Different ways and paradigms - a theoretical post
I know many bright people in software and high tech, both in business and in academia. I feel very privileged. My network gives me new ideas, helps me testing some of mine, and gives me valuable feedback. I try to learn and understand why people and companies do the things the way they do. I then try to utilize this learning in my own work.
In addition to talking to people, I find it good to have some models in my mind when trying to understand how people and companies operate. A good model helps to put things into perspective and answer many "why" questions.
I've found the model shown below usefull in many occasions. It models the way companies utilize open source. As an example, the model can help companies to set goals for their open source operations. They do not necessary need to get very deep if they are on the left hand corner. But, if they plan to laund a new project around what is important to them, they better prepare to invest in it!
OSSI project; http://www.coss.fi/web/coss/research/ossi Another important consideration is the approach towards software openness in general. Does openness have a value as such? Why? What value? Consequences?
We can think about two extremes and then many shades of gray in between. In one extreme we can find true free software advocates. Those who follow the thoughts of Richard Stallman and such. In the other extreme are people and companies who keep everything closed. They never reveal their source code and only give a limited license for the end users to use the binary package they provided. I bet you have you favorite company or individual to think of.
Then, somewhere in between are various ways of sharing code, opening APIs, and giving more rights to the user. Foundations, such as the LIMO foundation and the Symbian Foundation are examples of these. Also, products, such as our internet tablets that have some portions of the code free and some parts closed --- for various reasons -- fall into between these extreams.
Closed ---> Free
I wonder if anybody has tried to develop something similar to the one shown above. Academia? Anybody.? How would you label the extremes? How would you label colored boxes form A to E? Why? More or less boxes? Why would somebody choose a specific box for their products? What are the consequences? Is this even relevant? (I think it is!)
I have something like above often in my mind when thinking about products, individuals, and companies. But has anybody really tried to develop it further? I'd like to the map reality against the models. A good model can help us to explain and understand why things are the way they are. Also, as an example, reading Free As In Freedom helps you to understand the far right corner. It is a good book for those interested in understanding why people feel so strongly about free software -- why it is so fundamental, un-compromized, and clear in many people's minds. There are also good books explaining the far left corner; e.g. from the man himself!
Anyway, please, let me know if you have something more tangible -- papers, publications or thoughts. I'd like to know.
P.S.
See you at the OSiM. Also, do not forget the Maemo Summit.
posted by Ari Jaaksi at 5:52 PM
25 Comments:
Things are not so simple as a one dimensional progression.
For example, the N810 (Maemo) is not open. Part of the platform is open, but not the UI. Contributors are not included in the release process, etc.
It would be interesting to see how different players embrace openness, but I don't think it can be measured as 80% open, or such.
Instead you have organizations which have different parts, and each part can be identified as open, or not.
Microsoft: closed everything
LIMO: open deliverables, closed organization
Ubuntu: open organization, platform and products, closed organizational tools (Launchpad)
GNOME: everything open
FSF: everything open
Some players might be very open in certain regard, but closed on another, maybe a matrix would make more sense.
I would put additionally a "Steve" on the far left end of model.jpg, the "actively-closed-and-restricted" approach.
Of course, I don't mean "this is good, that is evil" sort of thing. It is more like Eloi vs. Morlocks being the target group :)
Ari,
You ask for links to academic papers discussing classification of approaches to open and free software.
A good place to start is probably the paper by Siobhán O’Mahony and Joel West, mentioned here, which discusses two dimensions of openness in managing an open source community: "transparency (letting others watch) and accessibility (letting others have a say)."
how about running windows mobile 6 on this unit-nokia have created this great device with below average software
Hi Ari,
I had a similar approach in my talk at the FrosCon, two weeks ago. Take a look at my last slide in:
http://www.slideshare.net/smancke/mobile-development-platforms-presentation-568185
In my opinion, the biggest problem in the arising mobile platforms are not the license issues, but the incompatibilities due to the lack of standardisations. Platforms do not necessary stop to be proprietary, if they only change their licenses.
Maybe a good representation would be a 2-dimensional graph
x-axis: available api's & standardisations
y-axis: freedom in the license
Oh, the link got corrupted:
Slides
On how to represent such things, a few months ago, I imagined it as an "open-source triangle":
http://www.maemopeople.org/index.php/jaffa/2008/04/20/maemo_org_what_next
There are lots of things that claim to be "open". I can think of at least three.
1. As David Wood graciously notes, one way we identified was the ability to watch and the ability to participate.
2. The same paper also talks about what part of the community is open: the IP (right to use), the development (right to influence the direction of the code) and the governance (right to make decisions like who's in charge).
