Vuetify has been an excellent companion to me.
I absolutely suck as a designer. I can hide behind libraries like Vuetify and unleash my web app creativity on this poor planet.
So, it was with lot of excitement that I looked forward to Vuetify 2.0, which had been sometime in the making.
Meanwhile, my lack of interest in following up on beta news meant that I merrily continued to use Vuetify without a care of the world. That’s the reason the many breaking changes came as rather shocking news. My enterprisey application background, and a somewhat pseudo-hobbyist mindset for a few applications did not help one bit.
Vuetify 2.0 release documentation of breaking changes did not look daunting at first. I just ignored a small ‘expander’ for maximum effect at a later time.
I was not quite jumping with joy when I actually started upgrading my first few applications to v2.0. There were so many changes to do that I couple of my newer projects forward to focus here. The ESLint plugin is helpful but not to a great extent. Even a simple compile after installing Vuetify 2.0 starts crying about all the things that I had to do to get the application to start. At that time, I have not even started thinking about the minor things - colors, typography, impact to some of my other styling plugins, and the like.
While I absolutely understand my place in the universe and how Vuetify has the freedom to choose its path, I will be 10x more careful about the choices I make in the future. I am also overjoyed that I did not pick Quasar at a time when it was doing rapid changes (before v1) - that would have probably made me suicidal.
It is now, and only now, I can see what Angular must have been to the hapless sods working over there (though to be fair - these changes are not close to the scale of changes seen in Angular). To make matters somewhat more complicated, Vuetify 1.5x is supported till Y2020 - that implies I have to migrate more than few applications, a couple of SaaS applications to the new framework.
While changes are good for an ecosystem and you can’t make progress unless you break something - the potential impact of making changes in a complex environment under tight budget looks disenchanting. I am now nervously scourging the Internet on how Vue 3 will impact my applications.
Though evident in hindsight, I have a new-found respect for multiple projects that struggle to manage changes. I appreciate the way Virk is handling changes in AdonisJSv5.0 and the way NestJS is taking the community along (though I have not been an active user of the latter).
And now, more than ever, the so-slow-to-change ASP.NET and its razor syntax look all the more enticing.