WinMerge future?

Main development forum.

WinMerge future?

Postby kimmov » Sat Sep 05, 2009 7:25 pm

I've been very much inactive last few weeks. The reason is I have new role/position at daily work and I'm very busy with all kinds of stuff. I simply don't have time now, or in following months to concentrate to WinMerge development.

Immediate effects are:
  • I won't be able to do WinMerge stable release(s). Perhaps I can do some experimentals since they are lot less work. But then, if there are no changes, what to release...
  • I can do some patch reviews, but it will take time. Don't expect me to reply to bug reports or feature requests. Or mailing lists.
Longer term I'm considering just abandoning current WinMerge as it is. I'm a lot more interested in working on fresh start based on QT framework. I feel the time amount I can use for WinMerge is much more better spent in developing new than trying to workaround garbage we've added over the years.
kimmov
 
Posts: 562
Joined: Thu Sep 11, 2008 8:51 pm
Location: Finland

Re: WinMerge future?

Postby gerundt » Mon Sep 07, 2009 7:45 am

To be true, WinMerge is currently only you as developer, some translators, David with the manual and me as webmaster. So we live or die with your decision. :(

For the users mean the QT fork a long time without a new WinMerge version. That is not so nice, but on the other side get there later hopefully a better WinMerge. Maybe we can force the work at the QT version, but can still add patches (if we get) to WinMerge and do sometimes experimentals builds? So the people see we are still alive? And we should also publish our plans add the website!

I will support you with the QT and normal version, like I can!

Greetings,
Tim
gerundt
Site Admin
 
Posts: 193
Joined: Wed Sep 24, 2008 8:47 am
Location: Germany

Re: WinMerge future?

Postby kimmov » Mon Sep 07, 2009 6:50 pm

gerundt wrote:To be true, WinMerge is currently only you as developer, some translators, David with the manual and me as webmaster. So we live or die with your decision. :(

Which is exactly the problem there is and why I'm forced to think about alternatives. Current situation is getting way too painful for me and for users. I have no time to fix bugs and at the same time try to improve WinMerge. All other related stuff already takes way too much of my time.

I don't really expect much improvements if we just continue as is. We get couple of fixes now and then and perhaps couple of new features. But nobody really improves things. So I'm afraid the project will die eventually this way - too much old cruft is piling up.

gerundt wrote:For the users mean the QT fork a long time without a new WinMerge version. That is not so nice, but on the other side get there later hopefully a better WinMerge. Maybe we can force the work at the QT version, but can still add patches (if we get) to WinMerge and do sometimes experimentals builds? So the people see we are still alive? And we should also publish our plans add the website!


Yes, it is sad situation. But I'm not sure it is worse than current - look at list of open bugs and feature requests. Many of those bugs would require lots of effort to fix. Heck, nobody has really improved our GUI for years and we are GUI program?! Small tweaks, new options and such but really no any big improvements. Archive support is basically dead.

The thing is - with QT we already have lots of that time-wasting coding done by the framework. So the effort is not at all so big as it may sound. And if we just decide to drop some features it won't be so big work at all. Lots of time in current development is going to trying to fix architecture and dealing with all the limitations of current implementation (and frameworks).

If I have to say some times, its months of course, more than 6 for sure. But might be less than year to have "some kind of new generation WinMerge". No feature-parity with current but completely usable for sure. And if done with some care adding new features will be a lot easier - and taking we start extensively writing unit tests it will be stable from the beginning.

gerundt wrote:I will support you with the QT and normal version, like I can!

Thanks!

I just hope few other people feel like I feel about this project and are ready to participate. Now there are no obstacles like MFC and you can use Linux in development.
kimmov
 
Posts: 562
Joined: Thu Sep 11, 2008 8:51 pm
Location: Finland

Re: WinMerge future?

Postby denisbradford » Wed Sep 09, 2009 10:21 am

What you say makes perfect sense to me - seems a shame for such a cool utility to be available for Windows only. I'd love to help with doc for the QT effort any way I can, including both writing and doc infrastructure.
denisbradford
 
Posts: 46
Joined: Thu Sep 25, 2008 12:24 pm
Location: U.S.A.

