when choice is no choice at all

Well, since Richard started it 😉

First of all, I don’t think Richard is really wrong with his options -> choice -> competition -> better software argument. Being a free-market capitalist myself I can certainly appreciate that.  However, I see what, to me anyway, seem like weak points in this argument.

  1. options -> choice – this really is only effective when the user can distinguish between options and make an informed decision. An uninformed choice becomes basically a “coin toss” or random choice. Therefor nothing was gained, and in fact much may have been lost if the random choice turns out badly. What ends up happening is what I see on IRC all the time: user says, “Should I use program X or program Y?” and then the “support” person says, “I don’t know, try them both and see which one you like better. Linux is about choice.” There is two problems with this. First, the user shouldn’t have to be installing and testing lots of software to be productive and do what they want. Second, the user often times doesn’t know what to look for when “choosing”. There may be important technical differences that the user doesn’t pick up on that might affect them later.
  2. choice -> competition – this is really only effective if the “competitors” know about each other and that competition actually does take place. FLOSS and Linux are in general big enough that it’s easy to not find “competitors”.  I’ve repeatedly seen, and even been involved with, projects that were later found to be pretty much duplicative of what somebody else was doing. It happens quite a bit. Ideally, you combine forces and end up with more developer power, but often people don’t want to give up control or the idea they had, so you continue on with parallel projects that don’t talk to each other. This can even happen at the distro level.
  3. competition -> better software – this only is really effective if there really is competition based on who has the better software. I see a lot of “the project with the flashier website, wins” or social reasons why some software is “competitive”. It is also only really effective if the competing projects are close enough that an innovation or improvement on one side can reasonably be done by the other. In other words, you need to have comparable “things” for there to be competition. If two pieces of software become so different, or so complex/large to be practically incomparable, then competition does not necessarily produce better software. I find this a lot with distos. It becomes very very difficult to adequately compare distros that people resort to comparing inane or cosmetic differences.

Overall, I’d like to say that my position right now would most accurately be:

Choice and competition are a hallmark of a state of development, but are not a necessarily a hallmark of a state of maturity.

When we present new users (and experienced ones alike) with so many “choices” that they cannot in a reasonably way or time decide between them, then a choice is really no choice at all. Instead it is a burden and a confusion and makes the Linux community appear immature. So while the mantra “Linux is about choice” may be true and necessary for now, it don’t really feel like it should be our end goal.

As always, feel free to disagree, I’m always trying to gain a better understanding of the Linux community and FLOSS development.

Advertisements

8 thoughts on “when choice is no choice at all

  1. Good points. And you didn’t even touch the most problematic ones yet. That there are very limited amounts of (good quality) resources available. Splitting those will lead into no project getting the optimal ones. It’s not just about coders (they are the easiest to replace as pure coders) but the people who can actually weed out good requirements from bad, inspire people, have visions and communicate them, think of actually good usability, and wrap up the whole thing as an actual product. Those people are very, very few.

    Also there is this constant hum of nit picking, hair splitting and “my irrelevant technology is better than your irrelevant technology” sort of stuff going on (that’s what ie. LKML is nearly completely about – to exaggerate a little bit). All that really inhibits any chances of having actually working and solid strategy processes in place, which is the largest problem Linux is facing actually. If these people worked together more, it would in many ways make that more possible. No, it wouldn’t make it happen as still 99.9% people would be absolutely clueless, but it would at least reduce one obstacle.

  2. I figure, distributions are a mechanism to deal with making informed choices and acting on them. The Ubuntu Technical Board is responsible for choosing which programs are included in main / by default, not the CC, after all.

    The real criticism to that chain is the link it’s missing: competition -> more or better choices. As users discover competing projects, and as projects discover each other, they have to justify their existence a bit. So they slowly drift into niches, as BSD has done. Bittorrent clients are also highly competitive, each focusing on various niches. Headless torrent clients, tyrannical leachers, RSS subscriptions, etc.

    Normally, one can only evaluate software through it’s use, but with open source, another avenue of inspection is opened. Does anyone use ohloh to aid in making choices between open source software?

  3. I am sorry Jordan, but you are way way wrong 🙂 That was just a joke by the way.

    Anyways, thanks for the follow up post. I am hoping that both of our posts can draw in some really great arguments, as this is something that really hasn’t been studied in the past. People would just blog this or that, kind of like what we have done. But it would be great in the end, to maybe come up with some sort of paper or thesis on this topic.

    One thing I find interesting, is you have a bunch of people trying to market Linux, but none of them has taken this into consideration. The reason I brought it up with you the other night really, or the conversation came up, was that I had heard similar arguments as yours from people at the university. I think having sustainable arguments from both sides could possibly help with the future of Linux marketing.

    Mr. Troll has a good point about the “limited amounts of good quality.” At the same time, I think the freedom of creating or forking at times allows people who have limited amounts of development skill to hone in on it.

