Thread: Does Free Fail?
View Single Post
Fontus is offline Fontus
2009-08-14 , 21:20
Posts: 34 | Thanked: 20 times | Joined on Jun 2009 @ Bulgaria
#11
Originally Posted by silvermountain View Post
Individual development is fail?
The vast majority of applications for the NIT seems to be developed by individuals.
While passion and being part of a community is wonderful - a single developer for an application is also a recipe for disaster.
I take your NIT is your first and only experience with free software. Yes, I don't like the current state of the software for Maemo too, but if you try some "normal" distribution of Linux (e.g. Ubuntu) you will see that great software can be made by free developers.

There several reasons for the current state of the Maemo software and one of them is that currently the Maemo community is small.

Originally Posted by silvermountain View Post
The person gets bored, moves on to something else, 'real life' takes over, he has to sell his device, freemantle looks shiny - you name it - and poof...the development is in 9/10 cases now dead.

Forget about future enhancements, upgrades, support when issues comes up.
This is not a problem for the good free software. When a good free software is orphaned there are always volunteers to take over its maintenance. When a proprietary software is orphaned, it is dead. When the maintainer of a free software does things people don't like, there will be someone to fork the project and resolve the issues. When a company does bad things with a proprietary software the users have no choice but to accept the company's decision.

Some years ago I followed the development of FreeDos. For a long time their COMMAND.COM didn't work properly. At some time the main developer of COMMAND.COM regretfully decided to abandon it. In result two people started to work simultaneously on COMMAND.COM and now it is much better than the proprietary versions in MS-DOS and DR-DOS.

Later the same thing happened to me. I have orphaned software that I have written and despite this this software is used on milions of computers and other people are improving it further even now.

Originally Posted by silvermountain View Post
1) Put pressure on Nokia to support the base applications that were part of the reason we [I] bought the device in the first place (Skype just one example).
Many people here will not do this even if they use Skype (I don't). It is better to ask for open hardware specifications.

Originally Posted by silvermountain View Post
2) Be open to selling/purchasing applications. There are tens of iPhone applications I can think of that I would pay for to have on my NIT.
People should be more willing to sell their applications. If you have committed to accept payment for your development three things can/will happen;
a) The developer, gains a sense of responsibility to maintain the application and not drop it as soon as something seems more fun,
Look what happened to most shareware programs. They became dead when their authors decided something else was more fun. On the other hand no free software with enough user-base has ever died unless better replacement has been found.

Originally Posted by silvermountain View Post
b) If the application is good and gains a user-base it may very well be possible and even desirable to hand it over/sell it to someone should the initial developer have to phase out,
Except in practice this doesn't happen.

Originally Posted by silvermountain View Post
4) Specifically to developers here: Once you have a sufficient user-base - meaning a lot of people that really enjoy and use your application. Start taking some responsibility if you are the only person developing it. What is your contingency plan if you get bored with the application/NITs next week?
I don't think developers should worry too much about this. Some other thing is more important - people at Nokia need to learn how to work properly with self-driven free software community. Yes, now they are doing this much better than 3 years ago, but there are still some things to do. I'd like to propose the following good practices of Debian:
  • The package maintainers should be able to officially declare a package as orphaned.
  • There should be some facilities to support a group of volunteers working for orphaned packages. Very often an important change in the operating system requires to modify a few lines in package, in other cases the bug report contains patch fixing it. This allows a small group (even one person) to keep a lot of packages alive.
  • Someone should check regularly for orphaned packages. For example at Debian (about 1000 developers) they are sending email to developers that have not done anything for some time.
  • At Debian the section of the packages (Editor, Games, Communication programs, etc.) is not determined by the individual developers. This requires someone with proper view of the whole project.
 

The Following 11 Users Say Thank You to Fontus For This Useful Post: