as free software under GPL or a similar license. The choice of an "infecting" viral license (like GPL, but not LGLP) is critical for allowing the development and marketing of commercial variants of the product, in parallel with the free software variant

Finally, some business ideas that crossed my mind until now:

Dual-license
This business model is based on releasing the software under dual license, GPL and a commercial license: specifically, a commercial license option can be added (in parallel with the free-software license described in the three-stage licensing model), such that application developers can purchase and embed P2P OS in their own closed source application for a fee (a one time fee, or royalty, etc).
The main idea here is that while closed-source P2P OS-based applications can be released (i.e. under a proprietary license) even in conjunction with a GPLed version of P2P OS, said close-source applications will require the presence of P2P OS as a separate service that must be run alongside the application (think of P2P OS as an independent "system tray" application that must be started for the application to run), i.e. the applications that will use the GPLed version cannot be self-contained (this restriction results from the viral nature of GPL which disallows embedding GPLed software into a close-source app). In contrast, the applications that will use the commercial license version of P2P OS will be allowed to embed [a binary library version of] P2P OS inside the applications themselves, thus enabling the creation of commercial stand-alone products.

App directory
Okay, this idea might be a bit far-fetched, but i'll write it down none-the-less: based on the fact that P2P OS is an Operating System, i.e. it is a software module that serves third-party applications, maybe a business can be made around an application directory, similar to the application store model of iOS or Android, except that the app directory will only list applications for a fee, without attempting to own/control a complete e-commerce solution (i.e. the app directory would act as a one-stop-shopping advertisement platform, directly and prominently linked from P2P OS' main website). Additionally, a number of derived businesses can also be imagined around the app directory, e.g. subcontracting application reviewers (which will themselves be subject to user rankings), etc.

Application development
The two business model ideas described above (application store and dual-licensing) are both based on using P2P OS itself for generating a revenue stream, but a business model can also be built by expanding into in-house development, and marketing, of closed-source P2P OS-based end-user applications that can be released under different licensing terms for different platforms. For example, a full-fledged P2P Communicator solution can be developed in-house and released for free (like in "free beer", but closed source) for Win/Mac/Linux, while mobile platforms variants can be sold for a fee (Android and BlackBerry 10 are prime candidates here). Should such in-house applications manage to attain high quality standards and address real market needs, they will be able to rip the benefits resulting from P2P OS' reputation as an established and trusted brand.
  • "Pro" connectivity services
    as explained in the P2P networking post, the P2P application server's functionality heavily depends on the reliability of an underlying network of relaying-capable nodes, and said network of nodes is inherently not [close enough to] 100% reliable in a pure P2P network (primarily because there is no guarantee that a sufficient number of relay nodes are online at a given time, as required by a given traffic pattern). In this context, a constant revenue stream for an application vendor could come from selling a "pro" connectivity service in conjunction with its applications, which would guarantee the end users a reliable network of fallback servers whenever the p2p-based main network is unable to cope with the network traffic.
    Furthermore, the availability of "pro" connectivity services in conjunction with P2P OS-based applications is valuable not only to the end users of an app (by providing them with reliable connectivity for a decent price), but also to P2P OS's reputation by establishing it not only as an enabler for free best-effort services, but also as a foundation for reliable "pro"-class services

Sponsorships & partnerships
Well, this is more of a wishful thinking than a business idea, but i still think it has some merit: namely, when P2P OS will be released as free software, and should it indeed turn into a unique and valuable piece of software, i don't think it's totally off the wall to imagine receiving some sponsorship offers from the likes of Fedora, Ubuntu, or even Mozilla (and in exchange display their logo as sponsors on the project's home page, on the application store, etc) . Or maybe some partnership programs could be put in place with free-sofware-based companies, like e.g. imagine providing a free-software P2P communications backbone for Firefox OS, Jolla, or even BB10 (and maybe they'd be willing to provide a network of fallback servers - see the paragraph above); now wouldn't that be cool or what?

Selling the software
I'd rather keep this option as a last resort, but it is a fallback option none-the-less if WebRTC (or a similar free-software project) will evolve towards providing a self-sustained P2P network (i.e. doing away with the reliance on dedicated relay servers): in this case, whether P2P OS got to be released as free software or not, maybe a large software company/organization will be interested in acquiring the software's copyrights and owning a WebRTC alternative in-house - and again i'm primarily thinking of free-software outfits like Fedora, Ubuntu, or KDE, but maybe even a "more traditional" company like e.g. Microsoft, Apple, Viber, or Yahoo might be an interested buyer.