    Really great arguments on your post and mine thus far.

  4. I think that in certain applications competition can lead to better software. Take the Instant Messenger wars for example. They’ve gotten better and better and better and then wham, skype drops the bomb with a nice multi platform voip app. There’s some pretty heavy competition there from MSN, AOL, Yahoo, that Google got in on it and now MySpace too, and look at yahoo’s advancements in photo and document sharing… and think about how google stepped it up with the 2.5 gigs of free space..

  5. Joshua, yes. All proprietary. What happened? It they did something that jldugger earlier didn’t realize either. They focused on the correct things and did some actual marketing (which is all about understanding the needs of users). Instead of that the open source projects go into several counter-productive modes such as:
    – You don’t like that? Feel free to submit a patch
    – We don’t care, our is technically better (note: nearly entirely irrelevant point when it comes to software products)
    – We are not here to compete, we are diversity – each to their own taste
    – etc

    You see, there is no actual competition. The fact that there are overlapping moot projects does not mean that there is competition. Even where there is, it is extremely lax because most of these people have no idea of what they are doing. They lack those skills to see the big picture instead of staring at irrelevant tiny points, and fail to deliver that complete product (of which 95% or so is NON-technology and NON-programming related in the end).

    The times that there has been actual competition, ever, in open source can be counted using the fingers of one hand.

  6. http://www.download.com/3120-20_4-0.html?qt=unzip&tag=srch

    161 matches for “unzip” on download.com.

    A fairly extreme example, but proof that choice is endemic if anyone can pick up a compiler and scratch an itch.

    In OpenSource it’s slightly better though, because there’s at least a good chance that a lot of their code will be calling standard libraries or using standard protocols (I’m thinking things like DBus, which are gaining widespread adoptance everywhere).

    Sure I can object to having to choose between rhythmbox and banshee, but someone else is going to make that choice for me, either by way of it being part of the GNOME Desktop, or by distributions installing one application for one thing.

    You can protect users from excessive choice without sacrificing freedom for developers to have fun, I think.

  7. I don’t understand what you are up to. Everybody who is developing FOSS in his free time is free to code whatever he likes (within legal bounds, of course). Overlapping projects are just part of the deal. You won’t get rid of them in this system.

    And they might actually be useful.
    Ask yourself the question, why you are using the email-client/desktop-environment/im-client/distro/… you are using right now? Do you really think this is the kind of *insertapphere* that would have emerged when it would have been the only kind of that application developed? Don’t assume that the design choices that would have been made would magically be the ones you prefer.

    There is not only good and bad software, there are good and bad solutions for the same problem. While we probably agree on this, we very probably won’t agree which ones in particular are the best. As mr. troll mentions, technical details also don’t matter here.
    So choices are subjective (call me master of the obvious), which leads me again back to that it might be a good thing to have multiple solutions. I agree that it makes choosing harder and that you might not end up with what is best for you. Excuse me for saying something elevated like that: I think, this is the price of freedom. There are those downsides to freedom, but you are better off with it that without.

    Some maybe controversial thoughts about competition:
    Contrarily to closed source apps, FOSS does not primarily compete on users. When you are selling software you can hire people to develop it. Those people don’t need to be privately motivated by what they do. But the survival of a free software project depends on the motivation of its developers. In my point of view, this is where the real competition is going on. By the way, the funny thing about darwinism is that the competitors don’t have to be aware of it. Many projects just die. Others not. Some even survive without its founders.
    In other words, FOSS-projects are memes.
    (I have to agree that things go a bit fuzzy when there is a commercial entity behind a free software project)

  8. You’re referring to the paradox of choice. Actually, the paradox has been oversimplified. Here’s a better explanation:
    http://www.igvita.com/blog/2007/01/18/the-non-paradox-of-choice/

    What can be seen here is that choice is good, but too much choice is bad. The book, “The long tail” mentions a further refinement. It states that too much *uninformed* choice is bad, but *informed* choice that states “this is generally the safest no-lose option and further guidance is available if you want to refine your choices”. Essentially, if you provide sane default choices with a rational way to optimize those choices, people can enjoy an unlimited number of choices without difficulty. From what I understand, this is the goal of GNOME (and Ubuntu) and the reason why some features take so long to design (since coming up with a non-intrusive intuitive guided choice system that yields results people actually want is a lot harder than just throwing options at the user).

    Amazon clearly demonstrates this concept. There are thousands of books on every subject, so why doesn’t Amazon go out of business because people are too confused to buy anything? It’s because those choices are guided and rated in multiple ways. The overall rating system, the comment system, the sorted by criteria, and the “people who bought this book also bought” features all help *quickly* evaluate the book and zero in on alternative choices until you fined the book that best suits your needs. People revel in the choices that Amazon gives because they’re organized.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s