Re: WinMerge future?

Postby kimmov » Thu Sep 10, 2009 4:53 pm

This all is of course in very early stage planning still. With lots and lots of open questions.

Something I've though of so far is:
  • start from latest diffutils version (2.8.1), make it compile as a library in both Win/Linux. Add needed API to get difflist. Unlike with current WinMerge we don't directly use diffutils internals which means some more code but also abstracts some current problems away.
  • build diffing and merging lib above diffutils lib. This lib will fix lots of things where current WinMerge has also problems. We don't do diff handling in any of GUI/document classes. And we can reuse some of the current code in here.

I will use QT tools and libraries as much as possible. It saves huge amount of time.

I don't want to use those sucky sf.net trackers, or even their slow SVN. I will use Mercurial and publish code in BitBucket. No repository there yet, I'll create one when I have something to commit.
kimmov
 
Posts: 562
Joined: Thu Sep 11, 2008 8:51 pm
Location: Finland

Re: WinMerge future?

Postby kimmov » Thu Sep 10, 2009 4:59 pm

Oh, and suggestions for the new name? I'm not sure we should call it WinMerge anymore? It might be confusing name for Linux users? But then, lots of Windows users know the name and they may not find the tool with new name...
kimmov
 
Posts: 562
Joined: Thu Sep 11, 2008 8:51 pm
Location: Finland

Re: WinMerge future?

Postby gerundt » Fri Sep 11, 2009 6:16 am

kimmov wrote:Oh, and suggestions for the new name? I'm not sure we should call it WinMerge anymore? It might be confusing name for Linux users? But then, lots of Windows users know the name and they may not find the tool with new name...


Ok, "Win" can also mean the neutral "Window" which every OS use, but of course think the most users at Microsoft Windows.

"qtMerge" is to technical and maybe developers think this program is only for comparing QT projects or something. Maybe replace "Win" with a word like "easy" (EasyMerge), "quick" (QuickMerge) and so on? Or use a translation of "Merge"? Unfortunately the german "zusammenführen" sucks! :mrgreen:

At the moment I like "WinMerge QT" most. We still use the name "WinMerge" everybody knows and geeks know it works on other platforms to because the "QT". And for other users "QT" is just the (code)name of a new version like "Windows XP" or "Windows Vista".
gerundt
Site Admin
 
Posts: 193
Joined: Wed Sep 24, 2008 8:47 am
Location: Germany

Re: WinMerge future?

Postby denisbradford » Fri Sep 11, 2009 2:11 pm

A lot of good names already taken, like DiffMerge and XDiffMerge. Too bad, because diff is WM's strongest suit, IMO.

So my suggestion: XWinMerge

As Tim says, 'windows' is understood generically in every GUI: MS doesn't own it. Plus, if it's a goal to preserve existing users, there's a lot of value to preserving the old name. XWinMerge is still familiar, just adds the X prefix to indicate that it's now cross-platform.
denisbradford
 
Posts: 46
Joined: Thu Sep 25, 2008 12:24 pm
Location: U.S.A.

Re: WinMerge future?

Postby kimmov » Fri Sep 11, 2009 2:37 pm

Yep, there are some good options. I think I'll stick with WinMergeQT as work name for now on.
kimmov
 
Posts: 562
Joined: Thu Sep 11, 2008 8:51 pm
Location: Finland

Re: WinMerge future?

Postby matthias1955 » Sun Sep 27, 2009 9:21 pm

Does your decision mean we stop all activites on Winmerge.
so you don't want new foldercompareclass etc..

Bugs can also be solved by some not perfect changes, but without breakdowns.

There a some tens solved bugreports still open. I can create a list to close.

As soon you have some code on WinMergeQT to share, I will support you so fare I can.
matthias1955
 
Posts: 162
Joined: Wed Dec 17, 2008 1:55 pm

Next

Return to Developers

Who is online

Users browsing this forum: No registered users and 1 guest