3. My 2003 paper talks about "opening parts" vs. "partly open." The newer work fleshes out partly open, but the idea of opening parts is still valid. Apple has opened the WebKit part of Safari but not all of Safari. As felipec says, some parts of Maemo are open and some are not.
The focus of the O'Mahony and West paper is on the tradeoff that firms face between controlling a community and getting people to join and contribute to a community. Conversely, the community wants the firm to survive and support the project but otherwise make the community as transparent and accessible as possible. It's not quite a zero-sum game, but definitely there are competing goals here.
Joel West
Regarding your 3 questions "Why would somebody choose a specific box for their products? What are the consequences? Is this even relevant? " I found during my work experience that the criteria to select the type of software (closed or open) was based on the application/product type, security level, required certifications and IPR (Intellectual Propriety)requirement from customers.
Commodity products such as phones are best fit to open source/free software. Interlocking systems for trains, airplanes in-board control systems, require heavy certification, security and are expected to live for long years are less fit for open source. In telecom we have also some issues related to open source software, and they are relared to the IPR requirements from some customers (and IPR infringement rules).
BTW, on the figure, most non-FSF types agree that open source is more open than free software.
The argument for your classification is that free software remains free in perpetuity. However, free software comes with restrictions whereas a BSD or MIT-licensed open source package essentially comes with none (and an Apache licensed package comes with very few).
Restrictions, by definition, make something less open, even if those restrictions are not objectionable to some people. When it comes to free speech, Signapore is inarguably less open than Switzerland, even if the restrictions have little practical consequence for many citizens.
I can super-impose an exponential decay curve on your chart from left to right. The y-axis representing the popularity (measured by units sold/ number of users).
@ anonymous,
>"I can super-impose an exponential decay curve on your chart from left to right. The y-axis representing the popularity (measured by units sold/ number of users)."
What about Firefox? That seems pretty popular! It would break the pattern of your downward curve.
@ lusospikes,
>"Commodity products such as phones are best fit to open source/free software..."
Why do you say that phones are commodity products? Just because they're high volume doesn't mean they're easy to create. There are aspects of modern phone software systems that need to meet demanding security or regulatory requirements.
On the other hand...
>"... Interlocking systems for trains, airplanes in-board control systems, require heavy certification, security and are expected to live for long years are less fit for open source"
It's not clear that open source is unsuited to software with high security requirements. There's lots of academic discussion on this point. See eg here.
Incidentally, these last two comments cancel out. The conclusion is that open source can be suited to mobile phones, even though there are strong security requirements on these phones.
// David Wood
@ david wood,
'What about Firefox? That seems pretty popular! It would break the pattern of your downward curve.'
I believe Firefox is what a Statistician might call an 'outlier'. I firmly believe 'closed' source is best route for targeting consumers and gaining mass apeal/ acceptance.
Is Firefox an 'outlier' (as anonymous suggests), meaning that its significant market share among consumers is unlikely to be repeated by other open source software? Or is it instead a start of a new trend?
The underlying question here, I think, is whether open source methods can be applied to creating products that have sufficient "usability" to appeal to ordinary consumers. It's an important question!
There's no doubt that open source methods originated among developers who gave a lower priority to this kind of usability. However, I believe this can change.
For a very interesting analysis of this point, I recommend the fine paper "Usability and Open Source Software" by David M. Nichols and Michael B. Twidale.
Alternatively see the more recent paper "Why Free Software has poor usability, and how to improve it" by Matthew Paul Thomas.
That leaves one more question: are there any other factors, beyond usability, where open source methods need to improve, before the output of open source software will have wider appeal among consumers?
// David Wood
David, others
Thank you for your comments so far. I need to check out the links you provided!
I agree with you that many open source projects have a lower priority on usability -- but not all. I claim that e.g the whole GNOME project is very focused on usability.
But it seems, as I said earlier (http://jaaksi.blogspot.com/2008/05/phones-no-ubuntu-yes.html), that closed systems seem to offer better usability. I claim that one reason is that usability is a hard dicipline that requires expertise, skills, resources and so forth. You need special talents and working methods to create good usability -- and for some reason open source project and those usability dudes do not seem to get togehter; at least not often enough.
So what other factors in addition to a good usability, you ask. I'd say marketing & distribution. As and example, now suddenly, when broadband operators push cheap Linux laptops together with a broadband subscription, they are selling in volumes. Earlier, normal people could not have an easy access to those product or could not understand why they'd need such products.
You need a "need" and you need an "availablilty". My mother won't ever visit sourceforge! Why would she?
