On 12/26/19, 12:35 AM, "Carlos Rovira" <[email protected]> wrote:
Hi Alex,
I think you already did a good job maintaining making think PAYG and
avoiding "just-in-case" cases. Things are under control right now. Users
should now be happy on that concrete Royale feature. So what I'm saying is
that we can focus on "end users value" since PAYG is already very good,
even in Jewel.
To me, that's like me saying "now that I'm 58 and mostly healthy, I can stop
exercising and being careful about what I eat". And for sure, I will have more
fun in the short-term.
What I'm afraid is that making PAYG reach new quotes of perfection is not
good for Royale now. Since we're loosing focus on getting that "end user
value"
If we ask users, that mostly does not participate, they will want us to get
1.0 ASAP, more documentation, get better look and feel for emulation (we
don't have anyone here that can make this happen for now), or get Jewel
DataGrid working completely and usable.
Following my analogy above: Yes, there are always demands from friends and
family. But I know that if I do not maintain my health, I will be of less
service to them in the future.
And probably they will want event system will always be one way, and not
make it depend on PAYG things, since will be more difficult for them to use.
Our job is to hide the details from those who don't care.
IMHO, we have a good balance right now. And Platform things should be
always the same. Moreover if it comes from AS3 and Flash that are part of
the base of technology people coming from that world expect to have out of
the box.
Today the problem is not about download bits, we already got a very low
foot pint thanks to your focus on that part (and for me is enough), our
problem is now in other places like ease of development, or make people
that are not working all days since years with Royale in lower the
knowledge the need to use the huge amount of beads we already have and that
they don't know we have.
First impressions matter. If HelloWorld is just as big as it was in Flex, and
you thought Flex was too big and slow, you won't want to use Royale.
As the Gall's law says we should evolve from what we have now. Maybe when
we have all that problems solved and have 1.0, we can continue iterating
things even better with more time, and people will be glad at this time. In
exchange, I think if continue refining the already good solution, we can
fail to make Royale ready for users and loose all this years. IMHO, instead
of writing long emails, we all should be writing more docs to make people
know Royale better and succeed doing at least some simple migration or app.
I'd never heard of Gall's law. I looked it up. My interpretation is that it
means that we should continue to be vigilant about PAYG and composition. PAYG
and composition is the simple system/pattern that allows us to manage complex
component sets.
If you want to be philosophical, I would say the philosophies I subscribe to
are "History repeats itself", "People who are not aware of history repeat those
same mistakes", "short-term thinking often results in long-term pain".
I am working as hard as I can to deal with "end user" issues. But we must try
to avoid the problems in the last mile. Attracting people to start using
Royale won't matter if they can't finish and get to production. It has to be a
balance of both.
My 2 cents,
-Alex
