Page 1 of 1

ER - talk not fork

PostPosted: Sun Dec 07, 2014 10:32 am
by Otto
Easyreport developement goes the wrong direction!

Talk not fork - only together we are strong


We must talk not fork - forking split in the developer community

Best regards,
Otto


http://en.wikipedia.org/wiki/Fork_(software_development)

In software engineering, a project fork happens when developers take a copy of source code from one software package and start
independent development on it, creating a distinct and separate piece of software. The term often implies not merely a development branch, but a split in the developer community

Re: ER - talk not fork

PostPosted: Sun Dec 07, 2014 4:04 pm
by Antonio Linares

Re: ER - talk not fork

PostPosted: Sun Dec 07, 2014 6:35 pm
by Otto
Hello Antonio,

the right one is an user base big enough that the program gets mature to use.
I installed Cristobal version. The version is very good looking but I get many errors. The 2nd last version from Timm is much more stable in production use.
It is many month since we have “the deal” and in practice nothing happened.

I think there must be one Fivewin version and the licenses should not allow forking.
Forking is always splitting up the user base. I think we need discussion and you Antonio as last instance who integrates features and functions.
Same is with Harbour/xHarbour/dbase xbase. If all those energy would be concentrated in a single product we would have progress and leadership.

Best regards,
Otto

Re: ER - talk not fork

PostPosted: Sun Dec 07, 2014 7:25 pm
by Antonio Linares
Otto,

When I made the deal with Timm I did not considered that Timm was using an older FWH version, and modified FWH classes. So when we got his code we found that we could not made it work with the current FWH version, as it was. We can no got back to an older FWH version, thats obvious.

Thats why we had to modify it to adapt it to current FWH version. In the process we decided to modify its look to make it more FiveDBU style alike. Manuel and Cristobal have been pushing it very hard, and I really thank them for that because I was quite tired after all the energy that I put to make the deal, besides my personal circunstances that still did not allow me to be 100% productive yet.

I don't see a fork anywhere. The current code is an evolution of Timm's code. It is a much better interface and implementation, and even if it has still bugs and it is a work in progress I think that we have a great tool that as soon as it start being used my many FWH users, it may become an standard in mid term. Its free and open source, and we care for it, what else can we ask for ? :-)

Supossing that this were a fork, forks are very healthy. When xHarbour started I was against it, later on I realized that many advantages implemented in xHarbour were finally ported to Harbour and more over, Harbour became much stronger than xHarbour. So to me the lesson is well learned: forks are very healthy. Ok, they split the users base, but its the natural evolution not only of the software, but the way life itself evolves (Darwin). Natural selection (usability, robustness, facts) finally decides the one that continues (see how Clipper remains very well alive evolved as Harbour). It works very well and thats why it survives :-) See FiveWin remaining alive for all these years. We left many competitors in the way (dBase, FoxPro, xBase++, etc.)

So there is nothing to be afraid of :-)