In the recent (ep. 38) FLOSS weekly netcast (as they call it) Mark Spencer, the founder of Asterisk, identifies a few things that help an open source project succeed. One is the size of the target group, which, in the case of Asterisk is "everybody". One is the portion of users that will be able to contribute, which explains why so many FOSS projects are infrastructure projects and one is the audiences need for customization, which is high for Asterisk, since, according to Spencer, people really want their telephony solutions to work in a specific way.
The portion of tinkerers may be lower for users of a gui or IM client than a compiler, but I think we are beginning to see that the shear numbers of users of FOSS will soon allow the success (and victory over propprietary competition) of any, even not-so-infrastructury project and they will bring about the year of the Linux desktop 2009.
Ari,
I have attempted to do something quite similar: compare and contrast community models by two axes:
- governance model (brownian motion vs single company controlled)
- license type used.
http://www.visionmobile.com/blog/2008/05/community-dynamics-in-mobile-open-source/
I think it provides good insight on the pattern of licenses and governance models used by the most succesful projects.
Andreas
Tempe,AZ(Sports Network)-Jameswow goldwow goldwow goldHarden had 18 points and nine rebounds to help No.14 Arizona Statewow goldwow goldwow goldedge Pac-10 rival Arizona,70-68,at Wells-Fargo Arena.
Rihards Kuksiks donated 17dofus kamasdofus kamaskamas dofuspoints for the Sun Devils(21-5,10-4 Pac-10),who have won five straightacheter dofusbuy kamasacheter kamasgames.Jeff Pendergraph chipped in 16 points and 12 boards in the win.
Nic Wise scored 17world of warcraft goldcheap wow goldwow orpoints and doled out seven assists for thewow power levelingbuy wow goldcheap wow goldwow power levelingWildcats(18-9,8-6),who saw theirdofus kamasbuy ffxi gilLord of the Rings Online goldseven-game win streak come to an end.Kyle Fogg and Jordan Hill ended with 14 points apiece in defeat.
Down by 11 late in thewow goldwow goldgame,Jamelle Horne hit a jumper to spark a 13-0wow goldwow goldwow goldArizona run that gave the Wildcats a 62-60 lead with 4:14 left.
The Sun Devilsdofus kamaskamas dofusacheter kamasdofus kamascapped a three-game homestand in perfectkamas dofusacheter kamasdofus kamaskamas dofusbuy kamasfashion Sunday.ASU owns an 11-2dofus kamaskamas dofusachat kamasrecord on its home floor.After knockingdofus kamaskamas dofusacheter des kamasoff UCLA to kick off the homestand, ASU defeated USC,65-53,a week ago...Arizona is 2-7 on the road...Arizona leads the all-time series,138-77.
WoW shares many wow gold of its features with previously launched games. Essentially, you battle with wow gold cheap monsters and traverse the countryside, by yourself or as a buy cheap wow gold team, find challenging tasks, and go on to higher aoc gold levels as you gain skill and experience. In the course of your journey, you will be gaining new powers that are increased as your skill rating goes up. All the same, in terms of its features and quality, that is a ture stroy for this.WoW is far ahead of all other games of the genre the wow power leveling game undoubtedly is in a league of its own and cheapest wow gold playing it is another experience altogether.
Even though WoW is a Cheap Wow Gold rather complicated game, the controls and interface are done in warhammer gold such a way that you don't feel the complexity. A good feature of the game is that it buy wow items does not put off people with lengthy manuals. The instructions bygamer cannot be simpler and the pop up tips can help you start playing the game World Of Warcraft Gold immediately. If on the other hand, you need a detailed manual, the instructions are there for you to access. Buy wow gold in this site,good for you, BUY WOW GOLD.
Weekends to peopleig2tmean that they can have a two-day wowgold4europe good rest. For example， people gameusdcan go out to enjoy themselves or get meinwowgoldtogether with relatives and friends to talk with each storeingameother or watch interesting video tapes with the speebiewhole family.
Everyone spends agamegoldweekends in his ownmmoflyway. Within two days,some people can relax themselves by listening to music， reading novels，or watchingogeworld films. Others perhaps are more active by playing basketball，wimming ormmorpgvipdancing. Different people have different gamesavorrelaxations.
I often spend weekends withoggsalemy family or my friends. Sometimes my parents take me on a visit to their old friends. Sometimesgamersell I go to the library to study or borrow some books tommovirtexgain much knowledge. I also go to see various exhibition to broadenrpg tradermy vision. An excursion to seashore or mountain resorts is my favorite way of spending weekends. Weekends are always enjoyable for me.
igxe swagvaultoforu wowgold-usaignmax wowgoldlivebrogame thsaleGoldRockU
good post :)
Post a Comment
<< Home