That was a deceptively funny answer! Got quite the chuckle from me, thank you.
Could you explain a little bit more about what porting an app from GTK to QT is like? It makes sense now that you mention it but I'd never thought of doing that until now.
I don't remember super well, but I think GTK to Qt was pretty 1:1. The code ended up being longer, as reproducing some of the functionality that GTK had pre-made involved writing custom classes. But Qt allows much more digging into the internals than GTK does, so this was almost always possible. And the documentation is very good.
The threading model was a bit different, requiring message passing to do anything GUI related in the main thread rather than just updating widgets from any old thread with a lock held. But message passing is probably a better way to do this anyway!
Working out how to package up icons into Qt resource files was also a little trickier than just using files on disk.
It took me about two months full time to port about 6000 lines of Python from pyGTK to PyQt4. It's not the most complex application though, and I didn't refactor anything else, I just ported to Qt as 1:1 as possible. This sounds like a long time now that I think about it, but it seemed necessary at the time, I don't remember slacking off. Also I was starting from scratch with PyQt.
But Qt is much more stable than GTK and has better cross-platform support. We ported from Qt4 to Qt5 without much effort and I haven't regretted the switch at all. PyQt is an excellent way to do cross-platform GUIs. We adopted it long before electron was a thing, and I remain an electron sceptic because I don't see much in it that I am missing in Qt.
I fear that this is really getting off track for the main and this might be in response to the wrong comment but is it really that surprising? (sadly)
In another comment of yours you mentioned choosing a pet or a random child? What about another, supposedly easier question, which would you prefer to save? A human life or $23? $23 might seem arbitrary but I picked it from my personal experience with a charity I volunteer with called ICM. It helps those with ultra poverty in the Philippines and I've had the great blessing (and curse) to undertake a tragic trip to the slums to meet these people. So I know the best I can offer is mine and ICM's word but that really can save a life.
I want you to know I don't bring this up to try and start a fight or be argumentative but this question plagued me when I returned home and to some extent still does. How can I buy a video game (random example), knowing that money I've just spent could have saved a life?
How do I, how do we, live like this? I'm an Australian and statistically, we're only giving less with each passing year, and what we do give? As on 2016 a measly $764. For comparison our minimum weekly wage is about ~$670. The least able of us (that are able at all) would only have to give a little over one of their weeks to match that.
Personally I'm the sort of crazy where I think of this almost every time I make a purchase and for the longest time it was killing me. The TL;DR of how I was able to answer this for myself involves a lot of reading, looking and thinking is basically summed up as follows. For most normal humans (myself included) we need to spend on ourselves to make us functional enough to care for others. We are sadly not infinite source of giving.
However I think society today has forgotten or changed enough that we no longer see this self healing as a means to enable us to help others but as the point of living, hence the ever decreasing percentages of people giving but I still have hope.
I'm sorry at this point I don't know exactly where I'm going with this at this but it's not that people value animal life above human life, it's that they value their own life and quality of life above others and as sucky as that answer is, I hope you find it help answer your questions in a manner that makes more sense.
Hey g3rv4 sorry to hear the news. Although I personally don't use slack, when I saw your last post (congrats on the double front page btw) I was super impressed by the stuff you'd added, sorry to see it go.
So this seems rather unfortunate now but the first thought I had on reading your initial post was "wow, I wish he'd come and do some similar stuff for riot[1], but unless slack shut him down there's no chance." However lo and behold here we are, so I thought I might as well reach out. If you haven't heard of it, Riot can be described as "an open source slack" although I think that's unfair on it, it certainly lacks certain features that slack has (although with its current rate of development this gap is shrinking) but also can do things slack just can't.
Full disclaimer I'm not a developer on riot (or matrix[2] the open protocol it implements), just a happy user so I can't speak with true authority but I think they'd be happy for you to make these sorts of changes to riot.
So I invite you to come check it out, I hope you find it compelling. At the very least riot has a means of bridging to slack (along with irc, gitter and more) which you might find helpful in dealing with slack (I know I do).
Hope your future endeavors so smoother, I look forward to seeing whatever you create next.
I am a very happy Riot/Matrix user. Commercial software has a place. I think for things like Photoshop it can make sense. It's a complicated and powerful system. I just can't justify closed source software for a chat application. They are fundamentally simple systems. We CAN build competitive open source alternatives. Eventually companies have to make money, and all the approaches I've seen for that go against the user.
Speaking as someone working on matrix and riot, we would love you to come and do BetterSlack style things to Riot - we are getting tantalisingly close to being “as good as Slack” (except entirely FOSS, decentralised, standards based and with E2E encryption) - and positively encourage folks to customise the apps as long as they don’t introduce bugs or harm the security model. Plus we can just accept changes straight into the app rather than injecting via an extension. Pleeease come help us make it better - #riot-dev:matrix.org would be the place to come :) And yes, we have bridging to slack too.
I'll be honest, I don't have a huge amount of experience using it, never to a company slack sort of thing and there may be some company policy about this sort of thing (security and all that) so I'd encourage checking there but I hope it works out!
Could you explain a little bit more about what porting an app from GTK to QT is like? It makes sense now that you mention it but I'd never thought of doing that until now.