[ Date Index ][
Thread Index ]
[ <= Previous by date /
thread ]
[ Next by date /
thread => ]
On Sunday 10 October 2004 8:06 pm, Dave Trudgian wrote:
I could not have been paid to write that code as open source.Simply untrue. Lots of people are paid to produce free software.It is not untrue... if you know anywhere that would have paid me to write open source code for an intranet application that produced recipes for ceramic bodies to customer specifications, allowing comparison with competitors prices, as open source I'd love to know about them!
OK, you've got a small number of potential users there but this hasn't been a problem before. There are plenty of SourceForge projects that have similar niche sizes. I wouldn't dismiss it completely - every project has to start somewhere. I apologise if you've thought of all this, but it doesn't hurt to spell it out in case others are considering something similar: 1. Maximise existing tools. When I started with GnuCash I was astounded by the number and diversity of existing libraries in Debian and on SourceForge. It takes lots of research but it could pay off. A lot of the code may well be written already. 2. Generalise the program backend. The front end may well be specific and customised, but the programmatic needs of the program are always more general. It has to have a data store, a UI, engine - lots of these things have RAD tools or prebuilt packages. Could you bolt it onto an existing backend - it sounds like it could use a database, bits of it could even be in PHP or Perl seeing as it's intranet based. That radically simplifies the project. 3. Get others involved - this might seem a small area when you consider only your locality but concentrate on keeping the specialisms in the front end and you will find others are working on the same goals for different front ends. It took me 3 months before I was able to start writing actual code (at least code that was good enough to still be in the current release). That kind of investment is simply not going to be funded in most cases.
Oh, they also use Microsoft software to do this.
That's just a question of export/import filters. If the Microsoft program is any good, it'll have a decent range and there may be a way of managing that.
I may be a fan of open source etc but where in St Austell was I going to earn enough money to afford my rent by writing open source? Nobody would pay me to do it full stop.
Can you do what a lot of other developers do and do without deadlines and fixed dates, do whatever you need to do for other paid employment and write this in other time? It's not easy but not impossible either. If you start the research and get something on a website like SourceForge you might be surprised at how much interest / help could be available.
I did not say that people cannot be paid to write open source code in general, I said *I* could not have been paid to write the code that I wrote as open source.
Fair enough, there's the old Perl adage though, TIMTOWTDI.
Also, I am thinking in the present, of what is possible now, not what could be possible if comapnies beleived in the ideas of the FSF.
I'm fully aware of how lucky I am with my circumstances, but (and this isn't meant to sound hard or elitist) I worked damn hard for my qualifications and if they can fund something that actually interests me and benefits a lot of other users, I count that as mitigation for a bad choice of career in the sixth form.
It does not employ many people though does it? Are there any figures of how many people earn a living from free software?
Good question - I'm not at all sure who would have that info. I dare say you could count the majority of RedHat employees. It gets a bit mirky with IBM and Novell because it's hard to isolate the funding strands. However, the fact that Debian perhaps doesn't employ many developers directly does indicate the success of alternative models - there are lots of ways of finding the funds to make ends meet that are outside code.
I have seen the publication to which you refer and it's something I find perplexing. I admit I've not read it completely but some of the arguments really didn't do it for me. Sorry I cant' give exact examples, was a little while ago and I don't have time atm. Suffice to say it didn't demolish my arguments in my mind.
You're welcome to peruse it at your leisure. I'm not due to be working in St. Austell for a while but we can probably sort something out. (I'll add it to the library when I get half a chance - only bought it myself on Wednesday!)
Looking at the site, I remember "Why "Free Software" is Better Than "Open Source"" well. I'm one who quite passionately beleives the opposite
I would think carefully about that. How are you going to protect your code for future open source developers if your licence is not GPL compatible? If you use a BSD style licence, you could find your code being lifted straight into a proprietary product. The question of free software and open source is more about the future than the present. It considers the code from the stand point of the NEXT generation of developers - if the code will be useful into the future, it is selling the future short to fail to provide access to modified versions.
, and RMS's essays really get to me a bit. I much prefer Eric Raymond's pragmatic perspective on things and applaud his "Show them the code! approach to getting people on board open source.
That's fine as that intro, I'm just as concerned with what happens tomorrow. Licences are not an instant thing, the effect of copyleft and the GPL is FAR more pervasive (as Microsoft acknowledge) and has benefits that some open source proponents frankly don't seem to have considered. -- Neil Williams ============= http://www.codehelp.co.uk/ http://www.dclug.org.uk/ http://www.isbn.org.uk/ http://sourceforge.net/projects/isbnsearch/ http://www.biglumber.com/x/web?qs=0x8801094A28BCB3E3
Attachment:
pgp00023.pgp
Description: PGP signature