Social networking sites like Youtube, Facebook, MySpace, and Twitter (the Web 2.0 bunch) are not making money. Recently, The Economist wrote about their business model which is, well, not working much:
Web 2.0 still had only one business model, advertising, and the Valley was refusing to admit that only one company (Google) with only one of its products (search advertising) had proved that the model really worked. The older internet firms, Yahoo! and AOL, were doing their best to grab a piece of the action. But the “next big things” were selling negligible advertising, often on one another’s sites. Not one of them has become an advertising success in its own right.
A suggested alternative is for them to make money through the interactions of their users (I don’t know why, but I find it a bit unsettling):
While today, these may not look like great businesses (which hasn’t stopped investors’ willingness to fund them), I’m convinced that the daily interactions of their vast memberships–and their users’ willingness to share their interests, tastes, relationships and intentions, and the massive amounts of data around users’ behavior–will eventually lead to substantial revenues and profits.
These discussions have got me wondering whether we might not be wrong in thinking of the sites in terms of how much money might be made from them. Don’t get me wrong, I use some of them, and I find them very useful, but I think that we should not throw away the idea that they might in fact not lend themselves to being turned into money-making tools.
UPDATE
I did not mean to imply that social networking sites should not be making money, and I did not wish to imply any distaste for money-making. As a person who uses them, and who would like to continue doing so, I would like them to make money so that they can continue operating. This post was meant to suggest that they probably would not make money because their model for generating revenue is largely based on advertising, which, as I noted, is currently not working. Another option would be to charge users for using the sites. I personally do not think this would work because people still view them as a sort of commons, therefore paying for their use might not exactly sit well with the users.
The issue here is that, if these sites don’t make money, who will pay to keep them running. Somebody has to pay for servers, software, maintenance, that sort of thing, and it’s hard to see convincing taxpayers that they ought to be state-supported.
Maybe they will be housed in bigger organisations? Google already owns Youtube, News Corp has MySpace, and Microsoft has invested quite some money in Facebook. They might be able to either derive some value from them, or just make owning them a point of pride. I was referring to their ability to make money, by themselves, as services.
YouTube is owned by Google, and I suspect that Twitter will be before long. Facebook has its own “director of monetization” whose full time and well-paid job is to figure out ways for Facebook to make money off of the interactions of its users. It has to figure this out or get bought by Google. Ditto Myspace. There is much more to “advertising” than this article makes out. Facebook has the potential to sell customized demographic data to anyone who needs it, which could certainly enhance advertising, but might also be valuable to corporations for all kinds of reasons.
I think the interesting question to ask is not whether these sites serve a different purpose than making money–they certainly do–but whether they can survive if it is not the people who use them who benefit from the making of money. If Facebook figures out a way to monetize the interactions of its 180 million people, but it is only the people from Facebook who benefit from that, then who would go one using the site?
“Zotero”:http://www.zotero.org/ has a “2.0 beta”:http://www.zotero.org/support/2.0_beta out with some neat social networking options (though I wish the developers would make some of their other “long promised functionality”:http://community.muohio.edu/blogs/darcusb/archives/2009/02/24/zotero-15 a priority, but hey, it’s free).
Are there any standard lines of thought regarding the financing of open source projects? For example, I assume the Zotero project will crawl to a stop if it ever looses institutional support and soft money (which is not to question the commitment of the developers – I realize that they give more than they get as is).
You ask the wrong question. If Zotero was no longer funded through CHNM, then yes, it might stop being developed actively by them, but if EndNote were no longer funded by Thomson-Reuters, it would hardly go on being developed either, would it? The difference is that in the case of Zotero, some other entity could easily and legally pick up the development, whereas EndNote would need to be legally purchased by some other company before its development continued.
I also think the 2.0 version does address the functionality you point to no (syncing?), or is there something else you want it to do? Maybe you should join the team and urge them to do it 🙂
I don’t know that my question is entirely off base. What I really was trying to get at was what while we know that an individual is often willing to trade their labor for prestige in the development of a relatively simple application, what do we know about what it takes to keep development of a complex project going long term? I suppose I should insert the caveat “user friendly” along with complex. TeX, R, and GRASS are all great success stories but they are also beyond casual users’ abilities.
The syncing functionality has been available for a while now. What I really want is some of the features offered by Scribe (listing as well as biography and listing capabilities) and BibTeX (hierarchical relationships). I’m just not techy enough to be of any real help to development.
Readers should really check out the group feature of the 2.0 beta. I’m not entirely convinced that people are going to be willing to share their bibliographies, but the feature certainly has promise.
i can’t tell you how much I hate the term “user friendly.” Most of the time it is used as you used it here, to suggest that there is some abstract universal metric of user-friendliness which free software fails to meet that proprietary software somehow does meet. Intuitively, I can guess what you are on about (e.g. I think Adobe Illustrator is much better organized at the interface level than Inkscape, but I can also understand why I have certain expectations about drawing programs based on my own experience that makes them appear this way). The issue is not whether software is “user-friendly” or not, the issue is whether a task is well defined enough that anyone skilled in that task can use software of any kind to accomplish it.
But to your point… it’s a mistake to assume that all free software programmers contribute their labor only for prestige. Many of them are paid these days, more so than ever. ergo, the distinction between free contributions from unconnected individuals and organizations which pay people to create software is a false one, because many organizations pay people to create free software (Firefox, MySQL, SuSe, Redhat, etc. all pay people based on various sources of funding, from venture capital to foundation support, to straight up revenue from contracts).
I understand where these misconceptions come from–most of the literature about free software out there gives a false impression of where things stand today… but it’s important to understand just how much free software is now a core part of any software/IT firm’s business plan today.
Interesting comment about “user friendly”: but I disagree. First, I don’t think tasks are necessarily prior to design. It seems to me that part of what software does is teach us to accomplish new tasks, and that well designed software facilitates this, whereas poorly designed software requires us to know in advance what it is we wish to do. (This is why I like using iMovie for teaching video editing. Although AVID and Final Cut Pro have many more features, I find iMovie is designed in such a way that students learn these new tasks on their own, with minimal assistance from me.) Secondly, I see tasks as being part of a larger work-flow, in which we use multiple pieces of software to process the same data. Well designed software isn’t just software that looks nice, but software designed with an understanding of how the tasks it is designed for fit within a larger data life-cycle. (iMovie makes it easy to import almost any content without having to understand about various data formats, conversion, etc as one previously needed to know. Final Cut Pro – although good in many other areas – is actually poorly designed, compared to AVID, for data management.) And third, well designed software encourages us to experiment by protecting us from making mistakes. (Again, iMovie protects one at almost every stage of the game, so that it is nearly impossible to loose data. This encourages one to play with the various features of the software and learn without fear.)
bq. i can’t tell you how much I hate the term “user friendly.” Most of the time it is used as you used it here, to suggest that there is some abstract universal metric of user-friendliness which free software fails to meet that proprietary software somehow does meet. Intuitively, I can guess what you are on about (e.g. I think Adobe Illustrator is much better organized at the interface level than Inkscape, but I can also understand why I have certain expectations about drawing programs based on my own experience that makes them appear this way). The issue is not whether software is “user-friendly” or not, the issue is whether a task is well defined enough that anyone skilled in that task can use software of any kind to accomplish it.
Respectfully, I don’t think any lexicographer would spend much time mulling over whether an entry for ‘user friendly’ requires more than one sense—Accessible to the non-specialist with a minimum of training. ex. “Automatic transmission vehicles are more user friendly than manual transmission.” It’s just a heavy drinker-type collocation that means something other than you would expect given the meaning of its individual parts.
I do get your point and am sympathetic to it. Just because automatics are easier to learn to drive than sticks doesn’t mean they are invariably better. Automatic transmissions are a costlier repair and they’ll never win a race against a comparable manual transmission-equiped vehicle.
Our exchange has brought to the surface two criteria—Is a tool fairly accessible to the general public without extensive training? and Is a tool task appropriate for the task at hand? I really don’t know enough about the open source community to know if bringing affordable, non-proprietary software to the masses is a generally agreed upon goal of theirs. If it is, then the former criteria matters a great deal and they need to be honest about what it takes to not scare people away. (By way of metaphor, I know that Judith Butler has a lot more to say about our world than Anne Coulter does and that Coulter’s formulates rhetorical strategies that appeal by gross oversimplification, but even Butler’s most complex argument could be made in a more accessible manner.) As to the second criterion, open source applications can obviously match the power of proprietary software. What can MS Access do that MySQL can’t?
bq. But to your point… it’s a mistake to assume that all free software programmers contribute their labor only for prestige. Many of them are paid these days, more so than ever. ergo, the distinction between free contributions from unconnected individuals and organizations which pay people to create software is a false one, because many organizations pay people to create free software (Firefox, MySQL, SuSe, Redhat, etc. all pay people based on various sources of funding, from venture capital to foundation support, to straight up revenue from contracts).
Huh, I had no idea. And I have so long found Firefox so superior to other browsers that I had forgotten it was open source. (Which strikes me as a sign of success, but which I suppose is a branding failure.)
Interesting discussion. In my own software use, I’m finding that context and habit matter a lot. So does the fact that both, together with the tasks that I am trying to accomplish, change as the research evolves.
Two cases in point: First, when I became interested in Social Network Analysis, I quickly learned that the academic standard in SNA software is UCINET. Then I encountered Pajek, which is better for larger datasets of the type I am analyzing. Asking around I heard many comments about how people didn’t like the Pajek interface. In retrospect, I can see how, if I were already used to UCINET, I would find having to change my habits annoying. In my case, however, both programs were equally new, and now that I’ve gotten used to Pajek, the thought of doing something in UCINET, where I would be back to stumbling around with an unfamiliar interface, is repulsive.
Second, I wanted to capture the network diagrams that Pajek produces. I stumbled on what looked like a nifty little program called Jing, which not only captures an image but also allows you to annotate it on the spot. Now, however, that I am moving into performing several similar analyses sequentially, I have found that, as I get tired, I am increasingly likely to have errors in my annotations (entering “1991” instead of “1981” for example). Since, however, Jing saves its output as .png, I am unable to alter the annotations. If I want to change what I typed after a file is saved, I have to go back and run the operation again.
That is why, just day before yesterday I began to explore Pajek’s export as EPS command, which allows me to use Illustrator to edit the output. I had been reluctant to plunge into Illustrator, which is very powerful but not, to me at least, intuitive to use. Thanks, however, to Lynda.com’s excellent “Illustrator CS2 Essentials” video tutorials, I have quickly come up to speed to the point that I can do what I need.
The point I am trying to make here is that “User-friendly” is very much in the eye of the beholder confronted with a specific task. Some programs are well-designed in a way that allows a novice to do easily the sorts of things that people usually want to do. Others are well-designed in a way that allows someone with a bit of experience and access to the right resources to get over the learning hump in short order, after which they become extraordinarily user-friendly to the people who know how to use them.
Does this make sense?
It makes sense, but I think tech-savvy people grossly underestimate the “learning hump.” Most people don’t know what a URL is. One of the top searches on Google is “Yahoo.” Many people I talk to think the “internet” is that blue “e” icon. For some people the command line is the ultimate GUI, but those people will always be a minority.
And, personally, I find that a good GUI can itself be a feature, greatly facilitating some tasks – to the point that I will sometimes sacrifice functionality for elegant design.
bq. a good GUI can itself be a feature, greatly facilitating some tasks
No question about that. It can also be aesthetically pleasing, which can be very important when working for hours on end. It’s like having a comfortable seat in a car or an Aeron chair to sit in while you work. At the end of the day, there is much less fatigue. Did I mention that we have switched to iWork for most of our translation business? Elegant is easier in more than one sense.
Blast from the past: Shortly after Macintosh was introduced, someone wrote a marvelous little book titled _A Macintosh is not a typewriter_. I was reading it when I noticed that the coordinator I was working with was still hitting the return key at the end of every line she typed. And we still occasionally get things with tables manually constructed one text box at a time.
Many excellent comments.
The human-computer/software interface is a complex object whose functional and aesthetic properties cannot be understood outside the necessary relation between user and the interface design. The usability of a particular piece of software (and its interface) will depend on the objectives of the user, the user’s expectations and intuitions, and the physical properties of the various sensor and effectors that users employ to manipulate the software (eyes, hands, and fingers for example), and human cognition in general.
Human-computer interaction specialists call this the affordance relation, a term coined in the late 70s by James Gibson, the ecological psychologist. A baseball, because of its size, shape, and mass affords me any number of things (tossing, dropping, trading) but does not afford me countless other things (climbing, eating, and so on). On the other hand, a baseball affords an ant climbing but not throwing. But of the many things that an object may afford some individual, only those affordances which are perceived or acted upon by that individual effectively matter. In interface design one attempts to build perceivable affordances that suggest the purpose of each element of the interface. I think it was Donald Norman which said that a door which requires a sign indicating which direction it opens is a failed design. One way of accomplishing this design goal is to align elements of the interface in a natural way with their targets, i.e. in a way that is not arbitrary and requires less cognitive processing to figure out (e.g. a natural spatial mapping between control knobs on a kitchen stove, and each gas unit (closest left knob with closest left unit)), or in a way that conforms with how the body performs some action (e.g. vertical door handles for pulling, horizontal for pushing).
All other things being equal, if a task takes more steps (or less time) to complete in one application than in another, then it is clearly inferior. Or is it? Objective measurements such as this do not always capture the user experience. If the steps are fewer but more frustrating to achieve, then the user may prefer the slower application. Even the experience of time may not be the same. Programmers using text-editors when making numerous small corrections tend to prefer to navigate to each error by hitting the arrow keys (at least within a small region of the document), and consistently report their belief that it is much faster than using the mouse to highlight each error and then correct it. Yet, experimental research has demonstrated that the opposite is true: the mouse is the faster method. Yet, for the same task, users of the mouse method almost always report that it took longer to complete the task. (sorry, I don’t have a citation for this right now, but if anyone is interested, I can dig it up). Employers might favor the objective measurement; their employees not so much.
Lots of anthropologists in the human-computer interaction field. Would be great to have one guest on this blog.
http://en.wikipedia.org/wiki/Human-computer_interaction has a nice section giving thirteen principles of display design.
Norman, Donald The Design of Everyday Things (2002) is an excellent and easy read.
The future lies very much within niche social networks, which ultimately will have far greater appeal for people within specific jobs or professions. Such niche networks are allowing people to interact and engage with each other at a much more meaningful level. What’s more, suppliers to specific professions can use niche networks to target their communications very much more accurately
Not only target their communications, but get a great source of feedback on their product or service. This is also spawning an entirely new industry in the creation of social networking software and social networking web design. Love them or hate them social networks have changed the global landscape, they are driving business and trends in a way that has never been done before. Used in the proper manner, social networking websites can be very profitable.