I first want to point out that the transitions that we dealt with in
Polymer 2 and Polymer 3 we mostly out of our control. The Web Components v1
spec we just different in some important ways, and we tried our best to
ease that transition in Polymer 2 with hybrid mode. Then HTML Imports was
clearly not going to be adopted while JS modules were implemented in every
browser, and Bower was deprecated and didn't have many important packages
available on it. We tried to mitigate those transitions by not changing the
Polymer 3 API at all and building Modulizer.
But there's really not much we can do about libraries that haven't made the
transition. There really isn't a way to import HTML Imports into JS, which
is one of the reasons it didn't get adoption. The difficulty of using
Polymer 2 components in Polymer 3 is the same difficulty that the entire
rest of the web development ecosystem has had with Polymer until Polymer 3.
The best thing going is that if the components you use are open source they
can be forked and converted, and again, the Polymer API itself didn't
change.
The good news is that the Web Components v1 specs are stable and
implemented in 3 major engines and in-development in the 4th. So we won't
have a Polymer 1 -> Polymer 2 like transition again. And JS modules are
implemented in all engines, and the whole ecosystem has coalesced around
npm, so we won't have a Polymer 2 -> Polymer 3 like transition again.
What we will have is much smoother, incremental transitions like Polymer ->
LitElement, where we can easily mix and match components from different
vendors or built with different helper libraries. The largest barrier there
is needing to manually listen for Polymer's {property}-changed events to
implement 2-way data-binding if you use that, and there are helpers for
that popping up. There's no application-wide blocker like incompatible
specs or package managers.
So I apologize for the pains, and we really have tried to minimize them.
The only way this could have been avoided though, would be if the Web
Component v0 specs, including HTML Imports, were implemented by all
browsers, and the ecosystem coalesced around Bower. That just didn't happen
though.
Cheers,
Justin
On Tue, Nov 27, 2018 at 8:35 AM <[email protected]> wrote:
> I share your frustration Joerm, as a small startup who started with
> Polymer 1 we still have some of our components in V1. It's messy business
> but V2 upgrade to 3 is even worst just like you mentioned, having issue
> with many components in V2 still. luckily we sort of invested some time and
> just made pretty much all the components ourselves but that comes with it's
> own cost. Overall, we're grateful for Polymer team's great work and yet
> expect a little bit more attention to the service they're providing
> otherwise they can easily send many people down some rabbit wholes and
> leave us in a ditch.
>
> Cheers
>
> On Tuesday, November 27, 2018 at 8:03:13 AM UTC-7, Joern Turner wrote:
>>
>> We're still developing Polymer 2 applications but considering the move.
>>
>> However there's one big questions that comes up. What to do with
>> third-party components that still rely on Polymer 2? I know there's the
>> Polymer Modularizer but shall we really convert third-party code so that it
>> works in our newly ported Polymer 3 apps? That doesn't feel very good as
>> you suddenly end up with maintaining foreign code and have to look after
>> updates and fixes for those components yourself.
>>
>> There are piles of good components around at webcomponents.org which are
>> still in Polymer 2 and you never know if they will ever be ported to
>> Polymer 3.
>>
>> Are there alternatives to the conversion?
>>
>> Another worry: what happens with the next major update? Will we need to
>> port all our our components (which are certainly hundreds already) with
>> every new major release? That really puts a huge burden on developers (and
>> project leads looking for the budget).
>>
>> I understand that the standard itself is still moving but these questions
>> should be considered before such breaking changes are set in place.
>>
>> Thanks for your thoughts on this,
>>
>> Joern
>>
>>
>>
>>
>> Follow Polymer on Google+: plus.google.com/107187849809354688692
> ---
> You received this message because you are subscribed to the Google Groups
> "Polymer" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/polymer-dev/f05dd191-3948-480e-b9e1-de473834eefb%40googlegroups.com
> <https://groups.google.com/d/msgid/polymer-dev/f05dd191-3948-480e-b9e1-de473834eefb%40googlegroups.com?utm_medium=email&utm_source=footer>
> .
> For more options, visit https://groups.google.com/d/optout.
>
Follow Polymer on Google+: plus.google.com/107187849809354688692
---
You received this message because you are subscribed to the Google Groups
"Polymer" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/polymer-dev/CAEKsHmBi0ZuYsaPZfEEhGm%2BcS5dsXz98%2BpOm4L8n_zR1AxS5Bg%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.
